Forecasting Post-Wildfire Vegetation Recovery in California using a Convolutional Long Short-Term Memory Tensor Regression Network

TL;DR

This paper introduces a novel ConvLSTM tensor regression model to predict post-wildfire vegetation recovery in California, achieving high accuracy and revealing recovery patterns through clustering techniques.

Contribution

It develops a new ConvLSTM tensor regression approach for predicting NDVI and recovery rates, integrating UMAP clustering for ecological trend analysis.

Findings

50% of predictions have an absolute error of 0.12 or less

75% of predictions have an error of 0.24 or less

Identified distinct recovery trends across wildfire regions

Abstract

The study of post-wildfire plant regrowth is essential for developing successful ecosystem recovery strategies. Prior research mainly examines key ecological and biogeographical factors influencing post-fire succession. This research proposes a novel approach for predicting and analyzing post-fire plant recovery. We develop a Convolutional Long Short-Term Memory Tensor Regression (ConvLSTMTR) network that predicts future Normalized Difference Vegetation Index (NDVI) based on short-term plant growth data after fire containment. The model is trained and tested on 104 major California wildfires occurring between 2013 and 2020, each with burn areas exceeding 3000 acres. The integration of ConvLSTM with tensor regression enables the calculation of an overall logistic growth rate k using predicted NDVI. Overall, our k-value predictions demonstrate impressive performance, with 50% of predictions exhibiting an absolute error of 0.12 or less, and 75% having an error of 0.24 or less. Finally, we employ Uniform Manifold Approximation and Projection (UMAP) and KNN clustering to identify recovery trends, offering insights into regions with varying rates of recovery. This study pioneers the combined use of tensor regression and ConvLSTM, and introduces the application of UMAP for clustering similar wildfires. This advances predictive ecological modeling and could inform future post-fire vegetation management strategies.