English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Accurate and comprehensive vegetation prediction methods are essential for effective agricultural planning and budgeting. Most existing vegetation prediction methods rely on sampling points rather than on overall spatiotemporal characteristics, making it difficult to accurately forecast vegetation changes. Hence, we built a neural network model with encoding and decoding modules based on a convolution gate recurrent unit (ConvGRU) and applied it to spatiotemporal normalized difference vegetation index (NDVI) predictions for the upper Heihe river basin (UHRB) in the Qilian Mountains, China. Based on MODIS NDVI raster data for the UHRB for 2000&#x2013;2020, the analysis of the region&#x0027;s spatiotemporal characteristics showed that the NDVI varied significantly over time. To avoid the gradient disappearance problem during ConvGRU model prediction, we proposed several numerical scaling methods for preprocessing the data before conducting training and prediction. We then constructed a spatiotemporal prediction model based on the ConvGRU, trained and predicted the numerically scaled data, and used various metrics to evaluate the model. The results showed that the grouping tan<italic>h</italic>-ln function fitting was the least erroneous, and the ConvGRU model using data scaled by this method performed well across various metrics according to the test-set results. Also, unlike traditional time series prediction methods, the model accounted for spatiotemporal correlation features, and the output data of the prediction model were continuous and intuitive. Therefore, the proposed method is suitable for predicting dynamic vegetation changes. The NDVI prediction trend analysis indicated that the vegetation in the UHRB should improve in the future.

ieee journal of selected topics in applied earth observations and remote sensing

Spatiotemporal Prediction of Alpine Vegetation Dynamic Change Based on a ConvGRU Neural Network Model: A Case Study of the Upper Heihe River Basin in Northwest China