Multiple Human Activities Research Articles

ToiletBuilder: A PU-Learning-Based Model for Selecting New Public Toilet Locations

With increasing expansion and urbanization of cities, the gap is constantly widening between the current provision of urban public toilets and the fast-growing toileting demand. Building new ones becomes a promising way to alleviate such issue. Nevertheless, where to build them in a city is challenging. Different from other location selection (e.g., commercial sites) problems, the selection of public toilet locations is hard to quantify and evaluate. On one hand, the toileting demand that determines whether the new public toilet is needed cannot be measured accurately. On the other hand, the modeling of the toileting demand is also complicated, being influenced by multiple factors, e.g., human mobility, human activity, and geographical characteristics. In this article, we propose a novel data-driven framework named ToiletBuilder to address it, which consists of three components, i.e., region identification, region representation, and region classification. Specifically, region identification obtains many reachable regions with the reasonable size. Region representation extracts city-specific features from multiple urban data to characterize location selection influencing factors for each region. A deep embedding model is further applied to learn a high-order and concise semantic representation. By labeling some regions with the true positive label (i.e., having public toilets served in these regions) in advance, region classification trains a positive-unlabeled (PU) learning model from these samples to identify unlabeled positive ones. Finally, we conduct extensive experiments based on four real-world data sets including road network, river network, taxi trajectory, and POI data, in the city of Chongqing, China. Results demonstrate the effectiveness of our proposed approach.

Spatial variation in biodiversity loss across China under multiple environmental stressors.

Biodiversity is essential for the maintenance of ecosystem health and delivery of the Sustainable Development Goals. However, the drivers of biodiversity loss and the spatial variation in their impacts are poorly understood. Here, we explore the spatial-temporal distributions of threatened and declining ("biodiversity-loss") species and find that these species are affected by multiple stressors, with climate and human activities being the fundamental shaping forces. There has been large spatial variation in the distribution of threatened species over China's provinces, with the biodiversity of Gansu, Guangdong, Hainan, and Shaanxi provinces severely reduced. With increasing urbanization and industrialization, the expansion of construction and worsening pollution has led to habitat retreat or degradation, and high proportions of amphibians, mammals, and reptiles are threatened. Because distributions of species and stressors vary widely across different climate zones and geographical areas, specific policies and measures are needed for preventing biodiversity loss in different regions.

Distribution pattern of macroinvertebrate community and its relationships with environmental factors in the Yongding River Basin

Human disturbances have altered global stream ecosystems in multiple ways. Understan-ding spatio-temporal dynamics of aquatic organism composition and their relationships with environmental factors play an important role in revealing changes in the structure and function of river ecosystems. The macroinvertebrates and environmental parameters were sampled in the Yongding River Basin at spring (March), summer (July), and autumn (November) in 2017. The results showed that chironomids were the dominant group in the Yongding River Basin, including Orthocladius, Chironomus, Glyptotendipes, and Polypedilum. Results of the cluster analysis showed significant spatial variations of macroinvertebrate community. Sampling sites were clustered into two groups based on the similarities of macroinvertebrate composition in the Yongding River. Sampling sites belonging to group 1 mainly located in the Sanggan River and Dongyang River, with Orthocladius sp.being the dominant species. Sampling sites belonging to group 2 mainly located in the Yang River, Guishui River and main section of the Yongding River, with Glyptotendipes sp. being the dominant species. The biomass, Shannon index, Margalef index, Pielou evenness index and taxa richness of group 2 were significantly higher than those of group 1. Higher values of biological indices indicated better ecological condition of the stream ecosystem. The density, biomass and diversity indices of summer were significantly lower than those of both spring and autumn. Results of Canonical correspondence analysis (CCA) showed that the spatio-temporal distribution of macroinvertebrate in the Yongding River was mainly influenced by water temperature, ammonium, pH, dissolved oxygen, and total phosphorus. From the perspective of hydrological characteristics, flow velocity and flow capacity were significantly correlated. The significant positive correlations between hydrological parameters of velocity and capacity and biological indices of density and EPT% in group 2 indicated the preference of fast flow to sensitive taxa of EPT (Ephemeroptera, Plecoptera, and Trichoptera). The negative correlations between flow capacity and biodiversity indices implied the adverse effects of high flow on macroinvertebrate biodiversity. Our results confirmed that the composition and biodiversity of macroinvertebrate were affected by both water environment and hydrology, which reflected the multiple human activities in aquatic ecosystems of the Yongding River. It is important to control water contamination and sustain environmental flow at the specific reaches for the restoration of Yongding River.

The Marine Plastic Litter Issue: A Social-Economic Analysis

The issue of marine plastic litter pollution is multifaceted, cross-sectoral, and ongoing in the absence of appropriate management measures. This study analysed the issue of marine plastic litter pollution in the context of the Descriptor 10 of the Marine Strategy Framework Directive and Good Environmental Status of the oceans and seas. The Driver-Pressure-State-Impact-Response (DPSIR) framework was used to assess the causes, effects, and management measures to changes in the marine environment resulting from marine plastics pollution. We noted that less than 10 peer-reviewed publications have applied the Driver-Pressure-State-Impact-Response (DPSIR) model to the issue of marine plastics pollution. Some basic needs such as food security, movement of goods and services, and shelter are also some of the major drivers of marine plastic pollution. The use of plastics is linked to multiple economic sectors (fisheries, agriculture, transport, packaging, construction) and other human activities. A significant amount of the resulting pressures came from the economic sectors for packaging and construction. State changes occurred at the environmental (contamination and bioaccumulation), ecosystem (ingestion of plastics, ghost fishing) and ecosystem service levels (supply of sea food, salt and cultural benefits), with possible loss of jobs and income being some of the observed impacts on human welfare. Responses as management measures, which are tailored to meet each component of the DPSIR framework, were identified. These included policies, regulations, technological advancement and behavioural change. The research acknowledges the issue of marine plastics pollution as a global environmental problem and recommends a trans-disciplinary approach, involving all types of stakeholders. Future research and analysis applying the DPSIR framework will be useful to provide the information necessary for the effective, adaptive management of litter pollution by marine plastics.

An integrated methodology for assessing ecological and economic impacts for marine management: A case study for abrasion and mobile fishing gear effects

Ensuring a balance of conservation and human activities within the marine environment requires complex decisions on how to achieve conservation objectives whilst considering the likely economic and social costs and benefits of such decisions. We assessed the risk of adverse effects from abrasion caused by bottom fishing, as a case study, to demonstrate an integrated spatial management methodology. Risk was quantified in the Galway Bay, Ireland region using a 1 km grid, as the ratio between mortality and recovery of benthic infauna exposed to abrasion effects estimated from fishing data. Risk was considered high if annual mortality exceeded recovery. A Bayesian network model, incorporating ecological and economic data, compared three management scenarios: i) business as usual (BAU), ii) exclusion of fishing vessels over 18 m within 6 nautical miles (68% of study area) (Length) and iii) a network of marine protected areas (MPAs) closed to all fishing vessels (11% of study area) (MPA). The BAU scenario indicated that risk was high for 85.6% of the study area. The length scenario resulted in a 14.6% reduction in high risk cells, affecting €0.9 M of landings, €0.7 M of indirect output and 19 full time equivalent (FTE) employees. The MPA scenario resulted in a 1.7% reduction of high risk cells. This scenario affected €0.17M in landings, €0.13M in indirect outputs and 3 FTE employees. We present a method which provides a quantitative, spatial, risk and economic impact assessment of the effects of human activities and pressures on the marine environment that assesses potential costs and benefits of management options to mitigate against these pressures. Future development of a mortality-recovery ratio for defined time periods would enable application across multiple habitats, human activities and pressures.

Differential Channel-State-Information-Based Human Activity Recognition in IoT Networks

In this article, we recognize multiple human activities in an Internet-of-Things (IoT) network using differential channel state information (CSI) of the available wireless fidelity (Wi-Fi) signals. Different human activities in the Wi-Fi environment lead to multipath fading, resulting in a change of CSI for each activity. This CSI is sensed by smart IoT devices, such as smartphones, tablets, and laptops for activity recognition. The use of differential CSI mitigates the offset and background noise. Another advantage of the proposed method is that it eliminates the requirement of traditional wearable activity recognition sensors, such as gyroscope, pedometers, and accelerometers. A long short-term memory (LSTM) model is used for automatic feature extraction and classification of human activities from the differential CSI. Training the LSTM model with the phase of differential denoised CSI significantly improves the classification accuracy. The results show a good tradeoff between model complexity and classification accuracy, thereby ensuring better performance as compared to the previous state-of-the-art methods.

Marine debris on a tropical coastline: Abundance, predominant sources and fate in a region with multiple activities (Fortaleza, Ceará, northeastern Brazil)

The aim of the present study was to perform the first assessment of the abundance and classification of marine debris as well as determine the sources, transport and fate of this debris on an urbanized coast with multiple human activities. More than 80% of the marine debris was composed of synthetic materials. The beached marine debris was classified according to size. Meso-debris accounted for the highest portion of contamination (55%), followed by macro-debris (25.1%) and small debris (19.9%). Contamination by debris, such as cotton swabs (31%) and lollipop sticks (36.8%) accounted for the largest portion of the small debris class. Human recreational activities were the predominant source of debris, followed by navigation/fishing activities, domestic activities and industrial/port activities. The assessment of the predominance of human activities and the results of the model revealed a larger contribution of debris from recreational activities on nearby beaches on the small to larger scale and that rivers exert less of an influence due to the fact that they do not flow the entire year.

Migration Corridors and Threats in the Gulf of Mexico and Florida Straits for Loggerhead Sea Turtles

Along migration corridors, animals can face natural and anthropogenic threats that differ from those in breeding and non-breeding residence areas. Satellite telemetry can aid in describing the timing and location of these migrations. We use this tool with switching state-space modeling and line kernel density estimates to identify migration corridors of post-nesting adult female loggerhead sea turtles (Caretta caretta, n = 89 tracks) that nested at five beaches in the Gulf of Mexico. Turtles migrated in both neritic and oceanic areas of the Gulf of Mexico with some exiting the Gulf. High-use migration corridors were found in neritic areas to the west of Florida and also in the Florida Straits. Repeat tracking of post-nesting migrations for eight turtles showed variability in track overlap, ranging from ~13-82% of tracks within 10 km of each other. Migration primarily occurred in July and August. We document the longest known post-nesting migration to-date of a wild adult female loggerhead of > 4300 km, along with an apparent stopover of about one month. Migration corridors overlaid on three spatially explicit anthropogenic threats (shipping density, commercial line fishing, and shrimp trawling) showed hotspots in the Florida Straits, off the northwest Florida coast and off the coast of Tampa Bay. Identifying where and at what intensity multiple human activities and natural processes most likely occur is a key goal of Cumulative Effects Assessments. Our results provide the scientific information needed for designing management strategies for this threatened species. Information about this loggerhead migration corridor can also be used to inform adaptive management as threats shift over time.

Linking threat maps with management to guide conservation investment

Stressors to marine ecosystems are increasing, driven by human activities in the sea and on land, and climate change. Cumulative impact maps highlight regions affected by multiple human activities, but efficient conservation investment requires linking dominant pressures to management actions that best address the particular drivers of impacts. We rebuild cumulative impact maps by stressor type (climate change, marine and land) at a global scale to evaluate the expected effectiveness of various management strategies for all coastal territories. Average cumulative impact from non-marine stressors (climate and land) was double those of marine impacts at a national level. The greatest climate impacts by country were in the waters of Pacific Island and Antarctic territories; in the Caspian Sea region and East-Asia for land impacts; and in the waters of European, East-Asian and Caribbean countries for marine impacts. We developed a conservation-effectiveness indicator for the 10 worst-impacted countries in each of the three stressor categories. The indicator considered common management tools for each stressor category: ecosystem-based adaptation and disaster risk reduction (climate), marine protected areas (marine) and integrated coastal management (land). Key disparities were found between broad-scale management of marine ecosystems and the dominant stressors, with existing management in tropical island nations likely insufficient to address intense impacts from climate change. These countries also typically had low performance on governance indicators, suggesting challenges in implementing new mitigation. We highlight trade-offs in making decisions for stressor mitigation and offer strategic guidance on identifying locations to target management of marine, land, or climate impacts.

Decadal loss of paddy fields driven by cumulative human activities in the Three Gorges Reservoir area, China

AbstractPaddy fields are valuable agricultural land resources in the Three Gorges Reservoir area of China but have been threatened by the expansion and intensification of human activities over recent decades. How paddy fields evolve under human disturbances remains poorly understood, which is critical for land use planning to compromise agricultural production and ecosystem conservation as limited land resource is available. In this study, the spatial and temporal patterns of change in paddy fields were depicted and principal driving forces were explained. Mapping of paddy fields with LANDSAT satellite images reveals a constant declining trend by 14.3% in the period 1990–2015. The gross loss area is 903.3 km2, with an annual average decreasing rate being 36.1 km2·yr−1. The transformation matrix between paddy fields and other land uses explicitly explains the impact trajectories of multiple human activities. Specifically, 39.6% (408.1 km2) of the gross loss was converted to built‐up lands, reflecting town translocation and human resettlement (involuntary population migration) proposed by construction of the Three Gorges Dam, as well as urbanization, industrial development and infrastructure expansion along with economic development. The area converted to forests accounts for 26.4% (238.2 km2), which could be ascribed to land abandonment and the reforestation campaign. 23.1% (208.7 km2) was converted to dry croplands by shifts in farmers' preference on crop selection following socioeconomic changes (e.g., voluntary rural population migration, land use policy adjustment, commercial farming enterprises expansion). 46.5 km2 was converted to water bodies, indicating the effect of impoundment of the Three Gorges Reservoir.

Spatial modelling of cumulative human pressure in the tropical coastscape of Zanzibar, Tanzania

Coastal landscapes and seascapes (‘coastscapes’) are under immense anthropogenic pressures emanating from multiple human activities. We applied spatial modelling to characterise the cumulative pressures in the coastscape of Zanzibar’s main island Unguja (hereafter Zanzibar), in the western Indian Ocean. We modified the European Union’s list of the human pressures in this context, conjointly covering Zanzibar’s marine and terrestrial environments. We applied a widely accepted algorithm for modelling cumulative human pressure. Spatial data were gathered from various sources and, where primary data were absent, model results were used as proxies. We used a hexagon lattice with a resolution of 0.5 km2 to map the anthropogenic pressures within the coastscape. Five typologies of coastscape were identified. A total of 283.55 km (68.86%) of the island’s coastscape is under moderate to high anthropogenic pressure, while the remaining 128.06 km (31.14%) is under low to moderate pressure. Cumulative human pressures are highest in the urbanised coastal areas, especially Zanzibar City, and in the growth zones. Principal components analysis was used to further explore the coexistence of separate human pressures and their contributions to the cumulative human pressure. Predominant pressures within the coastscape were litter, noise, the introduction of synthetic compounds, and inputs of organic matter. We recommend the use of these results for coastal-zone management in Zanzibar and for prioritisation of the management efforts.

Relative impacts of multiple human stressors in estuaries and coastal waters in the North Sea–Baltic Sea transition zone

The objectives of this study are 1) to map the potential cumulative impacts of multiple human activities and stressors on the ecosystems in the transition zone between the North Sea and Baltic Sea, for Danish waters 2) to analyse differences in stressor contribution between the European Union’s Marine Strategy Framework Directive (MSFD, off-shore waters) and Water Framework Directive (WFD, coastal waters), and 3) to assess the local relative importance of stressors for 14 areas along a land-sea gradient, from inner fjords or coastal areas to offshore waters. The mapping of cumulative impacts is anchored in 35 datasets describing a broad range of human stressors and 47 ecosystem components ranging from phytoplankton over benthic communities to fish, seabirds and marine mammals, which we combined by means of a widely used spatial human impact model. Ranking of the stressor impacts for the entire study area revealed that the top five stressors are: ‘Nutrients’, ‘Climate anomalies’, ‘Non-indigenous species’, ‘Noise’ and ‘Contaminants’. The gradient studies showed that some stressors (e.g. ‘Nutrients’, ‘Shipping’ and ‘Physical modification’) have a relatively higher impact within the fjord/estuarine systems whilst others (e.g. ‘Fisheries’, ‘Contaminants’ and ‘Noise’) have relatively higher impact in the open waters. Beyond mapping of cumulative human impacts, we discuss how the maps can be used as an analytical tool to inform ecosystem-based management and marine spatial planning, using the MSFD and WFD as examples.

Assessing the Ecological Sensitivity of Coastal Marine Ecosystems: A Case Study in Xiamen Bay, China

Coastal marine ecosystems are sensitive to anthropogenic stressors and environmental change. The Chinese Government proposed an ecosystem-based marine spatial planning scheme called the “Marine Ecological Red Line” (MERL), aimed at protecting ecologically sensitive areas. The assessment and mapping of ecosystem sensitivity provide important tools for regional MERL and setting conservation priorities. In this study, an integrated framework for the comprehensive sensitivity assessment of coastal marine ecosystems was proposed. This framework had two components: endogenous sensitivity and exogenous sensitivity. A weighted summation method was used along with a GIS-based spatial analysis to calculate regional sensitivity quantitatively. Taking Xiamen Bay in China as a case study, the approach integrated spatial data on the distribution of marine habitats and multiple human activities. An index system (including 26 indicators) was established for the case study. Five areas with different levels of sensitivity were delineated. In the study region of 1281 km2, areas with extremely high and high sensitivity covered 87 km2 and 235 km2, respectively, and were hypothesized to be priority areas of conservation and regulation. Areas with the highest sensitivity were located closer to Jiulongjiang Estuary and along the coastline of Xiang’an. Areas with lower sensitivity were also mapped in Tong’an Bay, representing possible areas that could accommodate future industrial or intensive human use. Comparative analyses between endogenous sensitivity and exogenous sensitivity allowed the presence of development-protection conflict zones to be identified. The results provide an important scientific basis for MERL decisions. In addition, targeted management strategies were proposed for Xiamen Bay. This study presents an operational approach to provide relevant scientific knowledge on the process of ecosystem-based marine spatial planning, facilitating policy-making decisions in sustainable coastal and marine management.

Environmental controls and anthropogenic impacts on deep-sea sponge grounds in the Faroe-Shetland Channel, NE Atlantic: the importance of considering spatial scale to distinguish drivers of change

AbstractDetermining the scale of anthropogenic impacts is critical in order to understand ecosystem effects of human activities, within the context of changes caused by natural environmental variability. We applied spatial eigenfunction analysis to disentangle effects of anthropogenic drivers from environmental factors on species assembly in the Faroe-Shetland Channel (FSC), in the northeast Atlantic. We found that the species assembly considered here was structured at both small and large spatial scales. Specifically, substrate types, distance to oil wells and pipelines, the presence of objects and demersal fishing (both static and mobile) appeared significant in explaining large spatial scale species assembly structures. Conversely, temperature and variance in temperature shaped the species community across smaller spatial scales. Mobile scavenger species were found in areas impacted by demersal fishing. Oil and gas structures seemed to provide a habitat for a range of species including the commercially important fishes Molva sp. and Sebastes sp. These results demonstrate how the benthic ecosystem in the FSC has been shaped by multiple human activities, at both small and large spatial scales. Only by sampling datasets covering several sites, like in this study, can the effects of anthropogenic activities be separated from natural environmental controls.

A new approach to separating the impacts of climate change and multiple human activities on water cycle processes based on a distributed hydrological model

The impacts of human activities on the natural water cycle have become increasingly severe due to population growth and economic development. Therefore, it is important to evaluate the contributions of different human activities (e.g., land-use change, industrial water use, and domestic water use) on different hydrological variables. However, existing attribution methods have shortcomings when attributing the impacts of different human activities; for instance, human activities may only be considered as one impact factor, and the total impact of all factors may not be guaranteed to equal 100% of the total change. A new approach, using multiple scenario simulations from a distributed model, was developed in this study to overcome these shortcomings. All potential contributions for one impact factor were calculated by fixing the other impact factors at different statuses, and the arithmetic mean of all potential contributions was presented as the final result. This method ensures that the sum of the contributions of all impact factors equals the total change in hydrological variables between two periods, which was demonstrated mathematically. A case study of the Wei River Basin was used to evaluate the proposed methodology, which attributed four impact factors (climate change, land-use change, industrial and domestic water use, and agricultural irrigation water use) to three hydrological variables (evapotranspiration, rainfall infiltration, and streamflow drainage into the Yellow River). The results showed that the new method could attribute the impacts of the four factors to the changes of the three annual average hydrological variables between the period 1956–1980 and the period 1981–2005; these four factors accounted for 373.3%, −33.3%, −133.3%, and −106.7% of the change in evapotranspiration, 129.8%, −2.2%, −3.5%, and −24.1% of the change in rainfall infiltration, and 75.2%, 4.4%, 13.4%, and 7.0% of the change in streamflow drainage into the Yellow River, respectively. Moreover, the sum of the relative contributions of the four impact factors to the three hydrological variables was equal to 100% at an annual time scale. To avoid the limitations of existing methods, we recommend adopting the newly proposed method for evaluating the impacts of climate change and multiple human activities on water cycle processes.

Wi-Multi: A Three-Phase System for Multiple Human Activity Recognition With Commercial WiFi Devices

Channel state information-based activity recognition has gathered immense attention over recent years. Many existing works achieved desirable performance in various applications, including healthcare, security, and Internet of Things, with different machine learning algorithms. However, they usually fail to consider the availability of enough samples to be trained. Besides, many applications only focus on the scenario where only single subject presents. To address these challenges, in this paper, we propose a three-phase system Wi-multi that targets at recognizing multiple human activities in a wireless environment. Different system phases are applied according to the size of available collected samples. Specifically, distance-based classification using dynamic time warping is applied when there are few samples in the profile. Then, support vector machine is employed when representative features can be extracted from training samples. Lastly, recurrent neural networks is exploited when a large number of samples are available. Extensive experiments results show that Wi-multi achieves an accuracy of 96.1% on average. It is also able to achieve a desirable tradeoff between accuracy and efficiency in different phases.

Evaluating Landuse Change along Thiruvananthapuram Coast, South West Coast of India Using Geo-spatial Techniques

Coastal zone across the world is typified with intensive multiple human activities. Historic maps and multi-temporal satellite images are helpful in detecting the change in land use over the years. An attempt is made to evaluate the land use/land cover change along the coastal zone of Thiruvananthapuram using Geographic Information System (GIS) and Remote Sensing techniques. Landuse scenarios for 1988, 2000 and 2010 have been generated using Survey of India Topographical Sheets, Landsat ETM+ Image and Quickbird Image respectively. ERDAS software is used for image processing and land-use change is analysed using ArcGis 10.2.1 Software. The reasons and implications of landuse change in the coast are also studied through field investigation and valid conclusions are drawn. Agricultural land, clustered settlement area, a settlement with vegetation, sandy area and rocky area is the broad landuse categories delineated in the coastal zone. The analysis shows that the agricultural land area has decreased from 9.34% to 3.89% while the area under clustered settlement increased from 5.75 % to 25.65% during 1988-2010. This study can provide base information on land use scenario of the coast while framing out the ICZM plan.

Predicting future invaders and future invasions

Invasive alien species are a great threat to biodiversity and human livelihoods worldwide. The most effective way to limit their impacts and costs is to prevent their introduction into new areas. Identifying invaders and invasions before their occurrence would arguably be the most efficient strategy. Here, we provide a profiling method to predict which species-with which particular ecological characteristics-will invade, and where they could invade. We illustrate our approach with ants, which are among the most detrimental invasive species, as they are responsible for declines of numerous taxa, are involved in local extinctions, disturb ecosystem functioning, and impact multiple human activities. Based on statistical profiling of 1,002 ant species from an extensive trait database, we identify 13 native ant species with an ecological profile that matches that of known invasive ants. Even though they are not currently described as such, these species are likely to become the next global invaders. We couple these predictions with species distribution models to identify the regions most at risk from the invasion of these species: Florida and Central America, Brazil, Central Africa and Madagascar, Southeast Asia, Papua New Guinea Northeast Australia, and many islands worldwide. This framework, applicable to any other taxa, represents a remarkable opportunity to implement timely and specifically shaped proactive management strategies against biological invasions.

An Integrated Framework Combining Multiple Human Activity Features for Land Use Classification

Urban land use information is critical to urban planning, but the increasing complexity of urban systems makes the accurate classification of land use extremely challenging. Human activity features extracted from big data have been used for land use classification, and fusing different features can help improve the classification. In this paper, we propose a framework to integrate multiple human activity features for land use classification. Features were fused by constructing a membership matrix reflecting the fuzzy relationship between features and land use types using the fuzzy c-means (FCM) clustering method. The classification results were obtained by the fuzzy comprehensive evaluation (FCE) method, which regards the membership matrix as the fuzzy evaluation matrix. This framework was applied to a case study using taxi trajectory data from Nanjing, and the outflow, inflow, net flow and net flow ratio features were extracted. A series of experiments demonstrated that the proposed framework can effectively fuse different features and increase the accuracy of land use classification. The classification accuracy achieved 0.858 (Kappa = 0.810) when the four features were fused for land use classification.

Earth fissures developed within collapsible loess area caused by groundwater uplift in Weihe watershed, northwestern China

Earth fissures are geological hazards that frequently occur all over the world and greatly threaten the safety of people’s property and lives. This paper reports a case from a typical agricultural town in the Weihe watershed, northwest (NW) of China, Dali, which has been severely affected from earth fissures in recent years. A comprehensive survey was made on the geometric characteristics of the earth fissures developed in the study area through field investigations, trial trenches, boreholes, exploratory wells, and laboratory tests. The study indicates that fourteen nearly parallel fissures have developed in the area, with a fissure to fissure space of 140–350 m, and a maximum length of any of the given fissures of 500 m. The underlying faults, loess joints, and the Quaternary unconsolidated and collapsible sedimentary loess provide a favorable geological condition for the occurrence of the earth fissures. From 2003 to 2012, the groundwater level in the area rose about three meters due to agricultural irrigation and leakage of a water diversion canal. Collapsible deformation occurred in the loess, while seepage erosion expanded the original joints in the soil. The earth fissures occurred resulting from multiple geological environmental forces and human activities.