Evaluating the efficiency of coarser to finer resolution multispectral satellites in mapping paddy rice fields using GEE implementation

Abstract

Timely and accurate estimation of rice-growing areas and forecasting of production can provide crucial information for governments, planners, and decision-makers in formulating policies. While there exists studies focusing on paddy rice mapping, only few have compared multi-scale datasets performance in rice classification. Furthermore, rice mapping of large geographical areas with sufficient accuracy for planning purposes has been a challenge in Pakistan, but recent advancements in Google Earth Engine make it possible to analyze spatial and temporal variations within these areas. The study was carried out over southern Punjab (Pakistan)-a region with 380,400 hectares devoted to rice production in year 2020. Previous studies support the individual capabilities of Sentinel-2, Landsat-8, and Moderate Resolution Imaging Spectroradiometer (MODIS) for paddy rice classification. However, to our knowledge, no study has compared the efficiencies of these three datasets in rice crop classification. Thus, this study primarily focuses on comparing these satellites' data by estimating their potential in rice crop classification using accuracy assessment methods and area estimation. The overall accuracies were found to be 96% for Sentinel-2, 91.7% for Landsat-8, and 82.6% for MODIS. The F1-Scores for derived rice class were 83.8%, 75.5%, and 65.5% for Sentinel-2, Landsat-8, and MODIS, respectively. The rice estimated area corresponded relatively well with the crop statistics report provided by the Department of Agriculture, Punjab, with a mean percentage difference of less than 20% for Sentinel-2 and MODIS and 33% for Landsat-8. The outcomes of this study highlight three points; (a) Rice mapping accuracy improves with increase in spatial resolution, (b) Sentinel-2 efficiently differentiated individual farm level paddy fields while Landsat-8 was not able to do so, and lastly (c) Increase in rice cultivated area was observed using satellite images compared to the government provided statistics.

Figure 1

Map of the study area (Southern Punjab). The map is designed in ArcGIS Pro V2.9 software available at ESRI website (https://www.esri.com/en-us/arcgis/products/arcgis-pro/overview). The boundary shapefile used to draw these maps are available at Humdata website (https://data.humdata.org/dataset/cod-ab-pak).

Figure 2

Ground truth data for rice production in southern Punjab for year 2020 (Reported by Crop Reported Service, Government of Punjab). The map is designed in ArcGIS Pro V2.9 software available at ESRI website (https://www.esri.com/en-us/arcgis/products/arcgis-pro/overview). The data used in this map is available at CRS website (http://www.crs.agripunjab.gov.pk).

Figure 3

Methodology chart for rice crop classification.

Figure 4

Average NDVI temporal profile for rice using Sentinel-2, Landsat-8 and MODIS derived NDVI data and average Sentinel-1 backscatter coefficient (dB) temporal profile for rice.

Figure 5

Crops NDVI temporal variation using Sentinel-2 (a) and Average Monthly LST (Day and Night) along with CHIRPS precipitation temporal variation (b), for 2020 Kharif season in Southern Punjab region.

Figure 6

Training samples in the study area, (a) all samples, (b) ricefield samples, (c) non-rice samples, and (d) water samples. The map is designed in Google Earth Engine, which is an cloud computing browser based platform (https://earthengine.google.com).

Figure 7

Ricefield photographs, taken during training sample collection at different locations and timeperiod within the study area. The photographs (a–c) are taken by first author, during the field survey.

Figure 8

Rice crop map in southern Punjab for the year 2020 using Sentinel-2, Landsat-8 and MODIS datasets.

Figure 9

Comparing classified maps at a random location in Southern Punjab using VHR Google Earth imagery (a), Sentinel-2 10 m derived NDVI (b), rice classified map using Sentinel-2 (c), Landsat-8 (d) and MODIS (e).

Figure 10

Comparison of rice classified cultivated area from GEE and Crop statistics report (local agrarian data). The percentage of overestimation (Δ) by GEE classification relative to ground truth data is provided for each dataset.

Figure 11

Comparing our rice estimated area using three instruments with previously published study and crop statistics report for 2020. Note: in the Figure area is compared for Multan division, and call-outs in bar plots shows the percentage increase or decrease in area compared to crop statistics report. The Figure is designed in Photoshop V23.3 software provided by Adobe (https://www.adobe.com/products/photoshop).