Studying Spatiotemporal Changeability of Oak Forests in Zagros in Response to Rainfall Variation

The forests of the Zagros are one of the most important and fundamental treasures of the country, which plays a key role in providing water and soil resources in this region.The geographical distribution of different plant communities is dramatically dependent on climatic conditions. Changes in climatic elements, such as precipitation, can cause long-term and short-term reactions of various plant colonie. The main purpose of this study is to reveal the spatial changeability of Lorestan forest NDVI index in response to rainfall changes. The vegetation index of Lorestan was detected using Landsat 8 and 5 imageries during 2000-2017. The monthly and annual rainfall also has been obtained using accumulated monthly rainfall of 9 synoptic stations of Lorestan province. The Pearson correlation matrix has been used to analyze the relationship between annual variation of forest cover area and qnnual rainfall index. The results showed that the EVI> 0.4 threshold can be considered as the threshold of the province's forest cover. The correlation analysis showed that the 18-year time series of forest cover, was correlated with the spatial distribution of annual rainfall in Lorestan Province by 0.72 that is significant in 0.95 confidence level (P_value=0.05). Spatial analysis of the implementation of the greenness estimator model showed that the rainfall threshold of oak greenery (EVI> 0.4) was equal to 320 mm, above which the EVI index increased by 0.88 for each millimeter of rainfall growth of the studied oaks.

The geographical distribution of different plant communities is dramatically dependent on climatic conditions. Changes in climatic elements, such as precipitation, can cause long-term and short-term reactions of various plant colonies (Wang et al., 2017). The relationship between vegetation and climatic factors is so close that many researchers, including Coupon, have classified their climate systems based on the overall structure of plant communities (Guard and Prince, 1995). Climatic conditions and vegetation in each region have a two-way, intertwined relationship. In fact, climatic conditions determine the spatial distribution of species, growth period, phenological needs and even natural selection. On the other hand, they create a vegetative cover of a balanced microclimate in the heart of the region general climate, which sometimes differs significantly from the original climate. Borhan et al., 2015).

In this study, two categories of data were used. The first category of data, including the monthly rainfall data of the province during the oak tree growth period (i.e. periods when the average minimum temperature did not fall below 10C is presented. The source of this data was the monthly record of stationary precipitation during the statistical period of 2000-2017 by 9 synoptic stations of the province and the networked precipitation of GPCC climate base. The second group of data used in this study is related to the vegetation index, which was obtained from three red, infrared and blue bands of LANDSAT satellite on a monthly basis for the statistical period of 2000-2018. The Pearson spatial correlation model was used at the confidence level of 0.95 (P-value = 0.05) to analyze the relationship between spatial changes in the forest cover areas of Lorestan province and spatial distribution of precipitation based on annual and monthly scales.

In this study, forest areas, which were obtained by using threshold (EVI> 0.4), showed that the greenery of oaks in Lorestan province was the highest in peak mode (September) on an intra-year scale with rains 3 to 5 months ago. Such a delay has been observed in many other researches, including the work of Sedighifar et al. (2019), which examined the temporal and spatial response of the Hyrcanian forests of Mazandaran province to the climatic characteristics of the region, including precipitation. Secondly, the threshold of oak greenery in Lorestan province, which has been calculated using experimental sampling, has been 320 mm of rainfall, which indicates that the experimental greenery obtained is generally observed in areas where annual rainfall is more than 320 mm per year. The fitted model indicated that on an annual precipitation scale, an increase in each millimeter of precipitation (above 320 mm at the precipitation limit) would increase the EVI index by 0.88, a ratio of 0.46 to the confidence level. 0.95 has been significant. In terms of spatial changes in precipitation, the results showed that precipitation changes in oak habitats could significantly change the area of ​​oak green thresholds, so that according to the fitted linear model, each millimeter of precipitation changes in oak habitats (320 precipitation threshold). Millimeters will lead to a change of 1724 km in the area of ​​these areas with greenery above 0.4.

In this study, it was observed that precipitation of 5 months before the September of each year, ie precipitation from April to June, is the main controller of the greenness of healthy trees in the oak forests of the province. On the other hand, it was determined that the greenery threshold of these forests is 320 mm of annual rainfall. However, in some habitats, forest ash is also seen below this precipitation level. Due to this dependence of the province oak forests on precipitation, the increasing trend of anomalies and irregularities of precipitation is increasing in the future and the temporal and spatial changes of precipitation in the province will have an increasing trend provided that the continuation of the current management process and failure to take protective measures against the oak forests of the province, the process of destruction of these forests has intensified and in the near future the level of these forests in Lorestan province will greatly reduce.