Agriculture in the SW is a challenge due to high temperatures and limited water resources, problems that are expected to intensify with climate change. We think that measuring and modeling agriculture at the edge of plant physiological limits can be predictive of future agro-ecosystem conditions, as well as inform sustainable use of water and nitrogen (see Duval et al. 2018).

Sorghum precursors evolved in the high elevation, arid plains of East Africa, which is evident today as drought hardiness traits in modern varieties. Sorghum also produces a natural nitrification inhibitor, sorgoleone, from its roots. We are therefore interested in sorghum as a model for improving system-level N and water efficiency, studying plant control on microbial N cycling, and plant response to directional microbial manipulation with viruses.  

I am working with researchers at Sandia National Laboratory, the University of New Mexico and New Mexico State University to understand within plant C dynamics that lead to increased root architecture and potentially increased soil C. To that end we are working to develop micro-needle sensors to measure root exudates, and micro-GC's to measure volatile organic compounds (VOC's) that are indicative of plant stress. 

Longer-term goals with my sorghum research include quantifying "sustainability" in terms of inputs like nitrogen and water, and interacting with New Mexico producers to encourage sorghum grain production for human consumption. 



 

Microbes process organic matter, transform N, produce and consume GHG's at a sub-micrometer scale, with global implications for climate. Therefore, a major challenge in applied systems ecology research is how to work between scales. 

Examples of how heterogeneous crop root matrices can be

---

Weekly nitrous oxide fluxes from Los Lunas sorghum field. Halving the amount of irrigation water (fig A) reduced emissions early in the season, but the trend dissipated later in the summer. N2O flux from fully irrigated fields (fig B) showed a greater response to differences in N management. (Duval et al., in review)