Surface water quality deterioration is a serious problem in many rapidly urbanizing catchments in developing countries. There is currently a lack of studies that quantify water quality variation (deterioration or otherwise) due to both socio-economic and infrastructure development in a catchment. This paper investigates the causes of water quality changes over the rapid urbanization period of 1985–2009 in the Shenzhen River catchment, China and examines the changes in relation to infrastructure development and socio-economic policies. The results indicate that the water quality deteriorated rapidly during the earlier urbanization stages before gradually improving over recent years, and that rapid increases in domestic discharge were the major causes of water quality deterioration. Although construction of additional wastewater infrastructure can significantly improve water quality, it was unable to dispose all of the wastewater in the catchment. However, it was found that socio-economic measures can significantly improve water quality by decreasing pollutant load per gross regional production (GRP) or increasing labor productivity. Our findings suggest that sustainable development during urbanization is possible, provided that: (1) the wastewater infrastructure should be constructed timely and revitalized regularly in line with urbanization, and wastewater treatment facilities should be upgraded to improve their nitrogen and phosphorus removal efficiencies; (2) administrative regulation policies, economic incentives and financial policies should be implemented to encourage industries to prevent or reduce the pollution at the source; (3) the environmental awareness and education level of local population should be increased; (4) planners from various sectors should consult each other and adapt an integrated planning approach for socio-economic and wastewater infrastructure development.

Video data and algorithms have been driving advances in multi-object tracking (MOT). While existing MOT datasets focus on occlusion and appearance similarity, complex motion patterns are widespread yet overlooked. To address this issue, we introduce a new dataset called BEE24 to highlight complex motions. Identity association algorithms have long been the focus of MOT research. Existing trackers can be categorized into two association paradigms: single-feature paradigm (based on either motion or appearance feature) and serial paradigm (one feature serves as secondary while the other is primary). However, these paradigms are incapable of fully utilizing different features. In this paper, we propose a parallel paradigm and present the Two rOund Parallel matchIng meChanism (TOPIC) to implement it. The TOPIC leverages both motion and appearance features and can adaptively select the preferable one as the assignment metric based on motion level. Moreover, we provide an Attention-based Appearance Reconstruction Module (AARM) to reconstruct appearance feature embeddings, thus enhancing the representation of appearance features. Comprehensive experiments show that our approach achieves state-of-the-art performance on four public datasets and BEE24. Moreover, BEE24 challenges existing trackers to track multiple similar-appearing small objects with complex motions over long periods, which is critical in real-world applications such as beekeeping and drone swarm surveillance. Notably, our proposed parallel paradigm surpasses the performance of existing association paradigms by a large margin, e.g., reducing false negatives by 6% to 81% compared to the single-feature association paradigm. The introduced dataset and association paradigm in this work offer a fresh perspective for advancing the MOT field. The source code and dataset are available at https://github.com/holmescao/TOPICTrack.

In current literature, the first flush effect of urban runoff pollution has been studied and reported extensively. However, the effects of middle and final flushes on pollutant flushing were not given much attention. In addition, few previous studies have discussed the suitability of the widely used exponential wash-off model for describing the middle or final flush processes. In this paper, the Shiyan River catchment, a typical rapidly urbanizing catchment in China, is chosen as a study area to analyze the effects of first, middle and final flushes based on monitoring hydrographs and pollutographs. In order to simulate the middle and final flush processes observed in storm events, a new, realistically simple, parsimonious model (named as logistic wash-off model) is developed with the assumption that surface pollutant loads available for wash-off increase with cumulative runoff volume following a logistic curve. The popular exponential wash-off model and the newly developed model are used and compared in simulating the flush processes in storm events. The results indicate that all the three types of pollutant flushing are observed in the experiment; however, the first flush effect is weak, while the middle and final flush effects are substantial. The exponential model has performed well in simulating the first flush process but failed to simulate well the middle and final flush processes. However, the logistic wash-off model has effectively simulated all the three types of pollutant flush, and particularly, it has performed better in simulating the middle and final flush processes than the exponential model. © 2016 Elsevier B.V.