No abstract

No abstract

No abstract

Analyses of areal variations in the subsidence and rebound occurring over stressed aquifer systems, in conjunction with measurements of the hydraulic head fluctuations causing these displacements, can yield valuable information about the compressibility and storage properties of the aquifer system. Historically, stress-strain relationships have been derived from paired extensometer/piezometer installations, which provide only point source data. Because of the general unavailability of spatially detailed deformation data, areal stress-strain relations and their variability are r•ot commonly considered in constraining conceptual and numerical models of aquifer systems. Interferometric synthetic aperture radar (InSAR) techniques can map ground displacements at a spatial scale of tens of meters over 100 km wide swaths. InSAR has been used previously to characterize larger magnitude, generally permanent aquifer system compaction and land subsidence at yearly and longer timescales, caused by sustained drawdown of groundwater levels that produces intergranular stresses consistently greater than the maximum historical stress. We present InSAR measurements of the typically small-magnitude, generally recoverable deformations of the Las Vegas Valley aquifer system occurring at seasonal timescales. From these we derive estimates of the elastic storage coefficient for the aquifer system at several locations in Las Vegas Valley. These high-resolution measurements offer great potential for future investigations into the mechanics of aquifer systems and the spatial heterogeneity of aquifer system structure and material properties as well as for monitoring ongoing aquifer system compaction and land subsidence. Papernumber 2000WR900404. 0043-1397/01/2000WR900404 $09.00 mainder was met with water imported from Lake Mead. In the central part of the valley, declining heads in the aquifer system incorporating thick, highly compressible clay beds (aquitards) have led to subsidence rates of several centimeters per year during most of the 20th century, with resulting damage to structures and well casings. Differential subsidence has reactivated existing Quaternary faults and created new earth fissures [Bell and Price, 1991]. Recently, artificial recharge has become an increasingly important tool to mitigate the negative effects of land subsidence due to overdrafting of the aquifer system in Las Vegas [Pavelko et al., 1999].Subsidence in the Las Vegas area has previously been monitored using precise leveling surveys and a borehole extensometer installation, the Lorenzi site, that was installed in 1994

[1] We use land-subsidence observations from repeatedly surveyed benchmarks and interferometric synthetic aperture radar (InSAR) in Antelope Valley, California, to estimate spatially varying compaction time constants, t, and inelastic specific skeletal storage coefficients, S kv *, in a previously calibrated regional groundwater flow and subsidence model. The observed subsidence patterns reflect both the spatial distribution of head declines and the spatially variable inelastic skeletal storage coefficient. Using the nonlinear parameter estimation program UCODE we estimate compaction time constants between 3.8 and 285 years. The S kv * values are estimated by linear estimation and range from 0 to almost 0.09. We find that subsidence observations over long time periods are necessary to constrain estimates of the large compaction time constants in Antelope Valley. The InSAR data used in this study cover only a three-year period, limiting their usefulness in constraining these time constants. This problem will be alleviated as more SAR data become available in the future or where time constants are small. By incorporating the resulting parameter estimates in the previously calibrated regional model of groundwater flow and land subsidence we can significantly improve the agreement between simulated and observed land subsidence both in terms of magnitude and spatial extent. The sum of weighted squared subsidence residuals, a common measure of model fit, was reduced by 73% with respect to the original model. However, the ability of the model to adequately reproduce the subsidence observed over only a few years is impaired by the fact that the simulated hydraulic heads over small time periods are often not representative of the actual aquifer hydraulic heads. Errors in the simulated hydraulic aquifer heads constitute the primary limitation of the approach presented here.

No abstract

Nineteen percent of the global population may face a high probability of subsidence

A large proportion of the world's population lives on low-elevation (<10 m) land near the sea 1,2 , much of which is subject to subsidence due to natural and anthropogenic processes 3 . As of 2005, ~40 million people and assets worth 5% of global gross domestic product were exposed to a 1-in-100-year coastal flooding hazard 4 . By 2070, the exposed population is expected to grow more than threefold, and the value of property exposed is expected to increase to ~9% of the projected gross domestic product, with the USA, Japan and the Netherlands having the most exposure 4 . However, these estimates often rely only on projections of global average sea-level rise and do not account for vertical land motion (VLM), in terms of subsidence (downward VLM) or uplift (upward VLM) of the land surface. A different estimate of exposure could result when VLM is taken into account, particularly considering recent findings that the elevation of many coastal lowlands has, to date, been considerably overestimated 5 .The recent increase in global mean sea level (GMSL) has led to a present-day rate of rise of ~3.35 mm per year (ref. 6 ); GMSL rise since 1900 is mostly attributed to accelerated ice-mass loss of glaciers and ice sheets, plus the thermal expansion of ocean water 7 . However, the relative sea level (RSL), defined here as the elevation difference between the sea surface and the solid Earth 8 , excluding the dynamic sediment surface 9 , is of particular relevance for assessing the effects of sea-level change at any given location. RSL change is defined as the sum of geocentric sea-level change plus VLM 8 . Note that the sediment-accretion rate, which has sometimes been invoked as a term in the RSL equation 10 , merely affects local water depth, not RSL. VLM is driven by natural processes, such as glacial isostatic adjustment (GIA) [11][12][13] , tectonics and earthquakes 14,15 , and sediment consolidation, including natural compaction owing to sediment deposition (loading) [16][17][18][19] , as well as anthropogenic effects caused by peat oxidation following drainage [20][21][22][23][24] and the compaction of aquifer systems and hydrocarbon reservoirs accompanying the extraction of subsurface fluids 20,25,26 (fig. 1).These drivers can be divided into shallow processes affecting depths of less than ~25 m (for example, compaction of Holocene sediments) and deep processes (such as tectonics and compaction of pre-Holocene strata) 27 . VLM can be much greater than nearby geocentric sea-level rise alone and, in turn, GMSL rise, which is estimated, in part, based on tide-gauge records. Thus, knowing how much, where and why coastal land subsides and how its rate varies over time is essential to evaluating hazards associated with sea-level rise and estimating GMSL rise.