Estimating heat storage in urban areas using multispectral satellite data and machine learning

Abstract A satellite-derived hysteresis model is presented for estimate heat storage in urban areas. Storage heat flux, one of the dominant terms in the urban surface energy budget (USEB), is largely unknown despite its critical relationship to various urban environmental processes. This study introduces a novel technique for quantifying heat storage by relating multispectral satellite radiances and geophysical properties to ground-truth residual heat storage computed with flux instruments. Gradient-boosted regression trees serve as the method of maximizing the relationship between satellite data and flux measurements. Several flux networks are used to train and validate the model over varying land cover types, which strengthens the robustness of the model. The model performs well under variable weather conditions such as cloudy rainy days. In comparison with other studies, the RMSE and MAE values were found to be lower than some ground-to-ground studies, and is one of few satellite-derived methods that computes direct comparison over a range of different land cover types.

[1]  Zhi-hua Wang.  A new perspective of urban-rural differences: The impact of soil water advection , 2014 .

[2]  C. Grimmond,et al.  Intraurban Differences of Surface Energy Fluxes in a Central European City , 2006 .

[3]  J. Friedman.  Greedy function approximation: A gradient boosting machine. , 2001 .

[4]  Niko E. C. Verhoest,et al.  A review of spatial downscaling of satellite remotely sensed soil moisture , 2017 .

[5]  Amir Hossein Alavi,et al.  Machine learning in geosciences and remote sensing , 2016 .

[6]  Christian Feigenwinter,et al.  The urban heat island of Basel – seen from different perspectives , 2014 .

[7]  Ian B. Strachan,et al.  Development of the Surface Urban Energy and Water Balance Scheme (SUEWS) for cold climate cities , 2014 .

[8]  J. Friedman.  Stochastic gradient boosting , 2002 .

[9]  E. Shaviv,et al.  Evaluation of Urban Surface Energy Fluxes Using an Open-Air Scale Model , 2005 .

[10]  David J. Sailor,et al.  Risks of summertime extreme thermal conditions in buildings as a result of climate change and exacerbation of urban heat islands , 2014 .

[11]  K. Anandakumar.  A study on the partition of net radiation into heat fluxes on a dry asphalt surface , 1999 .

[12]  B. Lamb,et al.  Eddy covariance flux measurements of pollutant gases in urban Mexico City , 2009 .

[13]  J. M. Shepherd,et al.  Atlanta’s urban heat island under extreme heat conditions and potential mitigation strategies , 2010 .

[14]  Timothy R. Oke,et al.  Heat Storage in Urban Areas: Local-Scale Observations and Evaluation of a Simple Model , 1999 .

[15]  Matthias Schonlau,et al.  Boosted Regression (Boosting): An Introductory Tutorial and a Stata Plugin , 2005 .

[16]  Simon J. Hook,et al.  Synergies Between VSWIR and TIR Data for the Urban Environment: An Evaluation of the Potential for the Hyperspectral Infrared Imager (HyspIRI) , 2012 .

[17]  T. Oke,et al.  Urban heat storage derived as energy balance residuals , 1987 .

[18]  Timothy R. Oke,et al.  An objective urban heat storage model and its comparison with other schemes , 1991 .

[19]  Facundo Carmona,et al.  Development of a general model to estimate the instantaneous, daily, and daytime net radiation with satellite data on clear-sky days , 2015 .

[20]  J. Soares,et al.  Diurnal variation in stored energy flux in São Paulo city, Brazil , 2013 .

[21]  Elie Bou-Zeid,et al.  Heatwaves and urban heat islands: A comparative analysis of multiple cities , 2017 .

[22]  O. Bergeron,et al.  CO2 sources and sinks in urban and suburban areas of a northern mid-latitude city , 2011 .

[23]  Ximing Cai,et al.  Assessing interannual variability of evapotranspiration at the catchment scale using satellite‐based evapotranspiration data sets , 2011 .

[24]  Timothy R. Oke,et al.  Comparison of Four Methods to Estimate Urban Heat Storage , 2006 .

[25]  Jordi Inglada,et al.  Similarity measures for multisensor remote sensing images , 2002, IEEE International Geoscience and Remote Sensing Symposium.

[26]  David J. Sailor,et al.  A top-down methodology for developing diurnal and seasonal anthropogenic heating profiles for urban areas , 2004 .

[27]  Jay S. Golden,et al.  The Built Environment Induced Urban Heat Island Effect in Rapidly Urbanizing Arid Regions – A Sustainable Urban Engineering Complexity , 2004 .

[28]  S. Myint,et al.  Daytime cooling efficiency and diurnal energy balance in Phoenix, Arizona, USA , 2012 .

[29]  M. Kanda,et al.  Spatial Variability of Both Turbulent Fluxes and Temperature Profiles in an Urban Roughness Layer , 2006 .

[30]  J. Niemann,et al.  Evaluation of an empirical orthogonal function–based method to downscale soil moisture patterns based on topographical attributes , 2012 .

[31]  Margaret E. Gardner,et al.  Spectrometry for urban area remote sensing—Development and analysis of a spectral library from 350 to 2400 nm , 2004 .

[32]  ABI Imagery from the GOES-R Series , 2020 .

[33]  A. Siani,et al.  A canopy layer model and its application to Rome. , 2006, The Science of the total environment.

[34]  Uwe Stilla,et al.  Machine Learning Comparison between WorldView-2 and QuickBird-2-Simulated Imagery Regarding Object-Based Urban Land Cover Classification , 2011, Remote. Sens..

[35]  Gerhard Nahler,et al.  Pearson Correlation Coefficient , 2020, Definitions.

[36]  Nektarios Chrysoulakis,et al.  Urban energy exchanges monitoring from space , 2018, Scientific Reports.

[37]  J. Randerson,et al.  Continental-scale net radiation and evapotranspiration estimated using MODIS satellite observations , 2011 .

[38]  Fabio Del Frate,et al.  Urban Surface Temperature Time Series Estimation at the Local Scale by Spatial-Spectral Unmixing of Satellite Observations , 2015, Remote. Sens..

[39]  Martha C. Anderson,et al.  Monitoring daily evapotranspiration over two California vineyards using Landsat 8 in a multi-sensor data fusion approach , 2016 .

[40]  B. Lamb,et al.  Flux measurements of volatile organic compounds from an urban landscape , 2005 .

[41]  T. Oke,et al.  Comparison of modelled and «measured» heat storage in suburban terrain , 1994 .

[42]  Helmi Zulhaidi Mohd Shafri,et al.  Detailed intra-urban mapping through transferable OBIA rule sets using WorldView-2 very-high-resolution satellite images , 2015 .

[43]  Timothy R. Oke,et al.  Evaluation of the Town Energy Balance (TEB) Scheme with Direct Measurements from Dry Districts in Two Cities , 2002 .

[44]  Xiaotong Zhang,et al.  Estimating Surface Downward Shortwave Radiation over China Based on the Gradient Boosting Decision Tree Method , 2018, Remote. Sens..

[45]  E. Parlow.  The urban heat budget derived from satellite data , 2003 .

[46]  K. Gallo,et al.  Surface Energy Balance Measurements Above an Exurban Residential Neighbourhood of Kansas City, Missouri , 2009 .

[47]  N. Moussiopoulos,et al.  Satellite data based approach for the estimation of anthropogenic heat flux over urban areas , 2017 .

[48]  Zhe Zhu,et al.  Overall Methodology Design for the United States National Land Cover Database 2016 Products , 2019, Remote. Sens..

[49]  V. Masson,et al.  Investigating the Surface Energy Balance in Urban Areas – Recent Advances and Future Needs , 2002 .

[50]  T. Vesala,et al.  Revised eddy covariance flux calculation methodologies – effect on urban energy balance , 2012 .

[51]  C. Grimmond,et al.  Identification of Micro-scale Anthropogenic CO2, heat and moisture sources – Processing eddy covariance fluxes for a dense urban environment , 2012 .

[52]  Yasushi Yamaguchi,et al.  Analysis of urban heat-island effect using ASTER and ETM+ Data: Separation of anthropogenic heat discharge and natural heat radiation from sensible heat flux , 2005 .

[53]  Lingkui Meng,et al.  Downscaling SMAP soil moisture estimation with gradient boosting decision tree regression over the Tibetan Plateau , 2019, Remote Sensing of Environment.

[54]  Michael Dorman,et al.  Correcting Measurement Error in Satellite Aerosol Optical Depth with Machine Learning for Modeling PM2.5 in the Northeastern USA , 2018, Remote. Sens..

[55]  Peter L. Bartlett,et al.  Boosting Algorithms as Gradient Descent , 1999, NIPS.

[56]  W. Oechel,et al.  The Analytical Objective Hysteresis Model (AnOHM v1.0): methodology to determine bulk storage heat flux coefficients , 2017, Geoscientific Model Development.

[57]  Valéry Masson,et al.  A Physically-Based Scheme For The Urban Energy Budget In Atmospheric Models , 2000 .

[58]  Ismail Elkhrachy,et al.  Vertical accuracy assessment for SRTM and ASTER Digital Elevation Models: A case study of Najran city, Saudi Arabia , 2017, Ain Shams Engineering Journal.

[59]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[60]  T. Oke.  The urban energy balance , 1988 .

[61]  W. Oechel,et al.  Energy balance closure at FLUXNET sites , 2002 .

[62]  Stephen Tyree,et al.  Non-linear Metric Learning , 2012, NIPS.

[63]  Xiaoma Li,et al.  Relationships between land cover and the surface urban heat island: seasonal variability and effects of spatial and thematic resolution of land cover data on predicting land surface temperatures , 2013, Landscape Ecology.

[64]  E. Pardyjak,et al.  Toward understanding the behavior of carbon dioxide and surface energy fluxes in the urbanized semi-arid Salt Lake Valley, Utah, USA , 2011 .

[65]  Mathew Lipson,et al.  Efficiently modelling urban heat storage: an interface conduction scheme in an urban land surface model (aTEB v2.0) , 2017 .

[66]  M. Şahin,et al.  Modelling of air temperature using remote sensing and artificial neural network in Turkey , 2012 .

[67]  Gustavo Camps-Valls,et al.  Machine learning in remote sensing data processing , 2009, 2009 IEEE International Workshop on Machine Learning for Signal Processing.

[68]  Limin Yang,et al.  A new generation of the United States National Land Cover Database: Requirements, research priorities, design, and implementation strategies , 2018, ISPRS Journal of Photogrammetry and Remote Sensing.

[69]  Gensuo Jia,et al.  Satellite-based estimation of daily average net radiation under clear-sky conditions , 2014, Advances in Atmospheric Sciences.

[70]  T. Oke,et al.  The energy balance of central Mexico City during the dry season , 1999 .

[71]  Jungho Im,et al.  Estimation of daily maximum and minimum air temperatures in urban landscapes using MODIS time series satellite data , 2018 .

[72]  Timothy J. Schmit,et al.  Applications of the 16 spectral bands on the Advanced Baseline Imager (ABI). , 2018, Journal of Operational Meteorology.

[73]  Adriana Bernardi,et al.  An observational study of heat fluxes and their relationships with net radiation , 1982 .

[74]  Fabio Del Frate,et al.  Spatial distribution of sensible and latent heat flux in the URBANFLUXES case study city Basel (Switzerland) , 2017, 2017 Joint Urban Remote Sensing Event (JURSE).

[75]  Gerhard Tutz,et al.  Boosting techniques for nonlinear time series models , 2012 .

[76]  Enrique R. Vivoni,et al.  Downscaling soil moisture in the southern Great Plains through a calibrated multifractal model for land surface modeling applications , 2010 .

[77]  D. Sailor.  A review of methods for estimating anthropogenic heat and moisture emissions in the urban environment , 2011 .

[78]  P. Ramamurthy,et al.  Spatiotemporal variability in building energy use in New York City , 2017 .

[79]  Ayse Irmak,et al.  Estimation of Crop Coefficients Using Satellite Remote Sensing , 2009 .

[80]  Gautam Bisht,et al.  Estimation of net radiation from the MODIS data under all sky conditions: Southern Great Plains case study , 2010 .

[81]  Fabio Del Frate,et al.  Spatial Distribution of Sensible and Latent Heat Flux in the City of Basel (Switzerland) , 2018, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[82]  Jonathan Cheung-Wai Chan,et al.  Multiple Criteria for Evaluating Machine Learning Algorithms for Land Cover Classification from Satellite Data , 2000 .

[83]  C. S. B. Grimmond,et al.  An urban canyon energy budget model and its application to urban storage heat flux modeling , 1998 .

[84]  Y. Yamaguchi,et al.  Estimation of storage heat flux in an urban area using ASTER data , 2007 .

[85]  Isabelle Jobard,et al.  Validation of satellite and ground-based estimates of precipitation over the Sahel , 1998 .

[86]  J. Beringer,et al.  Impact of Increasing Urban Density on Local Climate: Spatial and Temporal Variations in the Surface Energy Balance in Melbourne, Australia , 2007 .

[87]  Jungho Im,et al.  Retrieval of Total Precipitable Water from Himawari-8 AHI Data: A Comparison of Random Forest, Extreme Gradient Boosting, and Deep Neural Network , 2019, Remote. Sens..

[88]  Haider Taha,et al.  Modifying a Mesoscale Meteorological Model to Better Incorporate Urban Heat Storage: A bulk-parameterization approach , 1999 .

[89]  M. Ueyama,et al.  Surface energy exchange in a dense urban built-up area based on two-year eddy covariance measurements in Sakai, Japan , 2017 .

[90]  Ronald G. Prinn,et al.  Urban Visible/SWIR surface reflectance ratios from satellite and sun photometer measurements in Mexico City , 2007 .

[91]  Ting Sun,et al.  Revisiting the hysteresis effect in surface energy budgets , 2013 .

[92]  Tie-Yan Liu,et al.  LightGBM: A Highly Efficient Gradient Boosting Decision Tree , 2017, NIPS.

[93]  F. Lindberg,et al.  Urban storage heat flux variability explored using satellite, meteorological and geodata , 2020, Theoretical and Applied Climatology.

[94]  Andrew S. Jones,et al.  A method to downscale soil moisture to fine resolutions using topographic, vegetation, and soil data , 2015 .

[95]  Simone Kotthaus,et al.  Energy exchange in a dense urban environment – Part II: Impact of spatial heterogeneity of the surface , 2014 .

[96]  Eberhard Parlow,et al.  Modelling the ground heat flux of an urban area using remote sensing data , 2007 .

[97]  Zhiqiu Gao,et al.  Attribution and mitigation of heat wave-induced urban heat storage change , 2017 .

[98]  M. Kerschgens,et al.  On the energetics of the urban canopy layer , 1990 .

[99]  B. Tsuang.  Ground Heat Flux Determination according to Land Skin Temperature Observations from In Situ Stations and Satellites , 2005 .

[100]  T. Oke,et al.  Turbulent Heat Fluxes in Urban Areas: Observations and a Local-Scale Urban Meteorological Parameterization Scheme (LUMPS) , 2002 .

[101]  Rick Mueller,et al.  The Multi-Resolution Land Characteristics (MRLC) Consortium - 20 Years of Development and Integration of USA National Land Cover Data , 2014, Remote. Sens..

[102]  T. Oke,et al.  Heat fluxes through roofs and their relevance to estimates of urban heat storage , 2000 .

[103]  C. S. B. Grimmond,et al.  Heat storage and anthropogenic heat flux in relation to the energy balance of a central European city centre , 2005 .

[104]  George Xian,et al.  An analysis of urban thermal characteristics and associated land cover in Tampa Bay and Las Vegas using Landsat satellite data , 2006 .

[105]  A. Just,et al.  Gradient Boosting Machine Learning to Improve Satellite-Derived Column Water Vapor Measurement Error , 2019 .