Influence of drought on canopy cover and plant dieback on an elevational gradient at Big Bend National Park

Background/Question/Methods

In 2011, Texas experienced the most severe single year drought ever recorded resulting in significant plant mortality throughout Big Bend National Park. We used this event to test how perennial plant response to extreme drought varied across communities at different elevations. In October 2010 and 2011, we measured plant cover by species at six elevations ranging from Chihuahuan desert (666-m) to oak forest in the Chisos mountains (1920-m). We asked the following questions: (1) Did susceptibility to drought differ among functional groups or by leaf traits?; (2) What was the relationship between elevation and dieback?; and (3) Was any relationship driven by turnover in functional groups or in leaf trait changes with elevation? At each elevation, species were identified and canopy cover was measured on randomly placed 50-m line-intercept transects. We included all woody and succulent species and categorized species by functional group. In 2011, the percent of canopy dieback for each species was estimated. In 2010, we collected leaves from 2-5 individuals per species per elevation to measure leaf traits including leaf mass per area (LMA). Living cover (total and by functional group) and dieback were analyzed across elevations using mixed-effect ANCOVA models with transects nested within elevation.

Results/Conclusions

Total plant cover for all species increased with elevation in both 2010 and 2011 from a transect average of ~20% cover at 666-m to >100% cover at 1920-m. Living canopy cover for all woody species decreased significantly between 2010 and 2011 (p<0.0001).  For all plants, there was more canopy dieback at lower elevations (p<0.001); live canopy cover decreased by 17% at 1920-m and by 83% at 666-m. Dieback for trees and shrubs decreased with elevation (p<0.05). However, dieback in subshrubs and succulents did not vary significantly among elevations. There was a significant relationship between LMA and dieback (p<0.05): transects with larger average LMA had less dieback than transects with smaller average LMA. These results indicate that the 2011 drought in Big Bend National Park had a large effect on communities at all elevations and this effect was consistent across growth forms with average dieback ranging from 5% at the highest elevation to 85% at lowest elevations. In 2011, the Panther Junction ranger station (1143-m) received 7.7 cm of precipitation or ~21% of the yearly average precipitation. Differences among elevations were driven largely by differences in precipitation and drought severity and not by turnover in growth forms.