Skip to main content
ARS Home » Plains Area » Mandan, North Dakota » Northern Great Plains Research Laboratory » Research » Publications at this Location » Publication #195496

Title: Tracking Palustrine Water Seasonal and Annual Variability in Agricultural Wetland Landscapes using Landsat from 1997 to 2005

Author
item BEERI, OFER - UND AEROSPACE
item Phillips, Beckie

Submitted to: Global Change Biology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/7/2006
Publication Date: 1/31/2007
Citation: Beeri, O., Phillips, R.L. Tracking Palustrine Water Seasonal and Annual Variability in Agricultural Wetland Landscapes using Landsat from 1997 to 2005. Global Change Biology 12:1-16.

Interpretive Summary: Surface waters for wetlands populating the PPR vary unpredictably with weather and subsurface flow, necessitating development of a monitoring system capable of identifying long-term trends and seasonal dynamics on basin and regional scales. Here, we address gaps in the knowledge of wetland landscapes with data not previously available. We developed a model for detecting seasonal advance and retreat of surface waters greater than 225 m2 in size for a 2500 km2 landscape with >96% overall accuracy. Given nine years of observations, we found dramatic variability in surface water coverage and water permanence categories within and between wetland basins. We found annual declines since 1997 in the number of wetland basins with water and the percentage of the landscape covered by surface water. Of the 19,047 maximum wetland basins with water observed over nine years, we found only 8,757 of these held water during 2005. Coincident with observed surface water changes over time was reduced snowfall, with hysteresis in the data likely due to temporal decoupling between surface and subsurface transient flow, although land use changes may have also contributed to declines. This satellite-based model provides a vehicle for current and future assessment of a) seasonal water coverage and persistence, b) hydroperiod linkages to wetland basin productivity, c) land-use change and effects on hydroperiod, and c) the strength of climatic forcing factors on individual wetlands and on wetland complexes.

Technical Abstract: Wetlands densely populate the ecoregion transecting the center of the Prairie Pothole Region (PPR) known as the Missouri Coteau and epicenter to the most productive waterfowl breeding habitat in North America. Basin water bodies in the PPR are highly vulnerable to any factors influencing regional hydrology, including changes to upland land-use and climate. These palustrine, depressional basin waters vacillate with regional drought and deluge, so hydroperiod modulates wetland productivity, habitat, and water quality functions. Wetland surface water observations are lacking for this ecoregion, largely due to intra-seasonal variability in water properties. We developed a spectral model for tracking seasonal waters, which we evaluated for a 2500 km2 landscape, that estimates seasonal and annual surface water variability for thousands of individual wetland basins in the Missouri Coteau ecoregion. Detection accuracy was assessed for water bodies greater than 225 m2 in size at 96%. We identified shifts in the distribution of water permanence classes within and between years for waters emerging in spring, mid-summer, and late-summer from 1997 to 2005 and identified a maximum of 19,047 basins with water (12% of the landscape) populating 2500 km2. For the 2005 growing season, we observed only 8,757 basins with water (6% of the landscape) for the same area. Declines were greatest for water bodies detected only in spring, suggesting a loss of those wetlands functioning to recharge groundwater stores early in the season. If landscape factors driving surface water coverage and wetland density are similar for the entire Missouri Coteau, we estimate the number of basins containing at least 225 m2 of water for this region declined from 577,600 to 266,000 between 1997 and 2005. Considering greater climate forcing in the future, additional water losses are expected, which would dramatically alter wetland productivity.