Table 6.

Prediction performance of the non-machine learning (naïve) and neural network models when tested on the evaluation dataset for all the model training sites.

Site	Model	Naive			FFNN			LSTM
	${\mathit{R}}^2$	$\mathit{MAE}$ $\left({\mathbf{m}}^3{\mathbf{m}}^{-3}\right)$	$\mathit{RMSE}$ $\left({\mathbf{m}}^3{\mathbf{m}}^{-3}\right)$	${\mathit{R}}^2$	$\mathit{MAE}$ $\left({\mathbf{m}}^3{\mathbf{m}}^{-3}\right)$	$\mathit{RMSE}$ $\left({\mathbf{m}}^3{\mathbf{m}}^{-3}\right)$	${\mathit{R}}^2$	$\mathit{MAE}$ $\left({\mathbf{m}}^3{\mathbf{m}}^{-3}\right)$	$\mathit{RMSE}$ $\left({\mathbf{m}}^3{\mathbf{m}}^{-3}\right)$
Baluderry	0.89	0.02	0.03	0.94	0.01	0.01	0.95	0.01	0.01
Stoughton	0.88	0.02	0.03	0.97	0.01	0.01	0.97	0.01	0.01
Waddeston	0.92	0.01	0.02	0.99	0.01	0.01	0.99	0.01	0.01