Adjustment of Sentinel-2 Multi-Spectral Instrument (MSI) Red-Edge Band Reflectance to Nadir BRDF Adjusted Reflectance (NBAR) and Quantification of Red-Edge Band BRDF Effects

A standardized catalogue of spectral indices to advance the use of remote sensing in Earth system research

Spectral Indices derived from multispectral remote sensing products are extensively used to monitor Earth system dynamics (e.g. vegetation dynamics, water bodies, fire regimes). The rapid increase of proposed spectral indices led to a high demand for catalogues of spectral indices and tools for their computation. However, most of these resources are either closed-source, outdated, unconnected to a catalogue or lacking a common Application Programming Interface (API). Here we present "Awesome Spectral Indices" (ASI), a standardized catalogue of spectral indices for Earth system research. ASI provides a comprehensive machine readable catalogue of spectral indices, which is linked to a Python library. ASI delivers a broad set of attributes for each spectral index, including names, formulas, and source references. The catalogue can be extended by the user community, ensuring that ASI remains current and enabling a wider range of scientific applications. Furthermore, the Python library enables the application of the catalogue to real-world data and thereby facilitates the efficient use of remote sensing resources in multiple Earth system domains.

The Classification of Reflectance Anisotropy and Its Application in Albedo Retrieval

The land surface albedo reflects the ability of the surface to reflect solar radiation and is a critical physical variable in the study of the Earth’s energy budget and global climate change. Algorithms for the retrieval of albedo usually require multi-angle measurements due to surface anisotropy. However, most of the satellites cannot currently provide sufficient and well-distributed observations; therefore, the accuracy of remotely sensed albedo is constrained. Based on the Moderate-Resolution Imaging Spectroradiometer (MODIS) Bidirectional Reflectance Distribution Function (BRDF) and albedo product (MCD43A1), this study proposed a method to further subdivide reflectance anisotropy and build an updated database of BRDF archetype, using both the Anisotropic Flat Index (AFX) and Perpendicular Anisotropic Flat Index (PAFX). The BRDF archetypes were used to fit the corresponding MODIS BRDF, and the optimal number of BRDF archetype categories was determined according to the tendency of fitting error. The effect of surface anisotropy and observation noise on albedo retrieval were explored based on simulated MODIS reflectance. Finally, the BRDF archetype A2P2 was taken as prior knowledge to retrieve albedo from a different number of MODIS observations, and the result was validated by the high-quality MODIS albedo product. The results show that the fitting error between BRDF archetypes and MODIS BRDF shows a rapid decline when introducing the PAFX in the classification process. A 3-by-3 matrix of BRDF archetypes, which occupy 73.44% and 70.13% of the total decline in the red and NIR band, can be used to represent the characteristics of reflectance anisotropy. The archetype A2P2 may be used as prior knowledge to improve the albedo retrieval from insufficient observations. The validation results based on MODIS observations show that the archetype A2P2-based albedo can reach root-mean-square errors (RMSEs) of no more than 0.03.

Harmonization of Multi-Mission High-Resolution Time Series: Application to BELAIR

High-resolution data are increasingly used for various applications, yet the revisit time is still low for some applications, particularly in frequently cloud-covered areas. Therefore, sensors are often combined, which raises issues on data consistency. In this study, we start from L1 to L3 data, and investigate the impact of harmonization measures, correcting for difference in radiometric gain and spectral response function (SRF), and the use of a common processing chain with the same atmospheric correction for Sentinel-2A/B, Landsat-8, DEIMOS-1, and Proba-V center cameras. These harmonization measures are evaluated step-wise in two applications: (1) agricultural monitoring, and (2) hydrological modelling in an urban context, using biophysical parameters and NDVI. The evaluation includes validation with in situ data, relative consistency analysis between different sensors, and the evaluation of the time series noise. A higher accuracy was not obtained when validating against in situ data. Yet, the relative analysis and the time series noise analysis clearly demonstrated that the largest improvement in consistency between sensors was obtained when applying the same atmospheric correction to all sensors. The gain correction obtained and its impact on the results was small, indicating that the sensors were already well calibrated. We could not demonstrate an improved consistency after SRF correction. It is likely that other factors, such as anisotropy effects, play a larger role, requiring further research.

Prediction and Uncertainty Capabilities of Quantile Regression Forests in Estimating Spatial Distribution of Soil Organic Matter

One of the core tasks in digital soil mapping (DSM) studies is the estimation of the spatial distribution of different soil variables. In addition, however, assessing the uncertainty of these estimations is equally important, something that a lot of current DSM studies lack. Machine learning (ML) methods are increasingly used in this scientific field, the majority of which do not have intrinsic uncertainty estimation capabilities. A solution to this is the use of specific ML methods that provide advanced prediction capabilities, along with innate uncertainty estimation metrics, like Quantile Regression Forests (QRF). In the current paper, the prediction and the uncertainty capabilities of QRF, Random Forests (RF) and geostatistical methods were assessed. It was confirmed that QRF exhibited outstanding results at predicting soil organic matter (OM) in the study area. In particular, R2 was much higher than the geostatistical methods, signifying that more variation is explained by the specific model. Moreover, its uncertainty capabilities as presented in the uncertainty maps, shows that it can also provide a good estimation of the uncertainty with distinct representation of the local variation in specific parts of the area, something that is considered a significant advantage, especially for decision support purposes.

Drone-Based Hyperspectral and Thermal Imagery for Quantifying Upland Rice Productivity and Water Use Efficiency after Biochar Application

Miniature hyperspectral and thermal cameras onboard lightweight unmanned aerial vehicles (UAV) bring new opportunities for monitoring land surface variables at unprecedented fine spatial resolution with acceptable accuracy. This research applies hyperspectral and thermal imagery from a drone to quantify upland rice productivity and water use efficiency (WUE) after biochar application in Costa Rica. The field flights were conducted over two experimental groups with bamboo biochar (BC1) and sugarcane biochar (BC2) amendments and one control (C) group without biochar application. Rice canopy biophysical variables were estimated by inverting a canopy radiative transfer model on hyperspectral reflectance. Variations in gross primary productivity (GPP) and WUE across treatments were estimated using light-use efficiency and WUE models respectively from the normalized difference vegetation index (NDVI), canopy chlorophyll content (CCC), and evapotranspiration rate. We found that GPP was increased by 41.9 ± 3.4% in BC1 and 17.5 ± 3.4% in BC2 versus C, which may be explained by higher soil moisture after biochar application, and consequently significantly higher WUEs by 40.8 ± 3.5% in BC1 and 13.4 ± 3.5% in BC2 compared to C. This study demonstrated the use of hyperspectral and thermal imagery from a drone to quantify biochar effects on dry cropland by integrating ground measurements and physical models.

Modeling Bidirectional Polarization Distribution Function of Land Surfaces Using Machine Learning Techniques

Accurate estimation of polarized reflectance (Rp) of land surfaces is critical for remote sensing of aerosol optical properties. In the last two decades, many data-driven bidirectional polarization distribution function (BPDF) models have been proposed for accurate estimation of Rp, among which the generalized regression neural network (GRNN) based BPDF model has been reported to perform the best. GRNN is just a simple machine learning (ML) technique that can solve non-linear problems. Many ML techniques were reported to work well in solving non-linear problems and consequently may provide better performance in BPDF modeling. However, incorporating various ML techniques with BPDF modeling and comparing their performances have never been well documented. In this study, three widely used ML algorithms—i.e., support vector regression (SVR), K-nearest-neighbor (KNN), and random forest (RF)—were applied for BPDF modeling. Using measurements collected by the Polarization and Directionality of the Earth’s Reflectance onboard PARASOL satellite (POLDER/PARASOL), non-linear relationships between Rp and the input variables, i.e., Fresnel factor (Fp), scattering angle (SA), reflectance at 670 nm (R670) and 865 nm (R865), were built using these ML algorithms. Results showed that taking Fp, SA, R670, and R865 as input variables, the performance of the four ML-based BPDF models was quite similar. The KNN-based BPDF model provided slightly better results, and improved the accuracy of the semi-empirical BPDF models by 9.55% in terms of the overall root mean square error (RMSE). Experiments of different configuration of input variables suggested that using multi-band reflectance as input variables provided better results than using vegetation indices. The RF-based BPDF model using all reflectances at six bands as input variables produced the best results, improving the overall accuracy by 6.62% compared with the GRNN-based BPDF model. Among all the input variables, reflectance at absorbing spectral bands—e.g., 490 nm and 670 nm—played more significant roles in RF-based BPDF modeling due to the domination of polarized partition in total reflectance. Fresnel factor and scattering angle were also important for BPDF modeling. This study confirmed the feasibility of applying ML techniques to more accurate BPDF modeling, and the RF-based BPDF model proposed in this study can be used to increase the accuracy of remote sensing of the complete aerosol properties.

Monitoring Invasion Process of Spartina alterniflora by Seasonal Sentinel-2 Imagery and an Object-Based Random Forest Classification

In the late 1990s, the exotic plant Spartina alterniflora (S. alterniflora), was introduced to the Zhangjiang Estuary of China for tidal zone reclamation and protection. However, it invaded rapidly and has caused serious ecological problems. Accurate information on the seasonal invasion of S. alterniflora is vital to understand invasion pattern and mechanism, especially at a high temporal resolution. This study aimed to explore the S. alterniflora invasion process at a seasonal scale from 2016 to 2018. However, due to the uncertainties caused by periodic inundation of local tides, accurately monitoring the spatial extent of S. alterniflora is challenging. Thus, to achieve the goal and address the challenge, we firstly built a high-quality seasonal Sentinel-2 image collection by developing a new submerged S. alterniflora index (SAI) to reduce the errors caused by high tide fluctuations. Then, an object-based random forest (RF) classification method was applied to the image collection. Finally, seasonal extents of S. alterniflora were captured. Results showed that (1) the red edge bands (bands 5, 6, and 7) of Sentinel-2 imagery played critical roles in delineating submerged S. alterniflora; (2) during March 2016 to November 2018, the extent of S. alterniflora increased from 151.7 to 270.3 ha, with an annual invasion rate of 39.5 ha; (3) S. alterniflora invaded with a rate of 31.5 ha/season during growing season and 12.1 ha/season during dormant season. To our knowledge, this is the first study monitoring S. alterniflora invasion process at a seasonal scale during continuous years, discovering that S. alterniflora also expands during dormant seasons. This discovery is of great significance for understanding the invasion pattern and mechanism of S. alterniflora and will facilitate coastal biodiversity conservation efforts.

Land Use/Land Cover Mapping Using Multitemporal Sentinel-2 Imagery and Four Classification Methods—A Case Study from Dak Nong, Vietnam

Information on land use and land cover (LULC) including forest cover is important for the development of strategies for land planning and management. Satellite remotely sensed data of varying resolutions have been an unmatched source of such information that can be used to produce estimates with a greater degree of confidence than traditional inventory estimates. However, use of these data has always been a challenge in tropical regions owing to the complexity of the biophysical environment, clouds, and haze, and atmospheric moisture content, all of which impede accurate LULC classification. We tested a parametric classifier (logistic regression) and three non-parametric machine learning classifiers (improved k-nearest neighbors, random forests, and support vector machine) for classification of multi-temporal Sentinel 2 satellite imagery into LULC categories in Dak Nong province, Vietnam. A total of 446 images, 235 from the year 2017 and 211 from the year 2018, were pre-processed to gain high quality images for mapping LULC in the 6516 km2 study area. The Sentinel 2 images were tested and classified separately for four temporal periods: (i) dry season, (ii) rainy season, (iii) the entirety of the year 2017, and (iv) the combination of dry and rainy seasons. Eleven different LULC classes were discriminated of which five were forest classes. For each combination of temporal image set and classifier, a confusion matrix was constructed using independent reference data and pixel classifications, and the area on the ground of each class was estimated. For overall temporal periods and classifiers, overall accuracy ranged from 63.9% to 80.3%, and the Kappa coefficient ranged from 0.611 to 0.813. Area estimates for individual classes ranged from 70 km2 (1% of the study area) to 2200 km2 (34% of the study area) with greater uncertainties for smaller classes.

Comparing Spectral Characteristics of Landsat-8 and Sentinel-2 Same-Day Data for Arctic-Boreal Regions

The Arctic-Boreal regions experience strong changes of air temperature and precipitation regimes, which affect the thermal state of the permafrost. This results in widespread permafrost-thaw disturbances, some unfolding slowly and over long periods, others occurring rapidly and abruptly. Despite optical remote sensing offering a variety of techniques to assess and monitor landscape changes, a persistent cloud cover decreases the amount of usable images considerably. However, combining data from multiple platforms promises to increase the number of images drastically. We therefore assess the comparability of Landsat-8 and Sentinel-2 imagery and the possibility to use both Landsat and Sentinel-2 images together in time series analyses, achieving a temporally-dense data coverage in Arctic-Boreal regions. We determined overlapping same-day acquisitions of Landsat-8 and Sentinel-2 images for three representative study sites in Eastern Siberia. We then compared the Landsat-8 and Sentinel-2 pixel-pairs, downscaled to 60 m, of corresponding bands and derived the ordinary least squares regression for every band combination. The acquired coefficients were used for spectral bandpass adjustment between the two sensors. The spectral band comparisons showed an overall good fit between Landsat-8 and Sentinel-2 images already. The ordinary least squares regression analyses underline the generally good spectral fit with intercept values between 0.0031 and 0.056 and slope values between 0.531 and 0.877. A spectral comparison after spectral bandpass adjustment of Sentinel-2 values to Landsat-8 shows a nearly perfect alignment between the same-day images. The spectral band adjustment succeeds in adjusting Sentinel-2 spectral values to Landsat-8 very well in Eastern Siberian Arctic-Boreal landscapes. After spectral adjustment, Landsat and Sentinel-2 data can be used to create temporally-dense time series and be applied to assess permafrost landscape changes in Eastern Siberia. Remaining differences between the sensors can be attributed to several factors including heterogeneous terrain, poor cloud and cloud shadow masking, and mixed pixels.

Developing Transformation Functions for VENμS and Sentinel-2 Surface Reflectance over Israel

Vegetation and Environmental New micro Spacecraft (VENμS) and Sentinel-2 are both ongoing earth observation missions that provide high-resolution multispectral imagery at 10 m (VENμS) and 10–20 m (Sentinel-2), at relatively high revisit frequencies (two days for VENμS and five days for Sentinel-2). Sentinel-2 provides global coverage, whereas VENμS covers selected regions, including parts of Israel. To facilitate the combination of these sensors into a unified time-series, a transformation model between them was developed using imagery from the region of interest. For this purpose, same-day acquisitions from both sensor types covering the surface reflectance over Israel, between April 2018 and November 2018, were used in this study. Transformation coefficients from VENμS to Sentinel-2 surface reflectance were produced for their overlapping spectral bands (i.e., visible, red-edge and near-infrared). The performance of these spectral transformation functions was assessed using several methods, including orthogonal distance regression (ODR), the mean absolute difference (MAD), and spectral angle mapper (SAM). Post-transformation, the value of the ODR slopes were close to unity for the transformed VENμS reflectance with Sentinel-2 reflectance, which indicates near-identity of the two datasets following the removal of systemic bias. In addition, the transformation outputs showed better spectral similarity compared to the original images, as indicated by the decrease in SAM from 0.093 to 0.071. Similarly, the MAD was reduced post-transformation in all bands (e.g., the blue band MAD decreased from 0.0238 to 0.0186, and in the NIR it decreased from 0.0491 to 0.0386). Thus, the model helps to combine the images from Sentinel-2 and VENμS into one time-series that facilitates continuous, temporally dense vegetation monitoring.

FORCE—Landsat + Sentinel-2 Analysis Ready Data and Beyond

Ever increasing data volumes of satellite constellations call for multi-sensor analysis ready data (ARD) that relieve users from the burden of all costly preprocessing steps. This paper describes the scientific software FORCE (Framework for Operational Radiometric Correction for Environmental monitoring), an ‘all-in-one’ solution for the mass-processing and analysis of Landsat and Sentinel-2 image archives. FORCE is increasingly used to support a wide range of scientific to operational applications that are in need of both large area, as well as deep and dense temporal information. FORCE is capable of generating Level 2 ARD, and higher-level products. Level 2 processing is comprised of state-of-the-art cloud masking and radiometric correction (including corrections that go beyond ARD specification, e.g., topographic or bidirectional reflectance distribution function correction). It further includes data cubing, i.e., spatial reorganization of the data into a non-overlapping grid system for enhanced efficiency and simplicity of ARD usage. However, the usage barrier of Level 2 ARD is still high due to the considerable data volume and spatial incompleteness of valid observations (e.g., clouds). Thus, the higher-level modules temporally condense multi-temporal ARD into manageable amounts of spatially seamless data. For data mining purposes, per-pixel statistics of clear sky data availability can be generated. FORCE provides functionality for compiling best-available-pixel composites and spectral temporal metrics, which both utilize all available observations within a defined temporal window using selection and statistical aggregation techniques, respectively. These products are immediately fit for common Earth observation analysis workflows, such as machine learning-based image classification, and are thus referred to as highly analysis ready data (hARD). FORCE provides data fusion functionality to improve the spatial resolution of (i) coarse continuous fields like land surface phenology and (ii) Landsat ARD using Sentinel-2 ARD as prediction targets. Quality controlled time series preparation and analysis functionality with a number of aggregation and interpolation techniques, land surface phenology retrieval, and change and trend analyses are provided. Outputs of this module can be directly ingested into a geographic information system (GIS) to fuel research questions without any further processing, i.e., hARD+. FORCE is open source software under the terms of the GNU General Public License v. >= 3, and can be downloaded from http://force.feut.de.

Mapping Plantations in Myanmar by Fusing Landsat-8, Sentinel-2 and Sentinel-1 Data along with Systematic Error Quantification

Forests in Southeast Asia are experiencing some of the highest rates of deforestation and degradation in the world, with natural forest species being replaced by cropland and plantation monoculture. In this work, we have developed an innovative method to accurately map rubber and palm oil plantations using fusion of Landsat-8, Sentinel 1 and 2. We applied cloud and shadow masking, bidirectional reflectance distribution function (BRDF), atmospheric and topographic corrections to the optical imagery and a speckle filter and harmonics for Synthetic Aperture Radar (SAR) data. In this workflow, we created yearly composites for all sensors and combined the data into a single composite. A series of covariates were calculated from optical bands and sampled using reference data of the land cover classes including surface water, forest, urban and built-up, cropland, rubber, palm oil and mangrove. This training dataset was used to create biophysical probability layers (primitives) for each class. These primitives were then used to create land cover and probability maps in a decision tree logic and Monte-Carlo simulations. Validation showed good overall accuracy (84%) for the years 2017 and 2018. Filtering for validation points with high error estimates improved the accuracy up to 91%. We demonstrated and concluded that error quantification is an essential step in land cover classification and land cover change detection. Our overall analysis supports and presents a path for improving present assessments for sustainable supply chain analyses and associated recommendations.

Experimental Demonstration of Remote and Compact Imaging Spectrometer Based on Mobile Devices

Imaging spectrometers show great potential for environmental and biomedical sensing applications. Selfie sticks, which are tools used to take photographs or videos, have gained global popularity in recent years. Few people have connected these two objects, and few people have researched the application of imaging spectrometers to perform scientific monitoring in point-of-use scenarios. In this paper, we develop a compact imaging spectrometer (35 g in weight, 18 mm in diameter, and 72 mm in length) that can be equipped on a motorized selfie stick to perform remote sensing. We applied this system to perform environmental and facial remote sensing via motorized scanning. The absorption of chlorophyll and hemoglobin can be found in the reflectance spectra, indicating that our system can be used in urban greening monitoring and point-of-care testing. In addition, this compact imaging spectrometer was also easily attached to an underwater dome port and a quad-rotor unmanned aerial vehicle to perform underwater and airborne spectral detection. Our system offers a route toward mobile imaging spectrometers used in daily life.