forest_puller is a python package for retrieving data concerning forests of European countries. This includes the amount of forested areas, the forest inventory (standing stock), the forest growth rates as well as the forest loss dynamics (disturbances).
This software is accompanied by a scientific publication that is available here: https://www.mdpi.com/1999-4907/12/2/176
There are several forest public data sources accessible online that provide these types of information in various forms and granularity. This package automates the process of scrapping these websites and parsing the resulting CSV tables or excel files.
Once forest_puller is installed you can easily access forest data through standard python pandas data frames.
Currently forest_puller provides data for the following 27 EU member states (past and present):
- Austria, Belgium, Bulgaria, Croatia, Cyprus, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Latvia, Lithuania, Luxembourg, Netherlands, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, Sweden, United Kingdom
Currently forest_puller caches and provides programmatic access to the forest-relevant data from these data sources:
- IPCC (https://tinyurl.com/y474yu9e)
- SOEF (https://dbsoef.foresteurope.org/)
- FAOSTAT (http://www.fao.org/faostat/en/)
- HPFFRE (https://doi.org/10.5061/dryad.4t880qh)
- FRA (http://www.fao.org/forest-resources-assessment/en/)
What other data source you would like to see added here? Contact the authors by opening an issue in the the tracker.
Since forest_puller is written in python, it is compatible with all operating systems: Linux, macOS and Windows. The only prerequisite is python3 (which is often installed by default) along with the pip3 package manager.
To check if you have python3 installed, type the following on your terminal:
$ python3 -V
If you do not have python3 installed, please refer to the section obtaining python3.
To check if you have pip3 installed, type the following on your terminal:
$ pip3 -V
If you do not have pip3 installed, please refer to the section obtaining pip3.
Simply type the following on your terminal:
$ pip3 install --user forest_puller
Or if you want to install it for all users of the system:
$ sudo pip3 install forest_puller
For instance, to retrieve the net carbon dioxide emission of Austria in 2017 that were due to coniferous forest land from the IPCC official data source, you can use the following python code:
# Import #
from forest_puller.ipcc.country import countries
# Get the country #
austria = countries['AT']
# Get the 2017 indexed dataframe #
at_2017 = austria.years[2017].indexed
# Print some data #
print(at_2017.loc['remaining_forest', 'Coniferous']['net_co2']) 904282.4970403439
To see what other information is available, you can of course display the column titles and row indexes of the data frame at hand:
print(at_2017.columns)
# Index(['area', 'area_mineral', 'area_organic', 'biomass_gains_ratio',
# 'biomass_losses_ratio', 'biomass_net_change_ratio', 'net_dead_ratio',
# 'net_litter_ratio', 'net_mineral_soil_ratio', 'net_organic_soil_ratio',
# 'biomass_gains', 'biomass_losses', 'biomass_net_change', 'net_dead',
# 'net_litter', 'net_mineral_soils', 'net_organic_soils', 'net_co2'],
# dtype='object', name='category')
print(at_2017.index)
# MultiIndex(levels=[['cropland_to_forest', 'grassland_to_forest',
# 'land_to_forest', 'other_land_to_forest', 'remaining_forest',
# 'settlements_to_forest', 'total_forest', 'wetlands_to_forest'],
# ['', 'Coniferous', 'Deciduous', 'Forest not in yield', 'Total']])To examine what countries and what years are available:
print(list(c.iso2_code for c in countries.values()))
# ['AT', 'BE', 'BG', 'HR', 'CZ', 'DK', 'EE', 'FI', 'FR', 'DE', 'GR',
# 'HU', 'IE', 'IT', 'LV', 'LT', 'LU', 'NL', 'PL', 'PT', 'RO', 'SK', 'SI',
# 'ES', 'SE', 'GB', 'ZZ']
print(list(y for y in austria.years))
# [1990, 1991, 1992, 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000,
# 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008, 2009, 2010, 2011, 2012,
# 2013, 2014, 2015, 2016, 2017]To get a large data frame with all years and all countries inside:
from forest_puller.ipcc.concat import df
print(df)To get a data frame that compares forest area between several sources:
from forest_puller.viz.area_comp import area_comp_data
print(area_comp_data.df)To get a data frame that compares gains (increments) and losses (harvest) between sources:
from forest_puller.viz.converted_to_tons import converted_tons_data
print(converted_tons_data.df)When you import forest_puller, we will check the $FOREST_PULLER_CACHE environment variable to see where to download and store the cached data. If this variable is not set, we will default to ~/.forest_puller and clone a repository there.
To access the same forest data directly from the IPCC website without the use of forest_puller, you would have to first select your country of interest from the CRF country table in a browser at this address.
Then you would have to manually download the zip file for that specific country through another page.
Next, you would have to uncompress the zip file and locate the xls file that concerns the year you are interested in.
Finally you would have to scroll to the right sheet in your spreadsheet software and find the pertinent cell.
This operation would have to be repeated for every country, and every year you are interested in.
With forest_puller you can easily display any information you want for all countries at the same time:
from forest_puller.ipcc.country import countries
category, key = ['total_forest', 'biomass_net_change']
biomass_net_change = {
k: c.last_year.indexed.loc[category, ''][key]
for k,c in countries.items()
}
import pprint
pprint.pprint(biomass_net_change){'AT': 1367857.0940855271,
'BE': 374245.08695361385,
'BG': 2192942.031982918,
'CZ': 387870.89395249996,
'DE': 12317598.87352293,
'DK': -216454.31026543948,
'EE': 320710.2459538891,
'ES': 8917649.261547482,
'FI': 6603815.0,
'FR': 15051831.9827214,
'GB': 2892518.0859005335,
'GR': 583205.0978272819,
'HR': 1477791.7578513895,
'HU': 1259385.5890665338,
'IE': 1069648.7636722159,
'IT': 5752883.095908434,
'LT': 2146933.309581986,
'LU': 101929.37461705346,
'LV': 1244965.2120000012,
'NL': 499021.93968,
'PL': 9353198.2907701,
'PT': 1536917.4736652463,
'RO': 5561343.4405591395,
'SE': 10185839.738999998,
'SI': 35391.09710503432,
'SK': 1184611.3471376207}
This data is provided by the "Ministerial Conference on the Protection of Forests in Europe" and is accessible at: https://dbsoef.foresteurope.org/
Three tables are provided for every country:
- Table 1.1a: Forest area
- Table 1.3a1: Age class distribution (area of even-aged stands)
- Table 3.1: Increment and fellings
It is accessed in a similar way to other data sources:
from forest_puller.soef.country import countries
country = countries['AT']
print(country.forest_area.indexed)
print(country.age_dist.indexed)
print(country.fellings.indexed)There is also a large data frame containing all countries concatenated together:
from forest_puller.soef.concat import tables
print(tables['forest_area'])
print(tables['age_dist'])
print(tables['fellings'])This data is acquired by picking the "All Data Normalized" option from the "Bulk download" sidebar at this address: http://www.fao.org/faostat/en/#data/FO
It is accessed in a similar way to other data sources:
from forest_puller.faostat.forestry.country import countries
country = countries['AT']
print(country.df)There is also a large data frame containing all countries concatenated together:
from forest_puller.faostat.forestry.concat import df
print(df)This data is acquired by picking the "All Data Normalized" option from the "Bulk download" sidebar at this address: http://www.fao.org/faostat/en/#data/GF
It is accessed in a similar way to other data sources:
from forest_puller.faostat.land.country import countries
country = countries['AT']
print(country.df)There is also a large data frame containing all countries concatenated together:
from forest_puller.faostat.land.concat import df
print(df)Diabolo was a project run by a consortium of 33 partners from 25 countries. Experts in the fields of policy analysis, forest inventory, forest modelling. 7 work packages.
Link: http://diabolo-project.eu/
One of the outcomes of the Diabolo project is the following publication:
Vauhkonen et al. 2019 - Harmonised projections of future forest resources in Europe
Abbreviated "hpffre". The authors used EFDM (mainly) to project forest area, growing stock, fellings and above ground carbon for European countries. There are several scenario outcomes.
The dataset is available at: https://doi.org/10.5061/dryad.4t880qh
It is accessed in a similar way to other data sources:
from forest_puller.hpffre.country import countries
country = countries['AT']
print(country.df)There is also a large data frame containing all countries concatenated together:
from forest_puller.hpffre.concat import df
print(df)FRA stands for "Forest Resource Assessment".
The data is acquired by picking the "CSV download" option from the page located at: http://countrystat.org/home.aspx?c=FOR&tr=3
The home page is at: http://www.fao.org/forest-resources-assessment/en/
Five datasets are provided in the package:
- "Forest characteristics (1 000 ha) by FRA categories"
- "Extent of forest and other wooded land (1 000 ha)"
- "Forest establishment total (ha/yr) by FRA categories"
- "Carbon stock (Million metric tonnes) by Forest/Other wooded land"
- "Biomass stock (Million metric tonnes) by Forest/Other wooded land"
FRA is an effort of the Food and Agriculture Organization of the United Nations (FAO)".
It is accessed in a similar way to other data sources:
from forest_puller.fra.country import countries
country = countries['AT']
print(country.df)There is also a large data frame containing all countries concatenated together:
from forest_puller.fra.concat import df
print(df)The forest_puller package can also generate several plots that enable the user to compare and visualize the data.
For instance here is are a series of graphs comparing the total reported forest area between data sources as seen in the forest_puller.viz.area submodule:
Another type of graph that can be produced is the comparison of gains and losses across several data-sources and across countries. This code is found in the forest_puller.viz.increments submodule and shows the five largest countries in terms of forest area.
With data from the SOEF source, we can also plot a breakdown of the growing stock volume genus composition of many countries across time. This code is found in the forest_puller.viz.genus_barstack submodule.
All the visualization produced by the package are assembled inside an auto-generated PDF report. You can download a cached version of this report here:
|
|
Comparison report |
If you encounter an error when using the forest_puller package, please open an issue on the tracker and we will get in contact with you.
More documentation is available at:
http://xapple.github.io/forest_puller/forest_puller
This documentation is simply generated from the source code with:
$ pdoc --html --output-dir docs --force forest_puller