CAN: Canadian bird species distribution data

In rspatial/disdat: Data for Comparing Species Distribution Modeling Methods

Canadian bird species distribution data

Description

Species occurrence data for 20 bird species from Ontario, a province in Canada (CAN), and associated environmental data. Full details of the dataset are provided in the reference below. There are four data sets with training (po and bg) and test (pa, env) data:

po (training data) includes site names, species names, coordinates, occurrence ("1" for all, since all are presence records), group (bird), and site values for 11 environmental variables (below).

bg (training data) has 10000 sites selected at random across the study region. It is structured identically to CANtrain_po, with "0" for occurrence (not implying absence, but denoting background in a way suited to most modelling methods) and "NA" for group.

env (testing data) includes group, site names, coordinates, and site values for 11 environmental variables (below), at 14571 sites. This file is suited to making predictions.

pa (testing data) includes group, site names, coordinates, and presence-absence records, one column per species. The sites are identical to the sites in env. This file is suited to evaluating the predictions made to env.

Raster (gridded) data for all environmental variables are available - see the reference below for details.

The reference system of the x and y coordinates is unprojected with Clarke 1866 ellipsoid . Latitude and longitude are in geographical coordinates using unknown datum based upon the Clarke 1866 ellipsoid (EPSG:4008).

The vignette provided with this package provides an example of how to fit and evaluate a model with these data.

Environmental variables:

Code	Description	Units	Type
alt	Digital elevation	m	Continuous
asp2	Aspect	ranges from -1 to 1 (sin transformation)	Continuous
ontprec	Annual Precipitation	mm	Continuous
ontprec4	April precipitation	mm	Continuous
ontprecsd	Precipitation Seasonality	dimensionless	Continuous
ontslp	Slope	degrees	Continuous
onttemp	Annual mean temperature	degrees C * 10	Continuous
onttempsd	Temperature standard deviation	dimensionless	Continuous
onttmin4	April minimum temperature	degrees C * 10	Continuous
ontveg	Vegetation, from Ontario Land Cover Database (OLC) vegetation map, derived from a mosaic of Landsat images.	5 classes: 1 = open forest & related; 2 = closed forest; 3 = open water, 4 = agriculture, 5 = human settlement	Categorical
watdist	Distance from Hudson Bay	m	Continuous

Source

Environmental predictors prepared by Falk Huettmann, Jane Elith and Catherine Graham. Species data: PO from the Ontario Nest Records database, Royal Ontario Museum (ROM) and supplied by M. Peck to Falk Huettmann; PA from Breeding Bird Atlas for Ontario, provided by M. Cadman to Falk Huettmann.

See the reference below for further details on source, accuracy, cleaning, and particular characteristics of these datasets.

References

Elith, J., Graham, C.H., Valavi, R., Abegg, M., Bruce, C., Ferrier, S., Ford, A., Guisan, A., Hijmans, R.J., Huettmann, F., Lohmann, L.G., Loiselle, B.A., Moritz, C., Overton, J.McC., Peterson, A.T., Phillips, S., Richardson, K., Williams, S., Wiser, S.K., Wohlgemuth, T. & Zimmermann, N.E., (2020). Presence-only and presence-absence data for comparing species distribution modeling methods. Biodiversity Informatics 15:69-80.

Examples

can_po <- disPo("CAN")
can_bg <- disBg("CAN")

can_pa <- disPa("CAN")
can_env <- disEnv("CAN")


# Or all in one list
x <- disData("CAN")
sapply(x, head)

disCRS("CAN")

rspatial/disdat documentation built on Feb. 14, 2023, 4:27 a.m.