Improved estimation of forage nitrogen in alpine grassland by integrating Sentinel-2 and SIF data
Abstract Nitrogen is an essential element for the growth and reproduction of vegetation in alpine grasslands and plays a vital role in determining the nutrient-carrying capacity of plants and maintaining the balance of forage nutrition supply and demand. In recent years, the widespread application o...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-05-01
|
| Series: | Plant Methods |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s13007-025-01389-2 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Abstract Nitrogen is an essential element for the growth and reproduction of vegetation in alpine grasslands and plays a vital role in determining the nutrient-carrying capacity of plants and maintaining the balance of forage nutrition supply and demand. In recent years, the widespread application of high-resolution multispectral satellites (i.e., Sentinel-2) equipped with multiple red-edge bands has proven an effective approach for estimating forage nitrogen content in alpine grassland habitats. In addition, solar-induced chlorophyll fluorescence (SIF), as a direct probe of vegetation photosynthesis, has become an effective indicator for estimating key growth parameters of green vegetation in recent years. However, it currently unknown whether integrating SIF and Sentinel-2 satellite data can further enhance the mapping accuracy of forage nitrogen content in alpine grassland. In this study, we integrates SIF products from TanSat and Orbiting Carbon Observatory-2 (OCO-2) satellites, Sentinel-2 Multi-Spectral Instrument (MSI) data with derived vegetation indices, and field observations across phenological stages (green-up stage, vigorous growth stage, and senescence stage) in northeastern Tibetan Plateau alpine grasslands to develop support vector machine (SVM), gaussian process regression (GPR), and artificial neural network (ANN) models for regional-scale forage nitrogen estimation. The results indicated that both the Sentinel-2 (V-R2 of 0.68–0.71, CVRMSE of 17.73–18.65%) and SIF data (V-R2 of 0.59–0.73, CVRMSE of 17.05–21.40%) individually yielded relatively accurate estimates of the forage nitrogen. The integrated model constructed using both spectral data types explained 69–74% of the variation in forage nitrogen content, with a CVRMSE ranging from 16.89 to 17.85%, which indicates that the synergy between Sentinel-2 and SIF data can slightly enhance the model’s estimation capability of forage nitrogen content. Thus, integrating Sentinel-2 and SIF data presents a potential solution for precisely measuring spatial distribution of forage nitrogen in alpine grassland at the regional scale. The proposed method provides a feasible framework for the spatiotemporal prediction of the key forage growth parameters of forage and offers a theoretical basis for determining the rational utilization of grassland resources and studying the nutritional balance between grassland and livestock. |
|---|---|
| ISSN: | 1746-4811 |