Towards Parameterization of Root-rock Hydrologic Interactions in the Earth System Model

The degree of carbon-climate feedback by terrestrial ecosystems is intimately tied to moisture availability for photosynthesis, transpiration and decomposition. Extreme weather will be the “new normal” in a warming climate. During downpours, the rapid penetration of rainwater through the soil and weathered bedrock will dampen delivery to channels, increase pore pressures that lead to land sliding, and leave behind reservoirs of plant-available moisture in rock fractures and the weathered rock matrix. In an extended drought, tree survival and hence the magnitude of carbon-climate feedback depends on the availability of subsurface moisture and the ability of plant roots to access that moisture.

Our proposed work is focused on CLM, the land module in the Earth System Model. The immediate goal is to develop and test algorithms of “rock moisture” and root-rock interaction that could improve the CLM in changing precipitation extremes.  The lack of relevant global datasets on the geology and hydraulic properties of the heterogeneous subsurface below the soil layer has necessitated many assumptions in current CLM algorithms.

We propose to employ a three-pronged approach.  The first aims to develop a 1D model of “rock moisture” and transpiration using the large volume of continuous (over 4 years) high-frequency (<30 minute) hydrologic data (water table of 12 wells, sapflow, soil moisture) from our hillslope study site in northern California.  The data also include repeated campaigns to measure the variations in “rock moisture” down to the water table.  We will start with CLM4.0 and incorporate a saprolite layer and several weathered rock layers between the soil mantle and the water table.  We will explore scenarios of hydraulic conductivity of the rock layers, with a mean value based on geology and a stochastic component that mimics flow through fractures.  We will also experiment with varying rooting depths of plant functional types as a function of hydroclimate and geology.  The model algorithm will be tested against transient response of water table and transpiration to winter storms and dry summers.  A second prong analyzes the high-frequency USGS data in the western US to quantify the fast and slow response of streamflow and water table to intense precipitation and extended droughts.  A third prong upscales the 1D model of rock moisture to the western US and assess its ability to simulate the transient dynamics in the USGS data.

Project Term: 
2013 to 2015
Project Type: 
University Project