Prepare and train a model to predict precipitation using selected atmospheric variables

[KishanthanKingston]

Description:
We aim to build a first predictive model where precipitation is the target variable, and all other relevant atmospheric variables serve as inputs. The objective is to evaluate whether the model can learn correlations between precipitation and the broader atmospheric state.

Proposed Input Variables:

• Precipitable water (PWAT)
• Convective available potential energy (CAPE)
• Temperature (T) at 500, 700 and 850 hPa
• Wind components (u, v) at 500, 700 and 850 hPa

Tasks:

• Validate availability of PWAT, CAPE, geopotential, T, u, v at 500/700/850 hPa
• Confirm final predictor list
• Implement preprocessing & normalization
• Train baseline model
• Evaluate model performance