Ying Ge

Existing Google Earth Engine-based retrieval workflows often use fixed normalized difference vegetation index thresholds and coarse atmospheric water vapor inputs, which may limit their adaptability to regional surface and atmospheric conditions. This study evaluates how these two intermediate parameters influence Landsat 8 land surface temperature retrieval over northeastern Egypt using the generalized single-channel algorithm. Atmospheric water vapor was derived from MERRA-2 and NCEP reanalysis products, while land surface emissivity was estimated using ASTER Global Emissivity Dataset data and an NDVI-threshold framework. Reanalysis-derived total precipitable water was first compared with MODIS MOD05_L2. MERRA-2 showed a stronger correlation with MOD05_L2, whereas NCEP produced lower bias and root mean square error. The retrieved land surface temperature was then compared with the Landsat 8 Collection 2 Level 2 LST product as an internal consistency check. Using MERRA-2 reduced the overall root mean square error from 1.3977 K to 1.2615 K relative to NCEP, although it also increased the magnitude of the negative bias. A grid search of 24 normalized difference vegetation index threshold combinations showed that retrieval consistency was sensitive to threshold selection in cropland areas, while desert and built-up areas were largely insensitive. The best overall consistency with the Landsat product was obtained using a soil threshold of 0.20 and a vegetation threshold of 0.75, with a root mean square error of 1.2507 K and a bias of −0.7236 K. External validation at the Baseline Surface Radiation Network Gobabeb station showed a slight improvement when using MERRA-2 instead of NCEP, with root mean square error decreasing from 4.726 K to 4.441 K. Overall, the results show that intermediate parameter choices can affect Landsat land surface temperature retrieval, but the optimized settings should be interpreted as region-specific and relative to the Landsat product because independent validation remains limited.

Accurate cropland parcel mapping is essential for food security and sustainable land management in arid Africa, yet it remains challenging in Egypt due to edge blurring, spectral confusion, and fragmented fields in medium-resolution imagery. A novel dual-encoder deep learning method that integrates multi-scale axial attention and boundary constraints (MAA-BCNet) is proposed for the precise extraction of croplands in Egypt from Sentinel-2 multispectral images. A dual-path encoder is designed to fuse CNN-based local textures with an RMT global branch using spatial decay attention for complementary feature extraction. A multi-scale axial attention module is introduced to capture anisotropic parcel structures for improved spectral–spatial discrimination, and a multi-directional gradient edge enhancement module is developed for explicitly preserving boundary integrity. A U-Net++ decoder is employed for dense multi-scale aggregation. Experimental results in Egypt demonstrate that MAA-BCNet achieves superior performance in delineating cropland parcels, particularly for irregular or fragmented croplands with complex landscapes and fuzzy boundaries. Compared with the widely used segmentation models such as DeepLabV3_plus, PSPnet, Link_net, FCN_resnet101, and U-Net++ under the same training and evaluation settings, our model has the best performance, with Recall, Precision, IoU, and F1-Score reaching 94.92%, 90.77%, 86.57%, and 92.80%, respectively. These advancements make MAA-BCNet suitable for cropland mapping of large areas of Egypt, with applications in precision agriculture and sustainable land management.

Citrus is an important commercial crop in many areas. The management and planning of citrus growing can be supported by timely and efficient monitoring of citrus-growing regions. Their complex planting structure and the weather are likely to cause problems for extracting citrus-growing regions from remote sensing images. To accurately extract citrus-growing regions, deep learning is employed, because it has a strong feature representation ability and can obtain rich semantic information. A novel model for extracting citrus-growing regions by UNet that incorporates an image pyramid structure is proposed on the basis of the Sentinel-2 satellite imagery. A pyramid-structured encoder, a decoder, and multiscale skip connections are the three main components of the model. Additionally, atrous spatial pyramid pooling is used to prevent information loss and improve the ability to learn spatial features. The experimental results show that the proposed model has the best performance, with the precision, the intersection over union, the recall, and the F1-score reaching 88.96%, 73.22%, 80.55%, and 84.54%, respectively. The extracted citrus-growing regions have regular boundaries and complete parcels. Furthermore, the proposed model has greater overall accuracy, kappa, producer accuracy, and user accuracy than the object-oriented random forest algorithm that is widely applied in various fields. Overall, the proposed method shows a better generalization ability, higher robustness, greater accuracy, and less fragmented extraction results. This research can support the rapid and accurate mapping of large-scale citrus-growing regions.

Ripley'sK‐function is a test to detect geographically distributed patterns occurring across spatial scales. Initially, it assumed infinitely continuous planar space, but in reality, any geographic distribution occurs in a bounded region. Hence, the edge problem must be solved in the application of Ripley'sK‐function. Traditionally, three basic edge correction methods were designed for regular study plots because of simplified geometric computation: the Ripley circumference, buffer zone, and toroidal methods. For an irregular‐shaped study region, a geographic information system (GIS) is needed to support geometric calculation of complex shapes. The Ripley circumference method was originally implemented by Haase and has been modified into a Python program in a GIS environment via Monte Carlo simulation (hereafter, the Ripley–Haase and Ripley–GIS methods). The results show that in terms of the statistical powers of clustering detection for irregular boundaries, the Ripley–GIS method is the most stable, followed by the buffer zone, toroidal, and Ripley–Haase methods. After edge effects of irregular boundaries have been eliminated, Ripley'sK‐function is used to estimate the degree of spatial clustering of cities in a given territory, and in this paper, we demonstrate that by reference to the relationship between urban spatial structure and economic growth in China.

In a migration network system, any change of regional characteristics may affect outflows and inflows from and to the region itself and potentially its neighbouring regions. To explore the impact of spatial spillovers on interprovincial migration flows in China, we proposed four spatial autoregressive interaction models by introducing different network structures to gravity models. The results indicate that the model taking account of dependence structure among origins, destinations and flows is more accurate than the other models. Population size is the most critical determinant in interprovincial migration from 2005 to 2010 in China and is greatly strengthened by spatial spillovers.

This paper proposes a convenient measure of the degree of distributional overlap, both parametric and nonparametric, useful in measuring the degree of Polarization, Alienation, and Convergence. We show the measure is asymptotically normally distributed, making it amenable to inference in consequence. This Overlap measure can be used in the univariate and multivariate framework, and three examples are used to illustrate its use. The nonparametric Overlap Index has two sources of bias, the first being a positive bias induced by the unknown intersection point of the underlying distribution and the second being a negative bias induced by the expectation of cell probabilities being less than the conditional expected values. We show that the inconsistency problem generated by the first bias, prevalent within this class of Goodness of Fit measure, is limited by the number of intersection points of the underlying distributions. A Monte Carlo study was used to examine the biases, and it was found that the latter bias dominates the former. These biases can be diluted by increasing the number of partitions, but prevails asymptotically nonetheless.

How does foreign direct investment (FDI) affect the wage levels of cities in a host country? This paper uses a panel dataset of Chinese cities to examine the impact of inward FDI on the urban real wage. The results suggest that the existence of FDI has a significant and positive effect on urban real wages and that this impact remains significant after controlling for other city characteristics. Cross-city variation of FDI intensity is one important factor contributing to intercity wage inequality.