Evaluation of climate change in northern Iran during the last four centuries by using dendroclimatology

Climate change is currently one of the most important environmental issues. Dendrochronology is frequently used to identify the climatic changes most closely associated with changes in tree-ring extent. We applied dendroclimatology to determine the climate changes in the Roodbar region of Iran during the last four centuries. The climatic index of De Martonne (aridity index), annual precipitation, and annual mean temperature were simulated by using dendroclimatology (tree-rings) and an artificial neural network (ANN). Dendroclimatology studies were carried out with the use of the Cupressus sempervirens species. A multilayer perceptron network was adopted for the ANN. Tree-ring width was the input variable for the simulation, whereas annual precipitation, annual mean temperature, and the aridity index were the outputs. After the training process, the network was validated. The validated network and tree-rings were used to simulate changes in climatic factors during the last four centuries. The climatic factors simulated by using dendroclimatology can be used in drought studies and climate change evaluation. The results showed that in the last four centuries, the climate of the study area changed from semiarid to arid, and its annual precipitation decreased significantly. The significant changes in climate were found to occur in the mid-twentieth century.