Plot-to field-scale root distribution data are relatively rare and difficult to measure with traditional methods. Nevertheless, these data are needed to accurately model root water uptake (RWU) processes within agronomic, hydrologic, and terrestrial biosphere models. New tools are needed to effectively observe root distributions and model dynamic root growth processes. In the past decade, geophysical tools have increasingly been used to study the vadose zone, and hydrogeophysical inversions have shown promise to noninvasively characterize water dynamics. In such an approach, the hydrology is modeled and hydrological data are inverted with the geophysical data, constraining the geophysical inversion results and decreasing uncertainty and the number of nonunique solutions. In this study, we developed and tested a coupled hydrogeophysical inversion approach that uses electrical resistivity data to estimate soil hydraulic, petrophysical, and root dynamic parameters. This builds on prior research that used either a coupled hydrogeophysical inversion to estimate soil hydraulic parameters only, or a hydrological inversion to estimate root distribution or root water uptake parameters. Our results indicate that under the conditions tested, this approach accurately captures root water dynamics and soil hydraulic parameters. This opens up opportunities to noninvasively image a variety of root distributions and soil systems, better understand the dynamics of RWU processes, and improve estimates of transpiration for systems models.Transpiration is the most important pathway for the exchange of water from Earth to the atmosphere, accounting for up to 80% of terrestrial evapotranspiration (Jasechko et al., 2013). Thus, disruptions to the plant community through climate and land-use changes will likely have serious implications for regional to global water balances. To predict and mitigate the effects of those changes, agronomic, hydrologic, and terrestrial biosphere models must accurately capture the exchange of water along transpiration pathways. Doing so requires understanding the underlying processes that drive such exchanges. The interdependent and dynamic nature of the factors controlling transpiration, and our inability to observe the processes directly, makes transpiration challenging to appropriately represent in these models.Transpiration is fundamentally controlled by root distributions and root water uptake (RWU) processes, yet due in part to a lack of dynamic root function data, these processes are often oversimplified in models (Warren et al., 2015). Such data are rarely available because it is challenging to observe roots in the field, especially changes with time (Cai et al., 2018). Direct approaches such as excavation and root windows are limited at the field scale and are very costly and labor intensive. These approaches are also not as feasible for deep roots associated with woody plants, such as trees (Maeght et al., 2013). Nondestructive methods are thus needed to understand plant functions as a response to ...

Biofuel crops, including annuals such as maize (Zea mays L.), soybean [Glycine max (L.) Merr.], and canola (Brassica napus L.), as well as high-biomass perennial grasses such as miscanthus (Miscanthus × giganteus J.M. Greef & Deuter ex Hodkinson & Renvoiz), are candidates for sustainable alternative energy sources. However, large-scale conversion of croplands to perennial biofuel crops could have substantial impacts on regional water, nutrient, and C cycles due to the longer growing seasons and differences in rooting systems compared with most annual crops. However, due to the limited tools available to nondestructively study the spatiotemporal patterns of root water uptake in situ at field scales, these differences in crop water use are not well known. Geophysical imaging tools such as electrical resistivity (ER) reveal changes in water content in the soil profile. In this study, we demonstrate the use of a novel coupled hydrogeophysical approach with both time domain reflectometry soil water content and ER measurements to compare root water uptake and soil properties of an annual crop rotation with the perennial grass miscanthus, across three growing seasons (2009-2011) in southwest Michigan, USA. We estimated maximum root depths to be between 1.2 and 2.2 m, with the vertical distribution of roots being notably deeper in 2009 relative to 2010 and 2011, likely due to the drought conditions during that first year. Modeled cumulative ET of both crops was underestimated (2-34%) relative to estimates obtained from soil water drawdown in prior studies but was found to be greater in the perennial grass than the annual crops, despite shallower modeled rooting depths in 2010 and 2011.

No abstract