A New Approach for Understanding Urban Microclimate by Integrating Complementary Predictors at Different Scales in Regression and Machine Learning Models

Climate change is a major contemporary phenomenon with multiple consequences. In urban areas, it exacerbates the urban heat island phenomenon. It impacts the health of the inhabitants and the sensation of thermal discomfort felt in urban areas. Thus, it is necessary to estimate as well as possible the air temperature at any point of a territory, in particular in view of the ongoing rationalization of the network of fixed meteorological stations of Météo-France. Understanding the air temperature is increasingly in demand to input quantitative models related to a wide range of fields, such as hydrology, ecology, or climate change studies. This study thus proposes to model air temperature, measured during four mobile campaigns carried out during the summer months, between 2016 and 2019, in Lyon (France), in clear sky weather, using regression models based on 33 explanatory variables from traditionally used data, data from remote sensing by LiDAR (Light Detection and Ranging), or Landsat 8 satellite acquisition. Three types of statistical regression were experimented: partial least square regression, multiple linear regression, and a machine learning method, the random forest regression. For example, for the day of 30 August 2016, multiple linear regression explained 89% of the variance for the study days, with a root mean square error (RMSE) of only 0.23 °C. Variables such as surface temperature, Normalized Difference Vegetation Index (NDVI), and Modified Normalized Difference Water Index (MNDWI) have a strong impact on the estimation model. This study contributes to the emergence of urban cooling systems. The solutions available vary. For example, they may include increasing the proportion of vegetation on the ground, facades, or roofs, increasing the number of basins and water bodies to promote urban cooling, choosing water-retaining materials, humidifying the pavement, increasing the number of public fountains and foggers, or creating shade with stretched canvas.

Download Full-text

Interpolation of Instantaneous Air Temperature Using Geographical and MODIS Derived Variables with Machine Learning Techniques

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi8090382 ◽

2019 ◽

Vol 8 (9) ◽

pp. 382 ◽

Cited By ~ 2

Author(s):

Marcos Ruiz-Álvarez ◽

Francisco Alonso-Sarria ◽

Francisco Gomariz-Castillo

Keyword(s):

Machine Learning ◽

Random Forest ◽

Linear Regression ◽

Multiple Linear Regression ◽

Air Temperature ◽

Cross Validation ◽

Daily Basis ◽

Machine Learning Algorithms ◽

Machine Learning Techniques ◽

Support Vector

Several methods have been tried to estimate air temperature using satellite imagery. In this paper, the results of two machine learning algorithms, Support Vector Machines and Random Forest, are compared with Multiple Linear Regression and Ordinary kriging. Several geographic, remote sensing and time variables are used as predictors. The validation is carried out using two different approaches, a leave-one-out cross validation in the spatial domain and a spatio-temporal k-block cross-validation, and four different statistics on a daily basis, allowing the use of ANOVA to compare the results. The main conclusion is that Random Forest produces the best results (R 2 = 0.888 ± 0.026, Root mean square error = 3.01 ± 0.325 using k-block cross-validation). Regression methods (Support Vector Machine, Random Forest and Multiple Linear Regression) are calibrated with MODIS data and several predictors easily calculated from a Digital Elevation Model. The most important variables in the Random Forest model were satellite temperature, potential irradiation and cdayt, a cosine transformation of the julian day.

Download Full-text

Measuring Urban Land Cover Influence on Air Temperature through Multiple Geo-Data—The Case of Milan, Italy

ISPRS International Journal of Geo-Information ◽

10.3390/ijgi7110421 ◽

2018 ◽

Vol 7 (11) ◽

pp. 421 ◽

Cited By ~ 7

Author(s):

Daniele Oxoli ◽

Giulia Ronchetti ◽

Marco Minghini ◽

Monia Molinari ◽

Maryam Lotfian ◽

...

Keyword(s):

Climate Change ◽

Land Cover ◽

Satellite Imagery ◽

Air Temperature ◽

Urban Areas ◽

Urban Land ◽

Mitigation Strategies ◽

Landsat 8 ◽

Local Climate ◽

Urban Land Cover

Climate issues are nowadays one of the most pressing societal challenges, with cities being identified among the landmarks for climate change. This study investigates the effect of urban land cover composition on a relevant climate-related variable, i.e., the air temperature. The analysis exploits different big geo-data sources, namely high-resolution satellite imagery and in-situ air temperature observations, using the city of Milan (Northern Italy) as a case study. Satellite imagery from the Landsat 8, Sentinel-2, and RapidEye missions are used to derive Local Climate Zone (LCZ) maps depicting land cover compositions across the study area. Correlation tests are run to investigate and measure the influence of land cover composition on air temperature. Results show an underlying connection between the two variables by detecting an average temperature offset of about 1.5 ∘ C between heavily urbanized and vegetated urban areas. The approach looks promising in investigating urban climate at a local scale and explaining effects through maps and exploratory graphs, which are valuable tools for urban planners to implement climate change mitigation strategies. The availability of worldwide coverage datasets, as well as the exclusive use of Free and Open Source Software (FOSS), provide the analysis with a potential to be empowered, replicated, and improved.

Download Full-text

Predictive modelling of soils’ hydraulic conductivity using artificial neural network and multiple linear regression

SN Applied Sciences ◽

10.1007/s42452-020-03974-7 ◽

2021 ◽

Vol 3 (2) ◽

Author(s):

Charles Gbenga Williams ◽

Oluwapelumi O. Ojuri

Keyword(s):

Neural Network ◽

Artificial Neural Network ◽

Hydraulic Conductivity ◽

Linear Regression ◽

Multiple Linear Regression ◽

Coarse Grained ◽

Soil Types ◽

Wide Range ◽

Artificial Neural ◽

Input Variables

AbstractAs a result of heterogeneity nature of soils and variation in its hydraulic conductivity over several orders of magnitude for various soil types from fine-grained to coarse-grained soils, predictive methods to estimate hydraulic conductivity of soils from properties considered more easily obtainable have now been given an appropriate consideration. This study evaluates the performance of artificial neural network (ANN) being one of the popular computational intelligence techniques in predicting hydraulic conductivity of wide range of soil types and compared with the traditional multiple linear regression (MLR). ANN and MLR models were developed using six input variables. Results revealed that only three input variables were statistically significant in MLR model development. Performance evaluations of the developed models using determination coefficient and mean square error show that the prediction capability of ANN is far better than MLR. In addition, comparative study with available existing models shows that the developed ANN and MLR in this study performed relatively better.

Download Full-text

Soil organic carbon storage by shaded and unshaded coffee systems and its implications for climate change mitigation in China

The Journal of Agricultural Science ◽

10.1017/s002185962100006x ◽

2021 ◽

pp. 1-8

Author(s):

Ziwei Xiao ◽

Xuehui Bai ◽

Mingzhu Zhao ◽

Kai Luo ◽

Hua Zhou ◽

...

Keyword(s):

Climate Change ◽

Linear Regression ◽

Organic Carbon ◽

Soil Organic Carbon ◽

Multiple Linear Regression ◽

Carbon Storage ◽

Soil Ph ◽

Soil Bulk Density ◽

Soil N ◽

And Storage

Abstract Shaded coffee systems can mitigate climate change by fixation of atmospheric carbon dioxide (CO2) in soil. Understanding soil organic carbon (SOC) storage and the factors influencing SOC in coffee plantations are necessary for the development of sound land management practices to prevent land degradation and minimize SOC losses. This study was conducted in the main coffee-growing regions of Yunnan; SOC concentrations and storage of shaded and unshaded coffee systems were assessed in the top 40 cm of soil. Relationships between SOC concentration and factors affecting SOC were analysed using multiple linear regression based on the forward and backward stepwise regression method. Factors analysed were soil bulk density (ρb), soil pH, total nitrogen of soil (N), mean annual temperature (MAT), mean annual moisture (MAM), mean annual precipitation (MAP) and elevations (E). Akaike's information criterion (AIC), coefficient of determination (R2), root mean square error (RMSE) and residual sum of squares (RSS) were used to describe the accuracy of multiple linear regression models. Results showed that mean SOC concentration and storage decreased significantly with depth under unshaded coffee systems. Mean SOC concentration and storage were higher in shaded than unshaded coffee systems at 20–40 cm depth. The correlations between SOC concentration and ρb, pH and N were significant. Evidence from the multiple linear regression model showed that soil bulk density (ρb), soil pH, total nitrogen of soil (N) and climatic variables had the greatest impact on soil carbon storage in the coffee system.

Download Full-text

A study of micromanufacturing process fingerprints in micro-injection moulding for machine learning and Industry 4.0 applications

The International Journal of Advanced Manufacturing Technology ◽

10.1007/s00170-021-07252-7 ◽

2021 ◽

Author(s):

Mert Gülçür ◽

Ben Whiteside

Keyword(s):

Machine Learning ◽

Linear Regression ◽

Multiple Linear Regression ◽

Injection Moulding ◽

Industry 4.0 ◽

Supervised Machine Learning ◽

Process Conditions ◽

Learning Models ◽

Micro Injection Moulding ◽

Machine Learning Models

AbstractThis paper discusses micromanufacturing process quality proxies called “process fingerprints” in micro-injection moulding for establishing in-line quality assurance and machine learning models for Industry 4.0 applications. Process fingerprints that we present in this study are purely physical proxies of the product quality and need tangible rationale regarding their selection criteria such as sensitivity, cost-effectiveness, and robustness. Proposed methods and selection reasons for process fingerprints are also justified by analysing the temporally collected data with respect to the microreplication efficiency. Extracted process fingerprints were also used in a multiple linear regression scenario where they bring actionable insights for creating traceable and cost-effective supervised machine learning models in challenging micro-injection moulding environments. Multiple linear regression model demonstrated %84 accuracy in predicting the quality of the process, which is significant as far as the extreme process conditions and product features are concerned.

Download Full-text

Compréhension du microclimat urbain lyonnais par l’intégration de prédicteurs complémentaires à différentes échelles dans des modèles de régression

Climatologie ◽

10.1051/climat/202017002 ◽

2020 ◽

Vol 17 ◽

pp. 2

Author(s):

Lucille Alonso ◽

Florent Renard

Keyword(s):

Machine Learning ◽

Partial Least Square ◽

Least Square ◽

Light Detection And Ranging ◽

Landsat 8 ◽

Light Detection ◽

Changement Climatique ◽

Milieu Urbain ◽

Changements Climatiques

Le changement climatique est un phénomène majeur actuel générant de multiples conséquences. En milieu urbain, il exacerbe celui de l’îlot de chaleur urbain. Ces deux manifestations climatiques engendrent des conséquences sur la santé des habitants et sur la sensation d’inconfort thermique ressenti en milieu urbain. Ainsi, il est nécessaire d’estimer au mieux la température de l’air en tout point d’un territoire, notamment face à la rationalisation actuelle du réseau de stations météorologiques fixes de Météo France. La connaissance spatialisée de la température de l’air est de plus en plus demandée pour alimenter des modèles quantitatifs liés à un large éventail de domaines, tels que l’hydrologie, l’écologie ou les études sur les changements climatiques. Cette étude se propose ainsi de modéliser la température de l’air, mesurée durant 4 campagnes mobiles réalisées durant les mois d’été, entre 2016 et 2019, dans Lyon par temps clair, à l’aide de modèle de régressions à partir de 33 variables explicatives issues de données traditionnellement utilisées, de données issues de la télédétection par une acquisition LiDAR (Light Detection And Ranging) ou satellitaire Landsat 8. Trois types de régression statistique ont été expérimentés, la régression partial least square, la régression linéaire multiple et enfin, une méthode de machine learning, la forêt aléatoire de classification et de régression. Par exemple, pour la journée du 30 août 2016, la régression linéaire multiple a expliqué 89% de la variance pour les journées d’étude, avec un RMSE moyen de seulement 0,23°C. Des variables comme la température de surface, le NDVI ou encore le MNDWI impactent fortement le modèle d’estimation.

Download Full-text

Arc contacts ablation state assessment method based on machine learning multiple linear regression

10.1117/12.2611061 ◽

2021 ◽

Author(s):

Yijun Liu ◽

Daopin Chen ◽

Muxin Diao ◽

Guangyu Xiao ◽

Jing Yan ◽

...

Keyword(s):

Machine Learning ◽

Linear Regression ◽

Multiple Linear Regression ◽

Assessment Method ◽

State Assessment

Download Full-text

Machine learning for geographically differentiated climate change mitigation in urban areas

Sustainable Cities and Society ◽

10.1016/j.scs.2020.102526 ◽

2021 ◽

Vol 64 ◽

pp. 102526

Author(s):

Nikola Milojevic-Dupont ◽

Felix Creutzig

Keyword(s):

Climate Change ◽

Machine Learning ◽

Urban Areas ◽

Climate Change Mitigation ◽

Change Mitigation

Download Full-text

Correlation of air temperature and land surface temperature of Landsat-8 in cities of South Korea

10.5194/egusphere-egu2020-13886 ◽

2020 ◽

Author(s):

Sungwon Choi ◽

Donghyun Jin ◽

Noh-hun Seong ◽

Daeseong Jung ◽

Kyung-soo Han

Keyword(s):

Correlation Analysis ◽

Surface Temperature ◽

Air Temperature ◽

Land Surface ◽

Urban Areas ◽

Temperature Data ◽

Landsat 8 ◽

In Situ Data ◽

Urban Greenery ◽

Important Position

Recently, there are many problems in urban area such as urban thermal island phenomenon, changes in urban green area, changes in urban weather and various urban types. And surface temperature data have been utilized in many areas to identify these phenomena. This means that surface temperatures is an important position in urban greenery and weather. High temporal and spatial resolution satellite data are needed to continuously observe the phenomenon in urban areas. In addition, the surface temperature varies from type of indicator, topography, and various factors, so there is a limit to the in-situ data for observing changes throughout the city. Therefore, various organizations around the world are currently conducting surface temperature measurements using satellites. However, the use of data in clear pixel is essential for accurate surface temperature calculations using satellites, but the accuracy of results will be reduced if the data from in the pixel which conclude clouds.Therefore, we tried to solve these problems by analyzing the correlation between the air temperature data and the Landsat-8 LST data. The variables used in the correlation analysis are air temperature, Landsat-8 LST, NDVI and NDWI, and the study period is 2014 to 2016 and the study area is South Korea's five cities (Seoul, Busan, Daejeon, Daegu, Gwangju). For correlation analysis, the air temperature data points provided by the Korea Meteorological Administration and the Landsat-8 pixels were matched, and the correlation coefficient calculated by the correlation analysis was applied to the Landsat-8 satellite to calculate the LST. We validated by direct comparison the re-produced Landsat-8 LST with observed Landsat-8 LST. And the result of validation showed a high correlation of 0.9. It shows that compensation for the satellite's shortcomings from clouds by using the correlation between temperature and LST.

Download Full-text

Can climate adaptation solutions fix the urban heat island? An assessment of the thermal conditions during heat waves in Vienna impacted by climate change and urban development scenarios for the mid-21st-century

10.5194/egusphere-egu2020-8668 ◽

2020 ◽

Author(s):

Paul Hamer ◽

Heidelinde Trimmel ◽

Philipp Weihs ◽

Stéphanie Faroux ◽

Herbert Formayer ◽

...

Keyword(s):

Climate Change ◽

Urban Heat Island ◽

Heat Wave ◽

Air Temperature ◽

Urban Areas ◽

Climate Adaptation ◽

Heat Waves ◽

Heat Island ◽

Human Thermal Comfort ◽

Urban Heat

Climate change threatens to exacerbate existing problems in urban areas arising from the urban heat island. Furthermore, expansion of urban areas and rising urban populations will increase the numbers of people exposed to hazards in these vulnerable areas. We therefore urgently need study of these environments and in-depth assessment of potential climate adaptation measures.We present a study of heat wave impacts across the urban landscape of Vienna for different future development pathways and for both present and future climatic conditions. We have created two different urban development scenarios that estimate potential urban sprawl and optimized development concerning future building construction in Vienna and have built a digital representation of each within the Town Energy Balance (TEB) urban surface model. In addition, we select two heat waves of similar frequency of return representative for present and future conditions (following the RCP8.5 scenario) of the mid 21st century and use the Weather Research and Forecasting Model (WRF) to simulate both heat wave events. We then couple the two representations urban Vienna in TEB with the WRF heat wave simulations to estimate air temperature, surface temperatures and human thermal comfort during the heat waves. We then identify and apply a set of adaptation measures within TEB to try to identify potential solutions to the problems associated with the urban heat island.Global and regional climate change under the RCP8.5 scenario causes the future heat wave to be more severe showing an increase of daily maximum air temperature in Vienna by 7 K; the daily minimum air temperature will increase by 2-4 K. We find that changes caused by urban growth or densification mainly affect air temperature and human thermal comfort local to where new urbanisation takes place and does not occur significantly in the existing central districts.Exploring adaptation solutions, we find that a combination of near zero-energy standards and increasing albedo of building materials on the city scale accomplishes a maximum reduction of urban canyon temperature of 0.9 K for the minima and 0.2 K for the maxima. Local scale changes of different adaption measures show that insulation of buildings alone increases the maximum wall surface temperatures by more than 10 K or the maximum mean radiant temperature (MRT) in the canyon by 5 K.&#160; Therefore, additional adaptation to reduce MRT within the urban canyons like tree shade are needed to complement the proposed measures.This study concludes that the rising air temperatures expected by climate change puts an unprecedented heat burden on Viennese inhabitants, which cannot easily be reduced by measures concerning buildings within the city itself. Additionally, measures such as planting trees to provide shade, regional water sensitive planning and global reduction of greenhouse gas emissions in order to reduce temperature extremes are required.We are now actively seeking to apply this set of tools to a wider set of cases in order to try to find effective solutions to projected warming resulting from climate change in urban areas.

Download Full-text