Title: Contrasting above- and belowground sensitivity of three Great Plains grasslands to altered rainfall regimes Authors
Submitted to: Global Change Biology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 27, 2014
Publication Date: N/A
Interpretive Summary: Intensification of the global hydrological cycle with atmospheric warming is expected to increase inter-annual variation in precipitation amount and the frequency of extreme precipitation events. Although studies in grasslands have shown sensitivity of aboveground net primary productivity (ANPP) to both precipitation amount and event size, we lack equivalent knowledge for responses of belowground net primary productivity (BNPP) and NPP. We conducted a two year experiment in three US Great Plains grasslands – the C4-dominated shortgrass prairie (SGP; low ANPP) and tallgrass prairie (TGP; high ANPP), and the C3-dominated northern mixed grass prairie (NMP; intermediate ANPP) – to test three predictions: (1) both ANPP and BNPP responses to increased precipitation amount would vary inversely with mean annual precipitation (MAP) and site productivity, (2) increased numbers of extreme rainfall events during high rainfall years would affect high and low MAP sites differently, and (3) responses belowground would mirror those aboveground. We increased growing season precipitation by as much as 50% by augmenting natural rainfall via (1) many (11-13) small or (2) fewer (3-5) large watering events, with the latter coinciding with naturally occurring large storms. Both ANPP and BNPP increased with water addition in the two C4 grasslands, with greater ANPP sensitivity in TGP, but greater BNPP and NPP sensitivity in SGP. ANPP and BNPP did not respond to any rainfall manipulations in the C3-dominated NMP. Consistent with previous studies, fewer, larger (extreme) rainfall events increased ANPP relative to many small events in SGP, but event size had no effect in TGP. Neither system responded consistently above- and belowground to event size; consequently, total NPP was insensitive to event size. The diversity of responses observed in these three grassland types underscores the challenge of predicting responses relevant to C cycling to forecast changes in precipitation regimes even within relatively homogeneous biomes such as grasslands.
Technical Abstract: To assess primary productivity responses to increases in precipitation amount and altered rainfall patterns, we conducted an experiment in 2011 and 2012 in shortgrass prairie (SGP; C4 dominated; Central Plains Experimental Grassland), northern mixed grass prairie (NMP; C3 dominated; Fort Keogh Livestock and Range Research Laboratory), and tallgrass prairie (TGP; C4 dominated; Konza Prairie Biological Station). Water was added to plots in two different patterns – in weekly small additions or a few large events added on top of existing storms - and compared to control plots which received ambient rainfall. To assess responses to watering amount regardless of pattern, treatments were pooled and compared with the control. The response variables measured were: aboveground net primary productivity (ANPP), belowground net primary productivity (BNPP; 0-15 cm and 15-30 cm), and soil moisture. Soil moisture at all sites was increased directly after water additions, but season long differences were only apparent in NMP. ANPP and BNPP responded positively in TGP and SGP to water additions applied May – August while NMP showed no response to water addition or watering pattern potentially reflecting the early growth dynamics of the extant vegetation community in this system. Sensitivity (g m-2 mm precip-1) of ANPP to water addition was highest in TGP while sensitivity of BNPP and NPP was highest in SGP. Additions coming in a few large events caused an increase in ANPP at SGP relative to the same amount of water applied in many small events while pattern had no effect on ANPP in TGP. Conversely, BNPP did not respond to watering pattern at SGP but showed the highest response to water added in a few large events in TGP. These contrasting responses above- and belowground resulted in no response of total NPP to watering pattern at any site. Implications of this study include the importance of incorporating BNPP into future studies focusing on precipitation effects on grasslands and consideration of dominant vegetative functional composition into cross-biome patterns of sensitivity to altered precipitation regimes.