Use this URL to cite or link to this record in EThOS:
Title: Towards advanced analytic strategies for estimating air temperature through remote sensing
Author: Schneider Dos Santos, Rochelle
ISNI:       0000 0004 8499 9893
Awarding Body: University College London (UCL)
Current Institution: University College London (University of London)
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Restricted access.
Access from Institution:
Urbanisation leads to a vegetation replacement by man-made surfaces, and an increase in anthropogenic heat generation. These factors create the Urban Heat Island (UHI) phenomenon, which is typified by increased air temperature (Ta) in urban areas relative to rural locations. Ta is used in the fields of public health to quantify deaths attributable to heat, where the UHI presence can exacerbate exposure to heat during summer periods. Understanding the spatial patterns of Ta in urban contexts is challenging due to the lack of a good network of weather stations. This study aims to: (i) determine the most suitable model to predict daily summer Tmax at 1 km2 resolution between 2006 and 2017 using Earth Observation satellite data, and (ii) estimate the mortality risk attributable to heat at Middle Super Output Area (MSOA). Linear regression and five machine learning methods (Decision Tree, Random Forest, Gradient Boosting, Support Vector Machine and Neural Network) were investigated to predict London's Tmax, using a data set from 56 meteorological stations, five satellite products, three airborne Light Detection And Ranging (LiDAR) features, three geospatial features and one temporal feature. The work is novel in four aspects: (i) it develops a framework to apply advanced statistical methods to estimate Tmax in cities with uneven coverage of weather stations, (ii) it investigates for the first time the predictive power of the gradient boosting algorithm to estimate Tmax for an urban area, (iii) it includes three built environment features (building density, height and volume) in combination with satellite data to predict Tmax, and (iv) it estimates the risk fraction attributable to heat in London at MSOA level, using Tmax data predicted from satellite-based machine learning methods. The research provides: (i) benefits to public health researchers to improve the estimation of mortality risk attributable to high temperatures and (ii) assistance to inform the decision-making process towards the prioritisation of actions to mitigate heat-related mortality amongst the vulnerable population.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available