The Monsoon Rainfall Manipulation Experiment (MRME) is designed to understand changes in ecosystem structure and function of a semiarid grassland caused by increased precipitation variability, by altering rainfall pulses, and thus soil moisture, that drive primary productivity, community composition, and ecosystem functioning. The overarching hypothesis being tested is that changes in event size and frequency will alter grassland productivity, ecosystem processes, and plant community dynamics. Treatments include (1) a monthly addition of 20 mm of rain in addition to ambient, and a weekly addition of 5 mm of rain in addition to ambient during the months of July, August and September. We predict that soil N availability with interact with rainfall event size to alter net primary productivity during the summer monsoon. Specifically, productivity will be higher on fertilized relative to control plots, and productivity will be highest on N addition plots in treatments with a small number of large events because these events infiltrate deeper and soil moisture is available longer following large compared to small events.