Multi‐century stasis in C3 and C4 grass distributions across the contiguous United States since the industrial revolution

Abstract

AbstractAimsUnderstanding the functional response of ecosystems to past global change is crucial to predicting performance in future environments. One sensitive and functionally significant attribute of grassland ecosystems is the percentage of species that use the C4 versus C3 photosynthetic pathway. Grasses using C3 and C4 pathways are expected to have different responses to many aspects of anthropogenic environmental change that have followed the industrial revolution, including increases in temperature and atmospheric CO2, changes to land management and fire regimes, precipitation seasonality, and nitrogen deposition. In spite of dramatic environmental changes over the past 300 years, it is unknown if the C4 grass percentage in grasslands has shifted.LocationContiguous United States of America.MethodsHere, we used stable carbon isotope data (i.e. δ13C) from 30 years of soil samples, as well as herbivore tissues that date to 1739 CE, to reconstruct coarse‐grain C3 and C4 grass composition in North American grassland sites to compare with modern vegetation. We spatially resampled these three datasets to a shared 100‐km grid, allowing comparison of δ13C values at a resolution and extent common for climate model outputs and biogeographical studies.ResultsAt this spatial grain, the bison tissue proxy was superior to the soil proxy because the soils reflect integration of local carbon inputs, whereas bison sample vegetation across landscapes. Bison isotope values indicate that historical grassland photosynthetic‐type composition was similar to modern vegetation.Main conclusionsDespite major environmental change, comparing modern plot vegetation data to three centuries of bison δ13C data revealed that the biogeographical distribution of C3 and C4 grasses has not changed significantly since the 1700s. This is particularly surprising given the expected CO2 fertilization of C3 grasses. Our findings highlight the critical importance of capturing the full range of physiological, ecological and demographical processes in biosphere models predicting future climates and ecosystems.