Lagged mortality among tree species four years after an exceptional drought in east Texas

Abstract

AbstractIn 2011, east Texas experienced the worst drought on record causing extensive tree mortality. Initial mortality estimates for 2012 varied among tree genera. A rapid damage assessment (RDA) estimated that 65.5 (± 7.3) million trees died as a result of the drought in this region one year post‐drought. However, this estimate was untested against established monitoring networks. Moreover, pests and physiological damage can elevate tree mortality multiple years beyond a drought event. Since the RDA was unable to quantify multi‐year trends, it remained unclear whether these drivers caused increased tree mortality in east Texas beyond one year post‐drought and how different species responded over time. To address these questions, we compared total 2012 standing dead tree (SDT) estimates (i.e., drought‐killed plus all other SDT excluding harvested or salvaged trees) derived from the RDA and U.S. Forest Service Forest Inventory and Analysis (FIA) data for east Texas. Total SDT estimates did not significantly differ between the RDA (120.5 ± 8.5 million) and FIA (108.4 ± 8.7 million). Furthermore, total SDT estimates for the four most common genera (Pinus, Quercus, Liquidambar, Ulmus), which comprised over 80% of all species, did not significantly differ between the RDA and FIA. Additionally, we used logistic regression and FIA data from east Texas for 2011 through four years post‐drought (2012–2015) to examine temporal trends in plot‐level drought‐ and pest‐driven tree mortality of seven key species (Pinus taeda, Pinus echinata, Quercus nigra, Quercus stellata, Quercus falcata, Liquidambar styraciflua, Ulmus alata) from the four most common genera. At the plot‐level, drought‐driven mortality was immediate for the three Quercus species (notably Q. falcata) and L. styraciflua which significantly increased in 2012 while P. taeda mortality was delayed, not increasing significantly until 2013. Pest‐driven mortality increased from 2013 to 2015 for all species, with the highest mortality observed in Q. falcata and lowest in P. taeda and U. alata. This study affirms the validity and value of independent sampling efforts to quantify mortality immediately following major disturbance and also demonstrates the need for longer‐term species‐level assessments beyond the initial year post‐drought to account for differential impacts from drought and pests.