A multi-scale time-series dataset of anthropogenic heat from buildings in Los Angeles County

Description

The dataset contains hourly Anthropogenic heat (AH) from buildings in Los Angeles County, based on weather data from 2018. The hourly AH is aggregated at three spatial resolutions: 450m x 450m grid, 12km x 12km grid, and census tract. The AH is broken down into three components: building envelope surface convection, heating, ventilation, and air conditioning (HVAC) system heat release, and zone exfiltration and exhaust air heat loss. The dataset is created with the physics-based EnergyPlus building energy models to calculate individual buildings' AH considering WRF-UCM simulated microclimate conditions. Please refer to the paper "A multi-scale time-series dataset of anthropogenic heat from buildings in Los Angeles County" for more information about the data generation workflow and the data validation procedure.

The data set contains two folders: the "output_data" folder holds the simulation results (EP_output and EP_output_csv), building metadata (building_metadata.geojson and building_metadata.csv), aggregated heat emission and energy consumption time-series data (hourly_heat_energy), and geographical data (geo_data) associated with the GEOID referenced in heat and energy consumption data. The "input_data" folder contains the raw data used to generate files in the "output_data" folder as well as data sets used in the validation.

The code repository (https://github.com/IMMM-SFA/xu_etal_2022_sdata) holds the processing scripts for data curation, validation, and visualization. 

Methods (English)

The grid-level heat emission data for LA County is generated with the workflow shown in this figure. First, a GeoJSON is compiled with building geometry and characteristics like use type, size (total floor area), and vintage (construction year). Next, the microclimate data is generated with WRF version 4.2.1 for each of the 62 12x12 km grids. Then the building location is mapped to the microclimate grids. In this step, a 12x12 km grid cell geometry is created around each grid point. Buildings are mapped to the closest grid point by checking which grid cell contains the building centroid. There are around 1.7 million buildings in LA county. It is not feasible (e.g., it is hard to get detailed data for each building and it is computationally intensive) to build detailed building energy simulation models for each of the 1.7M buildings. Instead, we use 54 prototype building models of certain type-vintage combinations to represent real buildings. Building energy simulations are run in EnergyPlus, for the 1667 prototype-WRF climate grid combination, using weather files (in EnergyPlus epw format) converted from the WRF simulated microclimate data. The EnergyPlus simulation step outputs heat emission and energy consumption for the prototype-WRF climate grid combinations. We applied the energy use intensities (EUIs), represented in energy use per building's total floor area, of the prototype buildings to the real buildings with the same use type and vintage. The simulation results are thus scaled up to the real LA building stock. Finally, a validation step is performed by comparing the simulated annual building energy use with other pre-simulated model output for the same climate region, and against a few measured data sources including the CEC California Energy Consumption Database, and Energy Atlas. In the end, the heat emission results are aggregated into three spatial resolutions for buildings in LA county.

Technical info

For a more detailed description of the file structure, see the code repository: the input data section and the output data section.

Files