Soil Organic Carbon Mapping Using LUCAS Topsoil Database and Sentinel-2 Data: An Approach to Reduce Soil Moisture and Crop Residue Effects

A critical systematic review on spectral-based soil nutrient prediction using machine learning

The United Nations (UN) emphasizes the pivotal role of sustainable agriculture in addressing persistent starvation and working towards zero hunger by 2030 through global development. Intensive agricultural practices have adversely impacted soil quality, necessitating soil nutrient analysis for enhancing farm productivity and environmental sustainability. Researchers increasingly turn to Artificial Intelligence (AI) techniques to improve crop yield estimation and optimize soil nutrition management. This study reviews 155 papers published from 2014 to 2024, assessing the use of machine learning (ML) and deep learning (DL) in predicting soil nutrients. It highlights the potential of hyperspectral and multispectral sensors, which enable precise nutrient identification through spectral analysis across multiple bands. The study underscores the importance of feature selection techniques to improve model performance by eliminating redundant spectral bands with weak correlations to targeted nutrients. Additionally, the use of spectral indices, derived from mathematical ratios of spectral bands based on absorption spectra, is examined for its effectiveness in accurately predicting soil nutrient levels. By evaluating various performance measures and datasets related to soil nutrient prediction, this paper offers comprehensive insights into the applicability of AI techniques in optimizing soil nutrition management. The insights gained from this review can inform future research and policy decisions to achieve global development goals and promote environmental sustainability.

Carbon farming: Are soil carbon certificates a suitable tool for climate change mitigation?

Increasing soil organic carbon (SOC) stocks in agricultural soils removes carbon dioxide from the atmosphere and contributes towards achieving carbon neutrality. For farmers, higher SOC levels have multiple benefits, including increased soil fertility and resilience against drought-related yield losses. However, increasing SOC levels requires agricultural management changes that are associated with costs. Private soil carbon certificates could compensate for these costs. In these schemes, farmers register their fields with commercial certificate providers who certify SOC increases. Certificates are then sold as voluntary emission offsets on the carbon market. In this paper, we assess the suitability of these certificates as an instrument for climate change mitigation. From a soils' perspective, we address processes of SOC enrichment, their potentials and limits, and options for cost-effective measurement and monitoring. From a farmers' perspective, we assess management options likely to increase SOC, and discuss their synergies and trade-offs with economic, environmental and social targets. From a governance perspective, we address requirements to guarantee additionality and permanence while preventing leakage effects. Furthermore, we address questions of legitimacy and accountability. While increasing SOC is a cornerstone for more sustainable cropping systems, private carbon certificates fall short of expectations for climate change mitigation as permanence of SOC sequestration cannot be guaranteed. Governance challenges include lack of long-term monitoring, problems to ensure additionality, problems to safeguard against leakage effects, and lack of long-term accountability if stored SOC is re-emitted. We conclude that soil-based private carbon certificates are unlikely to deliver the emission offset attributed to them and that their benefit for climate change mitigation is uncertain. Additional research is needed to develop standards for SOC change metrics and monitoring, and to better understand the impact of short term, non-permanent carbon removals on peaks in atmospheric greenhouse gas concentrations and on the probability of exceeding climatic tipping points.

Remote Sensing of Surface and Subsurface Soil Organic Carbon in Tidal Wetlands: A Review and Ideas for Future Research

Tidal wetlands, widely considered the most extensive reservoir of soil organic carbon (SOC), can benefit from remote sensing studies enabling spatiotemporal estimation and mapping of SOC stock. We found that a majority of the remote-sensing-based SOC mapping efforts have been focused on upland ecosystems, not on tidal wetlands. We present a comprehensive review detailing the types of remote sensing models and methods used, standard input variables, results, and limitations for the handful of studies on tidal wetland SOC. Based on that synthesis, we pose several unexplored research questions and methods that are critical for moving tidal wetland SOC science forward. Among these, the applicability of machine learning and deep learning models for predicting surface SOC and the modeling requirements for SOC in subsurface soils (soils without a remote sensing signal, i.e., a soil depth greater than 5 cm) are the most important. We did not find any remote sensing study aimed at modeling subsurface SOC in tidal wetlands. Since tidal wetlands store a significant amount of SOC at greater depths, we hypothesized that surface SOC could be an important covariable along with other biophysical and climate variables for predicting subsurface SOC. Preliminary results using field data from tidal wetlands in the southeastern United States and machine learning model output from mangrove ecosystems in India revealed a strong nonlinear but significant relationship (r2 = 0.68 and 0.20, respectively, p < 2.2 × 10−16 for both) between surface and subsurface SOC at different depths. We investigated the applicability of the Soil Survey Geographic Database (SSURGO) for tidal wetlands by comparing the data with SOC data from the Smithsonian’s Coastal Blue Carbon Network collected during the same decade and found that the SSURGO data consistently over-reported SOC stock in tidal wetlands. We concluded that a novel machine learning framework that utilizes remote sensing data and derived products, the standard covariables reported in the limited literature, and more importantly, other new and potentially informative covariables specific to tidal wetlands such as tidal inundation frequency and height, vegetation species, and soil algal biomass could improve remote-sensing-based tidal wetland SOC studies.

Satellite Imagery to Map Topsoil Organic Carbon Content over Cultivated Areas: An Overview

There is a need to update soil maps and monitor soil organic carbon (SOC) in the upper horizons or plough layer for enabling decision support and land management, while complying with several policies, especially those favoring soil carbon storage. This review paper is dedicated to the satellite-based spectral approaches for SOC assessment that have been achieved from several satellite sensors, study scales and geographical contexts in the past decade. Most approaches relying on pure spectral models have been carried out since 2019 and have dealt with temperate croplands in Europe, China and North America at the scale of small regions, of some hundreds of km2: dry combustion and wet oxidation were the analytical determination methods used for 50% and 35% of the satellite-derived SOC studies, for which measured topsoil SOC contents mainly referred to mineral soils, typically cambisols and luvisols and to a lesser extent, regosols, leptosols, stagnosols and chernozems, with annual cropping systems with a SOC value of ~15 g·kg−1 and a range of 30 g·kg−1 in median. Most satellite-derived SOC spectral prediction models used limited preprocessing and were based on bare soil pixel retrieval after Normalized Difference Vegetation Index (NDVI) thresholding. About one third of these models used partial least squares regression (PLSR), while another third used random forest (RF), and the remaining included machine learning methods such as support vector machine (SVM). We did not find any studies either on deep learning methods or on all-performance evaluations and uncertainty analysis of spatial model predictions. Nevertheless, the literature examined here identifies satellite-based spectral information, especially derived under bare soil conditions, as an interesting approach that deserves further investigations. Future research includes considering the simultaneous analysis of imagery acquired at several dates i.e., temporal mosaicking, testing the influence of possible disturbing factors and mitigating their effects fusing mixed models incorporating non-spectral ancillary information.

Estimation of Soil Organic Carbon Content in the Ebinur Lake Wetland, Xinjiang, China, Based on Multisource Remote Sensing Data and Ensemble Learning Algorithms

Soil organic carbon (SOC), as the largest carbon pool on the land surface, plays an important role in soil quality, ecological security and the global carbon cycle. Multisource remote sensing data-driven modeling strategies are not well understood for accurately mapping soil organic carbon. Here, we hypothesized that the Sentinel-2 Multispectral Sensor Instrument (MSI) data-driven modeling strategy produced superior outcomes compared to modeling based on Landsat 8 Operational Land Imager (OLI) data due to the finer spatial and spectral resolutions of the Sentinel-2A MSI data. To test this hypothesis, the Ebinur Lake wetland in Xinjiang was selected as the study area. In this study, SOC estimation was carried out using Sentinel-2A and Landsat 8 data, combining climatic variables, topographic factors, index variables and Sentinel-1A data to construct a common variable model for Sentinel-2A data and Landsat 8 data, and a full variable model for Sentinel-2A data, respectively. We utilized ensemble learning algorithms to assess the prediction performance of modeling strategies, including random forest (RF), gradient boosted decision tree (GBDT) and extreme gradient boosting (XGBoost) algorithms. The results show that: (1) The Sentinel-2A model outperformed the Landsat 8 model in the prediction of SOC contents, and the Sentinel-2A full variable model under the XGBoost algorithm achieved the best results R² = 0.804, RMSE = 1.771, RPIQ = 2.687). (2) The full variable model of Sentinel-2A with the addition of the red-edge band and red-edge index improved R² by 6% and 3.2% over the common variable Landsat 8 and Sentinel-2A models, respectively. (3) In the SOC mapping of the Ebinur Lake wetland, the areas with higher SOC content were mainly concentrated in the oasis, while the mountainous and lakeside areas had lower SOC contents. Our results provide a program to monitor the sustainability of terrestrial ecosystems through a satellite perspective.

Using Sentinel-2 Images for Soil Organic Carbon Content Mapping in Croplands of Southwestern France. The Usefulness of Sentinel-1/2 Derived Moisture Maps and Mismatches between Sentinel Images and Sampling Dates

In agronomy, soil organic carbon (SOC) content is important for the development and growth of crops. From an environmental monitoring viewpoint, SOC sequestration is essential for mitigating the emission of greenhouse gases into the atmosphere. SOC dynamics in cropland soils should be further studied through various approaches including remote sensing. In order to predict SOC content over croplands in southwestern France (area of 22,177 km²), this study addresses (i) the influence of the dates on which Sentinel-2 (S2) images were acquired in the springs of 2017–2018 as well as the influence of the soil sampling period of a set of samples collected between 2005 and 2018, (ii) the use of soil moisture products (SMPs) derived from Sentinel-1/2 satellites to analyze the influence of surface soil moisture on model performance when included as a covariate, and (iii) whether the spatial distribution of SOC as mapped using S2 is related to terrain-derived attributes. The influences of S2 image dates and soil sampling periods were analyzed for bare topsoil. The dates of the S2 images with the best performance (RPD ≥ 1.7) were 6 April and 26 May 2017, using soil samples collected between 2016 and 2018. The soil sampling dates were also analyzed using SMP values. Soil moisture values were extracted for each sample and integrated into partial least squares regression (PLSR) models. The use of soil moisture as a covariate had no effect on the prediction performance of the models; however, SMP values were used to select the driest dates, effectively mapping topsoil organic carbon. S2 was able to predict high SOC contents in the specific soil types located on the old terraces (mesas) shaped by rivers flowing from the southwestern Pyrénées.

Spatial prediction of soil organic carbon stocks in an arid rangeland using machine learning algorithms

Assessing the role of machine learning (ML) models concerning environmental predictors on spatial variation of soil organic carbon stocks (SOCS) in arid rangelands is very necessary. This study was conducted to explore the variability of surface SOCS in rangeland in the west of Iran using ML approaches. A number of 33 environmental predictors derived from Sentinel-2B and DEM were used. The optimal soil sampling (n = 80) position was determined by Latin hypercube sampling (cLHS) method. Robust and popular random Forest (RF), cubist (CB) along with random forest-ordinary kriging (RF-OK), and cubist-ordinary kriging (CB-OK) hybrid ML models were applied to the prediction of SOCS. Ten-fold CV was implemented for modeling performance and uncertainty map. According to data analysis, the maximum, minimum, and average values of SOCS are 44.50, 10.50, and 20.50 (ton. ha^-1) at the surface depth (0-30 cm), respectively. In general, normalized and standardized height covariates had a higher effect related to other predictors. On the other hand, two remote sensing (RS) indices, including salinity ratio (salinity) and GNDVI index, had a better impact on SOCS variability. The external validation of model performance indicated that RF-OK with (R² = 0.75, RMSE = 6.33 ton. ha^-1) with the high and low uncertainty range (3.33-9.50 ton. ha^-1) was the outperformed ML model in compare with other models as RF (R² = 0.65, RMSE = 7.38 ton. ha^-1), CB-OK (R² = 0.56, RMSE = 9.22 ton. ha^-1), and CB (R² = 0.33, RMSE = 10.42 ton. ha^-1). In general, the hybrid models improved the accuracy of RF and CB with increased 0.11 until 0.23 of R², and 1.05 to 1.2 (ton. ha^-1) decreased RMSE of model's prediction. Hence, we conclude that the topographic attributes (especially normalized and standardized height) were the most critical factors in controlling surface SOCS in arid rangelands when combining with robust RF ML model, and optimized soil sampling methods like RF-cLHS can prepare acceptable soil properties maps.

Earth Observation Data-Driven Cropland Soil Monitoring: A Review

We conducted a systematic review and inventory of recent research achievements related to spaceborne and aerial Earth Observation (EO) data-driven monitoring in support of soil-related strategic goals for a three-year period (2019–2021). Scaling, resolution, data characteristics, and modelling approaches were summarized, after reviewing 46 peer-reviewed articles in international journals. Inherent limitations associated with an EO-based soil mapping approach that hinder its wider adoption were recognized and divided into four categories: (i) area covered and data to be shared; (ii) thresholds for bare soil detection; (iii) soil surface conditions; and (iv) infrastructure capabilities. Accordingly, we tried to redefine the meaning of what is expected in the next years for EO data-driven topsoil monitoring by performing a thorough analysis driven by the upcoming technological waves. The review concludes that the best practices for the advancement of an EO data-driven soil mapping include: (i) a further leverage of recent artificial intelligence techniques to achieve the desired representativeness and reliability; (ii) a continued effort to share harmonized labelled datasets; (iii) data fusion with in situ sensing systems; (iv) a continued effort to overcome the current limitations in terms of sensor resolution and processing limitations of this wealth of EO data; and (v) political and administrative issues (e.g., funding, sustainability). This paper may help to pave the way for further interdisciplinary research and multi-actor coordination activities and to generate EO-based benefits for policy and economy.

Sentinel-2 Recognition of Uncovered and Plastic Covered Agricultural Soil

Medium resolution satellite data, such as Sentinel-2 of the Copernicus programme, offer great new opportunities for the agricultural sector, and provide insights on soil surface characteristics and their management. Soil monitoring requires a high-quality dataset of uncovered and plastic covered agricultural soil. We developed a methodology to identify uncovered soil pixels in agricultural parcels during seedbed preparation and considered the impacts of clouds and shadows, vegetation cover, and artificial covers, such as those of greenhouses and plastic mulch films. We preserved the spatial and temporal integrity of parcels in the process and analysed spectral anomalies and their sources. The approach is based on freely available tools, namely Google Earth Engine and R Programming packages. We tested the methodology on the northern region of Belgium, which is characterised by small, fragmented parcels. We selected a period between mid-April to end-May, when active agricultural management practices leave the soil bare in preparation for the main cropping season. The spectral angle mapper was used to identify soil covered by non-plastic greenhouses or temporary soil covers, such as plastic mulch films. The effect of underlying soil on temporary covers was considered. The retrogressive plastic greenhouse index was used for detecting plastic greenhouses. The result was a high quality dataset of potential bare uncovered agricultural soil that allows further soil surface characterisation. This offered an improved understanding of the use of artificial covers, their spatial distribution, and their corresponding crops during the considered period. Artificial covers occurred most frequently in maize parcels. The approach resulted in precision values exceeding 0.9 for the detection of temporary covers and non-plastic greenhouses and a sensitivity value exceeding 0.95 for non-plastic and plastic greenhouses.

Multitemporal satellite imagery analysis for soil organic carbon assessment in an agricultural farm in southeastern Brazil

Soil organic carbon (SOC) plays a crucial role for soil health. However, large datasets needed to accurately assess SOC at high resolution across scales are labor-intensive, time-consuming, and expensive. Ancillary geodata, including remote sensing spectral indices (RS-SIs) and topographic indicators (TIs), have been proposed as spatial covariates. Reported relationships between SOC and RS-SIs are erratic, possibly because single-date RS-SIs do not accurately capture SOC spatial variability due to transient confounding factors in the soil (e.g., moisture). However, multitemporal RS-SI data analysis may lead to noise reduction in SOC versus RS-SI relationships. This study aimed at: i) comparing single-date versus multitemporal RS-Sis derived from Sentinel-2 imagery for assessment of topsoil (0-0.2 m) SOC in two agricultural fields located in south-eastern Brazil; ii) comparing the performance of RS-SIs and TIs; iii) using adequate RS-SIs and TIs to compare sampling schemes defined on different collection grids; and iv) studying the temporal changes of SOC (0-0.2 m and 0.2-0.4 m). Results showed that: i) single-date RS-SIs were not reliable proxies for topsoil SOC at the study sites. For most of the tested RS-SIs, multitemporal data analysis produced accurate proxies for SOC; e.g., for the Normalized Difference Vegetation Index, the 4.5th multitemporal percentile predicted SOC with an R² of 0.64; ii) The best TI was elevation (ranging from 643 to 684 m) with an R² of 0.70; iii) The multitemporal SI and elevation maps indicated that the different sampling schemes were equally representative of the topsoil SOC's distribution across the entire area; and iv) From 2012 through 2019, topsoil SOC increased from 19.3 to 24.1 g kg^-1. The ratio between SOC in the topsoil and subsoil (0.2-0.4 m) decreased from 1.7 to 1.1. Further testing of the proposed multitemporal RS-SI analysis is necessary to confirm its dependability for SOC assessment in Brazil and elsewhere.

Sentinel-2 and Landsat-8 Multi-Temporal Series to Estimate Topsoil Properties on Croplands

The spatial and temporal monitoring of soil organic carbon (SOC), and other soil properties related to soil erosion, is extremely important, both from the environmental and economic perspectives. Sentinel-2 (S2) and Landsat-8 (L8) time series increase the probability to observe bare soil fields in croplands, and thus, monitor soil properties over large regions. In this regard, this work suggests an automated pixel-based approach to select only pure soil pixels in S2 and L8 time series, and to make a synthetic bare soil image (SBSI). The SBSIs and the soil properties measured in the framework of the European LUCAS survey were used to calibrate SOC, clay, and CaCO3 prediction models. The results highlight a high correlation between laboratory soil spectra and the SBSIs median spectra, especially for the SBSI obtained by a three-year S2 collection, which provides satisfactory results in terms of SOC prediction accuracy (RPD: 1.74). The comparison between S2 and L8 results demonstrated the higher capability of the S2 sensor in terms of SOC prediction accuracy, mainly due to the greater spatial resolution of the bands in the visible region. Whereas, neither S2 nor L8 could accurately predict the clay and CaCO3 content. This is because of the low spectral and spatial resolution of their SWIR bands that prevent the exploitation of the narrow spectral features related to these two soil attributes. The results of this study prove that large S2 time series can estimate and monitor SOC in croplands using an automated pixel-based approach that selects pure soil pixels and retrieves reliable synthetic soil spectra.

Estimation of Soil Organic Carbon Contents in Croplands of Bavaria from SCMaP Soil Reflectance Composites

For food security issues or global climate change, there is a growing need for large-scale knowledge of soil organic carbon (SOC) contents in agricultural soils. To capture and quantify SOC contents at a field scale, Earth Observation (EO) can be a valuable data source for area-wide mapping. The extraction of exposed soils from EO data is challenging due to temporal or permanent vegetation cover, the influence of soil moisture or the condition of the soil surface. Compositing techniques of multitemporal satellite images provide an alternative to retrieve exposed soils and to produce a data source. The repeatable soil composites, containing averaged exposed soil areas over several years, are relatively independent from seasonal soil moisture and surface conditions and provide a new EO-based data source that can be used to estimate SOC contents over large geographical areas with a high spatial resolution. Here, we applied the Soil Composite Mapping Processor (SCMaP) to the Landsat archive between 1984 and 2014 of images covering Bavaria, Germany. Compared to existing SOC modeling approaches based on single scenes, the 30-year SCMaP soil reflectance composite (SRC) with a spatial resolution of 30 m is used. The SRC spectral information is correlated with point soil data using different machine learning algorithms to estimate the SOC contents in cropland topsoils of Bavaria. We developed a pre-processing technique to address the issue of combining point information with EO pixels for the purpose of modeling. We applied different modeling methods often used in EO soil studies to choose the best SOC prediction model. Based on the model accuracies and performances, the Random Forest (RF) showed the best capabilities to predict the SOC contents in Bavaria (R² = 0.67, RMSE = 1.24%, RPD = 1.77, CCC = 0.78). We further validated the model results with an independent dataset. The comparison between the measured and predicted SOC contents showed a mean difference of 0.11% SOC using the best RF model. The SCMaP SRC is a promising approach to predict the spatial SOC distribution over large geographical extents with a high spatial resolution (30 m).

Sentinel-2 Exposed Soil Composite for Soil Organic Carbon Prediction

Pilot studies have demonstrated the potential of remote sensing for soil organic carbon (SOC) mapping in exposed croplands. However, the use of remote sensing for SOC prediction is often hindered by disturbing factors at the soil surface, such as photosynthetic active and non-photosynthetic active vegetation, variation in soil moisture or surface roughness. With the increasing amount of freely available satellite data, recent studies have focused on stabilizing the soil reflectance by building image composites. These composites tend to minimize the disturbing effects by applying sets of criteria. Here, we aim to develop a robust method that allows selecting Sentinel-2 (S-2) pixels with minimal influence of the following disturbing factors: crop residues, surface roughness and soil moisture. We selected all S-2 cloud-free images covering the Belgian Loam Belt from January 2019 to December 2020 (in total 36 images). We then built nine exposed soil composites based on four sets of criteria: (1) lowest Normalized Burn Ratio (NBR2), (2) Normalized Difference Vegetation Index (NDVI) < 0.25, (3–5) NDVI < 0.25 and NBR2 < threshold, (6) the ‘greening-up’ period of a crop and (7–9) the ‘greening-up’ period of a crop and NBR2 < threshold. The ‘greening-up’ period was selected based on the NDVI timeline, where ‘greening-up’ is considered as the last date of acquisition where the soil is exposed (NDVI < 0.25) before the crop develops (NDVI > 0.25). We then built a partial least square regression (PLSR) model with 10-fold cross-validation to estimate the SOC content based on 137 georeferenced calibration samples on the nine composites. We obtained non-satisfactory results (R2 < 0.30, RMSE > 2.50 g C kg–1, and RPD < 1.4, n > 68) for all composites except for the composite in the ‘greening-up’ stage with a NBR2 < 0.07 (R2 = 0.54 ± 0.12, RPD = 1.68 ± 0.45 and RMSE = 2.09 ± 0.39 g C kg–1, n = 49). Hence, the ‘greening-up’ method combined with a strict NBR2 threshold allows selecting the purest exposed soil pixels suitable for SOC prediction. The limit of this method might be its coverage of the total cropland area, which in a two-year period reached 62%, compared to 95% coverage if only the NDVI threshold is applied.

Potential of Sentinel-2 Satellite Images for Monitoring Green Waste Compost and Manure Amendments in Temperate Cropland

Increasing attention has been placed on the agroecological impact of applying exogenous organic matter (EOM) amendments, such as green waste compost (GWC) and livestock manure, to agricultural landscapes. However, monitoring the frequency and locality of this practice poses a major challenge, as these events are typically unreported. The purpose of this study is to evaluate the utility of Sentinel-2 imagery for the detection of EOM amendments. Specifically, we investigated the spectral shift resulting from GWC and manure application at two spatial scales, satellite and proximal. At the satellite scale, multispectral Sentinel-2 image pairs were analyzed before and after EOM application to six cultivated fields in the Versailles Plain, France. At the proximal scale, multi-temporal spectral field measurements were taken of experimental plots consisting of 14 total treatments: EOM variety, amendment quantity (15, 30 and 60 t.ha−1) and tillage. The Sentinel-2 images showed significant spectral differences before and after EOM application. Exogenous Organic Matter Indices (EOMI) were developed and analyzed for separative performance. The best performing index was EOMI2, using the B4 and B12 Sentinel-2 spectral bands. At the proximal scale, simulated Sentinel-2 reflectance spectra, which were created using field measurements, successfully monitored all EOM treatments for three days, except for the buried green waste compost at a rate of 15 t.ha−1.

Prediction of Soil Organic Carbon under Different Land Use Types Using Sentinel-1/-2 Data in a Small Watershed

Soil organic carbon (SOC) is a key property for evaluating soil quality. SOC is thus an important parameter of agricultural soils and needs to be regularly monitored. The aim of this study is to explore the potential of synthetic aperture radar (SAR) satellite imagery (Sentinel-1), optical satellite imagery (Sentinel-2), and digital elevation model (DEM) data to estimate the SOC content under different land use types. The extreme gradient boosting (XGboost) algorithm was used to predict the SOC content and evaluate the importance of feature variables under different land use types. For this purpose, 290 topsoil samples were collected and 49 features were derived from remote sensing images and DEM. Feature selection was carried out to prevent data redundancy. Coefficient of determination (R2), mean absolute error (MAE), mean squared error (MSE), percent root mean squared error (%RMSE), ratio of performance to interquartile range (RPIQ), and corrected akaike information criterion (AICc) were employed for evaluating model performance. The results showed that Sentinel-1 and Sentinel-2 data were both important for the prediction of SOC and the prediction accuracy of the model differed with land use types. Among them, the prediction accuracy of this model is the best for orchard (R2 = 0.86 and MSE = 0.004%), good for dry land (R2 = 0.74 and MSE = 0.008%) and paddy field (R2 = 0.66 and MSE = 0.009%). The prediction model of SOC content is effective and can provide support for the application of remote sensing data to soil property monitoring.

Evaluation of Near Infrared Spectroscopy (NIRS) for Estimating Soil Organic Matter and Phosphorus in Mediterranean Montado Ecosystem

The Montado is an agro-silvo-pastoral ecosystem characteristic of the Mediterranean region. Pasture productivity and, consequently, the possibilities for intensifying livestock production depend on soil fertility. Soil organic matter (SOM) and phosphorus (P2O5) are two indicators of the evolution of soil fertility in this ecosystem. However, their conventional analytical determination by reference laboratory methods is costly, time consuming, and laborious and, thus, does not meet the needs of current production systems. The aim of this study was to evaluate an alternative approach to estimate SOM and soil P2O5 based on near infrared spectroscopy (NIRS) combined with multivariate data analysis. For this purpose, 242 topsoil samples were collected in 2019 in eleven fields. These samples were subjected to reference laboratory analysis and NIRS analysis. For NIRS, 165 samples were used during the calibration phase and 77 samples were used during the external validation phase. The results of this study showed significant correlation between NIRS calibration models and reference methods for quantification of these soil parameters. The coefficient of determination (R2, 0.85 for SOM and 0.76 for P2O5) and the residual predictive deviation (RPD, 2.7 for SOM and 2.2 for P2O5) obtained in external validation indicated the potential of NIRS to estimate SOM and P2O5, which can facilitate farm managers’ decision making in terms of dynamic management of animal grazing and differential fertilizer application.

Mapping Soil Organic Carbon for Airborne and Simulated EnMAP Imagery Using the LUCAS Soil Database and a Local PLSR

Soil degradation is a major threat for European soils and therefore, the European Commission recommends intensifying research on soil monitoring to capture changes over time and space. Imaging spectroscopy is a promising technique to create spatially accurate topsoil maps based on hyperspectral remote sensing data. We tested the application of a local partial least squares regression (PLSR) to airborne HySpex and simulated satellite EnMAP (Environmental Mapping and Analysis Program) data acquired in north-eastern Germany to quantify the soil organic carbon (SOC) content. The approach consists of two steps: (i) the local PLSR uses the European LUCAS (land use/cover area frame statistical survey) Soil database to quantify the SOC content for soil samples from the study site in order to avoid the need for wet chemistry analyses, and subsequently (ii) a remote sensing model is calibrated based on the local PLSR SOC results and the corresponding image spectra. This two-step approach is compared to a traditional PLSR approach using measured SOC contents from local samples. The prediction accuracy is high for the laboratory model in the first step with R2 = 0.86 and RPD = 2.77. The HySpex airborne prediction accuracy of the traditional approach is high and slightly superior to the two-step approach (traditional: R2 = 0.78, RPD = 2.19; two-step: R2 = 0.67, RPD = 1.79). Applying the two-step approach to simulated EnMAP imagery leads to a lower but still reasonable prediction accuracy (traditional: R2 = 0.77, RPD = 2.15; two-step: R2 = 0.48, RPD = 1.41). The two-step models of both sensors were applied to all bare soils of the respective images to produce SOC maps. This local PLSR approach, based on large scale soil spectral libraries, demonstrates an alternative to SOC measurements from wet chemistry of local soil samples. It could allow for repeated inexpensive SOC mapping based on satellite remote sensing data as long as spectral measurements of a few local samples are available for model calibration.

High-resolution digital mapping of soil organic carbon and soil total nitrogen using DEM derivatives, Sentinel-1 and Sentinel-2 data based on machine learning algorithms

Soil organic carbon (SOC) and soil total nitrogen (STN) are important indicators of soil health and play a key role in the global carbon and nitrogen cycles. High-resolution radar Sentinel-1 and multispectral Sentinel-2 images have the potential to investigate soil spatial distribution information over a large area, although Sentinel-1 and Sentinel-2 data have rarely been combined to map either SOC or STN content. In this study, we applied machine learning techniques to map both SOC and STN content in the southern part of Central Europe using digital elevation model (DEM) derivatives, multi-temporal Sentinel-1 and Sentinel-2 data, and evaluated the potential of different remote sensing sensors (Sentinel-1 and Sentinel-2) to predict SOC and STN content. Four machine-learners including random forest (RF), boosted regression trees (BRT), support vector machine (SVM) and Bagged CART were used to construct predictive models of SOC and STN contents based on 179 soil samples and different combinations of environmental covariates. The performance of these models was evaluated based on a 10-fold cross-validation method by three statistical indicators. Overall, the BRT model performed better than RF, SVM and Bagged CART, and these models yielded similar spatial distribution patterns of SOC and STN. Our results showed that multi-source sensor methods provided more accurate predictions of SOC and STN contents than individual sensors. The application of radar Sentinel-1 and multispectral Sentinel-2 images proved useful for predicting SOC and STN. A combination of Sentinel-1/2-derived predictors and DEM derivatives yielded the highest prediction accuracy. The prediction accuracy changed with and without the Sentinel-1/2-derived predictors, with the R² for estimating both SOC and STN content using the BRT model increasing by 12.8% and 18.8%, respectively. Topographic variables were the main explanatory variables for SOC and STN predictions, where elevation was assigned as the variable with the most importance by the models. The results of this study illustrate the potential of free high-resolution radar Sentinel-1 and multispectral Sentinel-2 data as input when developing SOC and STN prediction models.

Soil Organic Carbon Mapping from Remote Sensing: The Effect of Crop Residues

Since the onset of agriculture, soils have lost their organic carbon to such an extent that the soil functions of many croplands are threatened. Hence, there is a strong demand for mapping and monitoring critical soil properties and in particular soil organic carbon (SOC). Pilot studies have demonstrated the potential for remote sensing techniques for SOC mapping in croplands. It has, however, been shown that the assessment of SOC may be hampered by the condition of the soil surface. While growing vegetation can be readily detected by means of the well-known Normalized Difference Vegetation Index (NDVI), the distinction between bare soil and crop residues is expressed in the shortwave infrared region (SWIR), which is only covered by two broad bands in Landsat or Sentinel-2 imagery. Here we tested the effect of thresholds for the Cellulose Absorption Index (CAI), on the performance of SOC prediction models for cropland soils. Airborne Prism Experiment (APEX) hyperspectral images covering an area of 240 km2 in the Belgian Loam Belt were used together with a local soil dataset. We used the partial least square regression (PLSR) model to estimate the SOC content based on 104 georeferenced calibration samples (NDVI < 0.26), firstly without setting a CAI threshold, and obtained a satisfactory result (coefficient of determination (R2) = 0.49, Ratio of Performance to Deviation (RPD) = 1.4 and Root Mean Square Error (RMSE) = 2.13 g kgC−1 for cross-validation). However, a cross comparison of the estimated SOC values to grid-based measurements of SOC content within three fields revealed a systematic overestimation for fields with high residue cover. We then tested different CAI thresholds in order to mask pixels with high residue cover. The best model was obtained for a CAI threshold of 0.75 (R2 = 0.59, RPD = 1.5 and RMSE = 1.75 g kgC−1 for cross-validation). These results reveal that the purity of the pixels needs to be assessed aforehand in order to produce reliable SOC maps. The Normalized Burn Ratio (NBR2) index based on the SWIR bands of the MSI Sentinel 2 sensor extracted from images collected nine days before the APEX flight campaign correlates well with the CAI index of the APEX imagery. However, the NBR2 index calculated from Sentinel 2 images under moist conditions is poorly correlated with residue cover. This can be explained by the sensitivity of the NBR2 index to both soil moisture and residues.

Remote Sensing for Precision Agriculture: Sentinel-2 Improved Features and Applications

The use of satellites to monitor crops and support their management is gathering increasing attention. The improved temporal, spatial, and spectral resolution of the European Space Agency (ESA) launched Sentinel-2 A + B twin platform is paving the way to their popularization in precision agriculture. Besides the Sentinel-2 A + B constellation technical features the open-access nature of the information they generate, and the available support software are a significant improvement for agricultural monitoring. This paper was motivated by the challenges faced by researchers and agrarian institutions entering this field; it aims to frame remote sensing principles and Sentinel-2 applications in agriculture. Thus, we reviewed the features and uses of Sentinel-2 in precision agriculture, including abiotic and biotic stress detection, and agricultural management. We also compared the panoply of satellites currently in use for land remote sensing that are relevant for agriculture to the Sentinel-2 A + B constellation features. Contrasted with previous satellite image systems, the Sentinel-2 A + B twin platform has dramatically increased the capabilities for agricultural monitoring and crop management worldwide. Regarding crop stress monitoring, Sentinel-2 capacities for abiotic and biotic stresses detection represent a great step forward in many ways though not without its limitations; therefore, combinations of field data and different remote sensing techniques may still be needed. We conclude that Sentinel-2 has a wide range of useful applications in agriculture, yet still with room for further improvements. Current and future ways that Sentinel-2 can be utilized are also discussed.

Employing a Multi-Input Deep Convolutional Neural Network to Derive Soil Clay Content from a Synergy of Multi-Temporal Optical and Radar Imagery Data

Earth observation (EO) has an immense potential as being an enabling tool for mapping spatial characteristics of the topsoil layer. Recently, deep learning based algorithms and cloud computing infrastructure have become available with a great potential to revolutionize the processing of EO data. This paper aims to present a novel EO-based soil monitoring approach leveraging open-access Copernicus Sentinel data and Google Earth Engine platform. Building on key results from existing data mining approaches to extract bare soil reflectance values the current study delivers valuable insights on the synergistic use of open access optical and radar images. The proposed framework is driven by the need to eliminate the influence of ambient factors and evaluate the efficiency of a convolutional neural network (CNN) to effectively combine the complimentary information contained in the pool of both optical and radar spectral information and those form auxiliary geographical coordinates mainly for soil. We developed and calibrated our multi-input CNN model based on soil samples (calibration = 80% and validation 20%) of the LUCAS database and then applied this approach to predict soil clay content. A promising prediction performance (R2 = 0.60, ratio of performance to the interquartile range (RPIQ) = 2.02, n = 6136) was achieved by the inclusion of both types (synthetic aperture radar (SAR) and laboratory visible near infrared–short wave infrared (VNIR-SWIR) multispectral) of observations using the CNN model, demonstrating an improvement of more than 5.5% in RMSE using the multi-year median optical composite and current state-of-the-art non linear machine learning methods such as random forest (RF; R2 = 0.55, RPIQ = 1.91, n = 6136) and artificial neural network (ANN; R2 = 0.44, RPIQ = 1.71, n = 6136). Moreover, we examined post-hoc techniques to interpret the CNN model and thus acquire an understanding of the relationships between spectral information and the soil target identified by the model. Looking to the future, the proposed approach can be adopted on the forthcoming hyperspectral orbital sensors to expand the current capabilities of the EO component by estimating more soil attributes with higher predictive performance.

Multispectral Models from Bare Soil Composites for Mapping Topsoil Properties over Europe

Reflectance of light across the visible, near-infrared and shortwave infrared (VIS-NIR-SWIR, 0.4–2.5 µm) spectral region is very useful for investigating mineralogical, physical and chemical properties of soils, which can reduce the need for traditional wet chemistry analyses. As many collections of multispectral satellite data are available for environmental studies, a large extent with medium resolution mapping could be benefited from the spectral measurements made from remote sensors. In this paper, we explored the use of bare soil composites generated from the large historical collections of Landsat images for mapping cropland topsoil attributes across the European extent. For this task, we used the Geospatial Soil Sensing System (GEOS3) for generating two bare soil composites of 30 m resolution (named synthetic soil images, SYSI), which were employed to represent the median topsoil reflectance of bare fields. The first (framed SYSI) was made with multitemporal images (2006–2012) framed to the survey time of the Land-Use/Land-Cover Area Frame Survey (LUCAS) soil dataset (2009), seeking to be more compatible to the soil condition upon the sampling campaign. The second (full SYSI) was generated from the full collection of Landsat images (1982–2018), which although displaced to the field survey, yields a higher proportion of bare areas for soil mapping. For evaluating the two SYSIs, we used the laboratory spectral data as a reference of topsoil reflectance to calculate the Spearman correlation coefficient. Furthermore, both SYSIs employed machine learning for calibrating prediction models of clay, sand, soil organic carbon (SOC), calcium carbonates (CaCO3), cation exchange capacity (CEC), and pH determined in water, using the gradient boosting regression algorithm. The original LUCAS laboratory spectra and a version of the data resampled to the Landsat multispectral bands were also used as reference of prediction performance using VIS-NIR-SWIR multispectral data. Our results suggest that generating a bare soil composite displaced to the survey time of soil observations did not improve the quality of topsoil reflectance, and consequently, the prediction performance of soil attributes. Despite the lower spectral resolution and the variability of soils in Europe, a SYSI calculated from the full collection of Landsat images can be employed for topsoil prediction of clay and CaCO3 contents with a moderate performance (testing R2, root mean square error (RMSE) and ratio of performance to interquartile range (RPIQ) of 0.44, 9.59, 1.77, and 0.36, 13.99, 1.54, respectively). Thus, this study shows that although there exist some constraints due to the spatial and temporal variation of soil exposures and among the Landsat sensors, it is possible to use bare soil composites for mapping key soil attributes of croplands across the European extent.

Soil Organic Carbon Mapping Using Multispectral Remote Sensing Data: Prediction Ability of Data with Different Spatial and Spectral Resolutions

The image spectral data, particularly hyperspectral data, has been proven as an efficient data source for mapping of the spatial variability of soil organic carbon (SOC). Multispectral satellite data are readily available and cost-effective sources of spectral data compared to costly and technically demanding processing of hyperspectral data. Moreover, their continuous acquisition allows to develop a composite from time-series, increasing the spatial coverage of SOC maps. In this study, an evaluation of the prediction ability of models assessing SOC using real multispectral remote sensing data from different platforms was performed. The study was conducted on a study plot (1.45 km2) in the Chernozem region of South Moravia (Czechia). The adopted methods included field sampling and predictive modeling using satellite multispectral Sentinel-2, Landsat-8, and PlanetScope data, and multispectral UAS Parrot Sequoia data. Furthermore, the performance of a soil reflectance composite image from Sentinel-2 data was analyzed. Aerial hyperspectral CASI 1500 and SASI 600 data was used as a reference. Random forest, support vector machine, and the cubist regression technique were applied in the predictive modeling. The prediction accuracy of models using multispectral data, including Sentinel-2 composite, was lower (RPD range from 1.16 to 1.65; RPIQ range from 1.53 to 2.17) compared to the reference model using hyperspectral data (RPD = 2.26; RPIQ = 3.34). The obtained results show very similar prediction accuracy for all spaceborne sensors (Sentinel-2, Landsat-8, and PlanetScope). However, the spatial correlation between the reference mapping results obtained from the hyperspectral data and other maps using multispectral data was moderately strong. UAS sensors and freely available satellite multispectral data can represent an alternative cost-effective data source for remote SOC mapping on the local scale.