Ecosystems fix carbon from the atmosphere through photosynthesis, which is then allocated to aboveground plant structures, such as leaves and branches, or to belowground structures, such as roots. Our current understanding of the factors that control belowground carbon allocation is significantly weaker than aboveground allocation. Yet, in grasslands the amount of carbon that goes belowground each year is much larger than the amount of carbon allocated aboveground. Our rudimentary understanding of the controls of belowground carbon allocation and the ratio of aboveground to belowground carbon is a significant knowledge gap, as roots are a major input of organic material and nutrients into soil. Predictions of future carbon storage in these ecosystems hinge on our understanding of the effects of environmental availability on allocation of carbon belowground. This study addresses the questions: How does precipitation affect the above/belowground partitioning of carbon? During drought periods, are above and belowground structures equally affected, or are roots affected less than leaves and branches? Finally, is the effect of precipitation on carbon allocation constant or does it vary from deserts to humid grasslands? This project will install experiments in New Mexico, Colorado and Kansas to understand the effect of precipitation on the partitioning of carbon above versus belowground and the underlying mechanisms.

This proposal presents three novel hypotheses based on: (1) a plant-response mechanism, suggesting decreased belowground allocation with increasing water availability, and (2) a trophic-cascade mechanism, suggesting the opposite pattern derived from the differential sensitivity of root feeders and their predators to water availability. A final hypothesis (3) suggests that the magnitude of plant responses decreases from arid to humid grasslands while the magnitude of the trophic-cascade phenomenon increases. The trophic-cascade mechanism may be constrained by the abundance of belowground predators in arid grasslands, which is greater in humid ecosystems. The experimental design includes complementary field and microcosm experiments located in three different ecosystem types: Chihuahuan Desert Grassland, NM, Shortgrass Steppe, CO, and Tallgrass Prairie, KS. The field experiment includes additions and reductions of precipitation at each site. The microcosm experiment is based on monoliths subjected to 4 soil fauna treatments x 5 water manipulations. Soil fauna treatments consist of (1) soil devoid of fauna (just native bacteria and fungi), (2) defaunated soil inoculated with nematode root feeders, (3) defaunated soil inoculated with nematode root feeders and nematode predators, and (4) control. Microcosm tubes will be located in each of the water manipulation plots using individuals of the dominant grass species of each ecosystem type. This project will train a post-doctoral fellow and graduate and undergraduate students, including students from underrepresented groups. Collaboration with the Asombro Institute for Science Education, a non-profit organization that provides award-winning science education to more than 12,000 children and 4,500 adults annually in southern New Mexico and western Texas, will include establishing a demonstration project of a soil-observation window at the Chihuahuan Desert Nature Park. Partnerships with the "Ask A Biologist" program will result in an article on the interactions between plants and belowground organisms and how these interactions are modulated by water availability. The project will contribute special events associated with the effects of climate on belowground processes during a summer sustainability program that engages local teachers and students, and the "Managing the Planet" series aimed at the general public.