[hal-04654536] Mapping compost and digestate spreadings from Sentinel-2 and Sentinel-1 on a farm scale
According to few recent studies, exogenous organic matters (EOM) can be detectable on either emerging vegetation or bare soil using optical and radar remote sensing techniques. Nevertheless, these image processing approaches considered one single EOM, one season and/or year only and were limited to one surface condition prior to spreading. So far no method addressed the simultaneously tracking of both liquid and solid EOM applications using satellite imagery, for several years and surface conditions. Relying on Support Vector Machine (SVM) classifier, this study aimed to track applications of both composted manure and liquid digestate over three periods and years in agricultural fields on a farm scale with distinct surface conditions (grassland, winter crop, bare soil) in Nouzilly, France. Various combinations of covariates from Sentinel-2 and Sentinel-1 data served to train SVM in a bootstrapping approach in order to assess the uncertainty of map results. Classification performance was higher for pre- and post-application image pairs compared to post-application images alone and slightly improved when adding Sentinel-1 data. While the areal percentage of the highest uncertainty class covered less of 10% of the mapped area regardless of the year, the best models showed accuracies higher than 93% in 2020 and 2021. In 2019, the overall accuracy did not reach more than 79%, probably due to rainfall events and considerable time lags between the image pairs. This study underscores not only the potential of Sentinel-2 and 1 for monitoring EOM applications, but also the requirement of better understanding the spectral behaviour of the EOM spreadings, in line with a thorough characterization of the sequence of crop technical management.
ano.nymous@ccsd.cnrs.fr.invalid (Maxence Dodin) 19 Jul 2024
https://hal.science/hal-04654536v1
[hal-04553210] Uncertainty in Digital Soil Mapping at broad-scale: A review
With the needs of efficient acquisition of soil information, Digital Soil Mapping (DSM) has been greatly developed and widely applied for over the past two decades. The spatial estimates of soil properties produced with diverse methods over various study areas, have been often seen as the main output of DSM, as they play an important role in environmental modelling and policy. However, compared with the soil property maps, their prediction uncertainty is still less emphasized, which may potentially lead to mis-uses of results and inappropriate decisions if the uncertainty is not assessed, reported, and taken into account by end-users.In this communication, we present a preliminary review of the sources of prediction uncertainties in DSM coming from learning soil data (data source, sampling in space and time, measurements), covariates, and models. We also summarize the methods used to estimate the uncertainty, and to assess the reliability of the uncertainty estimates. We also consider the propagation of uncertainties when several soil attributes are combined to derive information and/or used as inputs for modelling. Furthermore, we discuss some strategies for mitigating the uncertainty, challenges, and future perspectives. This review aims to consolidate the understanding of DSM uncertainties and to contribute to reliable DSM practices, facilitating more informed decision-making in soil-related research and management. 
ano.nymous@ccsd.cnrs.fr.invalid (Qianqian Chen) 19 Apr 2024
https://hal.inrae.fr/hal-04553210v1
[hal-04654485] Temporal S2/S1 mosaics combined with environmental covariates for regional SOC mapping: lessons from la Beauce (France) and the Västra-Skaraborg (Sweden)
Satellite-based soil organic carbon content (SOC) mapping over wide regions is generally hampered by the low soil sampling density and the discrepancy of soil sampling collection dates. Some unfavorable topsoil conditions, such as high moisture, roughness, the presence of crop residues, the limited amplitude of SOC values and the limited area of bare soil when a single image is used, are also among the influencing factors. For two contrasted wide agricultural areas in boreal and temperate zones, Veskra-Skaraborg (Sweden) and la Beauce (France), this study compares approaches relying on Sentinel-2 (S2) temporal mosaics of bare soil (S2Bsoil) over ≥5 years jointly with Sentinel 1, SOC measurements data and other environmental covariates derived from digital elevation models and/or lithology maps and/or airborne gamma-ray data. Prediction models relied on quartile random forest, with 10 fold cross-validation, according to several datasets: i) “Sentinel-2”, the Sentinel-2 bands of a given S2Bsoil; ii) “terrain”, the terrain covariates (Digital Elevation Model and its derivatives, plus oblique geographic coordinates); iii), “Sentinel-2” plus “terrain”; iv) “all”, i.e. “Sentinel-2”, “terrain” and a selection of relevant Sentinel-2 spectral indices. Lessons from la Beauce (Urbina-Salazar et al., 2023) deal with (i) the dates and periods that are preferable to construct temporal mosaics of bare soils while accounting for soil moisture and soil management; (ii) which set of covariates is more relevant to explain the SOC variability. The models using all the covariates had the best model performance. Airborne gamma-ray thorium, slope and S2 bands (e.g., bands 6, 7, 8, 8a) and indices (e.g., calcareous sedimentary rocks, “calcl”) from the “late winter–spring” time series were the most important covariates in this model. Our results also indicated the important role of neighboring topographic distances and oblique geographic coordinates between remote sensing data and parent material. These data contributed not only to optimizing SOC mapping performance but also provided information related to long-range gradients of SOC spatial variability, which makes sense from a pedological point of view. Lessons from the Veskra-Skaraborg deal with the impact of percentile thresholding for temporal mosaicking of bare soils in relationship with soil moisture and cloud frequency. Performance decreased from R90 to R25. Models were highly predictive over both Plain (RPIQ≤1.3) or Till, yet with a slight improvement for Till (best RPIQ 1.4). Results confirm the differences in performances according to soilscape and agricultural system, and the complex interactions due to soil moisture in satellite-based soil property mapping. For both study áreas, spectral models alone were not well performing, but covariates such as morphometric layers slightly improved the prediction from temporal mosaics of bare soils. Reference: Castaldi, F., Koparan, M.H., Wetterlind, J., Žydelis, R., Vinci, I., Savaş, A.Ö., Kıvrak, C., Tunçay, T., Volungevičius, J., Obber, S., Ragazzi, F., Malo, D., Vaudour, E. 2023. Assessing the capability of Sentinel-2 time-series to estimate soil organic carbon and clay content at local scale in croplands, ISPRS Journal of Photogrammetry and Remote Sensing, 199, 40-60, https://doi.org/10.1016/j.isprsjprs.2023.03.016 Urbina-Salazar, D., Vaudour, E., Richer-de-Forges, A.C., Chen, S., Martelet, G., Baghdadi, N., Arrouays, D., 2023. Sentinel-2 and Sentinel-1 Bare Soil Temporal Mosaics of 6-year Periods for Soil Organic Carbon Content Mapping in Central France. Remote Sensing 15, 2410. https://doi.org/10.3390/rs15092410
ano.nymous@ccsd.cnrs.fr.invalid (Diego Urbina-Salazar) 19 Jul 2024
https://hal.inrae.fr/hal-04654485v1
[hal-04508449] Sentinel-2/1 Bare Soil Temporal Mosaics of 6-year Periods for Soil Organic Carbon Content Mapping in La Beauce, Central France
[...]
ano.nymous@ccsd.cnrs.fr.invalid (Diego Urbina-Salazar) 18 Mar 2024
https://hal.inrae.fr/hal-04508449v1
[hal-04507597] Influence of percentile reflectance thresholding in Sentinel-2 temporal mosaicking on regional SOC and clay prediction performances: case of the Västra-Skaraborg region (Sweden)
[...]
ano.nymous@ccsd.cnrs.fr.invalid (Diego Urbina-Salazar) 16 Mar 2024
https://hal.inrae.fr/hal-04507597v1
[hal-04283669] Using Machine-Learning Algorithms to Predict Soil Organic Carbon Content from Combined Remote Sensing Imagery and Laboratory Vis-NIR Spectral Datasets
Understanding spatial and temporal variability in soil organic carbon (SOC) content helps simultaneously assess soil fertility and several parameters that are strongly associated with it, such as structural stability, nutrient cycling, biological activity, and soil aeration. Therefore, it appears necessary to monitor SOC regularly and investigate rapid, non-destructive, and cost-effective approaches for doing so, such as proximal and remote sensing. To increase the accuracy of predictions of SOC content, this study evaluated combining remote sensing time series with laboratory spectral measurements using machine and deep-learning algorithms. Partial least squares (PLS) regression, random forest (RF), and deep neural network (DNN) models were developed using Sentinel-2 (S2) time series of 58 sampling points of bare soil and according to three approaches. In the first approach, only S2 bands were used to calibrate and compare the performance of the models. In the second, S2 indices, Sentinel-1 (S1) indices, and S1 soil moisture were added separately during model calibration to evaluate their effects individually and then together. In the third, we added the laboratory indices incrementally and tested their influence on model accuracy. Using only S2 bands, the DNN model outperformed the PLS and RF models (ratio of performance to the interquartile distance RPIQ = 0.79, 1.36 and 1.67, respectively). Additional information improved performances only for model calibration, with S1 soil moisture yielding the most stable improvement among three iterations. Including equivalent indices of the S2 indices calculated using soil spectra obtained under laboratory conditions improved prediction of SOC, and the use of only two indices achieved good validation performances for the RF and DNN models (mean RPIQ = 2.01 and 1.77, respectively).
ano.nymous@ccsd.cnrs.fr.invalid (Hayfa Zayani) 14 Nov 2023
https://hal.science/hal-04283669v1
[hal-04189398] Sentinel-2 satellite images for monitoring cattle slurry and digestate spreading on emerging wheat crop: a field spectroscopy experiment
This study is aimed to evaluate the utility of Sentinel-2 imagery for monitoring exogenous organic matter (EOM) applied on winter wheat crop, using two spatial scales: proximal and satellite. From proximal sensing, multi-temporal spectral field measurements were taken on experimental fields consisting of three treatments (cattle slurry, liquid and raw digestates) and a control throughout 46 days. From Sentinel-2 satellites, images were analysed before and after EOM application. For both sensing scales, EOM and vegetation indices were used. On any scale of observation, the digestates spread on emerging wheat were easily detectable in late winter, in contrast to spring spreading events which were hindered by the developed vegetation. The agglomerative hierarchical clustering from the EOM indices divided by EVI achieved to discriminate digestates at early and medium stages of vegetation growth. Our findings did not apply for cattle slurry, presumably because of both lower organic and dry matter contents. HIGHLIGHTS • Digestates spread on emerging wheat are detectable in late winter. • Developed vegetation constrains the detection of spring spreading events. • Spectral measurements did not separate the field with cattle slurry and the control. • The visible to near infrared bands are the most impacted after digestate spreading.
ano.nymous@ccsd.cnrs.fr.invalid (Maxence Dodin) 28 Aug 2023
https://hal.inrae.fr/hal-04189398v1
[hal-04350813] Sentinel-2 Imagery for Monitoring Exogenous Organic Matter Fertilizers on Winter Wheat Crop: Proximal and Satellite Approaches
The use of exogenous organic matter (EOM) fertilizers, such as digestate and cattle slurry, has gained increasing attention due to their potential to reduce reliance on synthetic fertilizers in agriculture. This study evaluated the utility of Sentinel-2 imagery for monitoring different liquid EOM fertilizers applied on winter wheat crop, using both proximal and satellite scales. At the proximal scale, spectral field measurements were taken of experimental fields consisting of three treatments (cattle slurry, liquid and raw digestates) and a control over 46 days. Field reflectance spectra were simulated into the MSI spectral bands of Sentinel-2. At the satellite scale, Sentinel-2 images were analyzed before and after EOM application for each experimental field. EOM and vegetation indices were used to monitor EOM application at both scales. The main findings of this study refer to digestates. Firstly, the spread of digestates on emerging wheat can be easily detected in late winter, up to 15 days after application. Secondly, the visible to near infrared bands are the most impacted the first days after spreading and the visible to red-edge bands are persistently impacted 15 days after spreading. Finally, the detection of spring spreading events is constrained or even hindered by developed vegetation. These findings did not apply to cattle slurry, which was hardly visible in the field and in Sentinel-2 images. This Sentinel-2-based approach can serve as a primer for further implementation over larger fields.
ano.nymous@ccsd.cnrs.fr.invalid (Maxence Dodin) 18 Dec 2023
https://hal.inrae.fr/hal-04350813v1
