Impacts of shrub-encroachment on above- and belowground processes
Encroachment of woody vegetation into grasslands leads to land degradation and is a common phenomenon occurring across arid and semi-arid regions globally. Encroachment creates islands of fertility under shrub canopies interspersed between patches of unvegetated nutrient-poor soil. Subsequent changes in response to encroachment will likely alter soil carbon and nitrogen cycles. This study experimentally determines how encroachment of creosote bush (Larrea tridentata) alters soil process and microbial function in Chihuahuan Desert vegetation. To do so, we transplanted plant-soil monoliths containing blue grama (Bouteloua gracilis) and intact soil microbial communities into substrate adjacent to creosote in encroached shrubland. Reciprocal transplants containing black grama (Bouteloua eriopoda) beneath canopies of creosote were moved into the grassland to determine if soil processes can be restored following encroachment. At each site control monoliths were excavated and reinstalled to determine impact of disturbance. We then measured pretreatment total C and N, microbial extracellular enzyme activities, and soil respiration was measured weekly from April to October 2014. We predicted carbon flux and microbial activity (1) will increase in blue grama monoliths moved into the creosote shrubland compared to control monoliths, and (2) decrease in black grama monoliths moved into the blue grama grassland compared to control monoliths.
Pretreatment analysis showed extracellular enzyme activity to be significantly higher in creosote shrubland compared to blue grama grassland for alkaline phosphatase (means = 76.5 and 52.0 µmol/h/g, respectively, P = 0.024) and marginally higher for beta-D-glucosidase (means = 114.3 and 84.2 µmol/h/g, respectively, P = 0.087), proxies for microbial phosphorous and carbon acquisition, respectively. In addition, total soil carbon and nitrogen were significantly higher (P = <0.0001) in the shrubland compared to the grassland. After monoliths were transplanted, growing season carbon dioxide flux was significantly higher in creosote monoliths moved from creosote into grassland compared to controls in both August (means = 11.3+1.3 and 7.8+0.7 µmol m-2 s-1, respectively; P = 0.02) and September (means = 7.3+1.1 and 3.6+0.7 µmol m-2 s-1, respectively; P = 0.002). Creosote monoliths transplanted to grassland had significantly higher fluxes than grassland monoliths transplanted to shrubland in September (means = 7.3+1.1 and 4.5+0.5 µmol m-2 s-1, respectively, P = 0.03). Although dieback of grasses occurred in nearly all monoliths, late season recovery also occurred in both treatments. Our results were contrary to our initial hypotheses in that higher responses occurred in monoliths transplanted from shrubland to grassland relative to soils moved from grassland to shrubland.