ABSTRACTMulti-scale landscape functions play a critical role in revealing intricate functional structures within large regions. However, previous studies on landscape functions have predominantly ...focused on a single macro or micro scale, impeding a holistic multi-scale understanding of the spatial distribution and heterogeneity of landscape functions. To address this gap, this study proposes a framework leveraging the power of big geodata to mine multi-scale landscape functions from parcel to entire urban agglomerations, as well as non-administrative divisions. Our study focuses on the Guangdong-Hong Kong-Macao Greater Bay Area (GBA) in China. Firstly, we integrated multi-source big geodata to derive parcel-scale landscape functions. Subsequently, we employed the Normalized Revealed Comparative Advantage index to derive landscape functions at broader scales, including towns, counties and cities. The effectiveness of our approach is validated through in-field investigations and comparisons with established policy planning positions. The outcomes not only offer distinctive planning insights at various scales but also highlight the versatility of big geodata in extracting landscape functions across scales. This study demonstrates that big geodata is adept at uncovering multi-scale landscape functions irrespective of administrative boundaries, providing valuable insights for fostering multi-scale regional coordinated development.
Display omitted
•GDV is found in low permeable HSU with surficial groundwater circulation.•GDV classes got validated with an ecohydrological index from botanical in situ data.•Differences in in situ ...vitality exist between Quercus species in GDV and non-GDV.
Groundwater dependent ecosystems (GDEs) are biodiversity hotspots and provide important ecosystem services. This study presents a novel multi-instrument concept for the local identification of groundwater dependent vegetation (GDV) in the Mediterranean. The concept integrates high-resolution Sentinel-2 remote sensing data with available geodata and requires in situ vegetation data for validation and calibration. The approach combines five criteria to identify GDV: 1) high vitality, and wetness during dry period, 2) low seasonal changes in vitality and leaf area, 3) low interannual changes in vitality, 4) high topographic potential of water accumulation and low water table depth, 5) high potential inflow dependency. Iso Cluster Unsupervised Classification (ICUC) was applied to identify GDV in the study area (Campania, Italy). Botanical field mapping was utilized for validating the remote sensing approach, as it exhibited significant differences between GDV and Non-GDV in terms of ecohydrological indicator values, leaf anatomy and phreatophyte coverage. According to a new simple ecohydrological rule set that considers phreatophyte cover and mean moisture value of non-phreatophyte species, 9% of vegetation plots are considered GDV and 33% likely GDV. 80% of all GDV derived from classification occur in hydrostratigraphic units (HSU) that are characterized by surficial groundwater circulation and low permeability. The overall accuracy of classifying likelihoods is 62.7%. For 14.6% of the plots, non-GDVs were classified as GDVs (false positives), and only one GDV plot has been classified falsely as non-GDV (false negative). Local results on GDV locations can be overlayed with aquifer use or aquifer reaction to climate change in order to identify GDV under threat and implement sustainable managements of groundwater resources.
Music can produce a positive effect in runners’ motivation and performance. Nevertheless, these effects vary depending on the user’s location, the emotions that she/he feels at each moment or the ...type of training session. In this paper, a context and emotion-aware system for the recommendation and playing of Spotify songs is presented. It consists in a location-based mobile application that interacts with a novel emotional wearable and a recommendation service that predicts the next song to be recommended. These predictions are performed by an intelligent system that combines artificial intelligent techniques with geodata and emotionally-annotated music. A wide variety of location-based services and music services available in Internet have been integrated into the recommender in order to support the decision-making process in a real environment. The final solution has been customized to be tested in the city of Zaragoza.
In the new digital scenario, new professional figures are needed to manage the spatial and environmental information: geoinformatics engineers are high level experts in technologies for measuring, ...georeferencing, managing, analyzing, visualizing and publishing spatial and time varying information, with a particular concern to environmental data. As the academic teaching is concerned, some universities in Europe propose courses in Geoinformatics. In Italy, Politecnico di Milano started in 2016 the first national MSc in Geoinformatics Engineering: this paper describes it.
•Convolutional network for predicting daily maps of the probability of a wildfire burn.•Convolutional networks demonstrate higher predictive accuracy and map quality.•Exploratory feature statistical ...importance metrics improves model transparency.
Wildfire continues to be a major environmental problem in the world. To help land and fire management agencies manage and mitigate wildfire-related risks, we need to develop tools for mapping those risks. Big geodata—in the form of remotely sensed images, ground-based sensor observations, and topographical datasets—can help us characterize the dynamics of wildfire related events. In this study, we design a deep fully convolutional network, called AllConvNet, to produce daily maps of the probability of a wildfire burn over the next 7 days. We applied it to burns in Victoria, Australia for the period of 2006–2017. Fifteen factors that were extracted from six different datasets and resulted into 29 quantitative features, were selected as input to the network. We compared it with three baseline methods: SegNet, multilayer perceptron, and logistic regression. AllConvNet outperforms the other three baseline methods in four of the six quantitative metrics considered. AllConvNet and SegNet provide smoother and more regularized predicted maps, with SegNet providing greater sensitivity in dificriminating less wildfire-prone locations. Input feature statistical importance was measured for all the networks and compared against logistic regression coefficients. Total precipitation, lightning flash density, and land surface temperature occur to be consistently highly weighted by all models while terrain aspect components, wind direction components, certain land cover classes (such as crop field and woodland), and distance from power lines are ranked on the lower end. We conclude that wild-fire burn prediction methods based on deep learning present quantitative and qualitative gains.
Distributed, physics-based hydrologic models require spatially explicit specification of parameters related to climate, geology, land-cover, soil, and topography. Extracting these parameters from ...national geodatabases requires intensive data processing. Furthermore, mapping these parameters to model mesh elements necessitates development of data access tools that can handle both spatial and temporal datasets. This paper presents an open-source, platform independent, tightly coupled GIS and distributed hydrologic modeling framework, PIHMgis (www.pihm.psu.edu), to improve model-data integration. Tight coupling is achieved through the development of an integrated user interface with an underlying shared geodata model, which improves data flow between the PIHMgis data processing components. The capability and effectiveness of the PIHMgis framework in providing functionalities for watershed delineation, domain decomposition, parameter assignment, simulation, visualization and analyses, is demonstrated through prototyping of a model simulation. The framework and the approach are applicable for watersheds of varied sizes, and offer a template for future GIS-Model integration efforts.
•A coupled GIS and distributed hydrologic modeling framework, PIHMgis was developed.•PIHMgis uses national geospatial dataset to setup, execute, and analyze simulations.•Procedural framework improves model-data integration using shared geodata model.