A model is generated to combine occurrence data from GBIF and satellite data from NEO to predict species in space using environmental variables. Ground observational data along with NASA satellite-based remote sensing data of environmental variables such as temperature, precipitation, aerosol optical thickness, and vegetation, etc are taken to forecast habitual places for a particular species around the globe as shown in Figure 1. However, our model can be generalized to all species for any environmental variable at any place of desired interest. Also, biodiversity is measured using two scales: Richness and Shannon index. Richness defines the number of different species present whereas the Shannon index defines the ratio of the population of a particular species to the total population.

In this model, statistical analysis for the environmental factor, for instance, the temperature is shown in detail. A histogram is made to know the fraction of the total occurrence of species at any temperature. This leads to an understanding of the approximate temperature range suitable for that particular species. Using satellite data of world temperature, the points are marked on the map for the species lying in that temperature range. Those points correspond that the probability of locating a particular species is relatively higher than other unmarked points (Figure 2(a),(b)).

After doing the same prediction for other environmental variables, an intersection is taken of marked points from all the maps to generate a new prediction map. This prediction map best represents the most probable habitable places for the chosen species. Figure 3 explains the intersection of different plots on the world map for wolf species.

Similarly, several previous year data (1950-2020) for a specific environmental variable (such as temperature) is taken from Globe Data and NOAA for future prediction of species using curve fitting techniques. Figure 4 depicts the future temperature using the curve fitting method. Here, as an example, the temperature is predicted for the next 60 years of Charleston Intl. Airport, SC US.

Since the average future temperature is now known in a particular place, it can be compared with the temperature range obtained earlier (Figure 2(b)). If the predicted temperature lies outside the temperature range, it can be configured that the probability of getting endangered or extinct is quite high for that particular species in the future. Through this space-time prediction model, one can optimize every step with more number of environmental variables to get better prediction. Figure 5 shows the combination of space and time model to make predictions for biodiversity.

In this way, we are trying to create awareness among the people who are trying to save our planet from natural as well as human activities but unable to do so due to lack of resources. Our model can help the various organizations and environmentalists to predict the future of our planet 'Earth'.