Python script that attempts to identify records in GWELLs having coordinates that need attention.
Installation of the script and requirements is easiest via conda:
git clone https://github.com/smnorris/gwells_locationqa
cd gwells_locationqa
conda env create -f environment.yml
conda activate gwells_locationqaMost required data are downloaded by the script but the ESA WorldCover download is a separate task.
To download tiff tiles for BC, merge, and warp to BC Albers, ensure awscli and gdal are available at the command line and run the script:
./get_esa_worldcover_bc.shThe provided script is for bash or similar but can be modified to run on windows with only a couple of changes.
For more info on the WorldCover dataset:
Run the job in several steps:
python gwells_locationqa.py download # download required data to /data folder (gwells and pmbc parcel fabric)
python gwells_locationqa.py geocode <GEOCODER_API_KEY> # reverse-geocode all wells (requires ~6hrs)
python gwells_locationqa.py qa # match wells pids to pmbc, match addresses, overlay with agriculture data sourcesOutput file is gwells_locationqa.csv. All wells with a lat/lon are included.
Output columns are as per the source well.csv file, with the following additions:
| Column | Description |
|---|---|
| nr_district_name | Natural Resource District |
| nr_region_name | Natural Resource Region |
| fullAddress | geocoder result |
| streetAddress | geocoder results parsed to match gwells address |
| civicNumber | geocoder result |
| civicNumberSuffix | geocoder result |
| streetName | geocoder result |
| streetType | geocoder result |
| isStreetTypePrefix | geocoder result |
| streetDirection | geocoder result |
| isStreetDirectionPrefix | geocoder result |
| streetQualifier | geocoder result |
| localityName | geocoder result |
| distance_geocode | distance from well to result of geocode (value 99999 indicates no result) |
| distance_to_matching_pid | distance from well to BC Parcel Fabric polyon with matching PID (0 indicates the well point intersects the matching PID polygon, NULL indicates no matching PID found) |
| score_address | Token Set Ratio score for matching wells address to reverse geocoded address (street number/name/direction) |
| score_location_description | Token Set Ratio score for matching wells location description to reverse geocoded full address |
| score_city | Token Set Ratio score for matching wells city to reverse geocoded locality |
| xref_ind | Indicates if one of these strings found in comments column: CROSS R,CROSS-R,REF'D,REFERENCED,REFD,XREF,X-R,X R |
| alr_ind | Indicates if the well is within the ALR as defined by WHSE_LEGAL_ADMIN_BOUNDARIES.OATS_ALR_POLYS |
| btm_label | BTM PRESENT_LAND_USE_LABEL at well location |
| esa_landclass | ESA WorldCover land class at well location |
See notebooks/locationqa.ipynb for an example of how to work through the various address matching scores.