Application of Cosmic-ray Soil Moisture Sensing to Understand Land-atmosphere Interactions in Three North American Monsoon Ecosystems

- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner
Wednesday, 7 January 2015
Adam P. Schreiner-McGraw, Arizona State University, Tempe, AZ; and E. R. Vivoni

Land-atmosphere interactions in the North American monsoon region reflect complex interactions of soil, vegetation, and weather patterns occurring over large scales. Anthropogenic activity has caused widespread occurrence of woody plant encroachment into desert grasslands in the region, potentially altering these patterns. In this contribution, we validate and apply a new technique to measure soil moisture at scales relevant to land-atmosphere interactions in an attempt to determine the effects of woody-plant encroachment on soil water pools. The instrument used is a cosmic-ray sensing soil moisture sensor (CRS sensor) that measures soil moisture over an area of 28 ha. Three of these sensors were installed in different ecosystems in the North American Monsoon region; two sites where woody plants have encroached, a Sonoran desert ecosystem in southern Arizona and a Chihuahuan desert ecosystem in southern New Mexico, and a grassland site resulting from deforestation and pasture establishment in Sonora, Mexico. In each case, landscape heterogeneity in the form of bare soil and vegetation patches of different types leads to a complex mosaic of soil moisture and land-atmosphere interactions. Historically, the measurement of spatially-averaged soil moisture at the ecosystem scale (on the order of several hundred square meters) has been problematic. Thus, new advances in measuring cosmogenically-produced neutrons present an opportunity for observational and modelling studies in these ecosystems. For validation purposes, the two sensors in woody plant encroached ecosystems were located in experimental watersheds with instrument networks to measure the full water balance. At each location, we used both a distributed network of point-scale soil moisture sensors and the water balance to calculate a spatially averaged value of soil moisture to compare to the CRS observations. The three estimates of soil moisture compared favorably with an average discrepancy of 15%. After validation, we used an exponential decay model to estimate soil residence time across all three sites to examine the effects of woody plant encroachment on soil water availability. This study highlights the importance of measurement scale when studying watershed processes and ties that to the effects of woody plant encroachment in semi-arid ecosystems.