This year's challenge, GeoLifeCLEF 2021, uses the same data as last year's one, GeoLifeCLEF 2020, with some modifications in the format to make it both smaller (~80Go instead if ~850Go) and easier to use.
The challenge relies on a collection of millions of observations of plants and animals in the US and France, coming from the
[iNaturalist](https://www.inaturalist.org/)
and
[Pl@ntNet](https://plantnet.org/)
citizen science platforms.
Each observation consists of a species name with the GPS coordinates where it was observed.
In addition, observations are paired with a set of covariates characterizing the landscape and environment around them.
Covariates include high-resolution remote sensing imagery, land cover data, and altitude, as well as traditional low-resolution climate and soil variables.
![Observation distribution in France](https://raw.githubusercontent.com/maximiliense/GLC/GLC21/images/distribution_FR.png)
![Observation distribution in the US](https://raw.githubusercontent.com/maximiliense/GLC/GLC21/images/distribution_US.png)
In more detail, each occurrence is paired with the following covariates:
Remote sensing imagery: 256mx256m RGB-IR patches centered at each observation
Format: 256x256 JPEG images, a color JPEG file for RGB data and a grayscale one for near-infrared
Resolution: 1 meter per pixel
Source: NAIP for US and IGN for France
Land cover data: 256mx256m patches centered at each observation
Format: 256x256 TIFF files with Deflate compression
Resolution: 1 meter per pixel
Source: NLCD for US and Cesbio for France
Altitude data: 256mx256m patches centered at each observation
Format: 256x256 TIFF files with Deflate compression
Resolution: 1 meter per pixel
Source: SRTMGL1 for US and France
Bioclimatic data: 19 low-resolution rasters
Format: GeoTIFF files with LZW compression
Resolution: 30 arcsec per pixel (~ 1 kilometer per pixel)
Source: WorldClim for US and France
Pedologic data: 8 low-resolution rasters
Format: GeoTIFF files with LZW compression
Resolution: 250 meters per pixel
Source: SoilGrids for US and France
Here is an example of the patches corresponding to observation
10561900
:
![Observation distribution in the US](https://raw.githubusercontent.com/maximiliense/GLC/GLC21/images/patches_sample.jpg)
A workthrough notebook describing the data structure and how to load it and visualize it is available in the
[Code tab](https://www.kaggle.com/tlorieul/geolifeclef2021-data-loading-and-visualization)
.
A complete description of how the dataset was built can be found in the
[associated paper](https://arxiv.org/abs/2004.04192)
.
A script to download all the data at once as well as data loading and visualization utilities are provided on the
[GitHub of the competition](https://github.com/maximiliense/GLC)
.
On this competition page, the observations data, the associated metadata, the rasters and pre-extracted features are provided.
Only a subsample of the patches are available here.
The complete set of patches can downloaded using those links:
RGB patches:
[France (12.8GB)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_fr_rgb.tar.gz)
[US (14.0GB)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_us_rgb.tar.gz)
near-infrared patches:
[France (10.7Go)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_fr_near_ir.tar.gz)
[US (12.7Go)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_us_near_ir.tar.gz)
land cover patches:
[France (0.4Go)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_fr_landcover.tar.gz)
[US (1.0Go)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_us_landcover.tar.gz)
altitude patches:
[France (3.1Go)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_fr_altitude.tar.gz)
[US (4.1Go)](https://lilacdn.azureedge.net/geolifeclef-2020/patches_us_altitude.tar.gz)