Each ENVI-met simulation creates a huge amount of data which is organized in different files and folders. In general, there are two different types of output files regardless their content:
In order to use advanced analysis functions like Time Series in LEONARDO, it is strongly recommended to store only one simulation run in the same folder and not to mix different model runs. While the files remain usable itself, the logic of the different applications might not be able to sort out the structure of your simulation data.
The huge amount of data generated by each simulation requires a strict concept of storage and labelling in order not to loose overview or to overwrite files. ENVI-met uses a three-level concept to organize the output data:
For simple Text files, Level 3 data are not available, so that the filename and filename extension provide information about the content of the files.
The uppermost level of file organisation starts by sorting the output files into subfolders of the selected output folders. Some folders are only existing in the full version of ENVI-met, others may be added in the course of time. All files in one folder are of the same structure and contain the same set of information, some of the output folders are further organized in sub folders. The following list summarizes the basic content of the folders. For more information on the file types, see sections below.
STATIC
) as well as dynamic data of meteorology along the building and building physics data like temperatures and fluxes (folder DYNAMIC
, not in ENVI-met LITE). In addition, summary files about the different buildings are included.STATIC
folder for constant properties and a DYNAMIC
folder with time-dependent properties of the plants.In addition, the BIOMet tool will create further folders, by default labled “Biomet”, but with a free choice of names. Future versions or additional modules of ENVI-met will create further folders and file, but the general concept of storage will be the same.
Sorting the output files into folders provides a first system of order in the output data. However, once a file has been moved out of its folder or if several simulations come together, this system is not unique and not persistent. As a solution, the ENVI-met file name generation scheme allows a direct identification of the simulation files and their content.
Each ENVI-met output filename consists of 3 parts:
urbanLayout
has been selected as basename._AT_
defines that this file holds atmospheric data. The different content ids are listed in the description of the output files categories and at the end of this page. _AT_
) at 08:00:01 model time on the 15th July (07) 2018. If the data in the file are not time sensitive, this part of the filename is empty. For pure text files, which might be processed in other programs or using the Python-based DataStudio, it is not possible to include metadata or other addtional data in the files except a header line defining the variables stored in the different colums of the file.
From ENVI-met Version 5 on, all text files are comma-seperated (CSV) files that can be processed directly in Python/Pandas or other software. In order to identify the contents of the file and suggest the correct scripts in DataStudio, the file extension is requried for these file types.
Level 1 and 2 provide a good information basis about the content of a simulations file and they where the only sources of information in ENVI-met versions prior to V4. However, files get moved and files get renamed. For that reason, starting from ENVI-met Version 4, each data file again holds detailed information about its data type in the metadata stored in the EDX information file, in the <data_content>
section. For all ENVI-met tools, this is the evaluated information when working with data files. All ENVI-met tools try to interpret data files coming from older ENVI-met versions, but this import process might not work in all cases.
The list below lists and links to the basic main files in binary format generated by the ENVI-met simulations.
This list gives an overview over the different text output files generated by ENVI-met. All files are CSV text files that can be loaded directly in to Python (Pandas). Not all files may be generated in each simulation. This list refers to ENVI-met V5 and newer.
ID | Content | Remark |
---|---|---|
.AT_1DR | Receptor Atmospheric data (Profile) | Vertical profile atmosphere at receptor at a given model time |
.AT_1DT | Receptor Atmospheric data (Time Series) | Vertical profile atmosphere at receptor for all model times of simulation in one file |
.FX_1DT | Receptor Surface data (Time Series) | Time series of ground surface state at receptor for all model times |
.SO_1DR | Receptor Soil data (Profile) | Vertical profile soil at receptor at a given model time |
.SO_1DT | Receptor Soil data (Time Series) | Vertical profile soil at receptor for all model times of simulation in one file |
.BLDG_status | Single Building Status (Time Series) | Actual status of individual buildings for all model times of simulation in one file (one for each building) |
.BLDG_statistics | Building Statistics | Summary of static parameters for the buildings (one for each building) |
.BLDG_list | Buildings Inventory | List of all buildings in model (all buildings in one file) |
.VEG_status | Vegetation Status (Time Series) | Actual status of 3D vegetation for all model times of simulation (only for observed plants) |
.VEG_list | Vegetation Inventory | List of all 3D plants and their key properties in the model |
.Inflow1D | Profile Atmosphere Model at inflow boundary | State of the 1D inflow boundary layer model at a given time up to 2500 m |
Please note, for all receptor files, you can produce output files with a higher resolution than the main output file intervall. All other files will be generated when the main model files are saved.
To import ENVI-met output text files with time series into Python Pandas, use:
import pandas as pd # this imports the (ENVI-met) text file "mysourcefile" into a Panda dataframe object. # Comma "," is selected as seperator sign, # the "DateTime" column is flaged to be handeld as date information # and also is set as the default index column for data processing df = pd.read_csv("mysourcefile", sep=",", parse_dates=['DateTime'], index_col=['DateTime'])
Each Time Series files now contains a Python-compatible DataTime column to allow a direct use of the different Date and Time functions supplied by Python.
The LEONARDO DataStudio automatically recognizes the content of the ENVI-met output text files and offers specific scripts matching the file types.