Making the leap from ponds to landscapes: Integrating field-based monitoring of amphibians and wetlands with satellite observations

Abstract

• Landsat and ground-based monitoring can be integrated to monitor wetlands and amphibians. • Landsat data supported characterizations of wetland hydrologic types and surface area trends. • Most Northern Range wetlands experienced changes in surface area over last 35 years. • Variation in snowmelt runoff was associated with changes in surface area for many wetlands. • Amphibian breeding occurred across all wetland hydrologic types. A unifying goal of wetland monitoring programs is to characterize how wetland flooding and drying regimes vary in space and time. In remote and mountainous regions, wetlands have been difficult and labor-intensive to monitor, especially at landscape scales. Here, we demonstrate how data from annual, ground-based monitoring of wetlands and amphibians in Yellowstone National Park (YNP) can be enhanced in time and spatial extent by Landsat-based observations of wetlands. Specifically, we applied spectral mixture analysis (SMA) to the Landsat time series to characterize seasonal variation in wetland surface water area over a 35-year period. This extended our monitoring record backward in time (increasing temporal extent) and allowed us to monitor wetlands outside the boundaries of our long-term monitoring catchments, thereby increasing spatial extent. Additionally, we reconstructed and characterized hydroperiod regimes of 427 wetlands across YNP’s Northern Range (NR) as ephemeral, intermittent, semi-permanent, and permanent; these categories relate to the local ecology of these sites. Across the NR, the mean summer surface water area in most wetlands declined, regardless of hydrologic regime. Declines in wetland surface water area were associated with variations in snowmelt runoff for approximately half of NR wetlands. The mean effect of runoff on wetland area was strongest for ephemeral wetlands and weakest for permanent wetlands, which were significantly larger than other wetland types. Ground-based observations from established, long-term monitoring sites (n = 37) showed that amphibians used all wetland hydroperiod regimes. Intermittent and permanent regime types contained sites where amphibian breeding was consistently detected and highlighted the importance of a portfolio of wetland types for amphibian conservation. Through the integration of ground-based and Landsat satellite datasets, we not only took the leap from ponds to landscapes, dramatically expanding the spatial and temporal extent of monitoring, but also successfully used SMA to characterize hydroperiod regimes and evaluate longer-term wetland surface water area trends across a range of wetland sizes. This work will allow wetland managers to place current conditions within longer-term trends, understand broader patterns of surface water change, analyze the relationship between wetland surface water trends and management actions, and calibrate models of climate impacts. Most importantly, the integration of field assessment and remote sensing technologies offers a strategy to detect changes in wetland extent and condition at scales ranging from individual ponds to landscapes.