Fire risk assessment based on vegetal covers evaluation by NDVI. The case of the wildfire of Rio Tinto in 2004 (Huelva, Spain)

Current wildfires generate a huge loss of natural resources, to assess their effect historical on natural coverage in the soil, a method based on the employment of a series of satellite resources of LANDSAT and MODIS analyzed by Google Earth Engine was developed. Cartography and graphics of the Differential Index of Vegetation Normalization, obtained from these satellite resources, have been used to analyze and monitor wildfires that occurred. The purpose of this study is to develop a method to evaluate the wildfire effect evaluating the impact that suffers soil areas and vegetal masses. For this purpose, a method based on satellite images, which allows for identifying coverage of natural conditions, their phenological phases, and their seasonal vulnerability to fire was developed and applied to the study area of Riotinto Mines wildfire in 2004. Tinto River wildfire was one of the highest in the south of Spain. For this reason, it was chosen to analyze the severity of the burned soil and its recovery. The methodology employed by Google Earth Engine and Index Normalized Vegetation Differential (NDVI) encompasses a process of obtaining satellite statistical data and processing these using Georeferenced Information Systems (GIS). The outcome has been a reproducible model to study other burned areas where changes in NDVI levels made it possible to identify a strong transformation of the vegetation cover after the fire. In particular, the degradation of vigorous vegetation cover was almost complete. The vegetation has taken years to recover, so in 2009, five years after the fire, the vegetation still did not reach the levels of pre-fire NDVI. The study carried out highlights the coverage of transitional forest-shrub and broadleaf forest since they are the ones that have received the greatest impact and those with the greatest difficulty in recovery, being the coverages that would have to receive greater post-fire monitoring and study. Furthermore, the sclerophyllous vegetation and other covertures are the easiest to light. On the other hand, the covers of natural grassland and sclerophyllous vegetation are those of faster recovery. Finally, it is highlighted that in the year of the fire, the transitional forest-shrub cover and sclerophyllous vegetation were located below their normal NDVI value for July, which could help the damage and fast extension of the wildfire. ACKNOWLEDGMENT  This study has been carried out thanks to FENIX 4.0 (PDC 2022-133157-I00) funded by MCIN/AEI/q0.13039/501100011033 and the Next Generation EU/PRTR; FENIX (PID2019-107257RB-I00) financed by MCIN/AEI//10.13039/501100011033/ FEDER “A way to build Europe” and ATLAS financed by Joint Programming Initiatives Cultural Heritage and Global Change (JPI CH). P. Ortiz thanks the University of Ca´Foscari (Venice, Italy) for her research stay and the Ministry of Universities (Spain Government) for the funds of the senior stay for researchers (PRX22/00544).