Skip to main content
ARS Home » Plains Area » Las Cruces, New Mexico » Range Management Research » Research » Publications at this Location » Publication #310885
Title: Application of spatial pedotransfer functions to understand soil modulation of vegetation response to climate

Author
item Levi, Matthew
item SCHAAP, MARCEL - University Of Arizona
item RASMUSSEN, CRAIG - University Of Arizona

Submitted to: Vadose Zone Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 6/25/2015
Publication Date: 9/22/2015
Citation: Levi, M.R., Schaap, M.G., Rasmussen, C. 2015. Application of spatial pedotransfer functions to understand soil modulation of vegetation response to climate. Vadose Zone Journal. 14(9):1-14.

Interpretive Summary: We applied spatial predictions of physical soil properties to a pedotransfer function (Rosetta) to predict hydraulic properties at high resolution in a semiarid landscape of southeastern Arizona. Estimated soil properties explained patterns of vegetation dynamics derived from Landsat across 7 years. Antecedent precipitation was more important for explaining the relationships between modeled soil properties and vegetation response than the amount of monsoon precipitation. Linking digital soil mapping with Rosetta led to predictions of hydraulic soil properties that were more closely related to vegetation dynamics compared to data available in the SSURGO soil database.

Technical Abstract: A fundamental knowledge gap in understanding land-atmosphere interactions is accurate, high resolution spatial representation of soil physical and hydraulic properties. We present a novel approach to predict hydraulic soil parameters by combining digital soil mapping techniques with pedotransfer functions and demonstrate that simple derived quantities are related to observed spatial patterns in ecosystem production during the North American Monsoon. Landsat reflectance and elevation data were used to predict physical soil properties at a 5 m spatial resolution for a semiarid landscape of 6,265 ha using regression kriging. Resulting soil property maps were applied to the Rosetta pedotransfer function to predict saturated hydraulic conductivity and water retention parameters from which approximate water residence times were derived. Estimated idealized residence time for water lost to the deeper vadose zone and evapotranspiration corresponded to vegetation response. Antecedent precipitation was more important for explaining the relationships between modeled soil properties and vegetation response than the amount of monsoon precipitation. Increased spring precipitation prior to the monsoon produced stronger negative correlations with hydraulic conductivity and stronger positive correlations with plant available water. Modeled water residence times explained the patterns of vegetation and landscape morphology validating our approach as a method of producing functional spatial pedotransfer functions. Linking digital soil mapping with Rosetta led to predictions of hydraulic soil properties that were more closely related to vegetation dynamics compared to data available in the SSURGO soil database.