Algorithm 1.

2-Step Optimization Workflow for Data Fusion Process.

1. Input: 2D input maps ($X_{climate}$) for each of the 5 climate features, 2D input maps ($X_{SEG}$) compacting all 11 socioeconomic-governmental factors, size of lead-time ($T_{lead}$) for Sub-Scenarios (a-f), values of model's hyperparameters ($H_1,H_2,\dots \dots$), threshold values for precision metric ($P_{T,1},P_{T,2}$), validation dataset, testing dataset

2. Output: 2D output maps ($Y$) representing spatial classes distribution for all 251 countries with defined lead-time using the proposed approach

3. for each$X_{climate}$ ← Step I of Optimization Process

do

a. initialize training of deep learning prediction model with user-defined $H_1,H_2,\dots \dots$ values & $X_{climate}$with selected Sub-Scenario (a-f) for defined $T_{lead}$ in Step I optimization process

b. derive $Y$ predictions on validation and testing datasets via trained model with defined $T_{lead}$ & $H_1,H_2,\dots \dots$ values

c. model evaluation of predictions on testing dataset via $P$ score

if$P$ ≥ $P_{T,1}$

Proceed

d. store $P$ score with corresponding $H_1,H_2,\dots \dots$ values & type of Sub-Scenario for defined $T_{lead}$

4. select optimal $X_{climate}$ with highest $P(>P_{T,1})$ score with corresponding $H_1,H_2,\dots \dots$ values & type of Sub-Scenario for defined $T_{lead}$

5. do

a. fuse optimal $X_{climate}$ with $X_{SEG}$

b. initialize training of deep learning prediction model with selected $H_1,H_2,\dots \dots$ values & type of Sub-Scenario for defined $T_{lead}$ from (4) ← Step II of Optimization Process

c. derive $Y$ predictions on validation and testing datasets with trained model

d. model evaluation of predictions on testing dataset via $P$ score

if$P$ ≥ $P_{T,2}$

Proceed

store best prediction model for near real-time predictions

Else

re-train prediction model with additional hyperparameters ($H_1,H_2,\dots \dots$) tuning

6. end