Climate change is affecting the planet due to lack of information regarding its temperatures, precipitation and decrease in the surface area of the snow-capped mountains, which causes a high impact on the melting of the snow-capped mountains, for this reason it is important to have predictive models of multivariate regressions that allow forecasting the surface area of snow-capped mountains over time. In the investigation, the multivariate regression model was used for each snowfall under study, having as regressive variables the temperature, precipitation and as a dependent variable the surface area of the snowfall with data obtained through Landsat 4–5, 7 and 8 images during the years 2010–2020. A model of the snow-capped Huaytapallana was obtained in the form y = 34.274 − 2.3197 X1 + 0.135 X2 with p-value = 0.034 of significance and the snow-capped Coropuna with a model of the shape y = 101.3487− 6.720 X1 + 0.331 X2 having p-value = 0.036 being significant for your prediction. The mean of the highest surface area was Coropuna (53.92 km2) with a SD of 14.94, and the lowest was Verónica (16.16 km2) with a SD of 37.77. The Huaytapallana mountain showed the least reduction in surface area, decreasing by 7%, while the Verónica mountain was reduced by 52%, being the most affected of the 4 snow-capped mountains, due to the melting under study during the 10 years of data analyzed.