Monitoring Cropland Dynamics of the Yellow River Delta based on Multi-Temporal Landsat Imagery over 1986 to 2015

Abstract

Natural deltas can provide human beings with flat and fertile land to be cultivated. It is important to monitor cropland dynamics to provide policy-relevant information for regional sustainable development. This paper utilized Landsat imagery to study the cropland dynamics of the Yellow River Delta during the last three decades. Multi-temporal Landsat data were used to account for the phenological variations of different plants. Several spectral and textural features were adopted to increase the between-class separability. The robust random forest classifier was used to generate the land cover maps of the Yellow River Delta for 1986, 1995, 2005 and 2015. Experimental results indicated that the proposed methodology showed good performance with an average classification accuracy of 89.44%. The spatial-temporal analysis indicated that the cropland area increased from 467.6 km 2 in 1986 to 718.5 km 2 in 2015 with an average growth rate of 8.65 km 2 /year. The newly created croplands were mainly due to the reclamation of grassland and bare soil while the losses of croplands were due to abandoned cultivation and urban sprawl. The results demonstrate that a sustainable perspective should be adopted by the decision makers in order to simultaneously maintain food security, industrial development and ecosystem safety.

Integrating Multitemporal Sentinel-1/2 Data for Coastal Land Cover Classification Using a Multibranch Convolutional Neural Network: A Case of the Yellow River Delta

“…Land cover types, photographs, and global positioning system (GPS) locations were recorded for each sampling site. According to the field survey and previous studies [8,[39][40][41][42], there were 11 land cover categories in this study: forest, grassland, salt marsh, shrubs, tidal flat, bare soil, clear water, turbid water, irrigated farmland, dry farmland, and built up. Landscape descriptions for each land cover category are shown in Table 1.…”

Mapping of Vegetation Using Multi-Temporal Downscaled Satellite Images of a Reclaimed Area in Saemangeum, Republic of Korea

“…The random forest classifier (RFC) proposed by Brieman [32] was applied to image classification because it is widely used in a range of fields and often yields good results [19,[33][34][35]. The random forest algorithm (RF) is a type of decision tree classification.…”

Land Use Change Impact on Flooding Areas: The Case Study of Cervaro Basin (Italy)

“…Different images can be processed separately, allowing variability at seasonal and directional scales to be accounted for [22][23][24][25][26]. Many methods for mapping large areas with a high level of accuracy have been studied [1,27], and one of the most widespread is the Artificial Neural Network (ANN) approach, which can simulate the decision making processes of the human brain.…”