Hydraulic redistribution (HR) of soil water through plant root systems has been demonstrated worldwide in seasonally-dry ecosystems, with clear implications for plant water use and landscape hydrology. Impacts of HR on soil microbial activities during dry seasons also have been hypothesized for decades and confirmed recently in the field. We combine modeling with measurements to explore the likely magnitude of HR's effects on microbe-controlled carbon and nutrient cycling at ecosystem scales in four seasonally-dry and well-instrumented ecosystems. These systems were chosen arrayed from Washington State to the Amazon, with annual rainfall spanning ~400 to 2000 mm, soil textures ranging from clay to sandy loam and loamy sand, and diverse vegetation types. Ameriflux towers at the four sites provided multi-year measurements of net ecosystem carbon exchange, energy, and water fluxes: US-Wrc (Wind River Crane, Pacific Northwest douglas fir/hemlock forest); US-SCf (Southern California Climate Gradient, oak-pine forest); US-SRM (Santa Rita Mesquite, semi-desert grassland with encroached mesquite); and BR-Sa1 (Santarem KM67, evergreen broadleaf primary tropical forest). At all sites, HR had been previously detected. For modeling, we incorporated a representation of HR into the Community Land Model Version 4.5 (CLM4.5).
Results/Conclusions
Including HR in CLM4.5 improved modeled predictions of measured water, energy, and carbon fluxes at the four Ameriflux sites. Modeled plant and microbial activities were differentially stimulated by upward HR, and diminished by downward HR, sometimes exacerbating nutrient limitation of ecosystem productivity. In two ecosystems (BR-Sa1 and US-SCf), when the fire module within CLM4.5 was engaged, upward HR moistened surface soil and vegetation sufficiently to reduce large modeled CO2 loss during dry season fire. Overall, our work illustrates that the ecological significance of HR in seasonally-dry ecosystems includes and extends beyond the direct effects of HR on stomatal conductance, evapotranspiration, and photosynthesis. Modeling suggests upward HR’s strongest effect on biosphere-atmosphere CO2 exchange may be through reduced spread of dry season fire in some ecosystems; this idea remains to be tested in the field. Modeling also indicates that both upward and downward HR can affect the system-scale balance of nutrient supply and demand, and CO2 fixation and respiration (plant and microbial), all of which interact as major determinants of system productivity and exchange of CO2 with the atmosphere.