A climate-spatial matrix growth model for major tree species in Lesser Khingan Mountains and responses of forest dynamics change to different representative concentration path scenarios
{"title":"A climate-spatial matrix growth model for major tree species in Lesser Khingan Mountains and responses of forest dynamics change to different representative concentration path scenarios","authors":"Qi Sheng, Zhaogang Liu, Lingbo Dong","doi":"10.3389/ffgc.2023.1309189","DOIUrl":null,"url":null,"abstract":"Climate change affects forest distribution, species composition, structure, and yield due to the sensitivity of forests to temperature, precipitation, and CO2. Therefore, for forest management decisions regarding climate change, it is crucial to explore the response of forest growth, mortality, and recruitment to future climate. We aimed to establish tree species’ responses by introducing variables such as climate, stand spatial structure parameters, and diversity indices.We produced fixed parameter transition matrix model (FM), climate-sensitive matrix growth model (CM) and climate-spatial matrix growth model (SCM) using data from 786 plots collected during the 7th (2010), and 8th (2015), Chinese National Forest Inventories in Heilongjiang Province, and long-term predictive performance of CM, SCM, and FM were compared using same data. The models were compared using tenfold cross-validation and long-term predictive performance analysis. To predict the response of major tree species in the Lesser Khingan Mountains to three future climate change scenarios (RCP2.6, RCP4.5, RCP8.5).The cross-validation results show small significant differences among the three models for short-term prediction (5 years), with the FM performing slightly better than the CM and the SCM. In contrast, for long-term projections (85 years), SCM outperformed FM and CM under three different RCPs, and SCM and CM under three representative concentration paths (RCPs), i.e., RCP2.6, RCP4.5, and RCP8.5, suggesting that rather different dynamics are more reliable, since climatic factors are taken into account which may significantly affect forest dynamics, while changes in stand spatial structure also affect the sensitivity of trees to climate, especially in long-term prediction interval, the results of this paper may provide a theoretical basis for optimizing forest management strategies under climate change.","PeriodicalId":12538,"journal":{"name":"Frontiers in Forests and Global Change","volume":"25 2","pages":""},"PeriodicalIF":2.7000,"publicationDate":"2023-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Frontiers in Forests and Global Change","FirstCategoryId":"97","ListUrlMain":"https://doi.org/10.3389/ffgc.2023.1309189","RegionNum":3,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"ECOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Climate change affects forest distribution, species composition, structure, and yield due to the sensitivity of forests to temperature, precipitation, and CO2. Therefore, for forest management decisions regarding climate change, it is crucial to explore the response of forest growth, mortality, and recruitment to future climate. We aimed to establish tree species’ responses by introducing variables such as climate, stand spatial structure parameters, and diversity indices.We produced fixed parameter transition matrix model (FM), climate-sensitive matrix growth model (CM) and climate-spatial matrix growth model (SCM) using data from 786 plots collected during the 7th (2010), and 8th (2015), Chinese National Forest Inventories in Heilongjiang Province, and long-term predictive performance of CM, SCM, and FM were compared using same data. The models were compared using tenfold cross-validation and long-term predictive performance analysis. To predict the response of major tree species in the Lesser Khingan Mountains to three future climate change scenarios (RCP2.6, RCP4.5, RCP8.5).The cross-validation results show small significant differences among the three models for short-term prediction (5 years), with the FM performing slightly better than the CM and the SCM. In contrast, for long-term projections (85 years), SCM outperformed FM and CM under three different RCPs, and SCM and CM under three representative concentration paths (RCPs), i.e., RCP2.6, RCP4.5, and RCP8.5, suggesting that rather different dynamics are more reliable, since climatic factors are taken into account which may significantly affect forest dynamics, while changes in stand spatial structure also affect the sensitivity of trees to climate, especially in long-term prediction interval, the results of this paper may provide a theoretical basis for optimizing forest management strategies under climate change.