1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

RECOMMENDATIONS FOR ANTICIPATING SEA-LEVEL RISE IMPACTS ON LOUISIANA COASTAL RESOURCES DURING PROJECT PLANNING AND DESIGN TECHNICAL REPORT

Louisiana Applied Coastal Engineering & Science (LACES) Division

24 January 2012

26

27 28 29 30 31 32 33 34 35 36

AUTHOR INFORMATION Kristin DeMarco, Coastal Resources Scientist 3 Jennifer Mouton, Coastal Resources Scientist Senior, DCL-B James W. Pahl, Ph.D., Coastal Resources Scientist Manager Office of Coastal Protection & Restoration Louisiana Applied Coastal Engineering & Science Division Applied Research & Development Section

37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57

Page intentionally left blank

ii

58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92

TABLE OF CONTENTS AUTHOR INFORMATION ............................................................................................................ i LIST OF TABLES ......................................................................................................................... iv LIST OF FIGURES ........................................................................................................................ v LIST OF ACRONYMS ................................................................................................................. ix 1. 2. INTRODUCTION AND OBJECTIVES OF THIS REPORT ................................................ 1 STATE OF THE SCIENCE.................................................................................................... 2 2.1. 2.1.1. 2.1.2. 2.1.3. 2.2. 2.2.1. 2.2.2. 2.3. 2.3.1. 2.3.2. 2.3.3. 2.4. 2.4.1. 2.4.1. 2.4.2. 3. 4. Techniques for Measuring Components of Sea Level.................................................. 3 Tide Gauges .............................................................................................................. 4 In-Situ Measurement ................................................................................................. 5 Satellite Altimetry Measurement .............................................................................. 7 Global Historical Sea-Level Rise.............................................................................. 8 Historical Regional Sea-Level Rise in the Gulf of Mexico .................................... 13 Global Projected Sea-Level Rise ............................................................................ 18 Projected Regional Sea-Level Rise in the Gulf of Mexico ..................................... 25 Gulf of Mexico Regional Sea-Level Rise Rate Recommended for CPRA Use ..... 25 Estimates of Historical Relative Sea-Level Rise in Coastal Louisiana .................. 27 Subsidence .............................................................................................................. 31 Marsh Vertical Accretion ........................................................................................ 34

Estimates of Historical Sea-Level Rise ............................................................................ 8

Projections of Future Sea-Level Rise ............................................................................. 18

Relative Sea-Level Rise in Coastal Louisiana ............................................................... 27

SUMMARY AND RECOMMENDATIONS....................................................................... 35 REFERENCES ..................................................................................................................... 38 Appendix A: Reviewer Comments and Responses ................................................................. 42 Appendix B: CPRA-LACES Technical Issues with the US Army Corps of Engineers Engineering Circular No. 1165-2-211 ...................................................................................... 49 Appendix C: Draft Southwest Coastal Feasibility Study, Wetland Accretion Summary ........ 53 Appendix D: Detailed Procedure for Incorporating Sea-Level Rise into Louisiana Coastal Project Planning and Design ..................................................................................................... 60

APPENDICES .............................................................................................................................. 42

iii

93 94

LIST OF TABLES Table and Legend Table 1. Comparison of derived acceleration constants in NRC (1987) and USACE (2009) for the generalized predictive GSLR equation E(t) = a*t + b*t2. Table 2. A summary of the current NOAA tide gauge stations shows that only two stations relevant to coastal Louisiana have a sufficient period of record to establish an RSLR trend. Table C1. Long-term accretion estimates (137Cs) from different marsh types in the Chenier and Delta Plains from different studies with large numbers of samples. (DP=Delta Plain; CP=Chenier Plain). Table C2. Long-term accretion estimates (137Cs) from different marsh types and habitats (interior vs. streamside). Compiled by Jarvis (2010). Table C3. A comparison of ranges of long-term accretion estimates (137Cs) from impounded and un-impounded brackish marsh sites in the Chenier Plain (summarized in Steyer 2008). Table C4. Summary statistics of recent elevation change data from the Coastwide Reference Monitoring Stations (CRMS) in freshwater (F), intermediate (I), brackish (B) and salt (S) marshes in the Chenier Plain.\ Table D1. Acceleration constants to be applied for sea level rise rates of 0.5 m, 1.0 m, and 1.5 m. Page Number 24

29

53

54 55

56

62

95

iv

96 97

LIST OF FIGURES Figure and Legend Figure 1. This sea-level curve for Surinam and French Guiana illustrates the 18.6-year lunar nodal cycle that must be accounted for in any SLR estimation. Figure 2. The confidence interval around sea-level rise projections increases rapidly as tide gauge periods of record decrease below sixty (60) years. Figure 3. A map of the Global Sea Level Observing System Long-Term Trends (GLOSS-LTT) network of tide gauges used for calculating global sea-level rise rates, illustrates the bias in station distribution in the northern hemisphere. Figure 4. The map of the transects used by NOAA Ships of Opportunity deploying eXtendable BathyThermographs (XBTs) for the Low Density and Frequently Repeated transects run by ships of opportunity under the NOAA Atlantic Oceanographic and Meteorological Laboratory to measure temperature across the global ocean. Figure 5. The spatial distribution of Argo Array floats for the 30-day period preceding 10 March 2011, illustrates the widespread spatial coverage of the network in the worlds oceans. Figure 6. Global sea-level estimates calculated from satellite altimetry (red line) diverged from MSL based on tide gauge records (blue line) in for about 10 years beginning in 1999 (Church & White 2011), for reasons that are uncertain. Figure 7. Satellite altimetry data from NOAA, accessed on 10 March 2011 (http://ibis.grdl.noaa.gov/SAT/SeaLevelRise/slr/slr_sla_gbl_free_txj1j2_90.pdf) , illustrate an overall global SLR rate (hTOT) higher than the 20th century average of 1.7 0.3 mm/yr. reported by Church and White (2006). Figure 8. Sea-level rise data from Church & White (2011) for the 1860-2009 time period was used to calculate an 1880-2009 linear trend of 1.5 mm/yr. (0.059 inches/yr.) and the accelerating quadratic trend discussed in the text of this report. Figure 9. Fitting a quadratic function to the 1880-2009 Church and White in press dataset available from CSIRO (http://www.cmar.csiro.au/sealevel/sl_data_cmar.html) results in a slightly better fit of the data than a simple linear regression. Page Number 3

10

12

13

98

99 100

LIST OF FIGURES (cont.) Figure and Legend Figure 10. The relative contributions of thermal expansion and bulk eustatic water volume to GSLR, as determined by subtraction of thermal expansion from GSLR to determine the eustatic contribution, changed between the 1993 and 2005. Figure 11. Satellite altimetry data from NOAA, accessed on 11 March 2011 (http://ibis.grdl.noaa.gov/SAT/SeaLevelRise/slr/slr_sla_gom_free_txj1j2_90.pd f), illustrate an overall SLR rate for the Gulf of Mexico (hGULF) lower than the GSLR trend calculated for the same time period (2.9 0.4 mm/yr.) as shown in Figure 8. Figure 12. A spatial depiction of the data from the TOPEX/Poseidon and Jason satellite altimeters clearly shows that regional rates of SLR can be very different at both the global scale (a) and within the Gulf of Mexico (b), covering the grey inset box). Figure 13. A graph of mean SLR trends for NOAA tide gauges in the Gulf of Mexico (here oriented with the eastern-most station on the left) show that both SLR and the variance around the trend are greater in the western Gulf of Mexico than in the east. Figure 14. Long-term stations in the western Gulf of Mexico in NOAAs National Water Level Observation Network (NWLON) vary significantly in the period of record. Figure 15. This graph from Zervas (2009) describes how the 95% confidence interval of the SLR trend determined from NOAA NWLON tide gauges is highly dependent on the age of the station. Figure 16. SLR rates for discrete points offshore of southern Louisiana show significant east-west variation, with SLR values highest offshore of the Balize Delta and trending lower moving west across the front of the Chenier Plain. Figure 17. Mitrovica et al. (2009) predicted the possible distribution of sealevel change (meters) in response to a collapse of the West Antarctic ice sheet accounting for the rotation of the Earth. Figure 18. A SLR budget figure from Bindoff et al. (2007) highlights the range of uncertainty surrounding relative contributions to observed global sea level rise. Page Number 14

15

16

17

17

18

19

20

21

101 vi

102 103

LIST OF FIGURES (cont.) Figure and Legend Figure 19. Extrapolation of the linear (black) regression for the Church & White (2011) data (as described in Figure 8) to predict global mean sea level in the 2109 results in a lower estimated sea level than if the quadratic (red) nonlinear regression is used. Figure 20. The rate of global sea-level rise during the 21st Century is modeled to increase, with the extent of acceleration dependent on the predicted temperature increases associated with the IPCC global climate change scenarios. Figure 21. Observed GSLR for the period 1993-2007 from satellite altimeters exceeded IPCC best estimate predictions of GSLR made in 1995 and 2001. Figure 22. NRC (1987) Curves as modified in USACE (2009) to account for different assumptions of the historical linear rate of global sea-level rise. Figure 23. This map predicts local sea level change relative to the global average for the 21st century, as calculated from the results of 16 global climate change models running the IPCC A1B climate scenario. Figure 24. CPRA-LACES recommendations for future global sea-level rise by 2100, compared to a range of values from recent published scientific literature. Figure 25. NOAAs tide gauge network in Louisiana does provide coverage of multiple geomorphic settings within the States coastal zone. Figure 26. RSLR trend lines for the Sabine Pass North (a) and Grand Isle (b) NOAA tide gauges illustrate the significance of geological stability on RSLR, and the difference between the more stable Chenier Plan (Sabine) and the less stable Deltaic Plain (Grand Isle). Figure 27. A map of subsidence developed by Britsch in 2007 illustrates the spatial variability in predicted subsidence rates in southern Louisiana. Figure 28. Map of projected subsidence ranges for south Louisiana generated by the Subsidence Advisory Panel for the Louisiana CPRA Master Plan 2012 Update. Figure D1. Data were derived by USGS from satellite altimetry data for center points of the analysis grid shown in Figure 12. Page Number 22

23

23

24

26

27

29

30

32

33

60

104 vii

105 106

LIST OF FIGURES (cont.) Figure and Legend Figure D2. Map of projected subsidence ranges for south Louisiana generated by the Subsidence Advisory Panel for the Louisiana CPRA Master Plan 2012 Update, following a meeting on 14 October 2010. Figure D3. Screen capture of the spreadsheet that LACES has drafted for calculating RSLR curves consistent with the four-step process recommended in the Technical Report. Page Number 62

63

107

viii

108 109

LIST OF ACRONYMS Acronym Description First Reference on Page # 1 31 37 2 5 4 8 2 34 2 4 22 4 34 4 2 1 12 5

CPRA CRMS EC GIA GOOS GLOSS GRACE GSLR HET MSL NOAA NRC NWLON PDT PSMSL RSLR SLR USACE XBT 110

Coastal Protection and Restoration Authority (of Louisiana) Coast-wide Reference Monitoring System Engineering Circular Glacial Isostatic Adjustment Global Ocean Observation System Global Sea Level Observing System Gravity Recovery and Climate Experiment Global Sea-Level Rise Habitat Evaluation Team Mean Sea Level National Oceanographic and Atmospheric Administration National Research Council National Water Level Observation Network Project Delivery Team Permanent Service for Mean Sea Level Relative Sea-Level Rise Sea-Level Rise US Army Corps of Engineers Extendable BathyThermograph

ix

111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132

Page intentionally left blank

133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172

1.

INTRODUCTION AND OBJECTIVES

Land changes in the coastal zone and high sea-level rise rates are exposing lowland areas to more frequent events of saltwater intrusion, flooding and rapid shoreline erosion, magnifying the negative effects of coastal storms and storm surge. Louisiana is particularly sensitive to sealevel rise (SLR) due to the unique geology and inherent nature of the Mississippi River Delta and Chenier Plains. There is a pressing need to integrate up-to-date SLR estimates into planning activities to anticipate coastal land loss patterns, protect coastal communities and adequately design restoration projects. Projections for future SLR and concurrent coastal vulnerability estimates are numerous and variable. State and regional planning efforts aimed at mitigating the impacts of SLR are iterating between policy development and implementation as projections are refined and confidence in future estimates increases. Adaptability and flexibility are key components to ensure the incorporation of the most current and accurate SLR projections into project planning and policy development for the Louisiana coast. In this document we synthesize historical SLR research, review the state of the current science and identify key principles and recommendations in determining and incorporating SLR into coastal restoration strategies, modeling efforts and project design. The objective of this document is to make technical recommendations for incorporating SLR into Coastal Protection and Restoration Authority of Louisiana (CPRA) planning and engineering of habitat restoration and storm protection projects. The document is structured to:

Deductively summarize the state of the science on the patterns of increase in the surface of the global ocean, regional Gulf of Mexico and local coastal waters, in order to recommend the rate(s) of anticipated SLR most appropriate for incorporating into project design, planning and analysis, and Describe how that recommended rate(s) of local sea-level rise should be combined with the present understanding of the highly variable spatial patterns in coastal landform subsidence and wetland vertical accretion to predict relative SLR at specific points in the Louisiana coastal zone.

It should be noted that SLR research released after August 2011 were not included in this version of the report. Because global, regional and local estimates of SLR are constantly changing as new data become available, this report will by necessity be revised via an iterative process. CPRA plans to update this report at a minimum every five years, or as major improvements in SLR understanding or changes in documented rates occur.

173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218

2.

STATE OF THE SCIENCE

Sea-level rise is caused by a variety of dynamic anthropogenic and natural factors. Estimates of past and projections of future SLR are dependent on the interplay between these factors. Challenges arise when attempts are made to identify changes in sea level that occur over varying time scales; it is important to isolate the long term historical trends from the background of regular natural cycles to make confident predictions about future trends. In any coastal zone, the actual rate of SLR is a combination of global sea-level rise (GSLR) and local coastal processes including natural cycles, glacial isostatic adjustment (GIA), subsidence, accretion and erosion of shorelines and coastal marshes. These influences result in a local rate of SLR, expressed as relative sea-level rise (RSLR) that may be very different from GSLR. While RSLR is more relevant for management purposes, it is necessary to first evaluate GSLR trends and then focus on local conditions in the Gulf of Mexico offshore of southern Louisiana to inform recommendations on estimating local RSLR. In discussing the current understanding of GSLR, this section will also detail the methodologies for measuring GSLR components, because this information has implications for local understanding of Gulf-specific SLR and RSLR. GSLR is the mean increase in sea surface elevation across all of the world oceans and is caused primarily by two factors: thermal expansion and freshwater influx. Thermal expansion, also called the steric component of GSLR, refers to the increase in total ocean volume resulting from increasing ocean temperatures. Freshwater influx into oceans, technically defined as the eustatic component of GSLR, causes a change in the mass of water in the ocean and is the result of melting glaciers, ice sheets and other land ice, and terrestrial runoff. While the term eustatic has been used in some technical literature and the popular press to describe sea level rise due to a total change in the global ocean, throughout this report we will only use that term to specifically refer to the change in water mass. To accurately determine the long-term historical trend of GSLR and begin making predictions of future patterns, researchers must first identify and remove the effects of natural forcings on the sea-level observation data to evaluate the steric and eustatic components only. Natural forcings are patterns or processes that can influence mean sea level (MSL); are typically cyclical in nature; and can be atmospheric, meteorological, or caused by alterations in the rotation of the Earth and moon. For example, changes in the orbital pattern of the Earth and the moon can impact sea level regionally. Natural cycles can change sea surface elevation over large oceans which will manifest as a short term trend in sea-level monitoring data. Therefore it is preferable to have a sea-level record long enough to encompass several periods of any natural cycles to properly account for and remove confounding patterns. Probably the most prominent natural forcing influencing coastal processes on a human timescale is the 18.6-year lunar nodal cycle and impacts coastal water levels and sedimentation rates. The 18.6 year cyclic process stems from interactions between the moons orbital path around the Earth and the orbital plane of the Earth around the sun; specifically, the migration of the intersection points between the two due to precession in the moons orbit. The peaks and valleys in this cycle can be observed astronomically as the major lunar standstills, which are the most northerly and southerly rising and setting of the moon. This cycle is predictable and observable 2

219 220 221 222 223 224 225 226 227 228 229 230 231

as a periodic function in sea surface elevation, all other factors being equal (Figure 1; Gratiot et al. 2008; Morris & Sundberg 2008). Relevant to this discussion is that at the peak or valley in the lunar nodal cycle, the rate of sea-level rise is flattest within the periodic SLR curve. As shown in Figure 1, the last major lunar standstill was evident in the low point of a de-trended sea-level curve in 2006 (http://www.umass.edu/sunwheel/pages/moonteaching.html), which means that the rate of SLR should be expected to increase through 2015. This cycle alters the rate and degree of sedimentation and accretion in tidally influenced wetlands, significantly influencing the morphology of estuaries and tidal basins (Jeuken et al. 2003) as well as adding to the interannual variability in the sea level surface. To make predictions about sea-level it is important to account for the 18.6-year lunar tidal cycle, and similar natural cycles, into any estimates of local SLR.

232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259

Figure 1. This sea-level curve for Surinam and French Guiana illustrates the 18.6-year lunar nodal cycle that must be accounted for in any SLR estimation. Figure from Gratiot et al. (2008).

Once the background effect of natural forcings is removed, the sea level budget (Willis et al. 2008; Leuliette & Miller 2009) can be expressed as hTOT = h STERIC + hMASS where hTOT is the global sea level rise rate, hSTERIC is the thermal expansion contribution and hMASS is the eustatic contribution due to freshwater influx. Many efforts to quantify the relative contributions of eustatic (hMASS) and steric (hSTERIC) components to the total sea-level budget have been made to try and determine which contribution dominates GSLR (Jevrejeva et al., 2008). As researchers attempt to close the budget, independent measurements of steric, eustatic and total sea level are checked against one another. This is not a trivial task and illuminating the relative contributions of thermal expansion and freshwater influx are critical for making projections of sea-level change into the future.
2.1. Techniques for Measuring Components of Sea Level

(Eqn. 1)

When attempting to close the sea level budget it is important to consider the measurement technique used to acquire the data. Currently sea-level data are gathered from three primary sources to determine the GSLR rate: tide gauges, in situ measuring devices and satellite altimetry readings. 3

260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304

2.1.1. Tide Gauges Globally, tide gauge data used for measurement of GSLR are typically acquired from the Permanent Service for Mean Sea level (PSMSL), established in 1933 and run by the UK Natural Environment Research Councils National Oceanography Centre (http://www.psmsl.org/about_us/). In the US, the National Oceanographic and Atmospheric Administration (NOAA) operate 128 long-term National Water Level Observation Network (NWLON) stations that monitor monthly mean sea-level data. Data are used to determine linear trends, average seasonal cycles, and interannual variability, including estimated errors, by measuring the height of sea surface relative to certain coastal benchmarks. In some areas these data go back hundreds of years. Tide gauge records have definite value due to the length of time evaluated, although regional and local corrections must be made to obtain a global mean for these trends. Tide gauges measure the RSLR at discrete locations and estimates of GSLR from tide gauge data are an average of all the tide gauge readings over a certain period of time (e.g. Church & White 2004). The length of tide gauge station record has to be sufficient to obtain a robust estimate of the historic relative mean sea-level change, especially if that rate will be used to predict future SLR. If the length of the record is too short it may be difficult to fully account for the impacts of interannual and decadal variations in sea level, such as the fore-mentioned 18.6-year lunar tidal cycle, resulting in misleading or erroneous sea level trends. The Intergovernmental Oceanographic Commission (2006) suggests that the duration of a tidal record should be at least two lunar nodal cycles (about 40 years) before being used to estimate a local relative sea-level trend, while Douglas et al. (2001) claims that the length of record should be at least three lunar nodal cycles (approximately 60 years) and have 85% coverage during that time period. At minimum, record lengths shorter than 40 years in duration could have significant uncertainty that can quickly outweigh any SLR projections of a few millimeters per year as the period of record decreases (Figure 2). USACE (2009) describes that if only long-term estimates ( 40 years) are available from the local tide gauge, local trends should be evaluated in a regional context, using nearby station records with adequate record lengths for the same time period to compare with the local data to determine validity. However, as we will discuss later in this report, there are significant limitations to that approach in coastal Louisiana. Tide gauges have traditionally been used for navigational purposes and consequently were historically placed in areas of heavy water traffic restricted to coastlines and the open ocean. This has led to an uneven spatial distribution of gauges, possibly hindering accuracy when determining long-term global sea-level trends. At present GSLR rate and acceleration statistics used by NOAA are calculated from approximately 190 stations known as the Global Sea Level Observing System (GLOSS) Long-Term Trends network (Figure 3). There are some significant limitations in that data, as NOAA recognizes the present station distribution is biased towards the northern hemisphere (http://www.gloss-sealevel.org/). Additionally, management of the data can become cumbersome and quality assurance /quality control of original long-term datasets can be problematic. While these and other limitations led some researchers (Grger and Plag 1993) to conclude tide-gauge data should not be used to determine GSLR trends, such data are very useful as checks to newer data collection methods as well as being the most appropriate, and often the only, method for obtaining estimates of historical RSLR prior to 1992. 4

305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328

Figure 2.The confidence interval around sea-level rise projections increases rapidly as tide gauge periods of record decrease below sixty (60) years. Dots indicate the approximate graph positions for the period of record of the NOAA tide gauges shown in Table 2 and Figure 18; see discussion in Section 3.3.1. Figure from Zervas (2009).

2.1.2. In-Situ Measurement In the US, NOAAs Global Ocean Observation System (GOOS) manages a fleet of volunteer observatory ships which use extendable probes, termed eXtendable BathyThermographs (XBTs), to measure ocean temperature (http://www.aoml.noaa.gov/goos/uot/xbt-what-is.php). The ships travel along set transects and take XBT readings; certain areas are sampled 18 times a year to gain insight on interannual and seasonal variability (Figure 4). XBTs are designed to fall through the water column at a known rate; the depth of the probe is not measured but inferred from the time launched by a fallrate equation provided by the manufacturer. Recently, researchers found systematic errors in the fall-rate equations which resulted in temperature readings being assigned to incorrect depths (Willis et al. 2007; Willis et al. 2009). These errors have since been corrected for in reports making GSLR measurements and provide measurement for the thermal expansion component (hSTERIC) of the sea level budget (Lyman et al. 2010). Argo is an international project started in 2000 as part of GOOS and is made up of over 3000 battery-powered floats. The floats collect temperature and salinity data from the upper 2000meters of the ice-free ocean. Upon reaching the ocean surface, satellites detect the floats position and the data are transmitted to a data assembly center. Argo floats measure temperature 5

329 330 331 332 333 334 335

Figure 3. A map of the Global Sea Level Observing System Long-Term Trends (GLOSS-LTT) network of tide gauges used for calculating global sea-level rise rates, illustrates the bias in station distribution in the northern hemisphere. Map from GLOSS (http://www.gloss-sealevel.org/, accessed 19 August 2010).

336 337 338 339 340 341 342 343

Figure 4. The map of the transects used by NOAA Ships of Opportunity deploying eXtendable BathyThermographs (XBTs) for the Low Density and Frequently Repeated transects to measure temperature across the global ocean. Note the absence of any XBT transects in the Gulf of Mexico. There is also no Gulf coverage from AOMLs High-Density XBT Transects (not shown, see http://www.aoml.noaa.gov/phod/hdenxbt/index.php). Figure from http://www.aoml.noaa.gov/phod/goos/ldenxbt/index.php, accessed 22 September 2010.

344 345 346 347 348 349 350 351 352

and salinity as a function of depth to describe how much the increase in MSL is steric in origin (hSTERIC), and how the steric signal is distributed over depth. Argo covers a large portion of the world ocean (Figure 5) and collects continuous data important for GSLR estimates. Researchers recently identified an error in a small portion of the Argo floats that led to a false cold bias from 2003-2006, which has since been corrected in current estimates (Willis et al. 2007). Conversely, there may also be a potential Argo warm bias due to the inability for floats to be placed in ice covered ocean areas (Roemmich & Gilson 2009).

353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375

Figure 5. The spatial distribution of Argo Array floats for the 30-day period preceding 10 March 2011, illustrates the widespread spatial coverage of the network in the worlds oceans. Note the limited coverage of Argo floats in the Gulf of Mexico. Map from http://www.argo.ucsd.edu/index.html.

2.1.3. Satellite Altimetry Measurement The Argo Array was named such to reflect the relationship between that network of floats and the Jason-1 and -2 satellite altimetry missions; data collected by the two methods are coupled to attain more accurate estimates of GSLR and to evaluate relative contributions. The Jason satellites followed the 1992 French-US joint TOPEX/Poseidon satellite mission to track sea level height with radar altimeters (http://topex-www.jpl.nasa.gov/technology/technology.html). The satellites measure total sea level and are currently active to obtain more widespread coverage. Specifically, Jason-1 measures the total sea level (hTOT) and can be used in concurrence with the Argo data (hSTERIC) to estimate the eustatic contribution (hMASS) to total sea level, using Equation 2. hMASS = hTOT[from Jason-1] - hSTERIC[from Argo] where all variables are as in Equation 1. (Eqn. 2)

376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421

A more recent satellite mission initiated in 2002 by NASA is the Gravity Recovery and Climate Experiment (GRACE), which measures the changes in mass of the world ocean and thus more directly measures hMASS (http://grace.jpl.nasa.gov/information/). Additionally, GRACE was used to determine the ice mass-loss for the Greenland and Antarctic Ice sheets (Khan et al. 2010; Velicogna, 2009). Theoretically, GRACE eustatic sea-level measurements (hMASS) and the Argo-based steric measurements (hSTERIC) should equal the total sea level measurement obtained from Jason-1(hTOT), i.e. hTOT[from Jason-1]= hMASS[from GRACE]+ hSTERIC[from Argo] (Eqn. 3)

On global scales, Argo and Jason together with satellite gravity measurements from GRACE, partition global sea-level rise into its steric and mass-related components (Willis et al., 2008; Cazenave et al, 2009; Leuliette and Miller 2009; Wunsch et al. 2007). Although satellite altimetry-based GSLR rates were in good agreement with those shown by tide gauge records (Figure 6; Ablain et al., 2009; Prandi et al. 2009) for the first six to seven years of the altimetry record, values between the two data sources deviated between 1999 and 2008 for reasons that were unclear as of the time that this report was drafted (Domingues et al., 2008; Church & White 2011). However, it is important to recognize that satellite altimetry data, only available since the end of 1992, only recently represent a full 18.6-year tidal period, let alone two or three periods as recommended above for defining a trend. Although the MSL budget has been balanced occasionally (Leuliette & Miller 2009), discrepancies have been identified by many regarding the relative contributions of the eustatic and steric components (Bindoff et al. 2007; Lombard et al. 2007; Willis et al. 2008). Differences in data processing and potential biases of collection methods could be the causes for this variance. Altimetry-derived measurements and GSLR estimations need to be processed to account for factors such as barometric pressure, glacial isostatic adjustment (GIA) and seasonality. It is also possible that some integral component to the driving processes of MSL change may not be understood. It is essential to understand these components to accurately make predictions for future rates of SLR. For instance, if we are confident that the eustatic contribution will be the dominant component to increases in MSL, we can expect the rate of increase to be nonlinear (Cazenave & Llovel 2010) and thus neither uniform across the globe nor steady to rise (Gomez et al. 2010).
2.2. Estimates of Historical Sea-Level Rise

2.2.1. Global Historical Sea-Level Rise It cannot be emphasized enough that estimating a rate for historical GSLR is highly dependent on the specific time period selected for measurement due to the changing contributions of the steric and eustatic components to total MSL (Jevrejeva et al. 2006). Much of the debate about the mathematical nature of historical SLR trends and defined accelerations or decelerations in the rate of SLR over time is due to differential statistical trends in data within specific time frames. For example, Church & White (2006) analyzed a large set of tide gauge data spanning 18701935 and calculated a historical linear GSLR rate of 0.7 0.4 mm/yr. (0.028 0.016 inches/yr.); however, a revised analysis of the same data in Church & White (2011) that limited the range 8

422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443

Figure 6. Global sea-level estimates calculated from satellite altimetry (red line) diverged from MSL based on tide gauge records (blue line) in for about 10 years beginning in 1999 (Church & White 2011), for reasons that are uncertain. The deviations lead to slightly different GSLR trends from 1993-2009 (Table 1).

from 1880-1935 calculated a linear GSLR trend of 1.1 + 0.7 mm/yr. (0.043 0.028 inches/yr). Similarly, while Ablain et al.s (2009) analysis of satellite altimetry data for the period 19932008 calculated a linear GSLR trend of 3.1 0.6 mm/yr. (0.12 0.02 inches/yr), NOAAs Laboratory for Satellite Altimetry reported an altimeter-based linear GSLR trend of 2.9 0.4 mm/yr. (0.11 0.02 inches/yr) for the 1992-2011 time period when accessed on 10 March 2011. In addition to the period of record, estimating a historical rate of global sea-level rise is also greatly dependent on both the number and spatial distribution of gauging stations selected for measurement. Much of the debate about calculating historical sea-level rise trends, and the recent debate over both linear vs. non-linear trends in the data and accelerations vs. decelerations in the rate of sea-level rise over time results from choices in which stations are chosen and over which specific time period they are analyzed (Houston and Dean 2011a, b and c; Rahmstorf and Vermeer 2011; Donoghue and Parkinson 2011).

444 445 446 447 448 449 450 451 452 453 454 455 456

Most estimates of the 20th century long term GSLR have been less than 2 mm/yr. (0.08 inches/yr.; e.g. Church et al.2004; Church and White 2006). In comparison, as introduced above, NOAA satellite altimetry data based on TOPEX/Poseidon, Jason-1 and Jason-2 technologies, accessed on 10 March 2011, calculated an overall 1993-2010 GSLR rate (hTOT) of 2.9 0.4 mm/yr. (0.11 0.02 inches/yr.; Figure 7). Data from several other altimeters are available, and when those data are incorporated into the analyses, different GSLR trend values can result. Data used in this report to document Louisiana nearshore water surface changes are only from the TOPEX/Poseidon, Jason-1 and Jason-2systems, and for sake of direct data comparability this discussion will only focus on that restricted dataset.

457 458 459 460 461 462 463

Figure 7. Satellite altimetry data from NOAA, accessed on 10 March 2011 (http://ibis.grdl.noaa.gov/SAT/SeaLevelRise/slr/slr_sla_gbl_free_txj1j2_90.pdf), illustrate an overall global SLR rate (hTOT) higher than the 20th century average of 1.7 0.3 mm/yr. reported by Church and White (2006).

10

464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509

Any calculation of a linear rate of historical GSLR by its nature assumes that rate has been and is constant. A generalized linear regression is commonly represented as y = mx + b where y is the dependent variable, m is the slope of the line, x is the independent variable, and b is the 0-intercept on the y-axis. Relevant to SLR calculations, and ignoring the concept of the y-axis intercept (which for SLR is arbitrary), the regression equation can be expressed as E(t) = a*t where E is GSLR at time t, and a is the rate of GSLR (slope of the line). However, examination of the historical data and recent literature indicate that GSLR since the late nineteenth century has not been linear, but instead has been accelerating since that time. While Figure 8 shows the data set by which Church & White (2006) calculated the frequentlycited 1.7 0.3 mm/yr.(0.067 0.012 inches/yr.) average GSLR for the 20th Century, in that paper as well as their revised analysis in Church & White (2011) they discussed that in the least any linear analysis needed to recognize several visually-obvious inflection points in the data in the mid-1930s, the 1960s and the 1980s.. However, Church and White (2006, 2011) also defined a non-linear, second-order polynomial (quadratic) fit to the full data set. An accelerated SLR scenario is simplistically modeled by the equation: E(t) = a*t + b*t2 where E, t, and a are as defined in Eqn. 5, and b is an acceleration factor. The actual calculations to determine projected sea-level rise based on the non-linear approach are more complex, and are described further in the Summary and Recommendations section of this document as well as the separate, more concise guidelines included as Appendix D to this report. When linear and quadratic functions are fitted to the Church & White (2011) data shown in Figure 8, both functions explain greater than 97% of the variation (Figure 9), and it could be argued that the linear is an acceptable representation of the data. However, the differences in forecasting a linear vs. a quadratic function into the future can become significant and the use of one prediction over another should be carefully considered. Visual examination of Figure 9 supports some concern with the linear approach, where it can be seen that the linear trend line shows a departure from (specifically an underestimate of) the observed sea-level data beginning around the beginning of the 1990s. Also, and perhaps most important for a discussion of recommendations to CPRA staff of how to calculate RSLR, a curvilinear historical GSLR curve 11 (Eqn. 6) (Eqn. 5) (Eqn. 4)

510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531

Figure 8.Sea-level rise data from Church & White (2011) for the 1860-2009 time period was used to calculate an 1880-2009 linear trend of 1.5 mm/yr. (0.059 inches/yr.) and the accelerating quadratic trend discussed in the text of this report. Also shown is the 1870-2001 data published in Church & White (2006).

more strongly supports a deviation from the US Army Corps of Engineers (USACE) method for determining RSLR (see Appendix B). Acceleration in the historical data was also documented by Jevrejeva et al. (2008) and Woodworth et al. (2009). Recent analyses also suggest that the relative contributions of eustatic and steric influences on GSLR are in flux. Bindoff et al. (2007) concluded that thermal expansion (hSTERIC) accounted for approximately 50% of the total observed GSLR from 1993 to 2003. In comparison, Lyman et al. (2006) proposed a slowing influence of thermal expansion in the 2003-2007 data. More recent data suggest that the freshwater influx from melting glaciers and ice sheets (hMASS) is the primary contributor to current GSLR (Antonov et al. 2005; Cazenave & Llovel 2010; Cazenave et al. 2008; Jevrejeva et al. 2008; Nicholls & Cazenave 2010). Concurrently, satellite altimetry observations estimate that the eustatic component of GSLR (ice sheet, glacial ice and land ice melting) may have been up to 80%, or 2.4 0.35 mm/yr. (0.094 0.014 inches/yr.), of the total observed increase in MSL from 2003-2007, and that thermal expansion has been slowing (Figure 12

100 MONTHLY MEAN SEA LEVEL (mm, zeroed to December 1982) 50 0 -50 -100 -150 -200 1880 1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020

y = 0.005x2 - 17.804x + 15743 R = 0.9834 y = 1.5379x - 3060.1 R = 0.972

532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556

YEAR Figure 9. Fitting a quadratic function to the 1880-2009 Church and White (2011) dataset available from CSIRO (http://www.cmar.csiro.au/sealevel/sl_data_cmar.html) results in a slightly better fit of the data than a simple linear regression. Blue points are the CSIRO data, the black line and statistics describe the linear regression, and the red line and statistics describe the second order (quadratic) function.

10; Cazenave & Llovel 2010). These data could have significant implications to how predictable and relevant future GSLR scenarios are to the Gulf of Mexico, given limits in the monitoring networks that define the GSLR components, as shown in Figures 4 and 5. 2.2.2. Historical Regional Sea-Level Rise in the Gulf of Mexico The extensive description of sea-level components and measuring technologies above was necessary to help frame the deductive approach in this report and address whether GSLR is valid to predict changes in the Gulf surface elevation throughout the rest of this century. This section will discuss the data available for the Gulf of Mexico, the trends in those data, and the important caveats on those data and data products that must be considered. Satellite altimetry data specific to the Gulf of Mexico do show a slightly lower rate of sea surface elevation change over the past twenty years (Figure 11) than the global calculations described above. This may be due to the subtropical and semi-enclosed nature of the Gulf. Because of these two factors, the Gulf could reasonably be assumed to be a warmer body of water than the adjacent Atlantic Ocean, especially with the dominant inflow to the Gulf coming from the even warmer Caribbean Sea through the Straight of Campeche, and therefore less susceptible to steric 13

Observed sea level rise: 3.3mm/yr

Thermal expansion: 0.8mm/yr

Residual: ocean mass change: 2.5mm/yr

557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581

Figure 10. The relative contributions of thermal expansion and bulk eustatic water volume to GSLR, as determined by subtraction of thermal expansion from GSLR to determine the eustatic contribution, changed between the 1993 and 2005. Specifically, the eustatic contribution became more important. Figure adapted from http://wcrp.ipsl.jussieu.fr/Workshops/SeaLevel/Orals.html.

influences on MSL than the colder open ocean. However, there are very few GLOSS (Figure 3) or Argo (Figure 5) resources and no XBT coverage (Figure 4) devoted to the Gulf of Mexico and thus efforts to separate either eustatic (hMASS) or temperature-induced (hSTERIC) components from the overall observed elevation changes are less confident. Global satellite altimetry data and recent publications (Merrifield & Merrifield 2009) demonstrate that observed GSLR has not been evenly distributed across the world ocean (Figure 12). Satellite altimetry-based estimates of SLR in the Gulf of Mexico (hTOT) are also highly variable, with the general pattern of higher rates of SLR in the center of the Gulf and lower rates in both the eastern and western margins. To complicate the issue further, there is evidence that the tidal hydrodynamics of the western Gulf of Mexico are different than for the eastern Gulf (Figure 13). Quoting Zervas (2009), for the same year range of data, the Pacific, western Gulf of Mexico, and Bermuda stations have wider error bars than stations in the Atlantic, eastern Gulf of Mexico, and the Caribbean The western Gulf of Mexico stations appear to 14

582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605

Figure 11. Satellite altimetry data from NOAA, accessed on 11 March 2011 (http://ibis.grdl.noaa.gov/SAT/SeaLevelRise/slr/slr_sla_gom_free_txj1j2_90.pdf), illustrate an overall SLR rate for the Gulf of Mexico (hGULF) lower than the GSLR trend calculated for the same time period (2.9 0.4 mm/yr.) as shown in Figure 8.

alternate between periods of higher and lower rates of sea-level rise in contrast to the steadier rates seen in the eastern Gulf of Mexico. Unfortunately there is no information regarding the exact cause of this difference in variation and trends across the coast, making it difficult to discern if the trend will impact future SLR rates. This issue will be pursued in future iterations of this document. Supporting tide gauge data must also be approached carefully, because only 6 of the 12 tide gauges in western and central Gulf of Mexico have been in operation for more than 50 years (Figure 14) leading to a high level of error when attempting to make predictions for sea-level trends within the 95% range of confidence (Zervas 2009; Figures 2 and 15). The final complication for determining historical change in the water surface of the northern Gulf of Mexico is also illustrated by the satellite altimetry data. The spatial grid for analyzing the altimetry data can be seen in the inset map (b) of Figure 12. When a value for SLR from the 1992-2010 satellite altimetry data is established at the center point for each grid cell, the data can be projected as shown in Figure 16. A strong east-west gradient in derived 1992-2010 Gulf of 15

606 607

(a)

608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623

(b)
Figure 12. A spatial depiction of the data from the TOPEX/Poseidon and Jason satellite altimeters clearly shows that regional rates of SLR can be very different at both the global scale (a) and within the Gulf of Mexico (b), covering the grey inset box). Figures from http://ibis.grdl.noaa.gov/SAT/SeaLevelRise/slr/map_txj1j2_wysiwyg.pdf

Mexico SLR values for the near-shore coastal waters in southern Louisiana is observable, with values ranging from a high southeast of the Balize Delta and consistently decreasing both westward to a low south of the mouth of the Sabine River and northward into the Lake Borgne and Lake Pontchartrain systems. The range of values is significant, with the highest derived SLR rates being 58% higher than the lowest values. In the absence of any specific data discounting this variation, assigning a single SLR value to Louisianas coastal waters at this time is not justified. 16

624 625 626 627 628 629 630

Figure 13. A graph of mean SLR trends for NOAA tide gauges in the Gulf of Mexico (here oriented with the eastern-most station on the left) show that both SLR and the variance around the trend are greater in the western Gulf of Mexico than in the east. Figure from Zervas (2009).

631 632 633

Figure 14. Long-term stations in the western Gulf of Mexico in NOAAs National Water Level Observation Network (NWLON) vary significantly in the period of record. Figure from Zervas (2009).

17

634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654

Figure 15. This graph from Zervas (2009) describes how the 95% confidence interval of the SLR trend determined from NOAA NWLON tide gauges is highly dependent on the age of the station. Some separation between eastern and western Gulf of Mexico NWLON stations is evident, emphasizing the issues of differential tidal hydrodynamics described in Figure 13.

2.3. Projections of Future Sea-Level Rise

2.3.1. Global Projected Sea-Level Rise The 18.6-year lunar tidal cycle reached a low point in 2006 (Figure 1), and as of early 2011 we are early in the next upswing (J. Morris, personal communication). It is therefore reasonable to expect that year-to-year changes in calculated SLR for the satellite altimetry era will increase as we proceed into steeper portions of the curve. If the slowing of the thermal expansion contribution to global MSL persists and the eustatic component continues to dominate the bulk of the rise, coastal Louisiana may experience some slowing in the rate of GSLR followed by shorter periods of rapid increase due to freshwater influx. This pattern may increase the error in future estimates for GSLR and will have to be evaluated carefully to determine potential impacts on projections.

18

655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682

Figure 16. SLR rates for discrete points offshore of southern Louisiana show significant east-west variation, with SLR values highest offshore of the Balize Delta and trending lower moving west across the front of the Chenier Plain. Data were derived by USGS from satellite altimetry data for center points of the analysis grid shown in Figure 12, covering the 1992.96-2010.01 period of record. Figure from the Project Effects Modeling Team addressing the 2012 revision of the Louisiana Integrated Master Plan.

The largest uncertainty in predicting SLR over the next century is how ice sheets will respond to changes in temperature (Allison et al. 2009). It has been suggested that in the continued presence of warming temperatures the feedback mechanisms which allow the ice sheets to persist may be disrupted to the point that state of balance would shift and the ice sheet would disintegrate or collapse (Vaughan 2008). The Western Antarctic Ice Sheet alone has the capacity to raise sealevel between 3 and 8 meters (10 and 26; Bamber et al. 2009; Bindoff et al. 2007; Gomez et al. 2009). Since satellite altimetry data have become available, researchers are able to more clearly investigate the dynamics of ice sheets; GRACE is specifically employed to detect ice-mass loss for ice sheets in some cases. Recent observations have shown that the rates of ice mass loss from 2002-2009 for both the Greenland and Antarctic ice sheets are increasing, implying that the ice sheet contribution to SLR is increasing, which is consistent with other observations (Velicogna 2009). Additionally, in the event of a full or partial collapse of the Antarctic ice sheet, spatial distribution of the sea-level increase would be non-uniform, and could lead to higher sea-levels in the Northern Gulf of Mexico than other parts of the global ocean (Figure 17; Gomez et al. 2009; Mitrovica et al. 2009).The increase may be regionally concentrated along the Pacific and Atlantic coasts of the United States, which may experience an MSL increase 25% greater than the global mean even in the event of a full collapse (Bamber et al. 2009).

19

683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705

Figure 17.Mitrovica et al. (2009) predicted the possible distribution of sea-level change (meters) in response to a collapse of the West Antarctic ice sheet accounting for the rotation of the Earth. The above values are in addition to an underlying effective eustatic rate, for example, if GSLR is 1 meter (including eustatic and steric effects), a partial collapse of the Antarctic ice sheet would result in a MSL increase of 1.3 meters in the Northern Gulf of Mexico.

Although Bindoff et al. 2007 forecasted a range of 0.18-0.59 meters (0.6-1.9) for GSLR by 2100, these values have since been described as too conservative because they did not account for ice sheet dynamics (Fssel 2009). While the IPCC Working Group III report did try to account for ice sheet dynamics, they only did so empirically, assuming that high temperature increases (in excess of 3C) could result in a GSLR of 2-7 m (6.6-23) at the century to millennia time scale due to a near or total melting of the glaciers and ice sheet (Bindoff et al. 2007). After investigating the ice sheet contribution, Pfeffer et al. (2008) stated that a predicted GSLR of greater than 2 meters (6.6) by 2100 was the maximum possible increase based on the glacial physics, and predicted a GSLR range by 2100 of 0.8 - 2 meters (2.6-6.6). Future GSLR scenarios will be highly dependent upon the relative contribution of ice sheets and glaciers to the eustatic component to the GSLR budget, which, as discussed above, contains significant uncertainties. Historical rates of GSLR are difficult to replicate, and predictions will likewise be uncertain (Figure 18).

20

706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731

Figure 18.A SLR budget figure from Bindoff et al. (2007) highlights the range of uncertainty surrounding relative contributions to observed global sea level rise.

Continuation of the existing historical trends in GSL from the past 130 years (as defined in Figure 9) give very different results depending on whether the linear or quadratic trend is used (Figure 19). The 2010-2101 increase in GSL based on the linear trend is approximately 15 cm (6), while the increase in GSL based on the aquatic trend is approximately 26 cm (10). The red line shown in Figure 19 was from a simple effort to project the Figure 8 quadratic regression; LACES is aware that appropriate statistical transformations are needed to forecast polynomial functions, and will pursue a more statistically proper projection in the future in order to more accurately predict future MSL with this approach. However, we maintain confidence in the principle that extension of a linear function defining historical data, be it the 1.5 mm/yr.(0.06 inches/yr.) shown in Figure 9 or 1.7 mm/yr.(0.07 inches/yr.) data as reported in Church and White (2006) is most likely overly-conservative and a risky assumption to the States coastal human and natural resources. Interestingly, extension of the linear 2.9 mm/yr.(0.11 inches/yr.) reported by NOAA (Figure 9), based on the satellite altimetry data, actually exceeds the extrapolation of the nonlinear trend extension (Figure 19), but may still represent an improper technique for forecasting future SLR. More current estimates for GSLR by 2100 range from 0.5 to 2meters (1.6 6.6; Grinsted et al. 2009; Pfeffer et al. 2008; Rahmstorf 2007; Vermeer & Rahmstorf 2009). Much weight recently has been placed on the predictions of Rahmstorf (2007), who modeled GSLR in response to the IPCC global climate change model scenarios (Figure 20) and predicted a range in GSLR of 0.521

MONTHLY MEAN SEA LEVEL (mm, zeroed to December 1982)

400 300 200 100 0 -100 -200 1880 1900 1920 1940 1960 1980 2000 2020 2040 2060 2080 2100 2120 YEAR

732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759

Figure 19. Extrapolation of the linear (black) regression for the Church & White (2011) data (as described in Figure 8) to predict global mean sea level in the 2109 results in a lower estimated sea level than if the quadratic (red) non-linear regression is used.

1.4 meters (1.6 4.6) by 2100, with 1 meter (3.3) being the most likely. Comparisons of empirical data from tide gauges and satellite altimeters from 1990-2006 to model predictions of IPCC scenarios have found that observed GSLR mirrored the highest rates of SLR predicted (Figure 21). This demonstrates that current rates of SLR tracked the highest past IPCC predictions, lending credibility to the higher end of GSLR predictions. A more recent publication by Vermeer & Rahmstorf (2009), supported by a temperature-based SLR model with a 98% correlation with observed data from 1880-2000, increased the predicted GSLR associated with the IPCC scenarios to 0.75-1.90 meters (2.5-6.2). The differences in the GSLR curves shown in Figure 19 resulted from changes in the predicted rate of acceleration of GSLR. The National Research Council (NRC 1987) modeled several scenarios of GSLR increase with variable acceleration rates using Equation 6. Acceleration constants (b from Equation 6) were back-calculated for a priori scenarios of 0.5, 1.0 and 1.5 meters (1.6, 3.3 and 4.9 respectively) GSLR by 2100 (Figure 22) shown by NRC Curves I, II and III, respectively. The 0.5-meter (1.6) increase was a minimum value and corresponded to a minimum acceleration of the rate of rise while the 1.5-meter(4.9) increase was considered to be a maximum value that would occur with a rapid acceleration. It is important to note that the numerical values for the acceleration constants derived by NRC (1987) are specific to that effort and to the assumption of 1.2 mm/yr. as the historical linear rate of GSLR as defined by the variable (a) in Equations 5 and 6. If E(t) from 22

760 761 762 763 764 765 766 767

Figure 20. The rate of global sea-level rise during the 21st Century is modeled to increase, with the extent of acceleration dependent on the predicted temperature increases associated with the IPCC global climate change scenarios. The colored dotted lines are individual scenario-specific predictions of GSLR increase, while the gray dashed lines reflect uncertainty surrounding the statistical fit of the model data. Figure from Rhamstorf (2007).

768 769 770

Figure 21. Observed GSLR for the period 1993-2007 from satellite altimeters exceeded IPCC best estimate predictions of GSLR made in 1995 and 2001. Figure from Pielke (2008).

23

771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788

Figure 22. NRC (1987) Curves as modified in USACE (2009) to account for different assumptions of the historical linear rate of global sea-level rise.

Equation 6 is held steady at 0.5, 1.0 or 1.5 meters by 2100 as shown in Figure 22, but the linear rate of GSLR (a) is corrected for additional information, then mathematically the acceleration constant (b) must be recalculated. This was done in USACE (2009), wherein the historical linear rate of GSLR was instead assumed to be 1.7 mm/yr. As shown in Table 1, this assumption resulted in different acceleration constants from those calculated in NRD (1987). The implication then is that any attempt to establish a predictive sea-level rise curve for future scenarios that uses historical linear values different from NRC (1987) or USACE (2009) must recalculate the appropriate acceleration constants.

Table 1. Comparison of derived acceleration constants in NRC (1987) and USACE (2009) for the generalized predictive GSLR equation E(t) = a*t + b*t2. See Equation 6 for parameter definitions. GSLR Scenario (meters by 2100) Acceleration Constants (b) (meters/yr2) NRC (1987) USACE (2009) (a) = 0.0012 meters/yr. (a) = 0.0017 meters/yr. 2.80 x 10-5 6.60 x 10-5 1.05 x 10-4 2.36 x 10-5 6.20 x 10-5 1.005 x 10-4

0.5 meters 1.0 meters 1.5 meters

789 24

790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832

Compared to the multitude of research performed to derive a global linear trend in SLR, for example GSLR 2.9 0.4 mm/yr. (0.11 0.02 inches/yr.) in Figure 7, there have been very few attempts to calculate an acceleration trend from SLR data (Woodworth et al., 2008; Church & White 2011). If the long term trend in acceleration (approximately 0.01mm/yr2) were to remain constant then sea level would rise 28-34 cm (11-13) by 2100 (Church & White 2004; Jevrejeva et al. 2008). Even though the majority of scientific effort to date has concluded that the rate of SLR is accelerating the claim has been made Houston & Dean (2011a) that although sea-level is rising, the rate is actually decelerating based on the data and time period evaluated. This conclusion has been the subject of vigorous debate in the literature (Rahmstorf and Vermeer 2011; Houston and Dean 2011b; Donoghue and Parkinson 2011; Houston and Dean 2011c). It appears that just as the rate (a) from Equation 6) of SLR is dependent upon the time period evaluated the acceleration of that rate (b from Equation 6) is as well. We infer from the literature review that the acceleration of SLR is the most difficult value to determine from the historic rate and that it changes with the period of record evaluated, ultimately determining the total amount of increase in MSL at any discrete point in time. While we choose to support the consensus that SLR is accelerating, this issue will be closely examined in the future to evaluate progress in this area of predicting GSLR. Additionally, the eustatic contribution to GSLR will most likely have the greatest impact on future SLR rates and that because of the great uncertainty in future ice sheet and glacier contributions, it is also the most uncertain contributor. 2.3.2. Projected Regional Sea-Level Rise in the Gulf of Mexico Unfortunately, there has been very little work done to specifically model the overall change in the Gulf of Mexico water surface for the rest of this century. Until these regional investigations are performed MSL changes must be primarily extracted from satellite altimetry data (Figures 12 and 16), which is less precise due to the wide coverage and the short period of record, or from the average of all the tide gauges, which can be less reliable due to the period of record. In the absence of location-specific sea-level budget analyses it is unclear how the relative contributions of the steric and eustatic components of SLR will manifest in the Gulf. Because we are less able to discriminate between hSTERIC and hMASS as components of an overall hTOT for the Gulf we must assume that a shift to a eustatic-driven SLR will affect the Gulf similarly to the global oceans as described above. Modeling work of global climate change scenarios included in the IPCC AR4 report does suggest that the Gulf of Mexico will respond similarly to the coastal ocean, and it is reasonable to assume that projections of GSLR are appropriate to carry into the Gulf (Figure 23). 2.3.3. Gulf of Mexico Regional Sea-Level Rise Rate Recommended for CPRA Use To calculate a predicted future sea-surface elevation offshore of coastal Louisiana, and based on the literature reviewed, LACES is recommending that CPRA staff assume that Gulf SLR will be 1 meter (3.3) by 2100, with a bounding range of 0.5 1.5 meters (1.6 4.9). However, it is recommended for project specific application that the appropriate acceleration constants (b) be calculated after the local Gulf historical linear trend (a) is derived, accounting for the substantial east-west variation in near-shore SLR observed through satellite altimetry (Figure 16).

25

833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855

Figure 23. This map predicts local sea level change relative to the global average for the 21st century, as calculated from the results of 16 global climate change models running the IPCC A1B climate scenario. Results suggest that sea level changes in the Gulf of Mexico during the next 90 years will not differ substantially from the global mean, and that the northern and western Gulf may in fact respond with a slightly less local sea level change than the global average. Figure from Meehl et al. (2007).

While this recommendation results from an independent assessment of the available data, it is generally conservative when compared to recent publications predicting future GSLR (Figure 24). This recommendation is also consistent with similar efforts ongoing in other states. The Miami-Dade County Climate Change Task Force (2010) recommended that all county agencies include SLR estimates into their planning documents accounting for a 0.46-m (1.5) rise in sea level by 2050 and an SLR of 0.9-1.5 m (3-5) by 2100. The first of three Maryland Coastal Program Technical Guidance reports for Dorchester County (Cole 2008) estimates 0.6-0.9 m (23) of SLR for the Chesapeake Bay region by 2100. This recommendation is only part of the overall prediction of future relative SLR, and must be combined with predictions of subsidence and marsh vertical accretion. Those two factors are described next, and the overall recommendation for estimating RSLR for project planning and design purposes will follow in the last section of this report.

26

856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876

Figure 24.CPRA-LACES recommendations for future global sea-level rise by 2100, compared to a range of values from recent published scientific literature. The red line indicates the primary LACES recommendation of 1 meter GSLR by 2100, with a bounding range of 0.5 and 1.5 meters (green lines). Figure from http://wh.er.usgs.gov/slr/sealevelrise.html.

2.4.

Relative Sea-Level Rise in Coastal Louisiana

2.4.1. Estimates of Historical Relative Sea-Level Rise in Coastal Louisiana Relative sea-level refers to the height of sea level as measured from a particular point or area on the earth's surface. Change in relative sea-level usually results from the interaction of two independent processes: 1) change in the absolute elevation of the earth's ocean (GSLR), and 2) local change (uplift or subsidence) in the absolute elevation of the land mass. Adding to Equation 6, RSLR can be represented by the generalized equation: E(t) = a*t + b*t2 + S (Eqn. 7)

where E, t1, t2,a and b are as defined by Eqn. 6, and S is rate of subsidence (or uplift in areas of glacial rebound).

27

877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918

The exact mathematic calculations of predicted future RSLR are more complex than that shown in Equation 7, however. Close examination of the NRC (1987) acceleration scenarios shown in Figure 22, and those carried through into USACE (2009), highlight the very important caveat that those scenarios assume a 1986 start to achieve the relevant 2100 MSL. As such, the mathematical values of the acceleration constants are dependent on that specific starting point, and the generalized equation last defined in Equation 6 needs to be altered at this point because of this specificity. Specifically, the generalized equation needs to be replaced by E (t2-1986)-E(t1-1986)=a([t2-1986]-[t1-1986])+b(([t2-1986]2-[t1-1986]2) + S([t2-1986]-[t1-1986]) where all parameters are as defined in Equation 7. All other factors held equal, new acceleration constants will have to be calculated if a more contemporaneous date is assumed as t0. Tide gauges directly measure RSLR for the water bottom on which the gauge is located. NOAA presently maintains 19 tide gauges in Texas, Louisiana and Mississippi relevant to the hydrodynamic environment of coastal Louisiana (Figure 25, Table 2). Of these only two have the long-term period of record discussed earlier to justify calculating an RSLR trend line: Grand Isle and Sabine Pass North (Figure 26a and b). The linear RSLR trend lines as calculated in Zervas (2009) are substantially different for the two stations, and illustrate the importance of geology on the physical stability of those stations. Sabine Pass North, in the more stable Chenier Plain of southwestern Louisiana, has a much lower mean RSLR linear trend than Grand Isle, which is located within the Mississippi River Deltaic Plan. Moreover, the RSLR trend for Sabine Pass North is more than triple the 1.79 mm/yr.(0.070 inches/yr.) average GSLR for the 1958-2006 period of record (data from Church and White in press, as shown in Figure 8), and the RSLR trend for Grand Isle is more than five times the 1.73 mm/yr. (0.068 inches/yr.) GSLR linear trend for the 1947-2006 period of record. This comparatively high rate of RSLR is caused primarily by land subsidence, unequivocally the most variable and significant contributor to relative sea-level rise in coastal Louisiana. There has been some discussion of the utility of the RSLR trend from the Eugene Island Station (Figure 26c). The period of record for that station is only 1939-1974, so it is neither possible for that data to inform a present understanding of RSLR for the central Louisiana coast nor serve as the basis for estimates of future RSLR. This is especially the case given that the Eugene Island period of record coincides with some of the highest documented rates of coastal land loss in coastal Louisiana (Barras 2009). It is, however, possible to compare the Eugene Island data to that from Sabine Pass North and Grand Isle where the periods of record overlap and thus allow for some inference of the historical comparative stability of the central coast versus the western and eastern margins of the coast.

(Eqn. 8)

28

919 920 921 922 923 924 925 926 927 928 929 930 931

Figure 25. NOAAs tide gauge network in Louisiana covers multiple geomorphic settings within the States coastal zone. Note that three stations are not shown on this map: Carrollton, Crescent City Air Gap, and Huey Long Bridge Air Gap. Note also that the Stouts Pass and Mesquite Point stations have been decommissioned. Figure from http://egisws01.nos.noaa.gov/website/co-ops/stations/viewer.htm.

Table 2. A summary of the current NOAA tide gauge stations shows that only two stations relevant to coastal Louisiana have a sufficient period of record to establish an RSLR trend. RSLR trends are not shown if not given by NOAA. NOAA has only calculated linear trends for t he Grand Isle and Sabine Pass North gauges.
Station Bay Waveland Yacht Club Calcasieu Pass Carrollton Crescent City Air Gap Cypremort East Bank Freshwater Canal Grand Isle Gulfport Huey Long Bridge Air Gap Lake Charles LAWMA, Amerada Pass Pilots Station East Port Fourchon Rainbow Bridge, TX Sabine Pass North Tesoro Marine Terminal USCG New Canal West Bank NOAA Station ID 8747437 8768094 8761955 8761847 8765251 8762372 8766072 8761724 8745557 8762002 8767816 8764227 8760922 8762075 8770520 8770570 8764044 8761927 8762482 Period of Record 1978-present 2008- present 1996-present 1984-present 2005- present 2003- present 2005- present 1947-present 1979-present 2009-present 2002- present 2005- present 2004- present 2003- present 1996-present 1958-present 2003- present 2005- present 2003- present RSLR Trend n/a n/a n/a n/a n/a n/a n/a 9.24 0.59 mm/yr.* n/a n/a n/a n/a n/a n/a n/a 5.66 1.07 mm/yr.* n/a n/a n/a

932 933 934

* RSLR trend calculated by Zervas (2009) for Grand Isle was for the time period 1947-2006, and the trend calculated for Sabine Pass North was for the time period 1958-2006.

29

935 936

(a)

937 938

(b)

939 940 941 942 943 944 945 946 947 948 949

(c)
Figure 26. RSLR trend lines for the Sabine Pass North (a) and Grand Isle (b) NOAA tide gauges illustrate the significance of geological stability on RSLR, and the difference between the more stable Chenier Plan (Sabine) and the less stable Deltaic Plain (Grand Isle). The RSLR trend for Sabine Pass North, 1958-2006, was 5.66 1.07 mm/yr. (0.22 0.042 inches/yr.). The RSLR trend for Grand Isle, 1947-2006, was 9.24 0.59 mm/yr. (0.36 0.023 inches/yr.).The RSLR trend line for the Eugene Island tide gauge (c) for the period of record 1939-1974 demonstrates a high rate of RSLR, 9.65 1.24 mm/yr. (0.38 0.049 inches/yr.) at that station in the central Louisiana coast. The station has been removed, so only comparisons with historically contemporary stations are possible. Figures from Zervas (2009).

30

950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994

The LCA Science & Technology Program is likewise finalizing a report that calculates linear RSLR trend lines for nineteen USACE-operated tide gauges in south Louisiana (Ayres in preparation). That information will be included in this report as an amendment to Table 2 and Figure 25. Once CPRA-LACES, most likely with NOAA assistance, has calculated linear trends for all of the tide gauges shown in Figure 25, we will establish an appendix for the total pool of NOAA and USACE tide gauges as a concise reference for CPRA project delivery teams. 2.4.1. Subsidence The tide gauge data emphasize the importance of being able to document the contributions of subsidence and accretion to the overall RSLR at discrete points in coastal Louisiana. Any effort to confidently incorporate potential SLR impacts on coastal wetlands into planning must account for the sum of factors influencing RSLR: 1) the change in the surface elevation of the Gulf of Mexico, which is the primary topic of this document; 2) local land surface elevation change, which in Louisiana is exclusively represented as subsidence; and 3) marsh vertical accretion, which can offset some SLR impacts. This report does not attempt to exhaustively review the last two topics, but will summarize relevant products from the state of the science that can inform CPRA activities. Subsidence is widely recognized as a significant driver of relative sea-level rise in southern Louisiana, and probably the principal driver in southeast Louisiana for the near-term. There are a number of independent factors that influence the rate of subsidence (Reed and Yuill 2009). At the local scale the dominant factor may vary and thus we also recognize that rates of subsidence are highly variable across the Louisiana coastal zone. However, our understanding of the exact rates of subsidence at the local level is very limited. Recent attempts have been made to acknowledge that spatial variability in subsidence rates and factor that variability into program and project planning. The hydraulic and hydrodynamic (H&H) model for the proposed Donaldsonville, Louisiana to the Gulf of Mexico Flood Control Project utilized a digitized version of a coarse-scale map developed by Del Britsch of the USACE New Orleans District Office (Figure 27). More recently, as part of the modeling effort informing the 2012 revision of the States Master Plan, a Subsidence Advisory Group met on 14 September 2010 to assemble a draft map of a range of subsidence values that the States coastal zone can expect through 2060 (Figure 28). The ranges in that map are significant, both within and across polygons. For the project-effects modeling being undertaken for the Master Plan revision, the two subsidence scenarios being analyzed are using values of 20% and 50% of the range from the minimum value of the individual polygons shown in Figure 28. Ongoing work by the Louisiana Geological Survey, commissioned by CPRA, will summarize our understanding of the geological framework underlying south Louisiana as well as provide an overview of historical rates of subsidence across the landscape. This work will further build on the summary report Understanding Subsidence in Coastal Louisiana (Reed &Yuill 2009), prepared for the State and the US Army Corps of Engineers by the LCA Science and Technology Program Office. It is hoped that this information, as well as continued monitoring data from the CWPPRA Programs Coastal Reference Monitoring System (CRMS)-Wetlands stations, will help to tighten the predicted ranges of subsidence shown in Figure 28. 31

995 996

(a)

997 998 999 1000 1001 1002 1003

(b) Figure 27. A map of subsidence developed by Britsch in 2007 illustrates the spatial variability in predicted subsidence rates in southern Louisiana. A portion of this map was digitized (b) for use in the hydrodynamic modeling for the Donaldsonville to the Gulf project. Figure (b) from CHT 2010.

32

1004 1005 1006 1007 1008

Figure 28. Map of projected subsidence ranges for south Louisiana generated by the Subsidence Advisory Panel for the Louisiana CPRA Master Plan 2012 Update, following a meeting on 14 October 2010.

33

1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049

2.4.2. Marsh Vertical Accretion Our understanding of marsh vertical accretion is likewise evolving. The ability of marshes to keep up with moderate levels of RSLR via accretion of both mineral and organic soil material has long been understood (see summary in Mitsch and Gosselink 2000). Typically coastal marshes have a range of optimum depth and within this range they will respond positively to a rise in sea-level until it reaches a certain, heretofore undetermined, rate of rise wherein marshes will no longer be able to accrete and will drown. Organic matter production in coastal marshes is directly related to the rate of RSLR, and that up to a critical collapse threshold marshes in coastal Louisiana have the potential to organically accrete and match substantial levels of RSLR. Consistent with the calculations of RSLR above, we can thus define marsh vertical accretion as MVA = f(E(t)) where MVA is the rate of marsh vertical accretion, and E(t) is as defined in Equations 5-8. The optimum depth and marsh collapse threshold is likely unique to each marsh type and will depend upon the ability to maintain elevation via vegetative growth. Note this differs significantly from macrotidal coastlines where marsh accretion is sediment driven. This information will be more difficult than the subsidence predictions to incorporate in program and project planning because accretion in marshes is influenced by natural cycles, spatially dependent on the species mix of the plant communities of interest and is largely dependent on initial elevation relative to the water surface. It would be desirable to generate a map that spatially describes potential accretion in wetlands and identify by wetland type the critical threshold point of MSL increase beyond which marsh elevation collapses. It is necessary to incorporate the 18.6-year lunar nodal cycles into any accretion models in order to reflect the potential for collapse under an accelerating RSLR scenario. The failure to account for dynamic rates of GSLR under future scenarios risks underestimating the inundation stress that marsh vegetation will see, possibly leading to overly-optimistic predictions of vertical accretion and marsh persistence. While this science is nascent at present, it promises to be a significant contribution to predicting local net RSLR in Louisianas coastal wetlands. A good example of a methodology for predicting marsh vertical accretion can be seen in the activities of the Project Delivery Team (PDT) for the Southwest Coastal Feasibility Study. The methodology adopted by that PDT and its partner Habitat Evaluation Team (HET) established a 7 mm/yr. (0.28 inches/yr.) threshold under which wetland vegetation will continue accreting organic matter (see Appendix C). Beyond that threshold the wetland is assumed to convert to open water. In the case of the analysis predicting future landscapes for that feasibility study, these assumptions were applied to cells in the geospatial grid (i.e. persistence as a wetland or conversion to open water of a specific grid cell in the model). (Eqn. 9)

34

1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095

3.

SUMMARY AND RECOMMENDATIONS

Louisiana is experiencing a higher rate of RSLR than other parts of the world because of naturally occurring regional land subsidence. In a recent USGS Coastal Vulnerability Assessment of the Northern Gulf of Mexico (Pendleton et al. 2010) virtually the entire Louisiana coastline is identified as having a coastal vulnerability index (CVI) risk ranking as very high. This assessment developed CVI rankings for virtually the entire US shoreline by focusing on six variables which strongly influenced coastal response to SLR: 1) geomorphology, 2) historical shoreline change rate, 3) regional coastal slope, 4) relative sea-level change, 5) mean significant wave height and 6) mean tidal range. Although these variables are difficult to separate the end result is clear: the Louisiana coastline is one of the most at risk shorelines in the Northern Gulf of Mexico to the impacts of GSLR, the effects of which are clearly manifest in the rate of RSLR. The scientific literature indicates that the rate of global SLR (GSLR) has been increasing steadily over the past several centuries. This may be seen in an increase from a 20th Century linear average based on tide gauge data of 1.7 0.4 mm/yr.(0.070 0.016 inches/yr.) to a linear estimate for the past eighteen years of 2.9 0.4 mm/yr.(0.11 0.016 inches/yr.) in the Gulf of Mexico based on satellite altimetry. Although direct comparison of the two techniques supports the validity of the altimeter readings, there is some concern regarding the short period of record for the altimetry data. However, evidence suggests that SLR for the available period of record is best represented as a single, non-linear function, which has important implications for relating RSLR and GSLR estimates, and especially for assumptions of the differential representing local land surface change. More important for CPRA planning purposes is the projection of future GSLR. Based on the available data, LACES recommends that any SLR modeling scenarios models for state restoration projects assume a 1-meter (3.3) MSL rise by 2100 compared to the late 1980sand should be bracketed by GSLR ranges of 0.5-1.5 meters (1.4-4.9) by 2100. The specific recommendation for factoring in the range of GSLR into local calculations of RSLR is given below. While GSLR is an important factor to consider, data has revealed that variations in sea levels exist in the many regions of the earths oceans and water bodies. This regional variation associated with the Gulf of Mexico is much more relevant to coastal Louisiana. This paper examines regional SLR variation and, based on available data, concludes that while it does seem appropriate to bring future global SLR scenarios into the Gulf of Mexico, there is a definable east-west gradient in recent historical SLR across the Louisiana coast that needs to be considered in project planning. However, GSLR presents only one piece of the puzzle when anticipating future sea level rise when planning and designing coastal restoration and protection projects. Subsidence and marsh vertical accretion represent two other factors which must be included and, in fact, may dominate land change dynamics and amplify the effects of SLR. They are critical for predicting the RSLR that the coastal wetland plant communities perceive, and both are subject to extremely high spatial and temporal variation across the Louisiana coastal zone. Although work on subsidence and marsh vertical accretion is continuing, this report gives instruction on how to determine 35

1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140

these values and how to include them in project design. CPRA has identified a number of technical uncertainties surrounding these recommendations that it plans to address through future research and development activities. It is CPRAs goal that these activities will help constrain the uncertainties associated with predicting the impacts of future increases in sea level and increase our confidence in planning and implementing projects to achieve sustainable coastal Louisiana. Recommendations for Calculating RSLR in Coastal Louisiana Based on the information presented to this point, it is our recommendation that when participating in project planning and design activities, local RSLR be calculated using the following procedure to populate the variables of the generalized RSLR equation E(t) = a*t + b*t2 + S. (Eqn. 7)

1. Use local observations of historical sea-level rise from contemporary satellite altimetry (Figure 16) just offshore of coastal Louisiana, in order to account for the substantial eastwest gradient in documented rates. Specifically, we recommend using an average of the three most proximate points shown in Figure 16. This is the rate of SLR (mm/yr.) and variable (a) from the generalized equation. 2. Calculate the acceleration constant that assumes a MSL increase of 1 meter (3.3) by 2100 as the most heavily-weighted project alternative, while also testing MSL increases of 0.5 meters (1.6) and 1.5 meters (4.9) to account for uncertainty in the literature. This provides the change in water levels over time at a project location. To localize further, 3. Add in local subsidence values obtained from the most proximate local source, which is variable (S)in Equation 7. In order to predict the persistence the coastal wetland, and specifically the persistence of the wetland surface or conversely marsh surface collapse and drowning, a fourth step is necessary. 4. Use the sum of #s 1-3 above to establish an inundation function, especially the rate of inundation for the period of analysis, in order to predict local responses of marsh vertical accretion as those models and data products become available. This can be inferred from scientific literature if no reliable data exist on site, or can be estimated from vegetation productivity models if available. As discussed in Section 2.4.1., predicting future RSLR must account for the acceleration constants (variable b) being specific to NRC (1987) acceleration scenarios having a starting point of 1986. Appendix D of this report shows specifically how the variables discussed feed into a refined version of Equation 8 that accounts for that specific starting point.

36

1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151

Note that this recommended process differs from that described by the US Army Corps of Engineers (USACE) Engineering Circular (EC) 1165-2-211 (USACE 2009). The EC mandates how the USACE must estimate RSLR during the planning and engineering of water resources projects in the coastal zone, and LACES staff recognize that USACE staff participating on a PDT with the State for projects cost-shared with USACE will have to run RSLR scenarios in accordance with the EC. However, because of a number of significant technical issues with the EC that are described in Appendix B, we recommend that State staff participating on those PDTs also require the RSLR scenarios described in this document be run in addition to the EC-defined scenarios, and that the 4-step recommendation described here supersede EC mandates on any projects that the State is pursuing without the Corps as a cost-share partner.

37

1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197

4.

REFERENCES

Ablain, M., A. Cazenave, G. Valledeau & S. Guinehut. (2009). A new assessment of the error budget of global mean sea level rate estimated by satellite altimetry over 1993-2008. Ocean Science5: 193-201. Allison, I., R. B. Alley, H. A. Fricker, R. H. Thomas & R. C. Warner. (2009). Review: Ice sheet mass balance and sea level. Antarctic Science21: 413-426. Antonov, J. I., S. Levitus & T. P. Boyer. (2005) Thermosteric sea-level rise, 1955-2003. Geophysical Research Letters, 32, L12602, doi:10.1029/2005GL023112. Bamber, J. L., R. E. M. Riva, B. L. A. Vermeersen & A. M LeBrocq. (2009). Reassessment of the potential sea-level rise from a collapse of the west Antarctic ice sheet. Science324: 901903. Bindoff, N.L., J. Willebrand, V. Artale, A, Cazenave, J. Gregory, S. Gulev, K. Hanawa, C. Le Qur, S. Levitus, Y. Nojiri, C.K. Shum, L.D. Talley and A. Unnikrishnan. (2007). Observations: Oceanic Climate Change and Sea Level. In: Climate Change 2007: The Physical ScienceBasis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Buis, A. 2011. Bump in the road. NASA Global Climate Change News, 23 August 2011. http://climate.nasa.gov/news/index.cfm?FuseAction=ShowNews&NewsID=570. Cazenave, A.,K. Dominh, S. Guinehut, E. Berthier, W. Llovel, G. Ramillien, M. Ablain & G. Larnicol. (2008). Sea level budget over 2003-2008: A reevaluation from GRACE space gravimetry, satellite altimetry and Argo. Global and Planetary Change65: 83-88. Cazenave, A. & W. Llovel. (2010). Contemporary sea level rise. Annual Reviews of Marine Science2: 145-173. Childers, D. L., J. W. Day & R. A. Muller. (1990). Relating climatological forcing to coastal water levels in Louisiana estuaries and the potential importance of El Nino-Southern Oscillation events. Climate Research1: 31-42. Church, J. A. & N. J. White. (2006). A 20th century acceleration in global sea-level rise. Geophysical Research Letters33:L01602, doi:10.1029/2005GL024826. Church, J.A., and N.J. White. (2011). Sea-level rise from the late 19th to the early 21st century. Survey of Geophysics online publication, 30 March 2011. Church, J. A., N. J. White, R. Coleman, K. Lambeck& J. X. Mitrovica. (2004). Estimates of the regional distribution of sea-level rise over the 1950-2000 period. Journal of Climate, 17, pp2609-2625. CHT LLC. 2010. Donaldsonville to the Gulf Barataria Bay Interior Drainage and Transport Model Study. Final Report by Computational Hydraulics and Transport LLC for New Orleans District, U.S. Army Corps of Engineers. 296 pages. Cole, W. D. (2008). Sea level rise: technical guidance for Dorchester County. Maryland Department of Natural Resources, Coastal Zone Management Division. 60 pp. www.dnr.state.md.us/dnrnews/pdfs/Dorchester.pdf Domingues, C. T., J. A. Church, N. J. White, P. J. Gleckler, S. E. Wijffels, P. M. Barker & J. R. Dunn. (2008). Improved estimates of upper-ocean warming and multi-decadal sea-level rise. Nature, 453, 19, pp1090-1094. Donoghue, J.F., and R.W. Parkinson (2011). Discussion of: Houston, J.R. and Dean, R.G., 2011. 38

1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243

Sea-Level AccelerationBased on U.S. Tide Gauges and Extensions of Previous GlobalGaugeAnalyses. Journal of Coastal Research, 27(3), 409417. Journal of Coastal Research27: 994-996. Douglas, B. C. (2001). Sea level Rise, History and Consequences. International Geophysics Series, Vol 75. Eds. B. C. Douglas, M. S. Kearney & S. P Leatherman. Academic Press, San Diego, CA. 250 pp. Eilperin, J. 2011. Weather cycles cause a drop in global sea level, scientists find. Washington Post, 25 August 2011. http://www.washingtonpost.com/national/health-science/weathercycles-cause-a-drop-in-global-sea-level-scientists-find/2011/08/25/gIQA6IeaeJ_story.html. Ferrigno, J.G., Cook, A.J., Mathie, A.M., Williams, R.S., Jr., Swithinbank, Charles, Foley, K.M., Fox, A.J., Thomson, J.W., and Sievers, Jrn, 2009, Coastal-change and glaciological map of the Palmer Land area, Antarctica: 19472009: U.S. Geological Survey Geologic Investigations Series Map I2600C, 1 map sheet, 28-p. text. Fssel, H. (2009). An updated assessment of the risks from climate change based on research published since the IPCC Fourth Assessment Report. Climate Change, Vol 97, pp 469-482. Gomez, N., J. X. Mitrovica, M. E. Tamisiea& P. U. Clark. (2009). A new projection of sea level change in response to collapse of marine sectors of the Antarctic ice sheet. Geophysical Journal International, doi: 10.1111/j.1365-246X.2009.04419.x Gratoit, N., E. J. Anthony, A. Gardel, C. Gaucherel, C. Proisy& T. J. Wells. (2008). Significant contribution of the 18.6 year tidal cycle to regional coastal changes. Nature Geoscience 1: 169-172. Grinsted, A., Moore, J. C., & Jevrejeva, S. (2009). Reconstructing sea level from paleo and projected temperatures 200 to 2100 ad.Climate Dynamics, 34(4), 461-472. Grger, M., and H. Plag. (1993). Estimations of a global sea level trend: limitations from the structure of the PSMSL global sea level data set. Global Planetary Change 8: 161-179. Houston, J.R. and R.G. Dean (2011a). Sea-level acceleration based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyses. Journal of Coastal Research27: 409-417. Intergovernmental Oceanographic Commission. (2006). Manual on sea level measurement and interpretation. Vol IV: An update to 2006. JCOMM Technical Report No. 31. 88 pp. Houston, J.R.,and R.G. Dean (2011b). Reply to: Rahmstorf, S. and Vermeer, M., 2011. Discussion of: Houston, J.R.and Dean, R.G., 2011. Sea-Level Acceleration Based on U.S. Tide Gauges andExtensions of Previous Global-Gauge Analyses. Journal of Coastal Research,27(3), 409417.Journal of Coastal Research27: 788-790. Houston, J.R.,and R.G. Dean (2011c). Reply to: Donoghue, J.F. and Parkinson, R.W., 2011. Discussion of: Houston, J.R. and Dean, R.G., 2011. Sea-Level Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyses. Journal of Coastal Research, 27(3), 409417. Journal of Coastal Research27: 997-998. Jevrejeva, S., A. Grinsted, J. C. Moore, & S. Holgate. (2006). Nonlinear trends and multiyear cycles in sea level records. Journal of Geophysical Research, 111, C09012, 11pp. Jevrejeva, S. A., J. C. Moore & A. Grinsted. (2008). Relative importance of mass and volume changes to global sea-level rise. Journal of Geophysical Research, 113, D08105, doi:10.1029/2007JD009208. Jeuken, M.C.J.L., Z.B. Wang, D. Keiller, I. Townend& G.A. Like. (2003). Morphological response of estuaries to nodal tide variations. International Conference on Estuaries and Coasts; Hangzhou, China. Khan, S. A., J. Wahr, M. Bevis, I. Velicogna & E. Kendrick. (2010). Spread of ice mass loss into 39

1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289

northwest Greenland observed by GRACE and GPS. Geophysical Research Letters, 37, L06501, doi:10.1029/2010GL042460. Kolker, A. S. & S. Hameed. (2007). Meteorologically driven trends in sea-level rise. Geophysical Research Letters, Vol 34, L23616, doi:10/1029/2007/GL03814. Lombard, A., D. Garcia, G. Ramillien, A. Cazenave, R. Biancale, J. M. Lemoine, F. Fletcher, R. Schmidt & M. Ishii. (2007). Estimation of steric sea level variations from combined GRACE and Jason-1 data. Earth and Planetary Science Letters, 254, pp 194-202. Leuliette, E. W. & L. Miller. (2009). Closing the sea level rise budget with altimetry, Argo, and GRACE. Geophysical Research Letters, Vol 36, L04608, doi:10.1029/2008GL036010. Lyman, J. M., S. A. Good, V. V. Gouretski, M. Ishii, G. C. Johnson, M. D. Palmer, D. M Smith & J. K. Willis. (2010). Robust warming of the global upper ocean. Naure, Vol 465, doi:10.1038/nature09043. Lyman, J. M., J. K. Willis & G. C. Johnson. (2006). Recent cooling of the upper ocean. Geophysical Research Letters, 33, L18604, doi:10.1029/2006GL027033. Meehl, G.A., T.F. Stocker, W.D. Collins, P. Friedlingstein, A.T. Gaye, J.M. Gregory, A. Kitoh, R. Knutti, J.M. Murphy, A. Noda, S.C.B. Raper,I.G. Watterson, A.J. Weaver and Z.-C. Zhao. (2007). Global Climate Projections. In: Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change[Solomon, S.,D. Qin, M. Manning, Z. Chen, M. Marquis, K.B. Averyt, M. Tignor and H.L. Miller (eds.)]. Cambridge University Press, Cambridge, UnitedKingdom and New York, NY, USA. Merrifield, M. A., S. T. Merrifield & G. T. Mitchum. An Anomalous Recent Acceleration of Global Sea Level Rise. Journal of Climate, Volume 22. http://dx.doi.org/10.1175/2009JCLI2985.1 Miami-Dade County Climate Change Advisory Task Force. (2010). Annual Report and Supplemental Recommendations. 49 pp. http://www.miamidade.gov/derm/climatechange/library/ccatf_2010_report.pdf Mitrovica, J. X., N. Gomez & P. U. Clark. (2009). The sea-level fingerprint of West Antarctic collapse. Science, Vol 323, pp 753. Mitsch, W. J. & J. G. Gosselink. (2000). Wetlands. John Wiley & Sons, Inc. New York, NY. 920pp. Morris, J. T. & K. Sundberg. (2008). Astronomical forcing of salt marsh biogeochemical cascades. American Geophysical Union, Fall meeting,. Morton, R.A., Bernier, J.C., and Kelso, K.W., 2009, Recent subsidence and erosion at diverse wetland sites inthe southeastern Mississippi delta plain: U.S. Geological Survey Open-File Report, 20091158, 39 p., plus app.(p. 41-221). Nicholls, R. J. & A. Cazenave. (2010). Sea-level rise and its impact on coastal zones. Science,Vol 328, pp 1517 1520. National Research Council. (1987). Responding to Changes in Sea Level: Engineering Implications. National Academies Press, Washington, D.C. 160 pp.http://www.nap.edu/catalog/1006.html Pfeffer, W. T., J. T. Harper & S. O. Neel. (2008). Kinematic constraints on glacier contributions to 21st-century sea-level rise. Science, Vol 321,pp 1340-1343. Pielke, R. A. Jr. (2008). Climate predictions and observations. Nature Geosciences1: 206. Prandi, P., A. Cazenave & M. Becker. (2009). Is coastal mean sea level rising faster than the global mean? A comparison between tide gauges and satellite altimetry over 1993-2007. 40

1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331

Geophysical Research Letters, Vol 36, L05602, doi:10.1029/2008GL036564. Rahmstorf, S. (2007). A semi-empirical approach to projecting future sea-level rise. Science, Vol315, pp 368-370. Rahmstorf, S., and M. Vermeer (2011). Discussion of: Houston, J.R. and Dean, R.G., 2011. SeaLevel Acceleration Based on U.S. Tide Gauges and Extensions of Previous Global-Gauge Analyses. Journal of Coastal Research, 27(3), 409417. Journal of Coastal Research 27: 784-787. Reed, D. J., & B. Yuill. (2009). Understanding subsidence in Coastal Louisiana. Ponchartrain Institute for Environmental Sciences, University of New Orleans, New Orleans, LA. 69 pp. Roemmich, D. & J. Gilson. (2009). The 2004-2008 mean and annual cycle of temperature, salinity, and steric height in the global ocean from the Argo Program. Progress in Oceanography, 82, pp 81-100. United States Army Corps of Engineers. (2009). Water resource policies and authorities incorporating sea-level change considerations in civil works programs. Circular No. 1165-2211. Vaughan, D. G. (2009). West Antarctic ice sheet collapse the rise and fall of a paradigm. (2009) Climate Change, DOI 10.1007/s10584-008-9448-3. Vermeer, M. S., & S. Rahmstorf. (2009). Global sea level linked to global temperature. Proceedings of the National Academy of Sciences 106: 21527-21532. Velicogna, I. (2009). Increasing rates of ice mass loss from the Greenland and Antarctic ice sheets revealed by GRACE. Geophysical Research Letters, 36, L19503, doi:10.1029/2009GL040222. Willis, J. K., D. Roemmich & B. Cornuelle. (2004). Interannual variability in upper ocean heat content, temperature, and thermosteric expansion on global scales. Journal of Geophysical Research, 109, C12036, doi:10.1029/2003JC002260. Willis, J. K., J.M. Lyman, G. C. Johnson & J. Gilson. (2007). Correction to Recent cooling in the upper ocean,. Geophysical Research Letters, 34, L16601, doi:10.1019/2007GL030323. Willis, J. K., D. P. Chambers & R. S. Nerem. (2008). Assessing the globally averaged sea level budget on seasonal to interannual timescales. Journal of Geophysical Research, 113, C06015, doi:10.1029/2007JC004517. Willis , J. K., J. M. Lyman, G. C. Johnson & J. Gilson. (2009). In situ biases and recent ocean heat content variability. Journal of Atmospheric and Oceanic Technology, 26, pp 846-852. Woodworth, P.L., N.J. White, S. Jevrejeva, S.J. Holgate, J.A. Church and W.R. Gehrels. 2009. Evidence for the accelerations of sea level on multi-decade and century timescales. International Journal of Climatology29: 777-789. Wunsch, C., R. M. Ponte & P. Heimbach, (2007). Decadal Trends in Sea Level Patterns: 1993 2004. Journal of Climate, Vol 20, pp 58895911. doi: 10.1175/2007JCLI1840.1 Zervas, C. (2001). Sea level variations of the United States 1854-1999. Technical Report NOS CO-OPS 36. 80 pp. Zervas, C. (2009). Sea level variations of the United States 1854-2006. Technical Report NOS CO-OPS 53. 194 pp.

41

1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378

APPENDICES
Appendix A: Reviewer Comments and Responses

CPRA-LACES released a draft document for internal LACES comment on 27 September 2010. Comments on 20100927CPRA LACES SLR Document FINAL DRAFT, 8 October 2010, from Honora Buras, Louisiana Office of Coastal Protection and Restoration I only have one technical comment, based on some information I ran across a while back that might be relevant. I was asked several years ago by Karims Ecological Review group to look into subsidence rates to use that were more updated than those based on Shea Penlands work (1988, 1990) using tide gauge data. There were some issues that I ran across related to the use of these data that I never saw any resolution of. This may be relevant or not to your paper, but I just wanted to bring it to your attention, since you reference these tide gauge data in numerous places. Sheas analysis was done before we were aware of the extent of coast-wide sinking of benchmarks that these tide gauges were tied to. It was never clear whether any of the records were later adjusted for this. Fortunately, the tide gauge with the longest record Grand Isle- is one of the most stable of all and apparently subsidence in this location is much less than in most of the coast (according to Dokkas work). However, even this gauge was actually a replacement for a different gauge in a slightly different location. This is not often mentioned when referencing the long-term record. Some said this older gauge was on the opposite side of a fault, thus differentially subsiding. Others disputed this. Only the earliest years of the record are at this other location. The datum used for all of these before NAVD88 was not very reliable. (I was once told that NGVD stood for No Good Vertical Datum.) Another problem with using tide gauge data for subsidence was that many of them are actually strongly influenced by localized runoff or river flow. I remember there was one near the mouth of the Atchafalaya that was obviously influenced by the heavier flows in flood years, I think this is the one at Eugene Island. Since this still is a reflection of water surface elevation, although likely somewhat localized, it may not be the issue in calculating SLR at the local level as it might be for subsidence alone. In fact, Sheas estimates really gave more of a RSLR than subsidence alone, so they worked for the purpose at the time and we had nothing better. This was also before GSLR was so well studied. Since I have not been involved in any of the efforts on subsidence since that time, I do not know if this question has already been addressed by the subsidence team or others. I have not seen it addressed in any white papers or proceedings from any of the subsidence working groups. The issue of sinking benchmarks, and the extent of the subsidence problem was very controversial in house at the time. This was right about the time the draft NOAA technical report 50 (Shinkle & Dokka 2004) on subsidence was being debated. You describe some localized east-west trends in SLR that need to be accounted for in addition to the GSLR rate. Even though you state this was based on altimetry data, is it possible that this trend is related to the subsidence trends that may be manifest in some of the data, especially if it is based on tide gauges? In looking at the map in figure 17, it appeared to have some correlation with the subsidence trends. While I recognize that the information presented in this paper is based on highly technical research and requires the use of terminology that may not be familiar to the reader, I believe the 42

1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424

paper, in some cases, focuses more on the details of the scientific debate than the relevance and may not be useful for the average CPRA employee attempting to understand and apply the information in project planning, design and implementation. I assume this is not meant to be a paper for publication in a climate change journal, but for practical use as applied science for our purposes. Even if meant to be published, white papers are usually much more simplified and summary documents that take the information gleaned from various sources and synthesize it into a format and language that is easy for most readers to understand, even if they have no training in the subject area. You do a good job of defining the terminology where it is more easily understood, but the language in the summaries is difficult to follow. The conclusions need to be explained or at least summarized in language that everyone from engineers, biologists, ecologists, geologists, and oceanographers to senior administrative officials can understand. For example, in the executive summary, I would not use the numbered bullet format in lines 3957 to explain such a complicated concept with such long, complicated sentences. (The same applies to the similar write-up on pp 25-26.) Bullets should be reserved for simplified concepts or points. These are paragraphs. If I was merely trying to figure out what rate of GSLR to use, and didnt care about all the background information that you used to derive it, I would have a hard time putting my finger on it easily. Provide the description in text then give an example of its use in one or more typical applications that are anticipated. Where the process is described for determining what number to use for SLR at specific locations, the format is especially awkward. It would be useful to separate the explanation of how the numbers are derived from the actual number to use and the formulas. The executive summary overall should be more of a very simplified explanation of how you determined a number or method to use and give an example of its use. The summary should also, in plain language, tell of the caveats and describe uncertainties and what type of future work is necessary or planned to resolve them. Additionally, I would have liked to see at least some discussion of potential implications of this rate of SLR and any associated uncertainty in how projects are designed and implemented. For example, under the scenario given, are there potential implications to using rocks on shorelines? Should we be abandoning some areas as unsustainable along the outer fringes of the coast? Should we spend so much time worrying about sculpting marsh to the exact intertidal elevation, or should we build in some additional height or heterogeneity of heights into our projects to account for it? You give an explanation of how to incorporate this information into determining marsh elevation, but also should give some examples of using this information for structures or other types of projects, not just marsh restoration. Remember we are in the realm of applied science, so show us how to apply this information in our work. One additional thing I would change throughout the text, for clarity, is how the rate of change is given. You present it as an overall increase in sea level by 2100, but the starting reference year is not necessarily given in each case (i.e. what is the time period of this 1m rise?). Therefore, it is hard to determine from this what annual rate is presented, and if it is the same for each reference. For example, if you state that we (or some other state) will use an increase of 1m by 2100, is that from this year or the year of the particular publication? Is it a linear rate? I think presenting annual rates in each case would be more useful, especially since this document is meant to be a living one and will be referenced in future years as well as this one. I also would 43

1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455

have liked to see the reasonable range of the uncertainty (sort of a margin of error) expressed somewhere in the summary, in addition to the number recommended for now. Response: We agree that the format and organization of the document could be improved. In order to make the document easier to use, we have simplified the process for determining the rate of GLSR into a step-by-step process that engineers and scientists can easily calculate. We have moved this information to the front of the document so it is easy to locate. We do feel, however, that it is important to retain the scientific background information for reference. Therefore, we have adjusted the organization of the document, corrected any formatting or typographical errors and moved the technical discussion to the back. Satellite altimetry works by orbiting satellites emitting signals that travel to the earth and are then reflected back to the satellite. The time it takes for the signal to travel that distance is then used to calculate the distance between the satellite and the surface. My continually emitting and receiving signals, the satellites can record the surface of the ocean. This allows us to monitor the sea surface height or sea level. Because this method of measurement does not require the use of a benchmark anchored to a land surface for reference, such as is used for tide gauge measurement, subsidence is not an issue. Therefore, the satellite altimetry is not influenced by subsidence so the east-west trend that is identified in the white paper is not related to the subsidence trend the commenter mentions. Subsidence can be a factor in tide gauge data; however, the east west GLSR trend depicted in the white paper is based on satellite altimetry data. While those questions are important to answer, this white paper was prepared to inform policy on sea level rise and, therefore, is limited to that discussion. We approached this paper by researching and synthesizing the current science related to GSLR, using that information to estimate the most likely GLSR over the next century, and how to best incorporate GSLR into CPRA project planning and design. There are many other factors other than just GLSR that must be factored into any specific project as the commenter points our; however, this white paper is limited to estimating a GLSR. The issues commenter mentions should be addressed during project design and not by this white paper.

44

1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489

CPRA-LACES released a draft document for internal CPRA comment on 23 November 2010. Comments on 20101123 CPRA LACES SLR Document FINAL DRAFT, 12December 2010, from Summer Martin, Louisiana Office of Coastal Protection and Restoration Comment on Line 189 through 191: Why would you want to remove [the influence of natural cycles on sea level rise estimates]? Its still a change in SL whatever the cause? Response: GLSR generally refers to the increase in MSL that is not natural and/or cyclical and is long term. We are primarily interested in this trend because it is assumed that natural systems, marshes in particular, are quite capable of persisting in the event of natural cycles. It is the long term change in sea-level against this background of natural forcings that will ultimately affect the ability for coastal areas to maintain elevation. Moreover, a change in MSL long term will significantly alter the degree to which natural cycles will affect an area. For example, if the 18.6 year lunar cycle tends to increase sea-level in an area by 2 cm over 10 years and the GSLR rate in that area is 1mm/yr. then the overall effect of the natural cycle in those 10 years will be 3 cm, and when the 18.6 year cycle ebbs, it will ebb by only 2 cm, leaving 1cm of overall increase in MSL. We believe that this was unclearly described in the section you referenced and have altered that section to clarify. Comment on Line 583 through 586: I dont understand this sentence; what is it you are recommending? Response: We are recommending that in order to account for the change in water level (not accounting for land change and RSLR at this point) at a specific project, that management teams assume that by 2100 sea-level in that area will increase by 1 meter and use the associated acceleration constant from the curve NRC II, 6.2 x 10-5and apply this to the yearly trend found from satellite altimetry or local tide gauges to determine the change in MSL for any other date, see equations 5 and 6 for clarification. We believe this was unclearly described in the section you referenced and have altered the text to clarify.

45

1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535

CPRA-LACES released a draft document for comment by select State agencies on 17 May 2011. Written comments were received from the Department of Natural Resources and the Department of Wildlife & Fisheries. Informal comments were also received from the Department of Environmental Quality and the Department of Transportation & Development, and the Governors Office of Homeland Security and Emergency Preparedness. Comments on 20110517 CPRA LACES SLR Document, 31May 2011, from Louis E. Buatt, Assistant Secretary, Office of Coastal Management, Louisiana Department of Natural Resources The Office of Coastal Management appreciates the opportunity to review the document referenced above. While we recognize that the document is being distributed for informational purposes to persons and groups with responsibilities and interests in the issue of sea-level effects on coastal resources, I do want to take this opportunity to provide feedback for your consideration in future iterations of the document at such time as it may be revised to reflect the progress of state of knowledge on the subject. I and my staff have found the document to be an excellent review of a complex and often controversial topic and appreciate the work you and your staff have done to provide a readable and comprehensive synopsis of the issue as it pertains to the Louisiana coast. I believe the white paper will be a useful tool to those working with issues affected by relative sea-level rise and appreciate the effort to discuss the related factors that distinguish these effects seen in Louisiana in the context of observed subsidence and vertical marsh accretion. I would offer only two comments for your consideration in future editions of the whitepaper. First, I believe the Executive Summary might be enhanced by a slightly expanded discussion of the issue as observed in Louisiana and the Gulf of Mexico and of the process being recommended in the paper. While readable to its intended technical audience, it will inevitably be read by many not versed in the jargon of SLR terminology, and who will read little else but the Executive Summary. Second, I believe it would be appropriate to mention in the context of planning and design of structural and non-structural projects and management measures of the Master Plan, that the State's Coastal Management Program, through its regulatory authorities, does have enforceable policies related to subsidence and inundation and the risks associated with them, whether caused by SLR or other phenomena. This could be useful particularly in planning exercises undertaken to implement the Master Plan. Finally, I am obliged to point out one very minor editing miscue which remains in the document as a result of the automated editing which we all now use. In line 937 of the document the word "of' appears when it seems obvious that the writer intended that the word be "or." Once again, let me offer kudos to you and your staff who have prepared an excellent and useful document dealing with a very complex and constantly changing subject. Response: We agree with the need for a revised discussion and have removed the Executive 46

1536 1537 1538 1539

Summary from the Technical Report. In its place we have created a stand-alone Summary of the Technical Report for Coastal Managers, which is more comprehensive but is written without the jargon of the earlier Executive Summary.

47

1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567

Comments on 20110517 CPRA LACES SLR Document, 17 June 2011, from Heather Warner-Finley, Louisiana Department of Wildlife and Fisheries Thanks very much for the opportunity to review this work. We think that RSLR is a critical parameter to consider and monitor relative to restoration efforts and protection and navigation planning. Its great that you guys are taking it seriously, and were particularly pleased that additional research / modeling is planned. Contribution of freshwater from melting glaciers and ice sheets the papers authors acknowledge that this phenomenon may be THE primary contributor to global sea level rise. These data could have significant implications to how predictable and relevant future GSLR scenarios are to the Gulf of Mexico given limits in the monitoring networks that define the GSLR components The paper then goes on to identify the acceleration constants that will be used and recommend use of a bracketed model to estimate sea level rise. Would it be valuable to try to estimate the uncertainty that is brought into the equation by the addition of freshwater from ice melt? Is that well accounted for in the high SLR scenario of 1.5m? Subsidence seems to be another variable that will require much more research and modeling. Nymans data cited in the SW Coastal study that a deteriorating marsh appeared to accrete at a higher rate is fascinating. This seems to introduce another large source of uncertainty in any attempt to predict how coastal projects will perform in the future.

We applaud you for beginning this work. Response: We have added language highlighting the uncertainty involved in the eustatic contribution to GLSR and, to support that language, included Figure 18.

48

1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614

Appendix B: CPRA-LACES Technical Issues with the US Army Corps of Engineers Engineering Circular No. 1165-2-211

The US Army Corps of Engineers released Engineering Circular (EC) 1165-2-211 on 1 July 2009 to guide Corps staff on how to account for local relative sea-level rise in water resource project planning and engineering. The recommendations in this report have been proposed because we recognize that there are several significant technical issues, with the EC that limit its use in southern Louisiana. While the EC is not a mandate per se for non-Corps entities such as the State to use during project implementation, it does become a required component of any water resources project for which the State is the local sponsor because it is Corps internal policy. While there is a requirement in the EC for Corps staff to account for specific RSLR scenarios in project planning, there is the flexibility for the local sponsor to add other scenarios for consideration. The fours-step process of accounting for RSLR described in this document represents such an alternative and, if adopted, internal CPRA policy. The EC process begins in Step 2 (page C1 of USACE 2009) by looking for local tide gauges that can serve as a data source for RSLR calculations. The EC requires that tide gauges are appropriate when the data period of record is greater than forty years, for the reasons of excessive confidence intervals with shorter data records discussed in Section 2.1.1. There are only two NOAA tide gauges that have that period of record, the Grand Isle (NOAA Station ID #8761724) and Sabine Pass North (NOAA Station ID #8770570) gauges. RSLR can be mathematically represented as E(t) = a*t + b*t2 + S*t (Eqn. B1)

where E(t) is MSL at time t, a is the observed rate of GSLR, b is the acceleration constant, and S is rate of subsidence (or uplift in areas of glacial rebound). For the Grand Isle gauge, NOAAs calculation of RSLR for the time period 1947-2006 was 9.24 0.59 mm/yr. (Zervas 2009). Referencing the variables in Eqn. 6, the 9.24 mm/yr. rate of RSLR is equal to (a + S), the linear rate of GSLR plus the subsidence function, with no acceleration constant because of the linear definition and no marsh vertical accretion function because the gauge is in open water. With this information it appears that we pass EC's step 2 and proceed to step 4, where it must be decided if the "... long-term gauges can be used ... [to] represent local ... conditions at [the] project site." Steps 4 and 5 essentially seek to establish if the physical gauge location is representative of the project location, which for coastal Louisiana would imply that the Sabine Pass North gauge is representative of the Chenier Plain and the Grand Isle gauge is representative of the Mississippi River Deltaic Plain. If we assume yes to both, we are then asked in Step 8 if there is a stable geologic platform within the same region as the project site. Our recommendation is that no, data from these two specific gauges in Louisiana cannot be used to characterize conditions throughout the Chenier and Delta Plains, because of the high degree of variability in subsidence rates illustrated in Figures 27 and 28 of the Technical Report. Fallback answers on the decision points of the EC are to "Consult with a tidal hydrodynamics expert." 49

1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660

The implication of this is that the geological complexities of coastal Louisiana render the EC approach to predicting SLR unusable. If for arguments sake we do to accept that the two gauges at Grand Isle and Sabine Pass are appropriate representatives for the Delta and Chenier Plains, respectively, the EC then instructs mathematically estimating subsidence by subtracting GSLR from the RSLR gauge record. To determine the rate of subsidence, the EC instructs an assumption that GSLR is a constant linear rate of 1.7 mm/yr., which was the overall linear rate of GSLR for the 20th century defined by Church & White (2006). Subtraction of the linear GSLR function of 1.7 mm/yr. from the linear RSLR function of 9.24 mm/yr. gives an assumed constant rate of subsidence at the gauge of 7.5 mm/yr. The first technical concern with the EC approach involves the determination of subsidence from the RSLR tide gauge data by subtraction of the linear rate of GSLR for the 20th century, and specifically the problem that there are two linear curves being compared even though they represent different time periods (1900-2000 for GSLR, 1947-2006 for RSLR). Examination of the Church & White (2006) dataset shows that while the overall GSLR rate for the 20th century was calculated at 1.7 0.3 mm/yr., there are two evident curves of different slope embedded in the overall 1870-2001 graph, with the 1936-2001 time period having a linear GSLR rate of 1.84 0.19 mm/yr. This discrepancy reiterates the importance we discussed in the main report of understanding the period of record of the SLR data. Data should only be compared where there is direct temporal overlap, which in the case of the Church & White vs. Grand Isle gauge data would be the time period 1947-2001, because there is no reason to expect that the actual slope of either line for the restricted time period would be equal to the slopes for either GSLR or RSLR for their full periods of record. To illustrate the difference, while the subtraction using the EC gave an assumption of 7.5 mm/yr. as described above, comparison of more comparable data for the rate of RSLR for the Grand Isle tide gauge of 9.85 0.35 mm/yr. (1947-1999; Zervas 2001) and GSLR for 1936-2001 equal to 1.84 0.19 mm/yr. gives the differential of 8.01 mm/yr., which by the EC would be the assumed subsidence rate. The implication here is a potential miscalculation of the rate of subsidence that will be carried out into the project lifespan planning horizon because of the assumption that subsidence will remain constant. Although the differential is minor, this example was illustrative only, since a direct comparison of 1947-2001 data is needed and the results would be expected to likewise differ. The next technical concern, however, is the assumption that the 20th century mean rate of GSLR is appropriate to compare with the tide gauge data. Specifically, both tide gauge and more recent satellite altimetry data indicate that the present rate of GSLR has accelerated beyond the 20thcentury rates determined by Church & White (2006). As shown in Figure 7 of the main report, the present linear rate of GSLR from the satellite altimetry data is 2.9 mm/yr. for the 1992-2010 period of record. The EC does not advise using data with periods of record shorter than 40 years, though it does say that If estimates based on shorter terms are the only option, then the local trends must be viewed in a regional context, considering trends from simultaneous time 50

1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706

periods from nearby stations to ensure regional correlation and to minimize anomalous estimates. The nearby stations should have long enough records (greater than 40 years) to determine reasonable trends, which can then be compared to the shorter, local sea-level records USACE 2009, Page B-4 This leaves pursuing projects in coastal Louisiana in a conundrum. If faith is placed in the global data showing an acceleration in the rate of GSLR during the past two decades, continuing to use the lower long-term GSLR rate will result in a difference calculation of subsidence that is probably greater than in reality. In the case of Church & White (2006) vs. satellite altimetry data, that difference would be as much as 1.2 mm/yr. The other option is to use adjacent gauges with longer-term data, but that option is of little use in southern Louisiana because of the recognition that there are very evident spatial differences in both observed SLR (Figure 16) and subsidence (Figure 27). Not using current estimates of the rate of GSLR also will underestimate future MSL in EC scenarios because the acceleration scenarios (b in Equation B1) are applied to the base rate of GSLR. Admittedly, some of these differences in calculated subsidence rates and magnitudes when carried out over a typical 50-year project life span are of questionable ecological or engineering significance. The underlying philosophical question is whether we accept a process that gets to a number thats close enough even though we recognize that the mathematics behind the calculations are flawed. However, the discussion to this point presupposes that continued use of a linear function for either GSLR or the tide gauge RSLR is appropriate. The remaining two technical issues have much more of a potential for significant differences in calculated depths of RSLR. The fact that the rate of GSLR has consistently increased across discrete periods of time between 1870 and 2010 suggests that even the historical rates of GSLR need to be assigned a non-linear function instead of a linear function. The data were not available as of the writing of this version of the report to discuss the difference in fit between a linear and exponential function for the RSLR tide gauge data at Grand Isle. Visual examination of the data, however, does suggest a more linear response than that shown for GSLR. This has extremely important implications because evidence of a more exponential GSLR rate and a more linear RSLR rate means that there is no mathematical possibility for subsidence to have been historically constant. In fact, if we retain subsidence defined as the difference between RSLR and GSLR, subsidence must have decreased during the period of record. This is in agreement with recent data from Morton et al. (2009) which suggests that reduced subsurface fluid withdrawal in the recent past compared to the 1950s-1970s may be leading to declining rates of subsidence. The assumption in the EC of a constant, subtraction-based calculation of subsidence rate carrying into the future then becomes invalid. Subsidence would decline over time, and the EC process would substantially overpredict RSLR for the projects period of analysis. While on the issue of subsidence, it is also a concern that the EC says to assume the subsidence rate calculated as discussed above across the entire coastal area for which the specific tide gauge is representative. Even with its existing limitations, Figure 27 of the main report illustrates our current understanding that subsidence varies across Louisianas coastal zone, and that using the 51

1707 1708 1709 1710 1711 1712 1713 1714 1715 1716 1717 1718 1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742

Grand Isle gauge to represent the deltaic plain and the Sabine Pass North gauge to represent the Chenier Plain is unrealistic. Any attempt to remove the subsidence calculations from the EC process and replace that data with a more spatially-explicit estimation of coastal zone subsidence pushes any SLR analysis closer to our recommendations. The final technical issue is concern over the future sea-level rise scenarios that are mandated in the EC process. Specifically, the EC requires that three future scenarios be examined: A continuation of the current linear rate of RSLR at the tide gauge for the project period of study (50 years), defined as the Low Scenario; An accelerated rate of SLR over the period of study, defined by the NRC (1987) Scenario I curve, which uses the acceleration constant of 2.36 x 10-5 mm/yr2 to accomplish a 0.5-meter rise in SLR by 2100, defined as the Intermediate Scenario; and An accelerated rate of SLR over the period of study, defined by the NRC (1987) Scenario III curve, which uses the acceleration constant of 1.005 x 10-4 mm/yr2 to accomplish a 0.5-meter rise in SLR by 2100, defined as the High Scenario.

As discussed in the main report, there is a strong argument for a non-linear (i.e. accelerating) trend for present and predicted future SLR. Although the argument has been made that consideration of the Low Scenario only increases the range of potential outcomes studied, with such a low probability for this scenario any attempt to model a future linear rate is a questionable use of limited time and financial resources. For planning and design scenarios we believe the NRC I curve to be the most appropriate low value for the future. Our recommendation in the main document was to plan for a 1-meter rise in MSL by 2100, based on the general consensus of the scientific literature. This is not one of the standard scenarios listed in the EC. Bracketing an analysis of SLR with the NRC (1987)-based acceleration constants for a 0.5- and 1.5-meter rise in MSL by 2100 does not allow us to estimate the impacts of a 1-meter GSLR unless the slope of the landscape is constant, which it is not. The EC does stipulate that The analysis may also include additional intermediate rates, if the project team desires. In joint state-Corps projects LACES and CPRA should require analyzing the 1meter projection of MSL increase by 2100.However, given the discussion above regarding the EC calculation and assumption of a linear subsidence rate, we recommend that CPRA only push for a 1-m SLR scenario using the 4-step process described in this report.

52

1743 1744 1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789

Appendix C: Draft Southwest Coastal Feasibility Study, Wetland Accretion Summary

VERTICAL SOIL ACCRETION ESTIMATES IN LOUISIANA MARSHES Background Coastal marshes can adjust vertically and maintain a dynamic equilibrium with Relative Sea Level Rise (RLSR) up to a certain rate (Morris et al. 2002). This active adjustment, to a great degree, is controlled by organic matter production and mineral sediment deposition (Turner et al. 2001, Nyman et al. 2006). While erosion with tides can remove surface detritus from the marsh, a large proportion of the organic matter storage that helps maintain the vertical elevation of the soil can be linked to belowground root production. Along the gulf coast, soil organic matter accumulation controls vertical accretion of the marsh, and mineral matter contributes less to the process of vertical adjustment. This is because organic matter occupies more than twice the volume of an equivalent mass of mineral matter (clay, silt, sand). Louisianas mature marshes (outside of the active deltas) rely on organic matter accumulation, through plant production, to adjust to relative sea level rise. Soil organic matter can be preserved or lost by reducing or oxidizing soil conditions. Persistent saturation favors long-term organic matter storage, while drought and moisture loss can cause oxidation, resulting in soil elevation loss. Thus, for coastal wetlands, long-term soil elevation is in a dynamic equilibrium with water-level variation. Also, there is an optimum elevation where plant production, organic storage, and soil elevation gain is optimized. A marsh at a high elevation in the tidal frame may have high plant production but organic matter oxidation is high, resulting in a static soil elevation. The converse situation, where marsh elevation is low and inundation is severe, plant production and organic matter accumulation are impaired. While these general relationships are recognized, quantitative relationships of soil elevation, hydrology, and plant production have not been fully developed for Louisiana wetlands. The purpose of this technical memorandum is to review the data pertaining to wetland vertical accretion in Louisiana to support the development of future landscape projections with sea-level rise. One of our objectives is to contrast geographic and coast-wide accretion patterns, but another objective is to understand the capacity of herbaceous coastal wetlands to accrete regardless of geography. Understanding the upper limit of wetland accretion is important since there is substantial sea level rise anticipated with a future coastal landscape. Data synthesis and results Most of the work done in Louisianas coastal marshes to understand longer-term (since 1963) accretion processes has been done using 137Cs. Shorter-term estimates of accretion use feldspar marker horizons placed on the soil and cored over time to measure the vertical accumulation of sediments. The marker horizon measurements can be coupled with vertical elevation measurements with a Rod Surface Elevation Table (RSET) to understand how much surface compaction occurs over time. Our goal here is to see if general patterns of accretion emerge between marsh types and provinces and provide suggestions for consideration. Here we present four tables of data summarized from various sources. 53

1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800

Longer- term, coast-wide patterns, Cesium 137 (Table C1): This is a comparison of delta and Chenier Plain accretion estimates from Nyman et al. (2006), which comprises some of his work with colleagues R. DeLaune and W. Patrick; they present information from 68 cores. Also in Table 1, Turner et al. 2000 present research in salt marshes of the Delta Plain with data from 52 cores. A study was led by USGS (Piazza et al. in press) that analyzed 48 cores from the Delta and Chenier Plain following Hurricanes Katrina and Rita.
Table C1. Long-term accretion estimates (137Cs) from different marsh types in the Chenier and Delta Plains from different studies with large numbers of samples. (DP=Delta Plain; CP=Chenier Plain).

Marsh type Nyman et al. 2006 Stable fresh brackish salt Deteriorating brackish salt Turner et al. 2001 salt Piazza et al. in press fresh brackish salt fresh brackish brackish mean median 1801 1802 1803 1804 1805 1806 1807 1808

Province Hydrology

Cores (n)

Accretion rate cm yr-1

DP/CP DP/CP DP DP/CP DP

14 12 12 8 22

0.82 (0.15) 0.88 (0.14) 0.59 (0.14) 0.96 (0.32) 0.98 (0.36)

DP

52

0.66 (0.21)

DP DP DP CP CP CP

natural natural natural natural natural impounded

15 5 10 5 3 10

0.57 (0.13) 0.72 (0.14) 0.64 (0.16) 0.57 (0.25) 0.65 (0.23) 0.38 (0.04) 0.70 0.66

1) Mean estimates of accretion average 0.70.2 cm yr-1 and fall within a range of 0.38-0.98 cm yr-1 regardless of marsh type or province. 2) There is coherence of salt marsh accretion in the Delta Plain regardless of study: 0.590.14 cm yr-1 (n=12, Nyman et al. 2006, from stable marshes) 0.660.21 cm yr-1 (n=52; Turner et al.) 0.640.16 cm yr-1 (n=10; Piazza et al. in press) 54

1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827

3) In the Delta Plain, deteriorating salt marshes exhibit higher rates of accretion (0.980.36

cm yr-1) than stable salt marshes (0.59 0.14 cm yr-1) (see Nyman et al. 2006). 4) Mean estimates of accretion have error terms of 0.15-0.3 cm yr-1. Thus, any comparison of mean accretion among marsh type or province is not likely to be considered significantly different enough to warrant separation. 5) Fresh marshes have the capacity to accrete at equal or higher rates than salt marshes, regardless of geological province (individual estimates of fresh or salt marshes accreting at 1.0 cm yr-1 are not uncommon).
6) In the Chenier Plain, data from Piazza et al. in press suggest that accretion could be significantly lower for impounded marshes; this needs further investigation (see Table 3).

Longer-term, coast-wide patterns, Cesium 137 (Table 2): Jarvis (2010) assembled a literature review of accretion estimates of streamside and interior marshes where available. The Jarvis (2010) summary presented estimates from various radiometric and physical measurement techniques, but only the 137Cs estimates are used here.
Table C2. Long-term accretion estimates (137Cs) from different marsh types and habitats (interior vs. streamside). Compiled by Jarvis (2010).
Marsh type F I B S F S S F S S S F S S B I 0.56 (0.11) 0.78 0.78 0.65 1.1 1.1 Interior marsh accretion rate (cm yr-1) 0.65 0.64 0.59 0.75 0.90 (0.10) 0.75 0.47 (0.09) 0.65 (0.18) 0.66 0.48 (0.09) 2.26 (0.09) >1.59 0.67 (0.49) 0.80 (0.17) 0.57 (0.10) 1.35 Streamside marsh accretion rate (cm yr-1) 1.06 1.35 1.40 1.35 0.99 (0.17) 1.10, 1.35 0.68 (0.17)

Province Delta Delta Delta Delta Delta Delta Delta Delta Delta Delta Delta Delta Delta Delta Chenier Chenier

Site Name Barataria B. Barataria B. Barataria B. Barataria B. Palmetto B. Barataria Lafourche P. Breton Sound Fourleague B. Old Oyster B. B. Chitigue Delta NWR Barataria B. Breton Sound Cameron P. Calcasieu L. mean median

Source
Hatton et al. 1983 Hatton et al. 1983 Hatton et al. 1983 Hatton et al. 1983 DeLaune et al. 19891 DeLaune et al. 19781 DeLaune et al. 19891 DeLaune and Pezeshki 20031 Baumann et al. 1984 Rybczyck and Cahoon 2002 Rybczyck and Cahoon 2002 Wilson and Allison 2008 Wilson and Allison 2008 Wilson and Allison 2008 DeLaune et al. 19891 DeLaune et al. 19831

1828 1829 1830

55

1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842

1) A consistent accretion pattern is detectable with streamside greater than interior marsh. 2) For interior marshes, mean and median accretion rates were 0.78 and 0.65 cm yr-1, respectively. Longer-term, Chenier Plain patterns, Cesium 137 (Table 3): Steyer (2008) summarized Chenier Plain accretion studies and made comparisons among and impounded and un-impounded (open) sites. He also examined how accretion may vary with distance inland from the Gulf of Mexico in the areas of Sabine basin and Rockefeller WMA.
Table C3. A comparison of ranges of long-term accretion estimates (137Cs) from impounded and unimpounded brackish marsh sites in the Chenier Plain (summarized in Steyer 2008).
Hydrology Impounded accretion rate cm yr
-1

Un-impounded accretion rate cm yr

-1

Source

0.33-0.64 0.31-0.38 0.16-0.38 0.31-0.60

0.27-0.58 0.46-0.54 0.28-0.35 0.26-0.50

Foret 1997 Foret 2001 Phillips 2002 Steyer 2008

1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866

1) Impounded and un-impounded marshes have the capacity to accrete up to 0.6 cm yr-1 in the Chenier Plain. 2) Sabine Basin: Steyer (2008) concluded accretion in the Sabine basin was significantly greater in un-impounded (0.510.05 cm yr-1) than impounded (0.360.02 cm yr-1) marsh. However, along a gradient inland from the coast (0-8 km) there was no discernable difference in accretion rate. 3) Rockefeller WMA: The opposite was true of the Rockefeller study area, where Steyer (2008) found a significant effect of distance inland on accretion which was highest close to the gulf (0.7-0.8 cm yr-1) and decreased (0.4-0.5 cm yr-1) 8 km inland. The impoundments only occurred far inland so that impoundment effects could not be adequately evaluated. Shorter-term, Chenier Plain patterns, marker horizon accretion (Table 4): This is a statistical summary of accretion (using feldspar), vertical elevation change, and subsidence rates measured through 2010 at more than 120 Coast-wide Reference Monitoring Stations in Southwest Louisiana. These CRMS data are available at www.lacoast.com/crms_viewer and were compiled by Tommy McGinnis, OCPR Lafayette Field Office. 1) Examining the recent CRMS data, mean vertical accretion among marsh types range from 0.68 to 1.01 cm yr-1. 2) Mean elevation change among these wetlands is less than 0.65 cm yr-1. 3) If we integrate all marsh types together, the mean accretion, elevation change, and 56

1867 1868 1869 1870 1871

Table C4. Summary statistics of recent elevation change data from the Coast-wide Reference Monitoring Stations (CRMS) in freshwater (F), intermediate (I), brackish (B) and salt (S) marshes in the Chenier Plain. Some stations have different record lengths, and the data are representative of conditions through the year 2010 (Data compiled from www.lacoast.com/crms_viewer by T. McGinnis, OCPR Lafayette field office). Subsidence (cm/yr.)
100 0.39 0.45 1.70

(ft., NAVD88)

Accretion (cm/yr.)
101 0.89 0.83 1.61

Marsh Elevation (ft., NAVD88) Marsh type

# stations mean median std deviation

Vertical Accretion (cm/yr.) S

9

Elevation Change (cm/yr.) F

20

Subsidence (cm/yr.) S
9

F
20

I
53

B
44

F
16

I
39

B
39

S
7

I
49

B
44

F
16

I
38

B
39

S
7

All marsh types combined

126 122

1.12

1.21

1.31

1.24

1.01

1.06

0.68

0.80

0.00

0.58

0.30

0.65

0.65

0.43

0.29

0.19

1.23

1.12

1.24

1.36

1.27

0.89

1.03

0.76

1.12

-0.01

0.58

0.38

0.66

0.38

0.42

0.46

0.58

1.27

0.53

0.52

0.34

0.40

1.08

2.30

0.93

0.76

1.79

1.26

0.68

0.73

1.32

2.40

1.01

1.01

0.46

1872

57

change (cm/yr.)
0.39 0.50 1.17

1873 1874 1875 1876 1877 1878 1879 1880 1881 1882 1883 1884 1885 1886 1887 1888 1889 1890 1891 1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902 1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918

subsidence rates are 0.89, 0.39, and 0.39 cm yr-1, respectively. 4) Shallow subsidence is an important process that affects surface elevation adjustment of Chenier Plain wetlands. General Conclusions The 137Cs dating technique is well accepted and has been used across the coast. The method is particularly useful for understanding the upper limits of vertical accretion and estimating the relative contributions of organic and mineral accretion among different marsh types. Examining the summary data presented by different investigators, marsh types from fresh-to-salt apparently have a similar capacity for high rates of accretion, on the order of 1.0 cm yr-1 (and even higher in the streamside habitats). Deteriorating salt marshes (22 cores, 0.98 cm yr-1, Nyman et al. 2006) seem to have an upper limit threshold of 1.0 cm yr-1. That high accretion is associated with marshes in a deteriorating landscape may be explained by their topographically low position, where deposition is favored and organic matter remains saturated. Nyman et al. 2006 suggest that Louisiana marshes can compensate for a RSLR rate up to about 1.0 cm yr-1. Submergence rates in excess of this would likely result in a reversion of emergent marsh to open water (also termed marsh collapse). There are limitations on what we can infer with accretion estimates. In the past, accretion estimates have been used to conclude whether the wetland is keeping up with sea level rise. A problem with this approach is that the elevation of the wetland (with respect to the mean tidal range or landscape) is unknown or ignored. The practical consequence is that the time to a critical submergence threshold (with respect to plant health) occurs sooner for a low lying marsh than that of a higher marsh. In other words, a higher or lower rate of accretion is needed to offset submergence simply depending on the marsh elevation. Moreover, a high accretion rate may be a symptom of a relatively low marsh elevation (compensational deposition in low areas) if we assume over a long time period that peat accretion is in equilibrium with mean water or tide level. The added problem that accretion cannot predict actual elevation change can be problematic (Cahoon 2006). General considerations can be made about long-term accretion estimates and wetland processes: 1) It is generally accepted that vertical soil elevation maintenance (peat accretion), is in dynamic equilibrium with mean water level (or sea level). 2) Theoretically, accretion rates should vary with elevation within a wetland system (this relationship is not well established for different Louisiana wetlands). Ultimately, the elevation of a wetland determines the length of time before it succumbs to a critical submergence. 3) Long-term accretion may not reliably predict the actual elevation change in high subsidence (shallow or deep compaction) environments. A long-term accretion estimate of 0.7 cm yr-1 captures the central tendency of all herbaceous marsh data that have been reviewed. Currently, there seems to be a lack of evidence to support applying a habitat specific accretion rate; that is, there is evidence of high accretion rates in both salt and fresh marshes. The long-term data show that Chenier Plain marshes have accreted over the last 50 years at rates of ~0.50.2 cm yr-1 while shorter-term data (CRMS) shows mean 58

1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931 1932 1933 1934 1935 1936 1937 1938 1939 1940 1941 1942 1943 1944 1945 1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964

accretion rates of ~0.8 cm yr-1 (Table 4). At 120 stations, median elevation change was 0.50 cm yr-1 (Table 4). This elevation gain corresponds well with the long-term RSLR rate of 0.56 cm yr-1 measured at Sabine Pass. In other words, wetland elevation gain should be approximately equivalent to contemporary RSLR rates. In general, comparatively high RSLR rates in the Delta Plain have produced greater vertical elevation change and accretion than that observed in the Chenier Plain. Without considering other stresses to wetland health, Chenier Plain marshes should be stable to RLSR rates on the order of 0.7 cm yr-1, as herbaceous wetlands in the Chenier Plain should respond similarly to increasing submergence as those of the Delta Plain.

Literature Cited Cahoon, D.R. 2006. A review of major storm impacts on coastal wetland elevations. Estuaries and Coasts. 29:889-898. DeLaune, R.D., R.H. Baumann, and J.G. Gosselink. 1983. Relationships among vertical accretion, coastal submergence, and erosion in a Louisiana gulf coast marsh. Journal of Sedimentary Petrology 53:147-157. DeLaune, R.D., J.H. Whitcomb, W.H. Patrick, Jr., J.H. Pardue, and S.R. Pezeshki. 1989. Accretion and canal impacts in a rapidly subsiding wetland. 137Cs and 210Pb techniques. Estuaries 12:247-259. Foret. J.D. 1997. Accretion, sedimentation, and nutrient accumulation rates as influenced by manipulations in marsh hydrology in the Chenier Plain, Louisiana. M.S. Thesis, Univ. of Louisiana, Lafayette. Foret. J.D. 2001. Nutrient limitation of tidal marshes on the Chenier Plain, Louisiana. Ph.D. Dissertation, Univ. of Louisiana, Lafayette. Jarvis, Jessie. 2010. Vertical accretion rates in coastal Louisiana: A review of the scientific literature: 1-14. Morris, J.T., P.V. Sundareshwar, C.T. Nietch, B. Kjerfve, and D.R. Cahoon. 2002. Responses of coastal wetlands to rising sea level. Ecology. 83:2869-2877. Nyman, J.A., R.J. Walters, R.D. DeLaune, and W.H. Patrick, Jr. 2006. Marsh vertical accretion via vegetative growth. Estuarine, Coastal and Shelf Science 69:370-380. Phillips, L.A. 2002. Vertical accretion and marsh elevation dynamics on the Chenier Plain, Louisiana. M.S. Thesis, Univ. of Louisiana, Lafayette. Piazza, S.C., G.D. Steyer, K.F. Cretini, C.E. Sasser, J.M. Visser, G. O. Holm, Jr., L. A. Sharp, D. E. Evers, and J. R. Meriwether. In press. Geomorphic and ecological effects of Hurricanes Katrina and Rita on Coastal Louisiana Marsh Communities: U.S. Geological Survey Open File Report. Steyer, G.D. 2008. Landscape analysis of vegetation change in coastal Louisiana following Hurricanes Katrina and Rita. Louisiana State Univ. Ph.D. Dissertation. 158p. Turner, R.E., E.M. Swenson, and C.S. Milan. 2001. Organic and inorganic contributions to vertical accretion in salt marsh sediments. Pgs. 583-595. In: M. Weinstein and K. Kreeger (eds.) Concepts and Controversies in Tidal Marsh Ecology. Kluwer Academic Publishing, Drodrecht, Netherlands.

59

1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Appendix D: Detailed Procedure for Incorporating Sea-Level Rise into Louisiana Coastal Project Planning and Design

The following is an example of how relative sea-level rise would be calculated using the recommendations in the main report and can be used as a stand-alone instruction manual. To predict the future relative sea-level rise for programmatic planning and project-level engineering and design, we recommend the following four-step procedure. To describe this more fully we use a hypothetical wetland restoration project on Marsh Island State Wildlife Management Area as an example, with a construction date of 2015 and a period of study that carries through 2100. The recommendations presented in the Technical Report begin with a three-step process to define relative sea-level rise (RSLR), represented by the generalized equation E(t) = (a*t + b*t2) + S*t (Eqn. D1)

where E is RSLR over the time increment t, a is the historical linear rate of global sea-level rise (GSLR) (Step 1), b is the acceleration constant for predicted GSLR (Step 2), and S is rate of subsidence (or uplift in areas of glacial rebound) (Step 3). Specifically, the equation illustrates that in order to calculate RSLR through 2065, the first component needed is a prediction of the change in sea (Gulf of Mexico) surface elevation. Subsidence is added to this value to define the total change in elevation of that location relative to the water surface. The first two steps of the recommendations apply to the change in elevation of the Gulf of Mexico sea surface, and define the values for the function in Eqn. D1 represented by (a*t + b*t2). However, for actual predictions of future RSLR at a specific location, we use a detailed equation that acknowledges that most sea-level predictions use a starting point in 1986 (the starting point for most GSLR scenarios). E(t2-1986)-E(t1-1986)=a([t2-1986]-[t1-1986])+b(([t2-1986]2-[t1-1986]2 where E, a and b are as defined for Figure D1, t1 is the initial year or first year of project, and t2 is the final year or last year of project. Although for our scenario the t1 and t2 values for the operational equation would be 2015 and 2100, respectively, we recommend building the GSLR scenario starting at 1986, to more properly define the acceleration constant (b) as well as providing a valuable check on the calculations out to 2100. E(2100-1986) - E(1986-1986) = a*([2100-1986]-[1986-1986]) + b*(([2100-1986]2[1986-1986]2), which simplifies to 60 (Eqn. D2)

2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023

E 2015-2100 = a*114 + b*12996.

(Eqn. D3)

This equation becomes the base operating equation for applying the values to the variables as described in our recommendations.
1. To begin the two-step process for estimating the future rate of change in the absolute sea

level of the Gulf of Mexico, we need to establish the current rate of sea-level rise, based on aggregate GSLR data from the satellite altimetry. As Figure D1 demonstrates, the rate of SLR varies across the coast of Louisiana from west to east by as much as 58%.

2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034 2035 2036 2037 2038 2039

Figure D1. Sea-level rise rates for points offshore of coastal Louisiana derived from 1992-2010 satellite altimetry as described in Figure 12 of the main report. The circle indicates the location of the hypothetical project at Marsh Island that is the basis of the example calculations in this appendix.

To account for the substantial east-west variation in near-shore sea-level rise observed in the recent satellite altimetry record, we recommend using the local observations of SLR from contemporary satellite altimetry most proximate to the site of interest. In the case of the hypothetical Marsh Island project, the two most proximate offshore SLR values range from 2.62 to 2.69 mm/yr and provide a mean value of 2.66 mm/yr (rounded, or 0.00266 m/yr), which is applied as the value for the variable (a) in Equation D3. Note that in the update shown below, the subsidence term is omitted, since the first two steps are only interested in calculating the anticipated change in the sea surface of the Gulf of Mexico, and not fully calculating RSLR. 61

2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062 2063 2064 2065 2066 2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084

E2015-2100 = 0.00266*114 + b*12996, which simplifies to E2015-2100 = 0.30324 + b*12996. (Eqn. D4)

2. We recommend that CPRA staff assume that Gulf sea-level rise will be 1 meter (3.3) by 2100 as the most heavily-weighted project alternative. To account for variability around that prediction given the debate in the scientific community, we recommend a bounding range of 0.5 1.5 meters (1.6 4.9) also be considered. These recommendations establish three a priori values for E. In doing so, and as a result of having defined (a) in the previous step, we must calculate the relevant acceleration constants (b) for each scenario. For the 0.5 meters scenario: 0.5 = 0.30324 + b*12996, which simplifies to b = 1.514 x 10-5. For the 1.0 meters scenario: 1.0 = 0.30324 + b*12996, which simplifies to b = 5.361 x 10-5. For the 1.5 meters scenario: 1.5 = 0.30324 + b*12996, which simplifies to b = 9.209 x 10-4. The two steps completed so far establish the values for (a) and (b) in GSLR predictive function shown in Equation D2. We can then use that function, with the included parameter values, to define the GSLR curve from 1986 to 2100. The importance in defining that curve is to calculate the annual incremental increases in Gulf sea-surface elevation. Since Steps 1 and 2 only provide an estimate of the elevation of the Gulf of Mexico at the target point in the future, these calculations must be combined with a prediction of local subsidence in order to craft an RSLR curve that will be used to predict marsh vertical accretion and overall persistence/collapse of marsh. We recommend: 3. Applying these calculations to either spatially-explicit empirical observations of subsidence or a map of predicted subsidence rates. In the case of our hypothetical Marsh Island project, the range of subsidence values identified by the Master Plan revision team for that area, Zone 15 on Figure D2, is 1-15 mm/yr. Applying the Master Plan most plausible scenario of 20% into each zone shown in Figure D2, we calculate a value of 3.8 mm/yr. Functionally, 3.8 mm are added onto the annual incremental GSLR defined after Steps 1 and 2, in order to define the annual incremental RSLR. This highlights some of the concern with Figure D2 as an operational product. If instead we were to use the coastal subsidence map generated by Britsch in 2007 (Figure 27 in the Technical

62

2085 2086 2087 2088

Figure D2. Map of projected subsidence ranges for south Louisiana generated by the Subsidence Advisory Panel for the Louisiana CPRA Master Plan 2012 Update, following a meeting on 14 October 2010.

63

2089 2090 2091 2092 2093 2094 2095 2096 2097 2098 2099 2100 2101 2102

Report), we would instead use a subsidence value of 0.5 feet per 100 years, which is equivalent to 1.524 mm/yr. The Britsch value is only 39% of that shown in Figure D2, which could have significant effects on the calculation of the annual RSRL increment that is applied to the marsh vegetation accretion model described in Step 4. Much more work will need to be done to refine the subsidence estimates of subsidence at the local level for predictive modeling purposes. Following steps 1-3 yields a RSLR curve for Marsh Island for each of the three acceleration scenarios discussed in Step 2. These curves will then be compared to marsh collapse values established for the Chenier Plan, as exemplified in Appendix C. For this example, we will only calculate the RSLR curve for the primary recommendation of a 1-meter GSLR by the year 2100. Part of the template spreadsheet that LACES has created for this purpose is shown in Figure D3 below.

2103 2104 2105 2106 2107 2108 2109 2110 2111 2112

Figure D3. Screen capture of the spreadsheet that LACES has drafted for calculating RSLR curves consistent with the four-step process recommended in the Technical Report. In this example, the historical linear GSLR rate (green) is combined with the acceleration constant associated with a 1-meter GSLR by 2100 (red) to establish annual incremental increases in Gulf sea-surface elevation (blue). That calculation is then combined with an estimate of subsidence from Figure D2 (brown) to establish a yearto-year incremental RSLR value (yellow).

64

2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124 2125 2126 2127 2128 2129 2130 2131 2132 2133

The annual incremental RSLR values for this example scenario range from 9.2 to 18.6 mm/yr (0.36 to 0.73 inches/yr), and the models predict a total increase in relative sea level of 0.613 meters (2.01 feet) between 2015 and 2065. In order to predict the persistence the coastal wetland, and specifically the persistence of the wetland surface or conversely marsh surface collapse and drowning, a fourth step is necessary. 4. Use the sum of #s 1-3 above to establish the annual incremental rate of inundation for the period of analysis to predict local responses of marsh vertical accretion; essentially local coastal elevation.. Marsh vertical accretion can be inferred from scientific literature if no reliable data exist on site, or can be estimated from vegetation productivity models if available. Appendix C of the Technical Report shows information of this type that was used by CPRA and U.S. Army Corps personnel conducting the Southwest Coastal Integrated Hurricane Protection and Coastal Restoration Feasibility Study. In this example, if we assume the 7 mm/yr (0.28 inches/yr) threshold for marsh plant vertical accretion used by the Southwest Coastal Feasibility Study team, the models predict that the marsh will not be able to keep up with projected RSLR. Theoretically the calculations show how much elevation needs to be added through the use of marsh creation or nourishment in order to accommodate continued marsh vertical accretion, and thus marsh persistence in the face of RSLR.

65