Effect of climate change and local management on aboveground carbon dynamics (1987–2015) in Zagros oak forests using Landsat time-series imagery

Abstract

Forest carbon stocks are a time-integrated manifestation of various phenomena and processes ranging from tree growth and mortality to natural and human disturbances. Understanding the effects of environmental and human activities is critically important in vulnerable ecosystems like arid and semi-arid forests, given climate variability coupled with historical human activities. Zagros forests are one of the largest vegetation communities in the Middle East. This region is highly affected by dust storms which are mainly a result of the loss of vegetation. This current study is an exploration of changes to aboveground carbon (AGC) density as affected by climate change (CC) and local management from 1987 to 2015 at 5-year intervals based on Landsat imagery and field data, analyzed with stochastic gradient boosting regression. Two sites with different local management practices were selected in Zagros coppice oak forests and AGC density was measured using field sampling. Four climate variables including mean annual temperature, rainfall, and evaporation were used to explore the effect of CC on AGC density dynamics over the study period. Results revealed an upward trend of AGC density in all intervals for both sites (from 6.36 to 7.5 ton ha−1) except for the 2010–2015 interval. The decrease of AGC density between 2010 and 2015 was likely related to dieback of oak trees due to different factors such as dust storms, pests, and other diseases. The trend of increasing AGC density in SarfiruzAbad forests was more than in Gahvareh forests, which is indicative of the young forest stand growth created by conservation and regeneration of damaged forests. Due to the large effect of local management on AGC density, our findings indicate that it was difficult to assess the effect of CC on AGC density. Our results suggest that other forests with a high degree of human disturbance, like Zagros forests, can play a role in mitigating global warming if the local management strategies are optimized. Additionally, these results demonstrate that in a region with no historical information about carbon stored in forests and woodlands, like most developing countries, Landsat-based AGC monitoring can provide information for forest managers and policy makers for understanding carbon accounting under human disturbances and CC, which could guide forest management strategies.