Mapping cropping intensity in Huaihe basin using phenology algorithm, all Sentinel-2 and Landsat images in Google Earth Engine

Development of a New Phenology Algorithm for Fine Mapping of Cropping Intensity in Complex Planting Areas Using Sentinel-2 and Google Earth Engine

“…Savitzky-Golay filtering [64] was used to smooth the NDVI dataset (the solid line in Figure 5). Since the growth period of most crops exceeds 90 days, we used a moving window of size 9 and a 2 order of filter [15]. The LSWI is more sensitive to surface moisture, so we did not smooth the LSWI time series.…”

“…At the same time, with the availability of satellite images with different resolutions greatly improved, considerable progress has been made in the production of cropping intensity distribution maps based on remote sensing [11,12]. Over the past few decades, cropping intensity maps have been drawn in different regions using satellite data, such as Brazil [13,14], China [5,15], India [16,17], Arizona [18]. From the selected data sources, the most commonly used satellite data for monitoring cropping intensity are MODIS data and Landsat data.…”

“…Vegetation phenology is a natural phenomenon that reflects the annual cycle of plants under the influence of biological and non-biological factors [25]. Previous studies have proved that crop phenology algorithms have great potential in agricultural remote sensing mapping, such as identifying the start of season (SOS) and the end of season (EOS) [15], identifying the peak growth (PG) and peak drought (PD) periods [42], calculating the length of the growth cycle [43], calculating the growth rate and decline rate of VIs' values [44], and assessing cropping intensity by identifying the number of peaks and troughs in the VIs' time series within a year [5,16,31,45]. A recent study used the NDVI threshold to distinguish between crops and natural vegetation, and used the LSWI threshold to identify bare soil to map the cropping intensity of seven representative grain producing areas in China successfully [46].…”

Mapping Winter Crops Using a Phenology Algorithm, Time-Series Sentinel-2 and Landsat-7/8 Images, and Google Earth Engine

“…However, due to the low temporal resolution of single satellites and the frequent clouds and cloud shadows, it is difficult to construct a complete and smooth time-series curve. In order to overcome the difficulty of the insufficient number of available images, many studies have begun to combine multiple sensors for research [8,[28][29][30][31]. The integration of multiple sensor data sources can improve the temporal resolution of images and reduce the interference of other factors such as clouds.…”

“…Third, a phenology based algorithm. Different crops have different phenological characteristics during specific periods [31,37]; extracting phenological characteristics can be used as a method to distinguish winter crops from other crops. At present, some phenological based algorithms have been developed, such as establishing a linear regression model of the Crop Proportion Phenology Index (CPPI) [21], calculating the growth rate and decline rate of VI values [16,38], extracting the VI time series curve slopes [39], identifying the peak growth (PG) and peak drought (PD) periods [40], a maximum curvature method which extracts phenology [41], etc.…”