Data Model¶
Todo
Update once new data model is agreed to
Data fed into aneris is assumed to be so-called wide-style panel
data. Within the code base, standard Pandas :code:`DataFrame`s are used. When
using the CLI, Excel spreadsheets or csvs are supported.
Variable Names¶
The most important data underlying harmonizations are timeseries. Timeseries are defined by a Variable and a number of Data columns for each timestep in the timeseries. Variable names are assumed to follow the IAMC style and have the form:
<prefix>|Emissions|<gas>|<sector>|<suffix>
prefix: a designation for the current study (e.g., “CEDS”)
gas: the emissions species (e.g., “BC”)
sector: the emissions sector (e.g., “Transportation”)
suffix: a designation for raw-model data (e.g., “Unharmonized”)
Importantly, model data variable names must match historical data variable names exactly.
Unharmonized IAM Data¶
Data from IAMs is expected to be in the following format with a sheetname “data”.
If overrides are provided, they are expected to be in the following formay with a sheetname “harmonization”.
Additionally, configuration parameters (described in Run Control and Configuration) can be set by two columns titled “Configuration” and “Value” in the harmonization sheet.
Historical Data¶
Historical data is expected to be in the following format
Regional Definitions¶
Data for regional mappings (countries to IAM regions) is expected to be in the following format
ISO Code |
Country |
Native Region Code |
5_region |
|---|---|---|---|
regiona |
Region A |
regionc |
R5LAM |
regionb |
Region B |
regionc |
R5LAM |
World |
World |
World |
World |
