Estimating Wheat Grain Yield Using Sentinel-2 Imagery and Exploring Topographic Features and Rainfall Effects on Wheat Performance in Navarre, Spain

Abstract

Reliable methods for estimating wheat grain yield before harvest could help improve farm management and, if applied on a regional level, also help identify spatial factors that influence yield. Regional grain yield can be estimated using conventional methods, but the typical process is complex and labor-intensive. Here we describe the development of a streamlined approach using publicly accessible agricultural data, field-level yield, and remote sensing data from Sentinel-2 satellite to estimate regional wheat grain yield. We validated our method on wheat croplands in Navarre in northern Spain, which features heterogeneous topography and rainfall. First, this study developed stepwise multilinear equations to estimate grain yield based on various vegetation indices, which were measured at various phenological stages in order to determine the optimal timings. Second, the most suitable model was used to estimate grain yield in wheat parcels mapped from Sentinel-2 satellite images. We used a supervised pixel-based random forest classification and the estimates were compared to government-published post-harvest yield statistics. When tested, the model achieved an R2 of 0.83 in predicting grain yield at field level. The wheat parcels were mapped with an accuracy close to 86% for both overall accuracy and compared to official statistics. Third, the validated model was used to explore potential relationships of the mapped per-parcel grain yield estimation with topographic features and rainfall by using geographically weighted regressions. Topographic features and rainfall together accounted for an average for 11 to 20% of the observed spatial variation in grain yield in Navarre. These results highlight the ability of our method for estimating wheat grain yield before harvest and determining spatial factors that influence yield at the regional scale.

Crop phenotyping in a context of global change: What to measure and how to do it

“…In comparison to Landsat‐8, which is the latest satellite launch of the Landsat mission, Sentinel‐2 has an improved spectral resolution (13 bands in Sentinel‐2 against 8 in Landsat‐8), temporal resolution (5 d revisit time in comparison to 15 d) and spatial resolution (10 m maximum resolution in Sentinel‐2 and 15 m in Landsat) (Segarra et al, 2020a). Implementation of Sentinel‐2 imagery has included estimation of grain yield in wheat (Toscano et al, 2019; Segarra et al, 2020b), maize (Battude et al, 2016), and potato (Gómez et al, 2019), and even observing within‐field yield variability (Hunt et al, 2019). Moreover, nitrogen‐related features of the plant such as canopy chlorophyll content have been calculated with Sentinel‐2 data (Clevers et al, 2017; Delloye et al, 2018) and the impacts of pests and diseases have also been monitored (Chemura et al, 2017; Zheng et al, 2018).…”

Modeling of Durum Wheat Yield Based on Sentinel-2 Imagery

“…However, corroborating our results, Lopresti et al [24] reported that prediction of wheat yield is best 30 days before harvest, after the stages of heading and flowering. Seggara et al [51] investigated the optimum period for winter wheat yield estimation in Spain and found that the predictions made on the heading stage outperformed the predictions on tillering or maturing stages, but they did not make any observations during the stage of flowering.…”

Estimating fall‐harvested alfalfa (Medicago sativa L.) yield using unmanned aerial vehicle–based multispectral and thermal images in southern California

“…2018), wheat (Segarra et al, 2020;Zhang et al, 2020), and rice (Arumugam et al, 2021;Fernandez-Beltran et al, 2021;Huang et al, 2013). In addition, several recent studies have also used satellite or UAV-acquired remote sensing images to develop alfalfa yield predicting models and reported promising results (Chandel et al, 2021;Dvorak et al, 2021;Feng et al, 2020).…”

“…Estimation of crop yield using remotely sensed images from UAVs and various satellite products has been well studied for cereals crops, including maize (Schwalbert et al., 2018), wheat (Segarra et al., 2020; Zhang et al., 2020), and rice (Arumugam et al., 2021; Fernandez‐Beltran et al., 2021; Huang et al., 2013). In addition, several recent studies have also used satellite or UAV‐acquired remote sensing images to develop alfalfa yield predicting models and reported promising results (Chandel et al., 2021; Dvorak et al., 2021; Feng et al., 2020).…”