Predicting quantiles of water quality from catchment characteristics

Water quality is often highly variable both in space and time, which poses challenges for modelling the more extreme concentrations. This study developed an alternative approach to predicting water quality quantiles at individual locations. We focused on river water quality data that were collected over 25 years, at 102 catchments across the State of Victoria, Australia. We analysed and modelled spatial patterns of the 10th, 25th, 50th, 75th and 90th percentiles of the concentrations of sediments, nutrients and salt, with six common constituents: total suspended solids (TSS), total phosphorus (TP), filterable reactive phosphorus (FRP), total Kjeldahl nitrogen (TKN), nitrate-nitrite (NO_x), and electrical conductivity (EC). To predict the spatial variation of each quantile for each constituent, we developed statistical regression models and exhaustively searched through 50 catchment characteristics to identify the best set of predictors for that quantile. The models predict the spatial variation in individual quantiles of TSS, TKN and EC well (66%–96% spatial variation explained), while those for TP, FRP and NO_x have lower performance (37%–73% spatial variation explained). The most common factors that influence the spatial variations of the different constituents and quantiles are: annual temperature, percentage of cropping land area in catchment and channel slope. The statistical models developed can be used to predict how low- and high-concentration quantiles change with landscape characteristics, and thus provide a useful tool for catchment managers to inform planning and policy making with changing climate and land use conditions.     

Determination of subcatchment and watershed boundaries in a complex and highly urbanized landscape

Urban development significantly alters the landscape by introducing widespread impervious surfaces, which quickly convey surface run‐off to streams via stormwater sewer networks, resulting in “flashy” hydrological responses. Here, we present the inadequacies of using raster‐based digital elevation models and flow‐direction algorithms to delineate large and highly urbanized watersheds and propose an alternative approach that accounts for the influence of anthropogenically modified land cover. We use a semi‐automated approach that incorporates conventional drainage networks into overland flow paths and define the maximal run‐off contributing area. In this approach, stormwater pipes are clustered according to their slope attributes, which define flow direction. Land areas drained by each cluster and contributing (or exporting) flow to a topographically delineated catchment were determined. These land masses were subsequently added or removed from the catchment, modifying both the shape and the size. Our results in a highly urbanized Toronto, Canada, area watershed indicate a moderate net increase in the directly connected watershed area by 3% relative to a topographically forced method; however, differences across three smaller scale subcatchments are greater. Compared to topographic delineation, the directly connected watershed areas of both the upper and middle subcatchments decrease by 5% and 8%, respectively, whereas the lower subcatchment area increases by 15%. This is directly related to subsurface storm sewer pipes that cross topographic boundaries. When directly connected subcatchment area is plotted against total streamflow and flashiness indices using this method, the coefficients of variation are greater (0.93 to 0.97) compared to the use of digital elevation model‐derived subcatchment areas (0.78 to 0.85). The accurate identification of watershed and subcatchment boundaries should incorporate ancillary data such as stormwater sewer networks and retention basin drainage areas to reduce water budget errors in urban systems.     

A machine learning approach for predicting computational intensity and domain decomposition in parallel geoprocessing

ABSTRACT

High performance computing is required for fast geoprocessing of geospatial big data. Using spatial domains to represent computational intensity (CIT) and domain decomposition for parallelism are prominent strategies when designing parallel geoprocessing applications. Traditional domain decomposition is limited in evaluating the computational intensity, which often results in load imbalance and poor parallel performance. From the data science perspective, machine learning from Artificial Intelligence (AI) shows promise for better CIT evaluation. This paper proposes a machine learning approach for predicting computational intensity, followed by an optimized domain decomposition, which divides the spatial domain into balanced subdivisions based on the predicted CIT to achieve better parallel performance. The approach provides a reference framework on how various machine learning methods including feature selection and model training can be used in predicting computational intensity and optimizing parallel geoprocessing against different cases. Some comparative experiments between the approach and traditional methods were performed using the two cases, DEM generation from point clouds and spatial intersection on vector data. The results not only demonstrate the advantage of the approach, but also provide hints on how traditional GIS computation can be improved by the AI machine learning.     

A land-use mix allocation model considering adjacency,intensity, and proximity

ABSTRACT

Land-Use Mix (LUM) refers to the strategy of integrating complementary functions within a building or area. While LUM has become a dominant approach in urban planning, its actual benefits and vision for spatial planning remain unclear. To clarify this issue, this study discerns the spatial features of land-use patterns depending on the compatibilities among land-use categories. Accordingly, this study introduces three LUM measures – adjacency, intensity, and proximity – to identify differences in the spatial distribution of land-use categories. Based on these measures, a land-use allocation model is developed to specify spatial patterns satisfying the given compatibilities. This model is tested by applying the concept of the neighborhood unit on a case study of normative land-use patterns subject to specified compatibilities. The results describe spatial features of four compatibility sets, including a set exhibiting a compatibility conflict between the same land-use pair and LUM measures when, for example, a given land-use pair is compatible in terms of intensity but incompatible in terms of proximity. Understanding the spatial features of a normative land-use pattern that satisfies various possible compatibilities will facilitate the incorporation of the LUM approach into local planning guidance and zoning ordinances.