Evaporative groundwater discharge in humid plains: The role of climate, vegetation, and farmers

JOBBÁGY, E G; NOSETTO, M D; CONTRERAS, S; JACKSON, R B; CALDERÓN, S

San Francisco

Congreso; American Geophysical Union. Fall Meeting; 2009

Evaporative groundwater discharge is, in most landscapes, restricted to riparian zones or depressions, yet, it can be a widespread hydrological feature of flat sedimentary regions with (sub)humid climate. We explored the interactive effects of climate, vegetation, and human decisions controlling evaporative discharge from shallow groundwater through (a) a conceptual model describing groundwater discharge vs. depth functions and their interaction with ecosystems attributes (b) field evaluations of the model in agricultural systems of the Pampas (Argentina), (c) numerical simulations under contrasting land uses and farming behaviours. (a) Although groundwater discharge (transpiration + soil evaporation + surface water evaporation) is assumed to increases as water tables raise, we propose that transpiration, the dominant evaporative water flux in humid climates, has an “optimum” response to water table depth. Groundwater transpiration declines when water tables are too deep to be accessed by roots or shallow enough to create anoxic conditions that inhibit plant activity. This behaviour would yield two attraction domains under fluctuating water table conditions: a stable one below the “optimum” zone, where water table raise enhances transpiration and prevents further elevation; and an unstable one above the “optimum” zone, where it inhibits transpiration, favouring further elevation until surface water evaporation regulates the system. Groundwater level vs. discharge functions vary with biotic attributes such as rooting depth, waterlogging tolerance of plants, leaf area and canopy roughness, and soil surface coverage; in interaction with soil properties and climate. (b) Two years of measurements of productivity, remote sensing of evapotranspiration, and frequent water table level/salinity records across topographic gradients in a sandy landscape, confirmed the “optimum” model proposed above. (c) We developed a simple 1-D code that captured the “optimum” response of transpiration to groundwater depth with algorithms for capillary transport and waterlogging effects. Scenarios with typical land uses for the Pampas (natural grasslands, alfalfa pastures, annual crops, tree plantations) revealed vegetation as a major control of water table depths and flooding. Simulations highlighted the biological role of groundwater supplying ecosystems with water transferred from excess to deficit periods. The simulation of contrasting farming decision rules regarding waterlogging (establish crops, pastures or nothing) and groundwater supply (intensify rotations or ignore opportunity) showed how behaviour could increase flooding risk or minimize it while getting maximum productive benefit from groundwater. Acknowledging groundwater-ecosystem feedbacks in (sub)humid sedimentary plains is crucial but not sufficient to predict hydrological behaviour. In increasingly managed/cultivated situations, human decisions could play a similarly important role, being a tool to cope with the hydrological impacts of climate change.