Downscaling Building Energy Consumption Carbon Emissions by Machine Learning

Downscaling of environmental indicators: A review

Environmental indicators at different scales are important for environmental management, daily life, and scientific research. Because of the lack of statistics below a national scale for many environmental indicators, scholars have developed various downscaling methods to obtain finer-scale and diverse forms of data for different environmental indicators. However, the existing downscaling methods for environmental indicators are diverse and fragmented. Here, we reviewed the downscaling methods by reclassifying the environmental indicators from a life cycle perspective into five categories: natural resources use and related attributes; material and energy consumption; environmental discharge; climate change; and environmental footprints. We first provide a general introduction to downscaling theory in the environmental field, including definitions, techniques, and evolution. We then elaborate on downscaling methods and make an inventory of the five categories of environmental indicators. We summarize the downscaling methods commonly applied to specific indicators, scale transformation, the strengths and limitations of corresponding methods, and provide specific examples. Next, we discuss ways to select or construct downscaling methods based on four principles: objective orientation, data accessibility, model feasibility, and model adjustment. Finally, we explore the future direction of downscaling and provide insights for improving downscaling for environmental indicators. In this review, we generalize and clarify the downscaling techniques for environmental indicators, which will help facilitate the appropriate selection of downscaling methods by researchers.

Accurate multi-objective prediction of CO2 emission performance indexes and industrial structure optimization using multihead attention-based convolutional neural network

The establishment of specific targets for the global carbon peaking and neutrality raises urgent requirements for prediction of CO₂ emission performance indexes (CEPIs) and industrial structure optimization. However, accurate multi-objective prediction of CEPIs is still a knotty problem. In the present study, multihead attention-based convolutional neural network (MHA-CNN) model was proposed for accurate prediction of 4 CEPIs and further provided the rational suggestions for further industrial structure optimization. The proposed MHA-CNN model introduces deep learning mechanism with efficient resolution strategies for training model overfitting, feature extraction, and self-supervised learning to acquire the adaptability for CEPIs. Multihead attention (MHA) mechanism plays important roles in influence weight interpretation of variables to facilitate the prediction performance of CNN on CEPIs. The MHA-CNN model presented its overwhelmingly superior performance to CNN model and long short-term memory (LSTM) model, two frequently-used models, in multi-objective prediction of CEPIs using 8 influence variables, which highlighted advantages of MHA module in multi-dimensional feature extraction. Additionally, contributions of influence variables to CEPIs based on MHA analyses presented relatively high consistency with the geographical distribution analyses, indicating the excellent capacity of the MHA module in variable weights identification and contribution dissection. Based on the more accurate prediction results by MHA-CNN than those by CNN and LSTM model, the increase in the tertiary industry and the decreases in the first and secondary industries are conducive to improvement of total-factor carbon emission efficiency and further enhancement of effective energy utilization in regions with inefficient carbon emissions. This study provides insights towards the critical roles of the proposed MHA-CNN model in accurate multi-objective prediction of CEPIs and further industrial structure optimization for improvement of total-factor carbon emission efficiency.

The Carbon Emission Characteristics and Reduction Potential in Developing Areas: Case Study from Anhui Province, China

Global warming and world-wide climate change caused by increasing carbon emissions have attracted a widespread public attention, while anthropogenic activities account for most of these problems generated in the social economy. In order to comprehensively measure the levels of carbon emissions and carbon sinks in Anhui Province, the study adopted some specific carbon accounting methods to analyze and explore datasets from the following suggested five carbon emission sources of energy consumption, food consumption, cultivated land, ruminants and waste, and three carbon sink sources of forest, grassland and crops to compile the carbon emission inventory in Anhui Province. Based on the compiled carbon emission inventory, carbon emissions and carbon sink capacity were calculated from 2000 to 2019 in Anhui Province, China. Combined with ridge regression and scenario analysis, the STIRPAT model was used to evaluate and predict the regional carbon emission from 2020 to 2040 to explore the provincial low-carbon development pathways, and carbon emissions of various industrial sectors were systematically compared and analyzed. Results showed that carbon emissions increased rapidly from 2000 to 2019 and regional energy consumption was the primary source of carbon emissions in Anhui Province. There were significant differences found in the increasing carbon emissions among various industries. The consumption proportion of coal in the provincial energy consumption continued to decline, while the consumption of oil and electricity proceeded to increase. Furthermore, there were significant differences among different urban and rural energy structures, and the carbon emissions from waste incineration were increasing. Additionally, there is an inverted "U"-shape curve of correlation between carbon emission and economic development in line with the environmental Kuznets curve, whereas it indicated a "positive U"-shaped curve of correlation between carbon emission and urbanization rate. The local government should strengthen environmental governance, actively promote industrial transformation, and increase the proportion of clean energy in the energy production and consumption structures in Anhui Province. These also suggested a great potential of emission reduction with carbon sink in Anhui Province.

Carbon Emissions Estimation and Spatiotemporal Analysis of China at City Level Based on Multi-Dimensional Data and Machine Learning

Carbon emissions caused by the massive consumption of energy have brought enormous pressure on the Chinese government. Accurately and rapidly characterizing the spatiotemporal characteristics of Chinese city-level carbon emissions is crucial for policy decision making. Based on multi-dimensional data, including nighttime light (NTL) data, land use (LU) data, land surface temperature (LST) data, and added-value secondary industry (AVSI) data, a deep neural network ensemble (DNNE) model was built to analyze the nonlinear relationship between multi-dimensional data and province-level carbon emission statistics (CES) data. The city-level carbon emissions data were estimated, and the spatiotemporal characteristics were analyzed. As compared to the energy statistics released by partial cities, the results showed that the DNNE model based on multi-dimensional data could well estimate city-level carbon emissions data. In addition, according to a linear trend analysis and standard deviational ellipse (SDE) analysis of China from 2001 to 2019, we concluded that the spatiotemporal changes in carbon emissions at the city level were in accordance with the development of China’s economy. Furthermore, the results can provide a useful reference for the scientific formulation, implementation, and evaluation of carbon emissions reduction policies.

A fine spatial resolution modeling of urban carbon emissions: a case study of Shanghai, China

Quantification of fossil fuel carbon dioxide emissions (CEs) at fine space and time resolution is a critical need in climate change research and carbon cycle. Quantifying changes in spatiotemporal patterns of urban CEs is important to understand carbon cycle and development carbon reduction strategies. The existing spatial data of CEs have low resolution and cannot distinguish the distribution characteristics of CEs of different emission sectors. This study quantified CEs from 15 types of energy sources, including residential, tertiary, and industrial sectors in Shanghai. Additionally, we mapped the CEs for the three sectors using point of interest data and web crawler technology, which is different from traditional methods. At a resolution of 30 m, the improved CEs data has a higher spatial resolution than existing studies. The spatial distribution of CEs based on this study has higher spatial resolution and more details than that based on traditional methods, and can distinguish the spatial distribution characteristics of different sectors. The results indicated that there was a consistent increase in CEs during 2000–2015, with a low rate of increase during 2009–2015. The intensity of CEs increased significantly in the outskirts of the city, mainly due to industrial transfer. Moreover, intensity of CEs reduced in city center. Technological progress has promoted the improvement of energy efficiency, and there has been a decoupling between the economic development and CEs in the city was observed during in 2000–2015.

Mapping Building-Based Spatiotemporal Distributions of Carbon Dioxide Emission: A Case Study in England

The spatiotemporal inventory of carbon dioxide (CO₂) emissions from the building sector is significant for formulating regional and global warming mitigation policies. Previous studies have attempted to use energy consumption models associated with field investigations to estimate CO₂ emissions from buildings at local scales, or they used spatial proxies to downscale emission sources from large geographic units to grid cells for larger scales. However, mapping the spatiotemporal distributions of CO₂ emissions on a large scale based on buildings remains challenging. Hence, we conducted a case study in England in 2015, wherein we developed linear regression models to analyze monthly CO₂ emissions at the building scale by integrating the Emissions Database for Global Atmospheric Research, building data, and Visible Infrared Imaging Radiometer Suite night-time lights images. The results showed that the proposed model that considered building data and night-time light imagery achieved the best fit. Fine-scale spatial heterogeneity was observed in the distributions of building-based CO₂ emissions compared to grid-based emission maps. In addition, we observed seasonal differences in CO₂ emissions. Specifically, buildings emitted significantly more CO₂ in winter than in summer in England. We believe our results have great potential for use in carbon neutrality policy making and climate monitoring.

Analysis of Influencing Factors and Trend Forecast of CO2 Emission in Chengdu-Chongqing Urban Agglomeration

Urban agglomeration is a primary source of global energy consumption and CO2 emissions. It is employed as a major means of modern economic and social activities. Analysis of the temporal and spatial characteristics of CO2 emissions in urban agglomerations and prediction of the future trends of CO2 emissions in urban agglomerations will help in the implementation of CO2 reduction policies within region-wide areas. So, based on that, this study contains four aspects. Firstly, it calculates the energy CO2 emissions of China’s Chengdu-Chongqing urban agglomeration. Secondly, it analyzes the time and space changes in the area by using ArcGIS. Then, the STIRPAT model is used to investigate the factors influencing CO2 emissions, and the elasticity coefficient of the influencing factors is estimated using the ridge regression method, and the important influencing factors are screened on the basis of the estimated results, which are then used as input features for prediction. Finally, a combined prediction model based on the improved GM (1, N) and SVR models is constructed, and then the optimal solution is found through the particle swarm optimization algorithm. It sets up different CO2 emission scenarios to predict the energy CO2 emission of the region and its cities. The results show that, first, the CO2 emissions of the Chengdu-Chongqing urban agglomeration have accumulated year by year, but by 2030, as predicted, it will not reach its peak. The spatial layout of CO2 emissions in this region is not expected to undergo major changes by 2030. Second, population, GDP, gas and electricity consumption, and industrial structure have served as important factors affecting energy CO2 emissions in the region. Third, on the basis of the prediction results for different scenarios, the CO2 emissions in the baseline scenario are low in the short term, but the CO2 emissions in the low-carbon scenario are low in the long run. This study also puts forward some policy recommendations on how to reduce CO2 emissions.