While terrestrial ecosystem productivity is generally thought to be nutrient-limited, evidence suggests that in arid and semi-arid environments, productivity may be co-limited by precipitation and nutrients. Numerous studies have shown that nitrogen limits productivity in semi-arid grasslands. However, we lack information on how other key limiting nutrients, including phosphorus and potassium, in addition to nitrogen, affect community composition and productivity over time. This information is critical, as community composition may shift in response to different nutrients than productivity, depending on the response of dominant vs. rare species. We used 6 years of data on a fully factorial experiment adding nitrogen (N), phosphorus (P) and potassium (K) to plots annually following Nutrient Network protocols, and monitored ANPP and community composition in permanently located 1-m2 plots semi-annually. We asked (a) Is primary production limited by all nutrients (N, P, and K), or just N? (b) Is primary production co-limited by nutrients and precipitation, and are these results consistent across plant functional groups (grass and forb)? (c) Does community composition respond to different nutrient treatments over time?
We found that total productivity was co-limited by N and precipitation, consistent with other studies in aridland ecosystems showing response to N addition only in wet years. Further, these results suggest that productivity is not limited by P or K. Species composition however, shifted in plots that received N and P together and in plots that received N, P, and K across all years, but only in the fall. Individual N and P treatment effects were not found in all years, but were present in both the spring and fall. While both productivity and community composition were responsive to nutrient additions, they were not responsive to the same nutrients, which has significant implications for our ability to predict how biodiversity and ecosystem function will respond to future nutrient inputs.