Assessing the capability of Modis fire detector products in identifying fires in Golestan State

The use of remote sensing is particularly important in identifying burned areas due to its extensive spatial coverage and the provision of information at different times. Today, Modis fire products are widely used for this purpose. The purpose of this study is to evaluate the capability of Modis MOD14, MOD14A2 (Terra), and MYD14, MYD14A2 (Aqua) fire detector products and to identify fire areas in Golestan state. First, a point map of all the images was generated, then to evaluate the accuracy of the fire products, the prepared point map for the products was compared with terrestrial reality data. If the location of each of the reported fires is consistent with the fires identified by the products, that location was correctly identified as the fire. Landsat images were used as a ground accuracy map to evaluate the accuracy of Modis images. The results showed that six regions identified by level 2 fire products and eight regions were detected by level 3 fire products were identified. The results show the accuracy of the images with a coefficient of R ^ 2 of 0.94 and a coefficient of RMSE of 426.12 ha. The studies conducted in this study show that to improve the performance of the text fire detection algorithm, this algorithm is proposed for the forests of Golestan province and following the conditions and characteristics of the fire area, its intensity, and area. Be developed to provide better results.