Regional Nonlinear Relationships Across the United States Between Drought and Tree‐Ring Width Variability From a Neural Network

Neural networks were previously applied to reconstruct climate indices from tree rings but showed mixed results in skill relative to more standard linear methods. A two‐layer neural network is explored for purposes of reconstructing summertime self‐calibrated Palmer Drought Severity Index (scPDSI) across the contiguous United States. Reconstructions using neural networks are more skillful than a linear approach at 75% of the gridboxes if evaluated by the coefficient of efficiency and at 54% when using the Pearson correlation coefficient. The increased reconstruction skill is related to the network capturing nonlinear growth‐climate relationships. In the Southwest, in particular, a nonlinear response function captures a diminishing sensitivity of growth to moisture under wetter conditions, consistent with alleviation of moisture stress. These results indicate somewhat less‐severe and more‐stable incidences of drought over the past two centuries in the U.S. Southwest. Plain Language Summary Drought reconstructions using tree‐ring records generally rely upon linear methods. The relationship between climate and tree growth can, however, involve nonlinearities, and so it is useful to explore methods capable of capturing nonlinear processes. We apply a neural network approach to drought reconstruction across the contiguous United States and show that it confers greater skill than a traditional linear method. To evaluate skill, both methods are trained on a subset of the data and then evaluated against another data set. Greater neural network skill is traced to this method accounting for a diminished growth response of trees to increasingly wet conditions, especially in the Southwest. Key Points Under the hypothesis that tree growth is nonlinearly related to climate, a neural network is implemented to reconstruct drought The neural network shows a modest increase in skill relative to a standard linear reconstruction method Accounting for nonlinearities, the magnitude of 19th century droughts in the United States Southwest are revised to be 20% less severe