Sight for Sorghums: Comparisons of Satellite- and Ground-Based Sorghum Yield Estimates in Mali

The advent of multiple satellite systems capable of resolving smallholder agricultural plots raises possibilities for significant advances in measuring and understanding agricultural productivity in smallholder systems. However, since only imperfect yield data are typically available for model training and validation, assessing the accuracy of satellite-based estimates remains a central challenge. Leveraging a survey experiment in Mali, this study uses plot-level sorghum yield estimates, based on farmer reporting and crop cutting, to construct and evaluate estimates from three satellite-based sensors. Consistent with prior work, the analysis indicates low correlation between the ground-based yield measures (r = 0.33). Satellite greenness, as measured by the growing season peak value of the green chlorophyll vegetation index from Sentinel-2, correlates much more strongly with crop cut (r = 0.48) than with self-reported (r = 0.22) yields. Given the inevitable limitations of ground-based measures, the paper reports the results from the regressions of self-reported, crop cut, and (crop cut-calibrated) satellite sorghum yields. The regression covariates explain more than twice as much variation in calibrated satellite yields (R2 = 0.25) compared to self-reported or crop cut yields, suggesting that a satellite-based approach anchored in crop cuts can be used to track sorghum yields as well or perhaps better than traditional measures. Finally, the paper gauges the sensitivity of yield predictions to the use of Sentinel-2 versus higher-resolution imagery from Planetscope and DigitalGlobe. All three sensors exhibit similar performance, suggesting little gains from finer resolutions in this system.

Download Full-text

Modeling of Durum Wheat Yield Based on Sentinel-2 Imagery

Agronomy ◽

10.3390/agronomy11081486 ◽

2021 ◽

Vol 11 (8) ◽

pp. 1486

Author(s):

Chris Cavalaris ◽

Sofia Megoudi ◽

Maria Maxouri ◽

Konstantinos Anatolitis ◽

Marios Sifakis ◽

...

Keyword(s):

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Vegetation Indices ◽

Model Development ◽

Wheat Yield ◽

Growth Period ◽

Yield Data ◽

Water Index ◽

Normalized Difference Water Index ◽

Sentinel 2

In this study, a modelling approach for the estimation/prediction of wheat yield based on Sentinel-2 data is presented. Model development was accomplished through a two-step process: firstly, the capacity of Sentinel-2 vegetation indices (VIs) to follow plant ecophysiological parameters was established through measurements in a pilot field and secondly, the results of the first step were extended/evaluated in 31 fields, during two growing periods, to increase the applicability range and robustness of the models. Modelling results were examined against yield data collected by a combine harvester equipped with a yield-monitoring system. Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI) were examined as plant signals and combined with Normalized Difference Water Index (NDWI) and/or Normalized Multiband Drought Index (NMDI) during the growth period or before sowing, as water and soil signals, respectively. The best performing model involved the EVI integral for the 20 April–31 May period as a plant signal and NMDI on 29 April and before sowing as water and soil signals, respectively (R2 = 0.629, RMSE = 538). However, model versions with a single date and maximum seasonal VIs values as a plant signal, performed almost equally well. Since the maximum seasonal VIs values occurred during the last ten days of April, these model versions are suitable for yield prediction.

Download Full-text

YIELD ESTIMATION OF SUNFLOWER PLANT WITH CNN AND ANN USING SENTINEL-2

ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-archives-xlvi-4-w5-2021-385-2021 ◽

2021 ◽

Vol XLVI-4/W5-2021 ◽

pp. 385-389

Author(s):

O. G. Narin ◽

A. Sekertekin ◽

A. Saygin ◽

F. Balik Sanli ◽

M. Gullu

Keyword(s):

Neural Network ◽

Vegetation Index ◽

Agricultural Land ◽

Crop Yields ◽

Decision Makers ◽

Yield Estimation ◽

Yield Data ◽

Artificial Neural Network Ann ◽

Agricultural Land Management ◽

Sentinel 2

Abstract. Due to food security and agricultural land management, it is crucial for decision makers and farmers to predict crop yields. In remote sensing based agricultural studies, spectral resolutions of satellite images, as well as temporal and spatial resolution, are important. In this study, we investigated whether there is a relationship between the Normalized Different Vegetation Index (NDVI) and Normalized Different Vegetation Index Red-edge (NDVIred) indices derived from the Sentinel-2 satellite. In addition, the efficiency of linear regression, Convolutional Neural Network (CNN), and Artificial Neural Network (ANN) techniques are examined with the use of indices in yield estimation. In this context, yield data of 48 sunflower parcels were obtained in 2018. The obtained results showed that both NDVI and NDVIred can be used to estimate the yield of sunflowers. The best results were obtained from the combination of the NDVI and the CNN technique with the RMSE equal to 20,874 Kg/da on 30 June 2018. Concerning the results, although there is not much superiority between the two indices, the best results were generally obtained from CNN as the method.

Download Full-text

Observing Sucrose Accumulation With Sentinel-1 Backscatter

Frontiers in Remote Sensing ◽

10.3389/frsen.2021.778691 ◽

2021 ◽

Vol 2 ◽

Author(s):

Nadja den Besten ◽

Susan Steele-Dunne ◽

Benjamin Aouizerats ◽

Ariel Zajdband ◽

Richard de Jeu ◽

...

Keyword(s):

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Growing Season ◽

Optical Observations ◽

Sucrose Accumulation ◽

Yield Data ◽

Sugarcane Plant ◽

Sugarcane Yield ◽

Optical Imagery ◽

The Impact

In this study the impact of sucrose accumulation in Sentinel-1 backscatter observations is presented and compared to Planet optical observations. Sugarcane yield data from a sugarcane plantation in Xinavane, Mozambique are used for this study. The database contains sugarcane yield of 387 fields over two seasons (2018-2019 and 2019-2020). The relation between sugarcane yield and Sentinel-1 VV and VH backscatter observation is analyzed by using the Normalized Difference Vegetation Index (NDVI) data as derived from Planet Scope optical imagery as a benchmark. The different satellite observations were compared over time to sugarcane yield to understand how the relation between the observations and yield evolves during the growing season. A negative correlation between yield and Cross Ratio (CR) from Sentinel-1 backscatter was found while a positive correlation between yield and Planet NDVI was observed. An additional modeling study on the dielectric properties of the crop revealed how the CR could be affected by sucrose accumulation during the growing season and supported the opposite correlations. The results shows CR contains information on sucrose content in the sugarcane plant. This sets a basis for further development of sucrose monitoring and prediction using a combination of radar and optical imagery.

Download Full-text

Using Sentinel-2 to Track Field-Level Tillage Practices at Regional Scales in Smallholder Systems

Remote Sensing ◽

10.3390/rs13245108 ◽

2021 ◽

Vol 13 (24) ◽

pp. 5108

Author(s):

Weiqi Zhou ◽

Preeti Rao ◽

Mangi L. Jat ◽

Balwinder Singh ◽

Shishpal Poonia ◽

...

Keyword(s):

Satellite Data ◽

Management Practices ◽

Growing Season ◽

Google Earth ◽

Large Field ◽

Zero Tillage ◽

Tillage Practices ◽

Spatio Temporal ◽

Sentinel 2 ◽

Smallholder Systems

Zero tillage is an important pathway to sustainable intensification and low-emission agriculture. However, quantifying the extent of zero tillage adoption at the field scale has been challenging, especially in smallholder systems where field sizes are small and there is limited ground data on zero tillage adoption. Remote sensing offers the ability to map tillage practices at large spatio-temporal scales, yet to date no studies have used satellite data to map zero tillage adoption in smallholder agricultural systems. In this study, we use Sentinel-2 satellite data, random forest classifiers, and Google Earth Engine to map tillage practices across India’s main grain producing region, the Indo-Gangetic Plains. We find that tillage practices can be classified with moderate accuracy (an overall accuracy of 75%), particularly in regions with relatively large field sizes and homogenous crop management practices. We find that models that use satellite data from only the first half of the growing season perform as well as models that use data throughout the growing season, allowing for the creation of within-season tillage maps. Finally, we find that our model can generalize well through time in the western IGP, with reductions in accuracy of only 5–10%. Our results highlight the ability of Sentinel-2 satellite data to map tillage practices at scale, even in smallholder systems where field sizes are small and cropping practices are heterogeneous.

Download Full-text

Data assimilation of remote sensing data for farm scale maize fertilization in northern Italy

10.5194/egusphere-egu2020-15941 ◽

2020 ◽

Author(s):

Calogero Schillaci ◽

Edoardo Tomasoni ◽

Marco Acutis ◽

Alessia Perego

Keyword(s):

Spatial Resolution ◽

Spatial Data ◽

Precision Agriculture ◽

Vegetation Index ◽

Vegetation Indices ◽

Northern Italy ◽

Yield Data ◽

Batch Mode ◽

Area Index ◽

Sentinel 2

To improve nitrogen fertilization is well known that vegetation indices can offer a picture of the nutritional status of the crop. In this study, field management information (maize sowing and harvesting dates, tillage, fertilization) and estimated vegetation indices VI (Sentinel 2 derived Leaf Area Index LAI, Normalized Difference Vegetation Index NDVI, Fraction of Photosynthetic radiation fPAR) were analysed to develop a batch-mode VIs routine to manage high dimensional temporal and spatial data for Decision Support Systems DSS in precision agriculture, and to optimize the maize N fertilization in the field. The study was carried out in maize (2017-2018) on a farm located in Mantua (northern Italy); the soil is a Vertic Calciustepts with a fine silty texture with moderate content of carbonates. A collection of Sentinel 2 images (with <25% cloud cover) were processed using Graph Processing Tool (GPT). This tool is used through the console to execute Sentinel Application Platform (SNAP) raster data operators in batch-mode. The workflow applied on the Sentinel images consisted in: resampling each band to 10m pixel size, splitting data into subsets according to the farm boundaries using Region of Interest (ROI). Biophysical Operator based on Biophysical Toolbox was used to derive LAI, fPAR for the estimation of maize vegetation indices from emergence until senescence. Yield data were acquired with a volumetric yield sensing in a combine harvester. Fertilization plans were then calculated for each field prior to the side-dressing fertilization. The routine is meant as a user-friendly tool to obtain time series of assimilated VIs of middle and high spatial resolution for field crop fertilization. It also overcomes the failures of the open source graphic user interface of SNAP. For the year 2018, yield data were related to the 34 LAI derived from Sentinel 2a products at 10 m spatial resolution (R2=0.42). This result underlined a trend that can be further studied to define a cluster strategy based on soil properties. As a further step, we will test whether spatial differences in assimilated VIs, integrated with yield data, can guide the nitrogen top-dress fertilization in quantitative way more accurately than a single image or a collection of single images.

Download Full-text

Capability of existing drought indices in reflecting agricultural drought in China

10.5194/egusphere-egu2020-8845 ◽

2020 ◽

Author(s):

Zhe Zhao ◽

Kaicun Wang

Keyword(s):

Soil Moisture ◽

Crop Yield ◽

Vegetation Index ◽

Meteorological Data ◽

Growing Season ◽

The Other ◽

Agricultural Drought ◽

Drought Indices ◽

Yield Data ◽

Prediction Ability

A variety of drought indices have been constructed to monitor agricultural drought using ground and satellite data. Our study aimed to evaluate the performance of drought indices to indicate agricultural drought in China. Seven drought indices of four types were selected over the main agricultural regions of China: indices based on regular meteorological data (DImet), indices based on vegetation index (DIvi), indices based on soil moisture (DIsm), and synthesized indices (DIsyn). The independent reference data used here included three aspects: soil moisture, vegetation photosynthesis and crop yield data. The latter two reference datasets were selected to check drought impact on agriculture. Drought indices with short timescales are more sensitive to topsoil moisture. Drought indices have different abilities to capture vegetation photosynthesis condition during the growing season. Expect for the Yangtze region and North China region during the wheat growing season, the DImet and DIsyn show significant positive correlations with the sun-induced chlorophyll fluorescence (SIF), while the other drought indices have weaker or no correlations. For crop yield, the prediction ability of the drought indices show a similar pattern with the results for vegetation photosynthesis but with relatively large uncertainty. Generally, our study show that DImet have better or equivalent performance than that of the other types of drought indices, and DIsyn show the widest applicability. Our study may shed light on agricultural drought research in the future.

Download Full-text

ESTIMATION OF MANGROVE FOREST ABOVEGROUND BIOMASS USING MULTISPECTRAL BANDS, VEGETATION INDICES AND BIOPHYSICAL VARIABLES DERIVED FROM OPTICAL SATELLITE IMAGERIES: RAPIDEYE, PLANETSCOPE AND SENTINEL-2

ISPRS Annals of Photogrammetry Remote Sensing and Spatial Information Sciences ◽

10.5194/isprs-annals-iv-3-29-2018 ◽

2018 ◽

Vol IV-3 ◽

pp. 29-36 ◽

Cited By ~ 9

Author(s):

Alvin Balidoy Baloloy ◽

Ariel Conferido Blanco ◽

Christian Gumbao Candido ◽

Reginal Jay Labadisos Argamosa ◽

John Bart Lovern Caboboy Dumalag ◽

...

Keyword(s):

Linear Regression ◽

Mangrove Forest ◽

Field Data ◽

Vegetation Index ◽

Prediction Models ◽

Vegetation Indices ◽

Satellite Systems ◽

Red Edge ◽

Simple Ratio ◽

Sentinel 2

Aboveground biomass estimation (AGB) is essential in determining the environmental and economic values of mangrove forests. Biomass prediction models can be developed through integration of remote sensing, field data and statistical models. This study aims to assess and compare the biomass predictor potential of multispectral bands, vegetation indices and biophysical variables that can be derived from three optical satellite systems: the Sentinel-2 with 10&thinsp;m, 20&thinsp;m and 60&thinsp;m resolution; RapidEye with 5m resolution and PlanetScope with 3m ground resolution. Field data for biomass were collected from a Rhizophoraceae-dominated mangrove forest in Masinloc, Zambales, Philippines where 30 test plots (1.2&thinsp;ha) and 5 validation plots (0.2&thinsp;ha) were established. Prior to the generation of indices, images from the three satellite systems were pre-processed using atmospheric correction tools in SNAP (Sentinel-2), ENVI (RapidEye) and python (PlanetScope). The major predictor bands tested are Blue, Green and Red, which are present in the three systems; and Red-edge band from Sentinel-2 and Rapideye. The tested vegetation index predictors are Normalized Differenced Vegetation Index (NDVI), Soil-adjusted Vegetation Index (SAVI), Green-NDVI (GNDVI), Simple Ratio (SR), and Red-edge Simple Ratio (SRre). The study generated prediction models through conventional linear regression and multivariate regression. Higher coefficient of determination (r2) values were obtained using multispectral band predictors for Sentinel-2 (r2&thinsp;=&thinsp;0.89) and Planetscope (r2&thinsp;=&thinsp;0.80); and vegetation indices for RapidEye (r2&thinsp;=&thinsp;0.92). Multivariate Adaptive Regression Spline (MARS) models performed better than the linear regression models with r2 ranging from 0.62 to 0.92. Based on the r2 and root-mean-square errors (RMSE’s), the best biomass prediction model per satellite were chosen and maps were generated. The accuracy of predicted biomass maps were high for both Sentinel-2 (r2&thinsp;=&thinsp;0.92) and RapidEye data (r2&thinsp;=&thinsp;0.91).

Download Full-text

Mapping Public Urban Green Spaces Based on OpenStreetMap and Sentinel-2 Imagery Using Belief Functions

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi10040251 ◽

2021 ◽

Vol 10 (4) ◽

pp. 251

Author(s):

Christina Ludwig ◽

Robert Hecht ◽

Sven Lautenbach ◽

Martin Schorcht ◽

Alexander Zipf

Keyword(s):

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Open Data ◽

Green Spaces ◽

Data Sets ◽

Urban Green ◽

Urban Green Spaces ◽

Urban Quality Of Life ◽

Shafer Theory ◽

Sentinel 2

Public urban green spaces are important for the urban quality of life. Still, comprehensive open data sets on urban green spaces are not available for most cities. As open and globally available data sets, the potential of Sentinel-2 satellite imagery and OpenStreetMap (OSM) data for urban green space mapping is high but limited due to their respective uncertainties. Sentinel-2 imagery cannot distinguish public from private green spaces and its spatial resolution of 10 m fails to capture fine-grained urban structures, while in OSM green spaces are not mapped consistently and with the same level of completeness everywhere. To address these limitations, we propose to fuse these data sets under explicit consideration of their uncertainties. The Sentinel-2 derived Normalized Difference Vegetation Index was fused with OSM data using the Dempster–Shafer theory to enhance the detection of small vegetated areas. The distinction between public and private green spaces was achieved using a Bayesian hierarchical model and OSM data. The analysis was performed based on land use parcels derived from OSM data and tested for the city of Dresden, Germany. The overall accuracy of the final map of public urban green spaces was 95% and was mainly influenced by the uncertainty of the public accessibility model.

Download Full-text

Status of Phenological Research Using Sentinel-2 Data: A Review

Remote Sensing ◽

10.3390/rs12172760 ◽

2020 ◽

Vol 12 (17) ◽

pp. 2760

Author(s):

Gourav Misra ◽

Fiona Cawkwell ◽

Astrid Wingler

Keyword(s):

Vegetation Index ◽

Temporal Frequency ◽

Short Wave ◽

Current State ◽

Red Edge ◽

Natural Grasslands ◽

Wide Range ◽

Normalised Difference Vegetation Index ◽

Satellite Sensors ◽

Sentinel 2

Remote sensing of plant phenology as an indicator of climate change and for mapping land cover has received significant scientific interest in the past two decades. The advancing of spring events, the lengthening of the growing season, the shifting of tree lines, the decreasing sensitivity to warming and the uniformity of spring across elevations are a few of the important indicators of trends in phenology. The Sentinel-2 satellite sensors launched in June 2015 (A) and March 2017 (B), with their high temporal frequency and spatial resolution for improved land mapping missions, have contributed significantly to knowledge on vegetation over the last three years. However, despite the additional red-edge and short wave infra-red (SWIR) bands available on the Sentinel-2 multispectral instruments, with improved vegetation species detection capabilities, there has been very little research on their efficacy to track vegetation cover and its phenology. For example, out of approximately every four papers that analyse normalised difference vegetation index (NDVI) or enhanced vegetation index (EVI) derived from Sentinel-2 imagery, only one mentions either SWIR or the red-edge bands. Despite the short duration that the Sentinel-2 platforms have been operational, they have proved their potential in a wide range of phenological studies of crops, forests, natural grasslands, and other vegetated areas, and in particular through fusion of the data with those from other sensors, e.g., Sentinel-1, Landsat and MODIS. This review paper discusses the current state of vegetation phenology studies based on the first five years of Sentinel-2, their advantages, limitations, and the scope for future developments.

Download Full-text

Outlier Detection at the Parcel-Level in Wheat and Rapeseed Crops Using Multispectral and SAR Time Series

Remote Sensing ◽

10.3390/rs13050956 ◽

2021 ◽

Vol 13 (5) ◽

pp. 956

Author(s):

Florian Mouret ◽

Mohanad Albughdadi ◽

Sylvie Duthoit ◽

Denis Kouamé ◽

Guillaume Rieu ◽

...

Keyword(s):

Outlier Detection ◽

Experimental Validation ◽

Vegetation Index ◽

Normalized Difference Vegetation Index ◽

Multispectral Images ◽

Vegetation Indexes ◽

Crop Development ◽

Different Types ◽

Selection Of ◽

Sentinel 2

This paper studies the detection of anomalous crop development at the parcel-level based on an unsupervised outlier detection technique. The experimental validation is conducted on rapeseed and wheat parcels located in Beauce (France). The proposed methodology consists of four sequential steps: (1) preprocessing of synthetic aperture radar (SAR) and multispectral images acquired using Sentinel-1 and Sentinel-2 satellites, (2) extraction of SAR and multispectral pixel-level features, (3) computation of parcel-level features using zonal statistics and (4) outlier detection. The different types of anomalies that can affect the studied crops are analyzed and described. The different factors that can influence the outlier detection results are investigated with a particular attention devoted to the synergy between Sentinel-1 and Sentinel-2 data. Overall, the best performance is obtained when using jointly a selection of Sentinel-1 and Sentinel-2 features with the isolation forest algorithm. The selected features are co-polarized (VV) and cross-polarized (VH) backscattering coefficients for Sentinel-1 and five Vegetation Indexes for Sentinel-2 (among us, the Normalized Difference Vegetation Index and two variants of the Normalized Difference Water). When using these features with an outlier ratio of 10%, the percentage of detected true positives (i.e., crop anomalies) is equal to 94.1% for rapeseed parcels and 95.5% for wheat parcels.

Download Full-text