The invasive annual cheatgrass releases more nitrogen than crested wheatgrass through root exudation and senescence.

Plant-soil feedbacks are an important aspect of invasive species success. One type of feedback is alteration of soil nutrient cycling. Cheatgrass invasion in the western USA is associated with increases in plant-available nitrogen (N), but the mechanism for this has not been elucidated. We labeled cheatgrass and crested wheatgrass, a common perennial grass in western rangelands, with (15)N-urea to determine if differences in root exudates and turnover could be a mechanism for increases in soil N. Mesocosms containing plants were either kept moist, or dried out during the final 10 days to determine the role of senescence in root N release. Soil N transformation rates were determined using (15)N pool dilution. After 75 days of growth, cheatgrass accumulated 30 % more total soil N and organic carbon than crested wheatgrass. Cheatgrass roots released twice as much N as crested wheatgrass roots (0.11 vs. 0.05 mg N kg(-1) soil day(-1)) in both soil moisture treatments. This occurred despite lower root abundance (7.0 vs. 17.3 g dry root kg(-1) soil) and N concentration (6.0 vs. 7.6 g N kg(-1) root) in cheatgrass vs. crested wheatgrass. We propose that increases in soil N pool sizes and transformation rates under cheatgrass are caused by higher rates of root exudation or release of organic matter containing relatively large amounts of labile N. Our results provide the first evidence for the underlying mechanism by which the invasive annual cheatgrass increases N availability and establishes positive plant-soil feedbacks that promote its success in western rangelands.