In the years shortly before his death, Dean Martens' research efforts were directed to investigations of land‑use effects on the cycling of C and N and related impacts on potential climate change. The land‑use investigations focused on grazing and associated growth and control of mesquite (Prosopis spp.) in semiarid riparian woodlands, grasslands, and mountainous forage areas. The growth of leguminous mesquite provides N inputs into N-limited semiarid soils, and using plant and soil isotope signatures, Dean ascertained that although mesquite litter is highly labile, limited rainfall in semiarid systems allowed soil C and N to accumulate. In the past, moisture limitations in southwestern ecosystems led researchers to believe that semiarid soils were not significant consumers or producers of trace gases and as such, these regions were traditionally overlooked in greenhouse gas inventories. Dean's work, however, confirmed the presence of a large, previously unreported methane sink in semiarid soils. While modeling the effects of soil moisture and temperature on ecosystem/atmosphere exchange of carbon dioxide, Dean found that, unlike more temperate systems, large differences in soil C content did not correlate with variations in carbon dioxide production unless soil moisture was limiting. Taken together, Dean's results suggested that soils of semiarid ecosystems could not be discounted in potential mitigation of climate change. In addition, his work confirmed the existence of a significant C sink in the San Pedro Riparian National Conservation Area in southeastern Arizona, supporting efforts to preserve this rare remnant of ecological diversity.