Search

Monitoring the coastline using remote sensing technologies has become a fundamental method in underpinning coastal management. Monitoring beach use in a spatio-temporal context provides coastal stakeholders with a breadth of knowledge on the areas of the beach that are important to users and may assist in the implementation of coastal defences and beach resilience measures. This paper provides a novel technique that utilises radar imagery in the tracking, feature extraction and subsequent classification of human beach users. Comparing the accuracy of several classification models, detected particle trajectories are classified into human and non-human and tracked over time. The resulting outputs enable beach user density to be quantified over various spatial and temporal scales.

In this paper, we introduce a framework to leverage satellite data through AI to build rich representations of urban form and function. We use a concept of Spatial signatures to characterise predominantly urban environment into data-driven classes based on both form and function composed of a large number of input data sources. Consequently, we explore the ability of Sentinel 2 satellite imagery and state-of-the-art AI models to capture the same classification of space using a single, regularly updated data source, including the conceptual questions of relationship between granular signature geometry and rigid raster grid of satellite data.

In this paper, we use the recent spatial signatures framework to develop morphosignatures, a characterisation of space based on urban form, and we apply them to the whole of Great Britain. Identification of recurring patterns in the built environment is deeply embedded within all schools of urban morphology. A long-standing challenge in this context has been the ability to scale the analysis to cover large regions. In effect, advancing the systematic study of urban morphology has been limited to the ability to deploy granular methods to large portions of the urban landscape. The recent rise of morphometrics, and its ability to scale while retaining detail, has given urban morphology a fundamentally new toolkit to analyse the form of urban fabric at metropolitan or even national extents. In this tradition, morphosignatures are conceptually defined as an aggregation of granular elements into contiguous areas based on similar nature. We adopt the enclosed tessellation cell (ETC), the result of combining street networks and other urban "delimiters" with building footprints, as the core spatial unit. For each ETC, we calculate a wide range of characters that capture aspects of their spatial organisation, from dimension and shape of individual features to their spatial distribution, intensity or connectivity. We then group these ETCs using the K-Means algorithm, effectively deriving a comprehensive typology of urban form. We employ this approach on the case of Great Britain, illustrating both the potential and limits of the analysis of urban form at scale. Our resulting classification identifies 19 types of morphosignatures that can be organised into three macro groups: countryside, suburban low density development, and dense city centres. The main contribution of the proposed method, with respect to traditional morphological studies, resides in the scalability of the analysis while retaining granularity. We argue national-scale classification such as that of the British morphosignatures allows us to ask fundamentally new questions and provide complementary answers to those asked in smaller scale studies.

This research applies a hierarchical k-means clustering method on the TF-IDF weighted 2019 cycling transactions from the Citi Bike bike-sharing system operating in New York City, with the primary goal of investigating the spatiotemporal usage pattern of its docking points. With a particular focus on bike-sharing stations in Manhattan, we classify 504 stations into four main clusters featuring heterogeneous dynamic usages, including leisure-oriented, residential- oriented, workplace-oriented, and off-peak oriented. We interpret each cluster based on their salient characteristics and anticipate possible future directions of this work.