Focal-TSMP: Deep learning for vegetation health prediction and agricultural drought assessment from a regional climate simulation

Abstract

Abstract. In this study, we investigate applying deep learning (DL) models on a regional climate simulation produced by the Terrestrial Systems Modelling Platform (TSMP Ground to Atmosphere G2A) for vegetation health modeling and agricultural drought assessment. The TSMP simulation is performed in a free mode and the DL model is used in an intermediate step to synthesize Normalized Difference Vegetation Index (NDVI) and Brightness Temperature (BT) images from the TSMP simulation over Europe. These predicted images are then used to derive different vegetation and drought indices like NDVI anomaly, BT anomaly, Vegetation Condition Index (VCI), Thermal Condition Index (TCI), and Vegetation Health Index (VHI). To ensure reliability and to assess the model applicability with different seasonality and spatial variability, we provide an analysis of model biases and uncertainties across different regions over the Pan-Europe domain. We further provide an analysis about the contribution of the input variables from the TSMP model components to ensure a better understanding of the model prediction. A comprehensive evaluation on the long-term TSMP using reference remote sensing data showed sufficiently good agreements between the model predictions and observations. While model performance varies on the test set between different climate regions, it achieves a mean absolute error (MAE) of 0.027 and 1.90 K° with coefficient of determination (R2) scores of 0.88 and 0.92 for NDVI and BT, respectively, at 0.11° resolution for sub-seasonal predictions. Our study could be used as a complimentary evaluation framework for climate change simulations with TSMP. Moreover, the developed DL model could be integrated with data assimilation and used for down-stream tasks, i.e., modelling the impact of extreme events on vegetation responses with different climate change scenarios. In summary, we demonstrate the feasibility of using DL on a TSMP simulation to synthesize NDVI and BT, which can be used for agricultural drought forecasting. Our implementation is publicly available at the project page (https://hakamshams.github.io/Focal-TSMP).

Focal-TSMP: deep learning for vegetation health prediction and agricultural drought assessment from a regional climate simulation

“…Meanwhile, others showed a different perspective of trends related to terrestrial vegetation from remote sensing products (Zhu et al, 2016;Kogan et al, 2020). This is usually explained as assessments are highly dependent on drought definition (Satoh et al, 2021;Reyniers et al, 2023) and extreme event attribution (Van Oldenborgh et al, 2021), i.e., the drought indicator that was chosen in the methodology and the variations in modeling platforms. In addition, prescribed vegetation assumptions exist in climate simulations which limit the modeling of atmospheric carbon effects or soil moisture deficiency on vegetation (Pirret et al, 2020;Pokhrel et al, 2021;Reyniers et al, 2023).…”

“…In addition, prescribed vegetation assumptions exist in climate simulations which limit the modeling of atmospheric carbon effects or soil moisture deficiency on vegetation (Pirret et al, 2020;Pokhrel et al, 2021;Reyniers et al, 2023). If we add to this the complex spatiotemporal response of vegetation to climate variability (Seneviratne et al, 2021;Jin et al, 2023), i.e., regional responses to climate have different dynamics and are more complicated than those on a global scale, we can conclude that predicting the vegetation state in response to drought under climate conditions still poses a major challenge. More precisely, in this study we predict satellite-based vegetation products from a free, evolving simulation based on the Terrestrial Systems Modeling Platform (TSMP) (Furusho-Percot et al, 2019a).…”