Scholarlink

<p>Aboveground Net Primary Production (ANPP) is a critical indicator of ecosystem productivity and health, heavily influenced by environmental and climatic factors. Monitoring ANPP in the fragile ecosystems of arid and semi-arid regions is vital for sustainable resource management. The objective of this study was to integrate biogeochemical modeling, machine-learning approaches, and climate change projections to assess historical dynamics and future trends of ANPP in Tehran Province, Iran. This study first estimated historical ANPP during 2000–2020 using the Carnegie–Ames–Stanford Approach (CASA) model driven by 16-day MODIS imagery with a spatial resolution of 250&#xa0;m. Ground validation was performed using data from 240 field sampling sites. Over this period, ANPP increased by an average of 51.71gCm⁻²yr⁻¹. To quantify the relationships between ANPP and environmental drivers, Generalized Additive Models (GAM), Artificial Neural Networks (ANN), and Support Vector Machines (SVM) were applied using selected climatic and topographic variables, with SVM showing the best predictive performance (training data: R² = 0.72, RMSE = 3.57; test data: R² = 0.78, RMSE = 3.39). Finally, the best model was coupled with CMIP6 climate projections to simulate future ANPP changes under SSP126, SSP370, and SSP585 scenarios for the period 2041–2070. The results indicate projected ANPP increases of 2.75%, 2.15%, and 1.7%, respectively, relative to 2020. These findings demonstrate the combined utility of remote sensing, machine-learning modeling, and climate scenario analysis for evaluating ANPP responses to climate change, and provide valuable insights for ecosystem management and climate adaptation planning in semi-arid regions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Modeling aboveground net primary production (ANPP) in Tehran Province: insights from remote sensing and climate change projections

  • 期刊论文

  • 发表日期：

    2026年3月28日

查看全文
  • Marzieh Asgari,
  • Mostafa Tarkesh,
  • Reza Jafari,
  • Mahshid Souri

摘要

Aboveground Net Primary Production (ANPP) is a critical indicator of ecosystem productivity and health, heavily influenced by environmental and climatic factors. Monitoring ANPP in the fragile ecosystems of arid and semi-arid regions is vital for sustainable resource management. The objective of this study was to integrate biogeochemical modeling, machine-learning approaches, and climate change projections to assess historical dynamics and future trends of ANPP in Tehran Province, Iran. This study first estimated historical ANPP during 2000–2020 using the Carnegie–Ames–Stanford Approach (CASA) model driven by 16-day MODIS imagery with a spatial resolution of 250 m. Ground validation was performed using data from 240 field sampling sites. Over this period, ANPP increased by an average of 51.71gCm⁻²yr⁻¹. To quantify the relationships between ANPP and environmental drivers, Generalized Additive Models (GAM), Artificial Neural Networks (ANN), and Support Vector Machines (SVM) were applied using selected climatic and topographic variables, with SVM showing the best predictive performance (training data: R² = 0.72, RMSE = 3.57; test data: R² = 0.78, RMSE = 3.39). Finally, the best model was coupled with CMIP6 climate projections to simulate future ANPP changes under SSP126, SSP370, and SSP585 scenarios for the period 2041–2070. The results indicate projected ANPP increases of 2.75%, 2.15%, and 1.7%, respectively, relative to 2020. These findings demonstrate the combined utility of remote sensing, machine-learning modeling, and climate scenario analysis for evaluating ANPP responses to climate change, and provide valuable insights for ecosystem management and climate adaptation planning in semi-arid regions.