One expected effect of global warming is that there will have more rain per rain event (e.g., storms will be more intense). Whether there will be more total precipitation is highly dependent on location - some locations likely getting more and others getting less. Knapp et al (2002) explore the question: " If total rainfall remains the same but storms are less frequent and more intense, how will carbon cycling and plant species diversity be affected in a grassland ecosystem?" The authors covered plots during natural rain events, captured the rain water and applied it to the plots in fewer (6-8 per growing season) events compared to the uncovered (control) plots (25-30 per year). But both types of plots received the same total precipitation over the season.
Results of the experiment showed that the increase in rainfall variability enhanced the diversity of native species. Plant water stress is increased and net photosynthesis, above-ground productivity, and soil CO2 flux all are altered. The authors cite references stating that 90% of all terrestrial vegetation types worldwide have more than 50% of their roots in the upper 30 cm of soil. Hence most biomes are at risk of being affected by projected increases in rainfall variablility even if total precipitation did not change. The authors conclude that changes in precipitation variability, in addition to changes in total seasonal precipitation, must be considered in assessing the impact of climate change on grassland ecosystems.
Reference
Knapp, A. K., and Co-authors, 2002: Rainfall variability, carbon cycling and plant species diversity in a mesic grassland. Science, 298, 2202-2205.