Food security is threatened by climate change worldwide; consequently, agriculture and farming livelihoods must adapt to new and unpredictable conditions. These conditions vary along spatial scales, and since agricultural yields are sensitive to microclimate conditions, a locally tailored data-driven approach may be helpful. Furthermore, limited agricultural resources like water and labor increasingly constrain food production. This research proposes a regional portfolio model for identifying crop choices and regional portfolio compositions that align with known and forecasted microclimate variation in temperature and humidity. The model will enable farmers to assess tradeoffs between the financial returns and agricultural production risks. The goal of this work is to provide new insights into agricultural planning in the face of climate risk and limited access to water and labor resources. Three steps are taken. Firstly, regional agricultural land is divided into farming subunits, with each representing a terroir characterized by temperature and humidity. Then a simulated yield coefficient is used to assess the effect of microclimate variables on the yield of the different crops in the portfolio of each subunit. Secondly, farming resource allocation, represented by water and labor, across crops and farming subunits is optimized to maximize the yield and associated financial return from farming across the agricultural region. Finally, a resilient agricultural planning model is developed based on the assumed data for regional microclimate and agricultural resources. The results of this research can be used by regional farmers as a reference for selecting crop portfolios and resource allocations to maximize overall profit.