气候变化领域集成服务门户(CCPortal): AgriCarbon-EO v1.0.1: large-scale and high-resolution simulation of carbon fluxes by assimilation of Sentinel-2 and Landsat-8 reflectances using a Bayesian approach

CCPortal

DOI	10.5194/gmd-17-997-2024
	AgriCarbon-EO v1.0.1: large-scale and high-resolution simulation of carbon fluxes by assimilation of Sentinel-2 and Landsat-8 reflectances using a Bayesian approach
	Wijmer, Taeken; Al Bitar, Ahmad; Arnaud, Ludovic; Fieuzal, Remy; Ceschia, Eric
发表日期	2024
ISSN	1991-959X
EISSN	1991-9603
起始页码	17
结束页码	3
卷号	17 期号:3
英文摘要	Soil organic carbon storage is a well-identified climate change mitigation solution. Quantification of the soil carbon storage in cropland for agricultural policy and offset carbon markets using in situ sampling would be excessively costly, especially at the intrafield scale. For this reason, comprehensive monitoring, reporting, and verification (MRV) of soil carbon and its explanatory variables at a large scale need to rely on hybrid approaches that combine remote sensing and modelling tools to provide the carbon budget components with their associated uncertainties at intrafield scale. Here, we present AgriCarbon-EO v1.0.1: an end-to-end processing chain that enables the estimation of carbon budget components for major and cover crops at intrafield resolution (10 m) and regional extents (e.g. 10 000 km 2 ) by assimilating remote sensing data (e.g. Sentinel-2 and Landsat8) in a physically based radiative transfer (PROSAIL) and agronomic models (SAFYE-CO2). The data assimilation in AgriCarbon-EO is based on a novel Bayesian approach that combines normalized importance sampling and look-up table generation. This approach propagates the uncertainties across the processing chain from the reflectances to the output variables. After a presentation of the chain, we demonstrate the accuracy of the estimates of AgriCarbon-EO through an application over winter wheat in the southwest of France during the cropping seasons from 2017 to 2019. We validate the outputs with flux tower data for net ecosystem exchange, biomass destructive samples, and combined harvester yield maps. Our results show that the scalability and uncertainty estimates proposed by the approach do not hinder the accuracy of the estimates (net ecosystem exchange, NEE: RMSE = 1.68 -2.38 gC m - 2 , R 2 = 0.87 -0.77; biomass: RMSE = 11.34 g m - 2 , R 2 = 0.94 ). We also show the added value of intrafield simulations for the carbon components through scenario testing of pixel and field simulations (biomass: bias = - 47 g m - 2 , - 39 % variability). Our overall analysis shows satisfying accuracy, but it also points out the need to represent more soil processes and include synthetic aperture radar data that would enable a larger coverage of AgriCarbon-EO. The paper's findings confirm the suitability of the choices made in building AgriCarbon-EO as a hybrid solution for an MRV scheme to diagnose agro-ecosystem carbon fluxes.
语种	英语
WOS研究方向	Geology
WOS类目	Geosciences, Multidisciplinary
WOS记录号	WOS:001190510000001
来源期刊	GEOSCIENTIFIC MODEL DEVELOPMENT
文献类型	期刊论文
条目标识符	http://gcip.llas.ac.cn/handle/2XKMVOVA/305937
作者单位	INRAE; Universite de Toulouse; Universite Toulouse III - Paul Sabatier; Centre National de la Recherche Scientifique (CNRS); Institut de Recherche pour le Developpement (IRD); INRAE; Universite Federale Toulouse Midi-Pyrenees (ComUE); Universite de Toulouse; Institut National Polytechnique de Toulouse
推荐引用方式 GB/T 7714	Wijmer, Taeken,Al Bitar, Ahmad,Arnaud, Ludovic,et al. AgriCarbon-EO v1.0.1: large-scale and high-resolution simulation of carbon fluxes by assimilation of Sentinel-2 and Landsat-8 reflectances using a Bayesian approach[J],2024,17(3).
APA	Wijmer, Taeken,Al Bitar, Ahmad,Arnaud, Ludovic,Fieuzal, Remy,&Ceschia, Eric.(2024).AgriCarbon-EO v1.0.1: large-scale and high-resolution simulation of carbon fluxes by assimilation of Sentinel-2 and Landsat-8 reflectances using a Bayesian approach.GEOSCIENTIFIC MODEL DEVELOPMENT,17(3).
MLA	Wijmer, Taeken,et al."AgriCarbon-EO v1.0.1: large-scale and high-resolution simulation of carbon fluxes by assimilation of Sentinel-2 and Landsat-8 reflectances using a Bayesian approach".GEOSCIENTIFIC MODEL DEVELOPMENT 17.3(2024).