Using neural networks and cellular automata for modelling intra‐urban land‐use dynamics

Empirical models designed to simulate and predict urban land‐use change in real situations are generally based on the utilization of statistical techniques to compute the land‐use change probabilities. In contrast to these methods, artificial neural networks arise as an alternative to assess such probabilities by means of non‐parametric approaches. This work introduces a simulation experiment on intra‐urban land‐use change in which a supervised back‐propagation neural network has been employed in the parameterization of several biophysical and infrastructure variables considered in the simulation model. The spatial land‐use transition probabilities estimated thereof feed a cellular automaton (CA) simulation model, based on stochastic transition rules. The model has been tested in a medium‐sized town in the Midwest of São Paulo State, Piracicaba. A series of simulation outputs for the case study town in the period 1985–1999 were generated, and statistical validation tests were then conducted for the best results, based on fuzzy similarity measures.     

Selective omission of road features based on mesh density for automatic map generalization

Selection of roads is an intractable generalization operation due to the difficulty in retaining the density difference and connectivity of a road network. This paper proposes a novel approach of selective omission for roads based on mesh density. The density of a road network and its local variations are calculated using meshes as units. Since maps at different scales usually reveal different densities, different density thresholds for road networks are determined on the basis of theoretical analysis and empirical study of mesh densities on maps at different scales. The selection process starts with the identification of the meshes that have a density beyond the threshold. The mesh with the largest density is first treated. Its bounding road segments are ordered according to their relative importance. The least important segment is eliminated. The remaining segments are then merged with the adjacent mesh, thus forming a new mesh. The selection procedure is repeated until none of the meshes has a density beyond the threshold. Such a process of eliminating road segments and merging meshes can ensure the road network connectivity. In this study, the meshes are classified depending on the types of road segment. For the different mesh types, their density thresholds are set to be different, which can be used as an indicator for the preservation of the density difference. This proposed approach considers topological, geometric and semantic properties of the road network. It was applied to two sets of road networks, and the results of selection are convincing. This methodology has now been adopted for the updating of 1:50,000 maps of China.     

Solutions for 4D cadastre – with a case study on utility networks

The increasing complexity and flexibility of modern land use requires that cadastres need to manage information on the third and temporal (fourth) dimension. This article considers the registration of legal space of utility networks in cadastre in this 3D + time (=4D) context. A requirement analysis in three countries that have methods to register utility networks complying with their legal, organizational and technical structure (Turkey, the Netherlands and Queensland, Australia) is the basis for three alternatives for 4D cadastre to register utility networks. The three alternatives are analysed with respect to legal, organizational and technical cadastral requirements. This article presents a case study and a prototype from the Netherlands. In this country by law utilities are considered to be real estate objects with obligatory registration of ownership and geometry. This study shows that the 3D space and separate temporal attributes approach (state-based model) is a very promising solution to maintain temporal changes of utility networks and that this approach is to be preferred above the current practice, where the 3D and temporal aspects are not considered when registering a network.     

Extended hybrid meshing algorithm for multiresolution terrain models

Hybrid terrains are a convenient approach for the representation of digital terrain models, integrating heterogeneous data from different sources. In this article, we present a general, efficient scheme for achieving interactive level-of-detail rendering of hybrid terrain models, without the need for a costly preprocessing or resampling of the original data. The presented method works with hybrid digital terrains combining regular grid data and local high-resolution triangulated irregular networks. Since grid and triangulated irregular network data may belong to different datasets, a straightforward combination of both geometries would lead to meshes with holes and overlapping triangles. Our method generates a single multiresolution model integrating the different parts in a coherent way, by performing an adaptive tessellation of the region between their boundaries. Hence, our solution is one of the few existing approaches for integrating different multiresolution algorithms within the same terrain model, achieving a simple interactive rendering of complex hybrid terrains.     

Comparison of an artificial neural network and a conceptual rainfall–runoff model in the simulation of ephemeral streamflow

ABSTRACT

The rainfall–runoff process is governed by parameters that can seldom be measured directly for use with distributed models, but are rather inferred by expert judgment and calibrated against historical records. Here, a comparison is made between a conceptual model (CM) and an artificial neural network (ANN) for their ability to efficiently model complex hydrological processes. The Sacramento soil moisture accounting model (SAC-SMA) is calibrated using a scheme based on genetic algorithms and an input delay neural network (IDNN) is trained for variable delays and hidden layer neurons which are thoroughly discussed. The models are tested for 15 ephemeral catchments in Crete, Greece, using monthly rainfall, streamflow and potential evapotranspiration input. SAC-SMA performs well for most basins and acceptably for the entire sample with R² of 0.59–0.92, while scoring better for high than low flows. For the entire dataset, the IDNN improves simulation fit to R² of 0.70–0.96 and performs better for high flows while being outmatched in low flows. Results show that the ANN models can be superior to the conventional CMs, as parameter sensitivity is unclear, but CMs may be more robust in extrapolating beyond historical record limits and scenario building.

EDITOR M.C. Acreman; ASSOCIATE EDITOR not assigned     

Artificial neural networks and empirical equations to estimate daily evaporation: application to Lake Vegoritis,Greece

ABSTRACT

Evaporation is one of the most important components in the energy and water budgets of lakes and is a primary process of water loss from their surfaces. An artificial neural network (ANN) technique is used in this study to estimate daily evaporation from Lake Vegoritis in northern Greece and is compared with the classical empirical methods of Penman, Priestley-Taylor and the mass transfer method. Estimation of the evaporation over the lake is based on the energy budget method in combination with a mathematical model of water temperature distribution in the lake. Daily datasets of air temperature, relative humidity, wind velocity, sunshine hours and evaporation are used for training and testing of ANN models. Several input combinations and different ANN architectures are tested to detect the most suitable model for predicting lake evaporation. The best structure obtained for the ANN evaporation model is 4-4-1, with root mean square error (RMSE) from 0.69 to 1.35 mm d^?1 and correlation coefficient from 0.79 to 0.92.

EDITOR M.C. Acreman

ASSOCIATE EDITOR not assigned     

Improvement of artificial neural networks to predict daily streamflow in a semi-arid area

ABSTRACT

The application of artificial neural networks (ANNs) has been widely used recently in streamflow forecasting because of their ?exible mathematical structure. However, several researchers have indicated that using ANNs in streamflow forecasting often produces a timing lag between observed and simulated time series. In addition, ANNs under- or overestimate a number of peak flows. In this paper, we proposed three data-processing techniques to improve ANN prediction and deal with its weaknesses. The Wilson-Hilferty transformation (WH) and two methods of baseflow separation (one parameter digital filter, OPDF, and recursive digital filter, RDF) were coupled with ANNs to build three hybrid models: ANN-WH, ANN-OPDF and ANN-RDF. The network behaviour was quantitatively evaluated by examining the differences between model output and observed variables. The results show that even following the guidelines of the Wilson-Hilferty transformation, which significantly reduces the effect of local variations, it was found that the ANN-WH model has shown no significant improvement of peak flow estimation or of timing error. However, combining baseflow with streamflow and rainfall provides important information to ANN models concerning the flow process operating in the aquifer and the watershed systems. The model produced excellent performance in terms of various statistical indices where timing error was totally eradicated and peak flow estimation significantly improved.

Editor D. Koutsoyiannis; Associate editor Y. Gyasi-Agyei     

Monthly streamflow forecasting using neuro-wavelet techniques and input analysis

ABSTRACT

Combinations of low-frequency components (also known as approximations) resulting from the wavelet decomposition are tested as inputs to an artificial neural network (ANN) in a hybrid approach, and compared to classical ANN models for flow forecasting for 1, 3, 6 and 12 months ahead. In addition, the inputs are rewritten in terms of the flow, revealing what type of information was being provided to the network, in order to understand the effect of the approximations on the forecasting performance. The results show that the hybrid approach improved the accuracy of all tested models, especially for 1, 3 and 6 months ahead. The input analyses show that high-frequency components are more important for shorter forecast horizons, while for longer horizons, they may worsen the model accuracy.     

Efficiently implementable algebra for distributed in-network spatial analysis

Existing sensor network query processors (SNQPs) have demonstrated that in-network processing is an effective and efficient means of interacting with wireless sensor networks (WSNs) for data collection tasks. Inspired by these findings, this article investigates the question as to whether spatial analysis over WSNs can be built upon established distributed query processing techniques, but, here, emphasis is on the spatial aspects of sensed data, which are not adequately addressed in the existing SNQPs. By spatial analysis, we mean the ability to detect topological relationships between spatially referenced entities (e.g. whether mist intersects a vineyard or is disjoint from it) and to derive representations grounded on such relationships (e.g. the geometrical extent of that part of a vineyard that is covered by mist). To support the efficient representation, querying and manipulation of spatial data, we use an algebraic approach. We revisit a previously proposed centralized spatial algebra comprising a set of spatial data types and a comprehensive collection of operations. We have redefined and re-conceptualized the algebra for distributed evaluation and shown that it can be efficiently implemented for in-network execution. This article provides rigorous, formal definitions of the spatial data types, points, lines and regions, together with spatial-valued and topological operations over them. The article shows how the algebra can be used to characterize complex and expressive topological relationships between spatial entities and spatial phenomena that, due to their dynamic, evolving nature, cannot be represented a priori.     

Comparison of five snow water equivalent estimation methods across categories

Snow water equivalent (SWE) is an important indicator used in hydrology, water resources, and climate change impact. There are various methods of estimating SWE (falling in 3 categories: indirect sensors, empirical models, and process‐based models), but few studies that provide comparison across these different categories to help users make decisions on monitoring site design or method selection. Five SWE estimation methods were compared against manual snow course data collected over 2 years (2015–2016) from the Dorset Environmental Science Centre, including the gamma‐radiation‐based CS725 sensor, 3 empirical estimation models (Sexstone snow density model, McCreight & Small snow density model, and a meteorology‐based model), and the University of British Columbia Watershed Model snow energy‐balance model. Snow depth, density, and SWE were measured at the Dorset Environmental Science Centre weather station in south‐central Ontario, on a daily basis over 6 winters from 2011 to 2016. The 2 snow density‐based models, requiring daily snow depth as input, gave the best performance (R² of .92 and .92 for McCreight & Small and Sexstone models, respectively). The CS725 sensor that receives radiation coming from soil penetrating the snowpack provided the same performance (R² = .92), proving that the sensor is an applicable method, although it is expensive. The meteorology‐based empirical model, requiring daily climate data including temperature, precipitation and solar radiation, gave the poorest performance (R² = .77). The energy‐balance‐based University of British Columbia Watershed Model snow module, only requiring climate data, worked better than the empirical meteorology‐based model (R² = .9) but performed worse than the density models or CS725 sensor. Given differences in application objectives, site conditions, and budget, this comparison across SWE estimation methods may help users choose a suitable method. For ongoing and new monitoring sites, installation of a CS725 sensor coupled with intermittent manual snow course measurements (e.g., weekly) is recommended for further SWE method estimation testing and development of a snow density model.