This dataset contains forecasts for the Finnish building stock for the years 2020, 2030, 2040, and 2050. The main data represents the estimated number of buildings by building stock year, type of the building, building period, municipality, and source of heating, as well as the estimated average gross-floor areas of buildings by municipality, type of building, and building period. See the end of the README for a detailed description of the dataset. ## Contents Here's a quick overview of the important contents of this dataset, and a brief description of the contents of each file. For a detailed description of the dataset, see the end of this README. 1. `datapackage.json`, the [Data Package](https://specs.frictionlessdata.io/data-package/) definition. 2. `forecasts2csvs.bat`: A script for processing the original raw `.xlsx` files into the translated `.csv` files using [BuildingStockForecasts.jl](https://vttgit.vtt.fi/flexib/BuildingStockForecasts). Mainly included for example purposes. 3. `import_finnish_building_stock_forecasts.json`, the [Spine Toolbox](https://github.com/Spine-project/Spine-Toolbox) importer specification. 4. `LICENSE` contains information about the permitted use of this dataset. 5. `data/average_floor_areas_m2.csv`: Contains estimated average gross-floor areas of buildings by `municipality`, `building_type` and `building_period`. 6. `data/building_periods.csv`: Contains each `building_period` with their start and end years as integers. 7. `data/frame_material_shares.csv`: Contains `municipality`-level data of building frame material shares across `building_types`. 8. `data/municipalities.csv`: Contains the list of included `municipality` IDs and names. 9. `data/number_of_buildings.csv`: Contains the building stock forecasts as estimated number of buildings by `building_stock_year`, `building_type`, `building_period`, `municipality`, and `heat_source`. ## Usage The dataset is formatted as a [Data Package](https://specs.frictionlessdata.io/data-package/), hopefully facilitating interoperability and reuse. However, this dataset was originally designed to be used via [Spine Toolbox](https://github.com/Spine-project/Spine-Toolbox), and thus contains the necessary definitions to easily set up an importer into a Spine Datastore: 1. Add a *Data Connection* item, and refer it to the `datapackage.json` file. 2. Use the *Add specification from a file* option from the toolbar, and select the `import_finnish_building_stock_forecasts.json` file to add the necessary *Importer* definition. 3. Add an *Importer* item, connect the above *Data Connection* to it, and select the above importer specification for it. 4. Connect the *Importer* to the desired *Spine Datastore* where you want the data to be imported to. ## License All rights reserved, see `LICENSE` for more information. ## Acknowledgements Original dataset and the building stock forecasts were created by Terttu Vainio at VTT. Topi Rasku merely gathered all the data in the original dataset *(lots of .xlsx files)* into fewer .csv files, as well as translated the data into English. This dataset was built for the [Academy of Finland](https://www.aka.fi/en) project [*"Integration of building flexibility into future energy systems (FlexiB)"*](https://akareport.aka.fi/ibi_apps/WFServlet?IBIF_ex=x_hakkuvaus2&CLICKED_ON=&HAKNRO1=332421&UILANG=en&TULOSTE=HTML) under grant agreement No 332421. [CSC – IT Center for Science, Finland](https://csc.fi/en/home) provides the [Fairdata.fi](https://www.fairdata.fi/en/) service for easy research metadata publication. # Dataset description This section hopes to describe the raw data in enough detail to give you a better idea of what is and isn't included in the dataset. Since significant parts of the raw data is owned by [the Finnish Digital Agency](https://dvv.fi/en/digital-and-population-data-services-agency), it is likely that only the descriptive metadata of this dataset can ever be published. However, access to this dataset for academic use can be requested from the copyright holders. In general, the dataset contains 5 key dimensions over which the data is organized: - `building_stock_year`: The year for which the building stock data is for. The years 2030, 2040, and 2050 are forecasts based on assumptions explained briefly below using forecasting models at VTT. The included building stock years are: - `2020`, `2030`, `2040`, `2050`. - `building_type`: The type of the building, based on the divisions in Official Statistics Finland, and the Building and Dwelling Register. The building types in the data include: - `apartment_block`, `communal_building`, `detached_house`, `office_building`, `service_building`, `terraced_house` - `building_period`: The time period during which the building was built. Division based on Official Statistics Finland data. The building periods in the data include: - `0000-1920`, `1921-1939`, `1940-1959`, `1960-1969`, `1970-1979`, `1980-1989`, `1990-1999`, `2000-2009`, `2010-2019`, `2020-2029`, `2030-2039`, `2040-2049` - `municipality` as a `(location_id, location_name)` pair: Finnish municipalities as they were at the end of the year 2020, except for those in Ahvenanmaa. - 294 municipalities in total. - `heat_source`: The primary heat source of the building, e.g. fuel or electricity. The included heat sources are: - `district_heating`, `light_oil`, `heavy_oil`, `electricity`, `natural_gas`, `coal`, `wood`, `peat`, `ground_source_heat`, `other` Historical information about the gross floor area of the Finnish building stock is based on the *"Official Statistics Finland (OSF): Buildings and free-time residences [e-publication]. ISSN=1798-6796. Helsinki: Statistics Finland [referred: 1.3.2021], access method https://www.stat.fi/en/statistics/rakke"* The municipality level data about the heat source of buildings is based on the *Building and Dwelling Register (BDR), https://dvv.fi/kiinteisto-rakennus-ja-paikkatiedot* that VTT has access to for research purposes. The forecast for the heat sources follows Finnish energy and climate strategy, that aims to move away from fossil fuels by 2030 in the majority of cases, and completely by 2050. ## Important notes on data quality and scope ### Missing data The municipalities in Ahvenanmaa are missing from the dataset! Furthermore, the frame material data for KU687 Rautavaara was missing from the underlying data, so data from the neighbouring KU686 Rautalampi is used in its place. ### Included `building_type`s This dataset was originally compiled for analysing the future Finnish building stock heating/cooling demand, as well as its potential flexibility for energy system analysis purposes. Thus, it only contains a select subset of the Finnish building stock. The detailed mapping is based on the *(now-outdated)* [1994 Classification of Buildings](https://www.stat.fi/en/luokitukset/rakennus/rakennus_1_19940101/) as follows: - `apartment_block`: Includes `A03 - Blocks of flats`. - `communal_building`: Includes `F - Buildings for institutional care`, `G - Assembly buildings`, and `H - Educational buildings` - `detached_house`: Includes `A01 - Detached and semi-detached houses`. - `office_building`: Includes `D - Office buildings`. - `service_building`: Includes `C - Commercial buildings`, and parts of `E - Transport and communications buildings` that are assumed to be heated *(e.g. some large transit stations)*. - `terraced_house`: Includes `A02 - Attached houses`. Building categories excluded from this dataset include: - `B - Free-time residential buildings` due to inconsistent occupation and heating patterns. - Parts of `E - Transport and communications buildings` that are assumed not to be heated *(e.g. parking garages)*. - `J - Industrial buildings` and `M - Agricultural buildings` due to challenges in separating heating demand from process energy demand. - `K - Warehouses` and `L - Fire fighting and rescue service buildings` due to inconsistent heated volumes. - `N - Other buildings` due to insufficient information on their nature. ## Individual data file descriptions ### `data/average_floor_areas_m2.csv` This file contains a pivot table of the estimated average gross-floor areas of the buildings in square meters for all `(building_type, (location_id, location_name), building_period)`. The first three columns define the `building_type` and `(location_id, location_name)`, while the `building_period` is pivoted into the headers of the rest of the columns, as indicated by the header row of the `.csv` file. The estimated average gross-floor area in square metres is contained in the pivoted columns 4-15. ### `data/building_periods.csv` This file contains flat tabular definitions for all the possible `building_period`s. As indicated by the header row in the `.csv` file, the first column defines the `building_period`, the second column contains the `period_start` year as an integer, and the third column contains the `period_end` year as an integer. ### `data/frame_material_shares.csv` This file contains pivoted tabular data about the frame materials used in the Finnish building stock for all `(building_type, (location_id, location_name), frame_material)`, where `frame_material` includes `concrete`, `brick`, `steel`, `wood`, and `other`. The first three columns define the `building_type` and `(location_id, location_name)`, while the rest contain the shares of the pivoted `frame_material`, as indicated by the header row in the `.csv` file. ### `data/municipalities.csv` This file simply contains flat tabular definitions for all the possible `(location_id, location_name)` pairs, as indicated by the header row in the `.csv` file. ### `data/number_of_buildings.csv` This file contains pivoted tabular data about the number of buildings in the Finnish building stock for all `(building_stock, building_type, building_period, (location_id, location_name), heat_source)`. The first 5 columns define the `building_stock, building_type, building_period, (location_id, location_name)`, while the rest contain the pivoted number of buildings per `heat_source`, as indicated by the header row in the `.csv` file.

2022