Statement for Record of S. Jeffress Williams
United States Geological Survey, U.S. Department of the Interior
Subcommittee on Water Resources and the Environment House Transportation and Infrastructure Committee on “Expert Views on Hurricane and Flood Protection and Water Resources Planning for a Rebuilt Gulf Coast”
October 20, 2005
On behalf of the Department of the Interior, I thank you for the opportunity to provide this statement to the Subcommittee on “Expert Views on Hurricane and Flood Protection and Water Resources Planning for a Rebuilt Gulf Coast.” I am a coastal-marine geologist, with over 30 years of research experience, including doing research in Louisiana for the past 20 years, for the U.S. Geological Survey (USGS) at the USGS Woods Hole Science Center located in Woods Hole, Massachusetts. My statement reflects results from a collaboration of research and science efforts by many in the USGS, by university scientists, and by our partners over the past 20 years. This statement provides a summary of research investigating the effects of geologic processes such as land subsidence, as well as the effects of human activities as both relate to coastal erosion, wetland loss, sea-level rise and the increased vulnerability of New Orleans infrastructure and ecosystems to natural hazards like hurricanes, flooding and future increased sea-level rise.
Over the past ~7,500 years, complex geologic processes have caused dramatic changes in the geography of the low-lying Mississippi River delta plain. The processes have produced more than six major shifts in the river channel; a sea-level change ranging from -400 feet at the end of the Ice Age, 20,000 years ago, to -15 feet about 7,500 years ago, when delta plain development started; and significant redistribution of sediments caused by frequent storm impacts that erode the relict deltas. The Louisiana Gulf Coast has been a dynamic environment for thousands of years, during which time the landscape has changed continuously.
We currently have a reasonably well-developed understanding of the geologic history of this region, and the effects that human activity has had on the evolution of the delta plain and the New Orleans landscape over the past 200 years.
Like many other low-elevation population centers, for example Venice, Shanghai, Bangkok, the Nile delta, the Mekong delta, and Bangladesh (Peck and Williams, 1992), the New Orleans region remains extremely vulnerable to natural hazards such as storm-surge flooding, as recently demonstrated by Hurricanes Katrina and Rita. In addition to the well-publicized damage to New Orleans and the other cities of the Gulf Coast, the recent storms have caused a significant loss of wetlands and marshes and massive coastal erosion throughout the entire region. USGS investigations conducted in response to Hurricane Katrina show that major parts of the Chandeleur barrier island beaches and dunes were eroded completely. Some wetland areas east of the river have lost 25 percent of their land area, and storm surge east of New Orleans and along the Mississippi coast was as high as 25 to 30 feet.
As destructive as Hurricane Katrina was, however, post-storm analyses suggest that as a Category 3 storm with a path east of the city, this was not the “big storm” predicted to devastate New Orleans. A strong Category 5 storm moving slowly along a path directly up the Mississippi River and slightly west of New Orleans would produce an even higher storm surge coupled with a significant increase in wind velocities. The predicted storm effects on New Orleans and adjacent urban areas would be far more destructive. Like other delta plain regions around the world, New Orleans continues to be particularly vulnerable to future near-term storm events and likely accelerated sea-level rise will make risks to New Orleans even greater.
Research by the USGS and others has provided a reasonably good understanding of the delta plain framework geology and how active geologic processes, such as subsidence, operate; but there are still significant data gaps which challenge our ability to quantify the various processes and to differentiate natural from human-caused processes. In order to mitigate coastal natural hazards in the future, we must continue to develop predictive models and improve our scientific understanding of all of the geologic processes acting on the Louisiana delta plain and New Orleans region.
When New Orleans was founded by the French and Spanish in the early 1700s, in a cypress swamp area between Lake Pontchartrain and a prominent crescent-shaped oxbow bend in the Mississippi River channel, the population was small, and the city was built on the natural levees of the Mississippi River about 5 to10 feet above sea level. This is the site of the current French Quarter, Garden District, and Uptown. The rest of the swamp was at or close to sea level. As the city developed into a trading center and river port during the mid-1800s to early 1900s, there was pressure to expand.
This expansion was facilitated by systematic city-wide land reclamation and forced drainage using a network of dredged canals and large pumping stations to move storm water to the Mississippi River and Lake Pontchartrain. At the same time, the natural levees were raised many times, and landfill and levees were built along the lakeshore by the 1930s as the city expanded north. The result of this construction was to alter the topography of the city, creating the current bowl-shaped configuration that prevents natural drainage. The highly effective system of canals and high-capacity pumps not only force-drained surface storm water, but also lowered the water table and dried out the organic-rich soils, which has led to their removal by oxidation and erosion processes.
From the early 1900s to today, these activities have resulted in the widespread loss of land elevation across the entire city due to compaction and oxidation of the soils. The result is that more than 50 percent of New Orleans is below sea level, some areas by as much as 10 feet. The only areas above sea level are either on the old levees or sandy linear ridges (e.g., Gentilly, Metairie ridges) marking old river channels or relict shoreline features. The geologic character and 200 year history of land reclamation are largely responsible for Hurricane Katrina’s ability to flood 80 percent of New Orleans. In planning for the future of the city following Hurricane Katrina, detailed information is needed on topics such as: high-resolution elevation data of the land surface from the 1700s to 2005; geotechnical characteristics of the soils at depth and their potential for subsidence; and geologic maps showing features that influence the land surface and development.
The biggest impact and primary geologic process driving Louisiana's coastal land loss as recognized by most scientists is land subsidence. There are three subsurface processes that contribute to subsidence: large regional-scale processes that result in crustal down warping; consolidation and compaction of soils resulting from both natural processes, such as dewatering of muddy sediments, and hydrocarbon extraction; and geologic faulting. While there is agreement that faults exist and several are active, there is considerable disagreement on the locations of the faults, whether they intersect the land surface, and especially the rate and frequency of fault movement (Lopez, 1991; Gagliano, 2005).
New Orleans and the entire delta plain to the south have changed a great deal over the past 200 years due to a complex combination of natural processes and anthropogenic activities that have had significant cumulative effects on the landscape. The result for the past two centuries has been a shift in the natural balance from net land-building deltaic processes to net land loss due to a variety of human alterations and natural processes.
As a result, Louisiana’s barrier islands erode more than 30 feet per year, and wetland loss has averaged 24 square miles each year over the past decade, a rate decrease from the 40 square miles per year observed in the period 1950s to 1970s. USGS studies suggest that wetland loss rates are down to 12 square miles per year (Williams, Penland and Sallenger, 1992; Barras et al., 2003). The primary natural and anthropogenic processes driving these changes are land subsidence (geological/faulting, local consolidation/ compaction), global sea-level rise, storms and floods, and human alterations to the Mississippi River and delta plain (river levees, dredged canals, land reclamation, oil and gas extraction, induced subsidence, dredged navigation channels, reduced sediment volumes).
The dramatic loss of Louisiana’s wetlands and barrier islands is well documented and recognized. Estimates of the contribution of human activities in driving land loss range between 10 percent and 90 percent. To further our understanding of the role of natural processes and multiple human factors, the USGS undertook a study to quantify and classify the processes causing wetland loss from the period 1932 to 1990. The results describe local processes only; important regional processes such as global sea-level rise, regional subsidence, and river flood controls were beyond the scope and not considered in the study.
While there is considerable debate as to exactly how much wetland loss is
attributable to specific causes, results published in two USGS reports demonstrate
that approximately 31 percent of coastal land loss is caused by natural processes,
and 69 percent is caused by a wide variety of human processes. These
findings indicate that the greatest impact is associated with subsurface fluid
(oil, gas, water) production, which could account for up to 36 percent of the
land loss over the 58-year period. Generally, human processes causing land
loss include a suite of activities such as extraction-induced (oil,
gas, water) land subsidence, dredged canals and channels, and altered surface
and subsurface hydrology. For further discussion, see http://pubs.usgs.gov/of/2000/of00-418/ofr00-418.pdf.
USGS-funded research led by Robert Morton on subsidence in the Louisiana delta plain conducted during the past five years has resulted in significant scientific findings. Regionally, the areas having the highest historical and geological subsidence rates coincide with the thickest modern deltaic sediments. However, the areas of highest historical subsidence rates (greater than 12mm per year) coincide closely with locations of producing oil and gas fields or faults. The lowest average subsidence rates were located between major hydrocarbon producing fields.
In addition, our scientific research shows that rapid interior wetland loss was caused primarily by subsidence rather than erosion of the marsh, as demonstrated by submerged marsh sediments that drowned in-place and are still preserved beneath water depths of up to 3 feet. Morton et al. (2002, 2003) show that historical subsidence rates, subsurface fluid production, and wetland loss are closely correlated temporally and spatially. Finally, these USGS studies demonstrate that mapped wetland loss rates have been substantially lower since the early 1990s and especially in the last decade than during previous decadal periods, 1950s to 1970s.
Land subsidence due to fluid extraction (oil, gas, ground water) has been demonstrated in many locations around the United States (e.g., Houston, TX, Wilmington and San Joaquin Valley, CA) and throughout the world (e.g., North Sea, Venezuela) and is well documented in the science and engineering literature (Nagel, 2001). The causes for the decreased rate of wetland loss in Louisiana from the 1990s to the present are still not certain, but may be the result of reduced subsurface fluid extraction activities across the region (Morton et al., 2003).
Current scientific methods allow modern subsidence trends to be evaluated on short time scales (less than 100 years) using tide gauges and benchmark re-leveling data and long time scales (greater than 100 years) using age-depth relationships of organic peat sediments. Although preliminary relative sea-level rise trends have been documented for many of the gauges using standardized regression analysis, the tide gauges with recently extended records require updating to achieve completion. Records provided by National Oceanic Atmospheric Administration/National Ocean Service stations need to be incorporated into the database to provide comprehensive estimates of relative sea-level rise.
Longer-term, time-dependent estimates of subsidence rates are provided by radiocarbon-dating and analysis of peat sediments comprising the wetlands. These peats, assumed to have been deposited at (or near) sea level, are encountered in the subsurface of the Louisiana coastal plain. The ages of these organic materials and their current depth allows an estimate of the rate of subsidence to be made. The ages are typically less than 5,000 years old, but may be as old as 10,000 years. Their present depths indicate long-term, average subsidence rates of 0 to 20 mm/yr. A comparison of relative sea-level changes indicated by multiple peat samples within a single core will potentially provide insight to acceleration or deceleration of relative sea level within the time scale of measurement provided by the varied age of the sampled peats.
Compaction and consolidation of the uppermost few hundred meters of deltaic sediment is commonly cited as an important contribution to subsidence. However, precise calculations of compaction rates require data that are not available for much of the Louisiana coastal plain, and generally have not been pursued rigorously until a recent USGS study conducted by Timothy Meckel. Other methods for estimating compaction in other coastal environments arrive at compaction rates in the range of 1 to 10 mm/yr.
Meckel’s study attempts to simulate sedimentation and compaction that may have occurred in the coastal plain over the last few thousand years with computational methods based on physical principles. These efforts incorporate geotechnical data from five fundamental depositional environments within the Louisiana coastal plain. The results of this study are currently under peer review prior to publication.
Compounding Louisiana's subsidence problem, there is a predicted increased rate of global sea level rise. Current rates indicate ~1-2 mm rise per year. The combination of subsidence and global sea-level rise for Louisiana results in a rate of relative sea level rise of about 1 cm per year or 3 feet per century, the highest rate of any coastal region in the world. The rise in global sea level of 19 inches by 2100 predicted by IPCC (2001) would effectively double the current rate. An improved understanding of the dynamics of subsidence and how it changes across the deltaic plain can be used to develop models that will project the effects of future rises in sea level and their potential impacts on future coastal restoration projects.
Two other major challenges remain for researchers responsible for providing the scientific data used to formulate public policy regarding wetland loss and coastal restoration in Louisiana. The first is to generate subsidence estimates for coastal plain areas that are not immediately adjacent to benchmarks and tide gauges where subsidence rates have been determined previously. The second challenge is to develop accurate computer models to forecast rates of future subsidence and areas of wetland loss.
In conclusion, complex geologic and other natural processes have changed the shape of the low-lying Mississippi River delta plain and the City of New Orleans for thousands of years. Recent human activity and development in the region have increased the complexity of the problem. Most scientists recognize land subsidence as the primary geologic process driving Louisiana’s coastal land loss. However, storm events like Hurricanes Katrina and Rita play a significant role in modifying the landscape through erosion and flooding. Continued subsidence will result in increased exposure of the people, land, and infrastructure in the region to storm events.
Thank you for the opportunity to present this statement for the record.
*Note: The information in this testimony is based on scientific research by USGS and other scientists reported in the scientific literature. In particular, contributions have come from Dawn Lavoie (USGS), Robert Morton (USGS), Shea Penland (University of New Orleans), Harry Roberts (Louisiana State University), Denise Reed (UNO), James Coleman (LSU), Jack Kindinger (USGS), John Barras (USGS), Virginia Burkett (USGS), Del Britsch (U.S. Army Corps of Engineers), as well as my own research. Complete references and copies of the scientific papers referred to in this testimony are available by request.