One of the biggest challenges today, is the need to feed a growing population. Considering climate change and the effect it may have on agriculture systems, the sustainable intensification of food production is a necessity. The objective of the research presented in this paper is to address the issues related to the large-scale stressors that can present a barrier to the current efforts for sustainable intensification of food production in Qatar. It relies on remote sensing imagery as the primary dataset owing to the ability of the satellite sensors to cheaply gather data with good areal coverage and spatial resolution, and with added benefit of repeated surveys from the satellite overpasses over a given area. The steps of the proposed methodology developed in this research include: (a) image processing for training and prediction using neural networks; (b) extraction of atmospheric and surface features that can directly affect greenhouse operations, e.g. those related to land and water resources, and energy requirements; (c) formulation of a mixed integer non-linear programming (MINLP) framework using surrogate modelling for the maximization of the crop productivity objective; (d) development of GIS database augmented with recommender systems. The results obtained so far are promising, with root mean squared percentage error of 11.93 for the crop water demand prediction, and it is envisaged that the proposed development of a smart geospatial framework with predictive capabilities, EWF optimization, and augmented GIS will facilitate for informed policy development aimed at the success of the food security programs in Qatar.