From fd26a28eae5bc93ac8600fdc8a7132787c0ac499 Mon Sep 17 00:00:00 2001 From: Austin Raney Date: Fri, 25 Oct 2024 12:34:34 -0400 Subject: [PATCH] fix!: correct site service crs to EPSG:4269 (NAD83) The nwis site service (and the USGS more broadly) uses the NAD83 crs. 64a575d introduced a bug that hard coded the nwis module's crs to EPSG:4236 -- Hu Tzu Shan 1950 (https://epsg.io/4236). I guarantee 64a575d meant to hard code _EPSG:4326_ -- WGS84. It appears it was fat fingered and missed in #143's PR review (it happens to the best of us:)). In either case, EPSG:4326 is still _technically_ the wrong crs. For example: https://waterservices.usgs.gov/nwis/site?sites=01013500&siteOutput=Expanded&format=rdb returns ``` USGS 01013500 ... NAD83 ``` for the field `coord_datum_cd -- Latitude-longitude datum`. So EPSG:4269 -- NAD83 is the correct crs. There is room for discussion if NAD83 vs WGS 84 really matters in the conterminous US however for Alaska errors in both the x and y direction can be upwards of 1 meter in either direction. BREAKING CHANGE: v1.0.10 is the only affected version. The GeoDataFrames returned from the following functions now correctly have a crs of EPSG:4269. dataretrieval.nwis.get_qwdata dataretrieval.nwis.get_discharge_measurements dataretrieval.nwis.get_discharge_peaks dataretrieval.nwis.get_gwlevels dataretrieval.nwis.get_stats dataretrieval.nwis.get_info dataretrieval.nwis.get_pmcodes dataretrieval.nwis.get_water_use dataretrieval.nwis.get_ratings dataretrieval.nwis.what_sites dataretrieval.nwis.get_dv dataretrieval.nwis.get_iv --- dataretrieval/nwis.py | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/dataretrieval/nwis.py b/dataretrieval/nwis.py index 60c132d..20c25ca 100644 --- a/dataretrieval/nwis.py +++ b/dataretrieval/nwis.py @@ -44,7 +44,8 @@ "water_use", "ratings", ] -_CRS = "EPSG:4236" +# NAD83 +_CRS = "EPSG:4269" def format_response(