Yeah with others variables, The idea of test data , you will test your algorithm and try to find out the prediction on test data and getting the accuracy, matrix confusion

How you create your test data :

Just training data : the training data is the flood extent in 2015 in southern Malawi

Testing data : The test data to measure the accuracy of your model is the flood extent in southern Malawi in 2019 : you need to take all features and target in this year