Comparative Analysis of Normalised Difference Spectral Indices Derived from MODIS for Detecting Surface Water in Flooded Rice Cropping Systems

Abstract

Identifying managed flooding in paddy fields is commonly used in remote sensing to detect rice. Such flooding, followed by rapid vegetation growth, is a reliable indicator to discriminate rice. Spectral indices (SIs) are often used to perform this task. However, little work has been done on determining which spectral combination in the form of Normalised Difference Spectral Indices (NDSIs) is most appropriate for surface water detection or which thresholds are most robust to separate water from other surfaces in operational contexts. To address this, we conducted analyses on satellite and field spectral data from an agronomic experiment as well as on real farming situations with different soil and plant conditions. Firstly, we review and select NDSIs proposed in the literature, including a new combination of visible and shortwave infrared bands. Secondly, we analyse spectroradiometric field data and satellite data to evaluate mixed pixel effects. Thirdly, we analyse MODIS data and Landsat data at four sites in Europe and Asia to assess NDSI performance in real-world conditions. Finally, we test the performance of the NDSIs on MODIS temporal profiles in the four sites. We also compared the NDSIs against a combined index previously used for agronomic flood detection. Analyses suggest that NDSIs using MODIS bands 4 and 7, 1 and 7, 4 and 6 or 1 and 6 perform best. A common threshold for each NDSI across all sites was more appropriate than locally adaptive thresholds. In general, NDSIs that use band 7 have a negligible increase in Commission Error over those that use band 6 but are more sensitive to water presence in mixed land cover conditions typical of moderate spatial resolution analyses. The best performing NDSI is comparable to the combined index but with less variability in performance across sites, suggesting a more succinct and robust flood detection method.

Assessing in-season crop classification performance using satellite data: a test case in Northern Italy

“…Rice is among all the best identified class, with CE and OE in the MCL classification lower than 2% after T3. This is due to specific rice crop features: i) in Northern Italy, the ricecultivated area is clustered in homogeneous regions and fields have an average size greater than 1.5 ha [Giuca et al, 2014], compared to other, more fragmented, agricultural districts within the study area, and ii) the distinct temporal signature of agronomic flooding [Xiao et al, 2005;Boschetti et al, 2014] and the single crop cycle typical of rice cultivations in temperate areas are relatively easy to track with multi-temporal spectral data. Maize, although being characterized by acceptable errors, is misclassified as Soybean at the beginning of the season, up to T3.…”

“…3). Three spectral VIs were calculated from each date of OLI 2013 dataset: Enhanced Vegetation Index (EVI) [Huete et al, 1997], Normalized Difference Flood Index (NDFI) [Boschetti et al, 2014], and Red Green Ratio Index (RGRI) [Gamon and Surfus, 1999]. Multi-temporal VIs, in particular Normalized Difference Vegetation Index (NDVI), have been extensively used in remote sensing of vegetation for their availability, simplicity and effectiveness in distinguishing phenological features of plant groups, including crops [e.g.…”

“…As NDVI can be subject to distortion due to environmental factors, for our analysis we preferred to use EVI, which includes a factor for correcting background effects and partially compensating for major atmospheric disturbances [Huete et al, 1997]. As a complement to EVI, we included NDFI, which is an index developed by Boschetti et al [2014] for the detection of surface water in flooded rice areas, modified for optimizing the performance of indices previously used by Xiao et al [2005] for mapping paddy rice cultivations. The RGRI was included due to its sensitivity to canopy greenness and light use efficiency in plant photosynthetic activity [Gamon and Surfus, 1999].…”

A Spatial Data Infrastructure Integrating Multisource Heterogeneous Geospatial Data and Time Series: A Study Case in Agriculture

“…The EO GD products are in raster format with distinct spatial and temporal resolution derived [33]), Enhanced Vegetation Index (EVI; [34]), Red Green Ratio Index (RGRI; [35]), and Normalized Difference Flood Index (NDFI; [36]). …”

Flood Mapping Based on Multiple Endmember Spectral Mixture Analysis and Random Forest Classifier—The Case of Yuyao, China

“…In general, the Normalized Difference Spectral Indices (NDSIs) is suitable for detecting the open water surface and the inundated areas. Boschetti et al [45] provided a useful review which summarized and compared dozens of NDSIs for detecting surface water in flooded rice fields, which also shed light on the inundated area detection. Moreover, the water related spectral indices are proposed as the combination of shortwave infrared and near infrared or visible spectral regions [45].…”