GEOSCAN Search Results: Fastlink


TitleEstimating variation of groundwater storage within the Great Lakes Water Basin from GRACE, soil moisture and lake levels
AuthorHuang, J; Halpenny, J
SourceGRACE Science Team Meeting, Joint International GSTM and DFG SPP Symposium, Abstract Book; 2007 p. 16
LinksAbstracts / Résumés
Alt SeriesEarth Sciences Sector, Contribution Series 20070326
PublisherGFZ (Potsdam, Germany)
MeetingGRACE Science Team Meeting, Joint International GSTM and DFG SPP Symposium; Potsdam; DE; October 15-17, 2007
Documentcomputer file
Mediaon-line; digital
NTS30L; 30M; 30N; 31C; 31D; 40; 41; 42C; 42D; 52A
AreaGreat Lakes; Lake Superior; Lake Michigan; Lake Huron; Lake Ontario; Lake Erie
Lat/Long WENS-93.0000 -76.0000 49.0000 41.0000
Subjectshydrogeology; geophysics; groundwater; groundwater resources; groundwater discharge; groundwater regimes; groundwater levels; surface waters; watersheds; water utilization; lake water; satellites; satellite imagery; remote sensing; Great Lakes Water Basin
AbstractComprised of Lake Superior, Lake Michigan, Lake Huron, Lake Erie, and Lake Ontario, the Great Lakes Water Basin (GLB) is one of the most important fresh water systems in the world. It covers an area of about 0.76 million square kilometers along the border of Canada and the United States. The total area of the five lakes is about 0.24 million square kilometres, accounting for roughly one-third of the total GLB area. Altogether they hold about 84% of North America's fresh surface water and about 21% of the world's fresh surface water. However it is important to note that water storage in the land area of the basin plays a similar role to the surface water of the five lakes in this system. The total land storage (snow, ice, soil moisture and groundwater) surrounding the five lakes provides a reservoir that plays an important role in replenishing the lakes. Although the water storage variation of the lakes is well-gauged, the temporal and spatial variation of groundwater over the entire region is poorly understood.

In this study we estimate variation of the groundwater in the GLB using the monthly GRACE gravity models (CSRRL04 and GFZRL04) for the period extending from 2002 to 2006. Both non-isotropic Gaussian filtering and least-squares fitting of spherical harmonic coefficients are used to extract a signal from the monthly models. In the Gaussian filtering method, 300 km and 450 km radii are used to produce mass estimates. In the least-squares fitting, linear and quadratic trends as well as annual and semi-annual terms are estimated for each coefficient. The coefficient terms are subsequently selected in terms of signal-to-noise-ratio (SNR), the ratio between the absolute value of a term and its a posteriori standard deviation. The GLDAS (snow, ice and soil moisture) models and water level of the lakes are used to separate groundwater storage variation.

Preliminary results show that the variation in soil moisture, snow and ice is the dominant contributor to the total water storage variation estimated from the monthly GRACE models over the GLB. The RMS Water-Thickness-Equivalent (WTE) variation from GRACE is about 3.4 cm while that of soil moisture, snow and ice is about 3.7 cm. The mean lake level responds to the GRACE estimates with a comparable magnitude (4.1 cm) but a phase delay of about 3 months. These results suggest that the lake water storage variation relates to the groundwater variation to a great extent over the GLB for the period of study. Water level data at a limited number of wells are used to assess groundwater estimates.