Search

Landslides pose significant and serious geological threat disasters worldwide, threatening human lives and property; China is particularly susceptible to these disasters. This paper focuses on Pengyang County, which is situated in the Ningxia Hui Autonomous Region of China, an area prone to landslides. This study investigated the application of machine learning techniques for analyzing landslide susceptibility. To construct and validate the model, we initially compiled a landslide inventory comprising 972 historical landslides and an equivalent number of non-landslide sites (Data sourced from the Pengyang County Department of Natural Resources). To ensure an impartial evaluation, both the landslide and non-landslide datasets were randomly divided into two sets using a 70/30 ratio. Next, we extracted 15 landslide conditioning factors, including the slope angle, elevation, profile curvature, plan curvature, slope aspect, TWI (topographic wetness index), TPI (topographic position index), distance to roads and rivers, NDVI (normalized difference vegetation index), rainfall, land use, lithology, SPI (stream power index), and STI (sediment transport index), from the spatial database. Subsequently, a correlation analysis between the conditioning factors and landslide occurrences was conducted using the certainty factor (CF) method. Three landslide models were established by employing logistic regression (LR), functional trees (FTs), and random subspace functional trees (RSFTs) algorithms. The landslide susceptibility map was categorized into five levels: very low, low, medium, high, and very high susceptibility. Finally, the predictive capability of the three algorithms was assessed using the area under the receiver operating characteristic curve (AUC). The better the prediction, the higher the AUC value. The results indicate that all three models are predictive and practical, with only minor discrepancies in accuracy. The integrated model (RSFT) displayed the highest predictive performance, achieving an AUC value of 0.844 for the training dataset and 0.837 for the validation dataset. This was followed by the LR model (0.811 for the training dataset and 0.814 for the validation dataset) and the FT model (0.776 for the training dataset and 0.760 for the validation dataset). The proposed methods and resulting landslide susceptibility map can assist researchers and local authorities in making informed decisions for future geohazard prevention and mitigation. Furthermore, they will prove valuable and be useful for other regions with similar geological characteristics features. [ABSTRACT FROM AUTHOR]

Sediment transport in coastal waters has an important impact on the siltation of port channels and changes in the estuary ecological environment. The southeast coast of China is often hit by typhoons, which can affect the suspended sediment concentration (SSC) in coastal waters. In this study, we used Geostationary Ocean Color Imager (GOCI) data to analyze SSC variations in Hangzhou Bay during Typhoon Maria (2018), and the influencing factors were also analyzed. The results showed that: (1) During the typhoon's transit, the SSC in Hangzhou Bay (HZB) increased by 200–800 mg/L, which was one-fold higher than the day before the typhoon. The variation of SSC on the south bank was noticeable, and the typhoon effect on SSC lasted for 2–3 days; (2) The wind speed and significant wave height (SWH) increased during the typhoon. In general, in the early stage of the typhoon, the SSC in HZB was affected by the wind, and in the interim and late period, SSC was influenced by the effect of wind and wave height; (3) Typhoon "Maria" accelerated the transport of sediment and land-based pollutants from land to sea; the effect of residual current and wind stress are the driving mechanisms for seaward sediment transport. However, mechanisms and driving factors of sediment transport in coast water are complex and diverse. The results of this study can help to understand the processes of riverbed erosion and deposition in Hangzhou Bay and adjacent waters. They are also significant for the study of nearshore hydrodynamic characteristics of typhoons and channel engineering. [ABSTRACT FROM AUTHOR]

Crops are usually planted on sloping land in mountainous areas due to limited suitable land area. This results in serious soil erosion and loss of nitrogen (N) and phosphorus (P) to land degradation and water eutrophication. It is important to adopt appropriate cultivation practices to change this situation. However, few long-term in situ measurements are available to assess the magnitude of effects of combined cultivation patterns on soil erosion and nutrient loss from sloping farmland with red soil, as well to quantify N and P losses through runoff and sediment transport. A field trial with the cash crop (CC) Nicotiana tabacum was carried out under natural rainfall conditions on sloping farmland with red soil in Yunnan, China during 2014–2017. Four cultivation patterns were applied. They included NVF (No fertilizer application + Vertical ridge + Film covered), OVF (Optimizing fertilizer application + Vertical ridge + Film covered), OHF (Optimizing fertilizer application + Horizontal ridge + Film covered), and OHFR (Optimizing fertilizer application + Horizontal ridge + Film removed). The first two treatments belonged to the vertical ridge (VR) group, and the remaining treatments belonged to the horizontal ridge (HR) group. Results indicated the HR group performed significantly better than the VR group, especially the OHFR treatment, in terms of the HR group producing average runoff (177.12–182.27 mm), sediment loss (2673.33–3309.17 kg·ha−1), and nutrient loss of total nitrogen (TN) (7.58–7.93 kg·ha−1), total phosphorus (TP) (1.00–1.09 kg·ha−1) through runoff, TN (3.53–4.72 kg·ha−1), TP (2.59–2.76 kg·ha−1) through sediment. TN was lost mainly through runoff transport, while TP was lost mainly through sediment transport. On average, the HR group decreased runoff, sediment, total N and P loss by 39% to 73% relative to the OVF treatment, whereas NVF treatment increased 3% to 30% of those (p < 0.05). Under four cultivation patterns, total dissolved nitrogen (TDN) was the dominant form, which accounted for 71–77% of TN. The average percentage of NO3−-N/TN was about 45–52%, much higher than NH4+-N/TN of around 8–10% in runoff. Total dissolved phosphorus (TDP) made up about 48–59% of TP in runoff. Redundancy analysis (RDA) showed that sediment, runoff, and soil pH were the three key factors controlling nutrient loss. In conclusion, OHFR is recommended because it consistently outperformed other patterns in terms of reducing runoff, sediment, and nutrient losses. [ABSTRACT FROM AUTHOR]

In this study, hybrid integration of MultiBoosting based on two artificial intelligence methods (the radial basis function network (RBFN) and credal decision tree (CDT) models) and geographic information systems (GIS) were used to establish landslide susceptibility maps, which were used to evaluate landslide susceptibility in Nanchuan County, China. First, the landslide inventory map was generated based on previous research results combined with GIS and aerial photos. Then, 298 landslides were identified, and the established dataset was divided into a training dataset (70%, 209 landslides) and a validation dataset (30%, 89 landslides) with ensured randomness, fairness, and symmetry of data segmentation. Sixteen landslide conditioning factors (altitude, profile curvature, plan curvature, slope aspect, slope angle, stream power index (SPI), topographical wetness index (TWI), sediment transport index (STI), distance to rivers, distance to roads, distance to faults, rainfall, NDVI, soil, land use, and lithology) were identified in the study area. Subsequently, the CDT, RBFN, and their ensembles with MultiBoosting (MCDT and MRBFN) were used in ArcGIS to generate the landslide susceptibility maps. The performances of the four landslide susceptibility maps were compared and verified based on the area under the curve (AUC). Finally, the verification results of the AUC evaluation show that the landslide susceptibility mapping generated by the MCDT model had the best performance. [ABSTRACT FROM AUTHOR]