Kiona Ogle, Jarrett J. Barber, Brandon M. Strange, Rohan D. Boone, Alicia M. Formanack, Drew M. P. Peltier
{"title":"识别与美国西部树木极端生长状态相关的气候条件","authors":"Kiona Ogle, Jarrett J. Barber, Brandon M. Strange, Rohan D. Boone, Alicia M. Formanack, Drew M. P. Peltier","doi":"10.1111/gcb.70317","DOIUrl":null,"url":null,"abstract":"<div>\n \n <p>Climate extremes—e.g., drought, atmospheric rivers, heat waves—are increasing in severity and frequency across the western United States of America (USA). Tree-ring widths reflect the concurrent and legacy effects of such climate extremes, yet our ability to predict extreme tree growth is often poor. Could tree-ring data themselves identify the most important climate variables driving extreme low- and high-growth states? How does the importance of these climate drivers differ across species and time? To address these questions, we explored the spatial synchrony of extreme low- and high-growth years, the symmetry of climate effects on the probability of low- and high-growth years, and how climate drivers of extreme growth vary across tree species. We compiled ring widths for seven species (four gymnosperms and three angiosperms) from 604 sites in the western USA and classified each annual ring as representing extreme low, extreme high, or nominal growth. We used classification random forest (RF) models to evaluate the importance of 30 seasonal climate variables for predicting extreme growth, including precipitation, temperature, and vapor pressure deficit (VPD) during and up to four years prior to ring formation. For four species (three gymnosperms, one angiosperm) for which climate was predictive of growth, the RF models correctly classified 89%–98% and 80%–95% of low- and high-growth years, respectively. For these species, asymmetric climate responses dominated. Current-year winter hydroclimate (precipitation and VPD) was most important for predicting low growth, but prediction of high growth required multiple years of favorable moisture conditions, and the occurrence of low-growth years was more synchronous across space than high-growth years. Summer climate and temperature (regardless of season) were only weakly predictive of growth extremes. Our results motivate ecologically relevant definitions of drought such that current winter moisture stress exerts a dominant role in governing growth reductions in multiple tree species broadly distributed across the western USA.</p>\n </div>","PeriodicalId":175,"journal":{"name":"Global Change Biology","volume":"31 7","pages":""},"PeriodicalIF":12.0000,"publicationDate":"2025-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Identifying the Climate Conditions Associated With Extreme Growth States in Trees Across the Western United States\",\"authors\":\"Kiona Ogle, Jarrett J. Barber, Brandon M. Strange, Rohan D. Boone, Alicia M. Formanack, Drew M. P. Peltier\",\"doi\":\"10.1111/gcb.70317\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div>\\n \\n <p>Climate extremes—e.g., drought, atmospheric rivers, heat waves—are increasing in severity and frequency across the western United States of America (USA). Tree-ring widths reflect the concurrent and legacy effects of such climate extremes, yet our ability to predict extreme tree growth is often poor. Could tree-ring data themselves identify the most important climate variables driving extreme low- and high-growth states? How does the importance of these climate drivers differ across species and time? To address these questions, we explored the spatial synchrony of extreme low- and high-growth years, the symmetry of climate effects on the probability of low- and high-growth years, and how climate drivers of extreme growth vary across tree species. We compiled ring widths for seven species (four gymnosperms and three angiosperms) from 604 sites in the western USA and classified each annual ring as representing extreme low, extreme high, or nominal growth. We used classification random forest (RF) models to evaluate the importance of 30 seasonal climate variables for predicting extreme growth, including precipitation, temperature, and vapor pressure deficit (VPD) during and up to four years prior to ring formation. For four species (three gymnosperms, one angiosperm) for which climate was predictive of growth, the RF models correctly classified 89%–98% and 80%–95% of low- and high-growth years, respectively. For these species, asymmetric climate responses dominated. Current-year winter hydroclimate (precipitation and VPD) was most important for predicting low growth, but prediction of high growth required multiple years of favorable moisture conditions, and the occurrence of low-growth years was more synchronous across space than high-growth years. Summer climate and temperature (regardless of season) were only weakly predictive of growth extremes. Our results motivate ecologically relevant definitions of drought such that current winter moisture stress exerts a dominant role in governing growth reductions in multiple tree species broadly distributed across the western USA.</p>\\n </div>\",\"PeriodicalId\":175,\"journal\":{\"name\":\"Global Change Biology\",\"volume\":\"31 7\",\"pages\":\"\"},\"PeriodicalIF\":12.0000,\"publicationDate\":\"2025-07-09\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Global Change Biology\",\"FirstCategoryId\":\"93\",\"ListUrlMain\":\"https://onlinelibrary.wiley.com/doi/10.1111/gcb.70317\",\"RegionNum\":1,\"RegionCategory\":\"环境科学与生态学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"BIODIVERSITY CONSERVATION\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Global Change Biology","FirstCategoryId":"93","ListUrlMain":"https://onlinelibrary.wiley.com/doi/10.1111/gcb.70317","RegionNum":1,"RegionCategory":"环境科学与生态学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BIODIVERSITY CONSERVATION","Score":null,"Total":0}
Identifying the Climate Conditions Associated With Extreme Growth States in Trees Across the Western United States
Climate extremes—e.g., drought, atmospheric rivers, heat waves—are increasing in severity and frequency across the western United States of America (USA). Tree-ring widths reflect the concurrent and legacy effects of such climate extremes, yet our ability to predict extreme tree growth is often poor. Could tree-ring data themselves identify the most important climate variables driving extreme low- and high-growth states? How does the importance of these climate drivers differ across species and time? To address these questions, we explored the spatial synchrony of extreme low- and high-growth years, the symmetry of climate effects on the probability of low- and high-growth years, and how climate drivers of extreme growth vary across tree species. We compiled ring widths for seven species (four gymnosperms and three angiosperms) from 604 sites in the western USA and classified each annual ring as representing extreme low, extreme high, or nominal growth. We used classification random forest (RF) models to evaluate the importance of 30 seasonal climate variables for predicting extreme growth, including precipitation, temperature, and vapor pressure deficit (VPD) during and up to four years prior to ring formation. For four species (three gymnosperms, one angiosperm) for which climate was predictive of growth, the RF models correctly classified 89%–98% and 80%–95% of low- and high-growth years, respectively. For these species, asymmetric climate responses dominated. Current-year winter hydroclimate (precipitation and VPD) was most important for predicting low growth, but prediction of high growth required multiple years of favorable moisture conditions, and the occurrence of low-growth years was more synchronous across space than high-growth years. Summer climate and temperature (regardless of season) were only weakly predictive of growth extremes. Our results motivate ecologically relevant definitions of drought such that current winter moisture stress exerts a dominant role in governing growth reductions in multiple tree species broadly distributed across the western USA.
期刊介绍:
Global Change Biology is an environmental change journal committed to shaping the future and addressing the world's most pressing challenges, including sustainability, climate change, environmental protection, food and water safety, and global health.
Dedicated to fostering a profound understanding of the impacts of global change on biological systems and offering innovative solutions, the journal publishes a diverse range of content, including primary research articles, technical advances, research reviews, reports, opinions, perspectives, commentaries, and letters. Starting with the 2024 volume, Global Change Biology will transition to an online-only format, enhancing accessibility and contributing to the evolution of scholarly communication.