| Tree mortality,strongly affected by climate,influences stands structure,community dynamics,and even ecosystem services.However,there is still a lack of comprehensive analysis of tree mortality of different tree heights and sizes(diameters at breast height,DBH)to climates over large areas.Here,we analyzed the Forest Inventory and Analysis data to characterize the mortality rate,cause of death,mortality depends on height and size,and how climate influences the tree mortality in the conterminous United States during 2012-2017 census.Our main research results are as follows:(1)Tree mortality rates and causes showed a clear spatial pattern.Midwest(mortality rate over 6%accounted for 21.8%),Northeast(the mortality rate over 6%accounted for 10.8%),and parts of Rocky Mountains(the mortality rate over 6%accounted for 12.5%)in CONUS show a spatial pattern of high mortality.The dead trees in the Rocky Mountains were mainly affected by fire and insect,while 47%and 63%of trees cannot observe the causes of death in the Midwest and Northeast.(2)The trees in low canopy levels became more vulnerable than in top canopy with increased aridity.The height difference between live and dead trees(△H decreases with the increase of the canopy level,which indicates the mortality risk decreases with the increase of the canopy level.The △H of each canopy decreased as the annual precipitation(PPTtotal)decreases,the maximum summer temperature(Tmax)rises,and the maximum vapor pressure deficit(VPDmean)rises,indicating the mortality risk of each canopy increased under increased aridity.In particular,the mortality risk of the low canopy increases faster than the top canopy,which indicates the high vulnerability of the low canopy.(3)The magnitude and concavity of the size-dependent mortality curve changed with different regions.The mortality curve magnitude shows the average proportion of dead trees,which increased with the increase of the absolute variability of the saturated vapor pressure deficit(VPDSD)(R2=0.135 p<0.05)and the relative variability of the saturated vapor pressure difference(VPDcov)(R2=0.10,p<0.05),decreased with the increase of PPTtotal(R2=0.26,p<0.001),the absolute variability of the precipitation(PPTSD)(R2=0.19,p<0.01),and mean annual temperature(Tmean)(R2=0.10,p<0.05).The mortality curve concavity shows the difference in the death of trees on different DBH,which decreased with the increase of VPDSD(R2=0.16,p<0.01),VPDcov(R2=0.16,p<0.01),and the relative variability of precipitation(PPTcov)(R2=0.20,p<0.01),and increased with the increase of PPTtotal(R2=0.23,p<0.001).Based on the regional shift of the curve concavity across states,we here proposed the competition-climate transition hypothesis/theorem as the fundamental underlying mechanism for the aridity-driven change of mortality curve concavity.The main causes of tree mortality changed from competition to climate with the increased aridity.In summary,this study analyzed the spatial patterns of mortality rate,causes of death,quantified the mortality depends on height and DBH,and further explored the impact of climate on tree mortality.Our result highlight the significant role of climate change in affecting tree mortality,particularly trees in low canopy.Future climate change is therefore likely to alter forest succession,composition,diversity,and myriad of ecosystem services more severe than we have perceived under evolving new mechanisms. |
