The problem of forest fires requires creating methodologies that allow evaluating and predicting the response that the ecosystem willhave to the impact of fire, in order to direct restoration actions in the areas that most require it. However, evaluating these areasdirectly in the field implies investment of resources (financial and personnel) which, along with time, are generally limited. For this,satellite images are a practical tool for the evaluation of large areas, or inaccessible areas, impacted by forest fires. In this work, thecorrelation presented by different variables measured in the field and derived from remote sensors, in relation to the naturalregeneration of pine that occurs in the La Primavera forest and in Sierra de Quila, Jalisco, was evaluated. The results showeddifferent variables to determine the predictive models of the natural regeneration of pine after the occurrence of a forest fire, beingthe fuels of 100 hours and 1000 hours, bark thickness and depth of burning, the variables taken directly in the field. that wereincluded in the models. While the burn area index, the regeneration index and the exposure, the variables taken by remote censorswere included in the predictive models. The models that showed a higher R² are those obtained by field variables for the tworegions. However, the model obtained only with remote sensor variables for La Primavera obtained an R² of 0.6083, Contrary toSierra de Quila where the model does not take any spectral index for the model, therefore it is advisable to establish a greaternumber of Sampling sites evenly distributed throughout the area affected by the fire, to improve the accuracy of the remote sensingmodels