simondedman / marspatauto Goto Github PK

Lunar phase can be important but for animals that might take cues from the light of the moon this may be an important covariate

I currently pull the data down from ERRDAP often for multiple locations across years. Since it doesn't let you do this all in one go (I think it only downloads one month at a time) I adapted code from Luke Miller https://github.com/millerlp/Misc_R_scripts/blob/master/coops_water_temp_retrieval_monthly_v3.R
Currently I use it to get hourly average barometric pressure

Can be used as a proxy to look a differences in night vs day time. I have been using the Global Horizontal Irradiance from the below website. Solar irradiance is recorded by 30 min bins

https://maps.nrel.gov/nsrdb-viewer/?aL=0&bL=groad&cE=0&lR=0&mC=28.616774693775938%2C-80.60737609863281&zL=11

Calculated from size at tag date & forward-projected on a daily basis.

https://noaa-nos-ofs-pds.s3.amazonaws.com/index.html

Accessing these on the Python side is not too hard with xarray and zarr, the R workflow is still not known to me however. An updating bucket with the last 3+ years of nowcast/forecast model results for many ocean variables.

Another hurdle may be the irregular grid models.

2019-11-03 used Secoora for sawfish project but not sure which dataset?

NCEI

NOAA NGDC (what I used, AKA: NOS SEDS).
Convert codes to names with Historic Codes pdf.
historic_codes.pdf
Convert names to substrate grain size with Ol Faithful:
Substrate Size Classes.xlsx
Needs processing in GIS: buffer parameters, intersection parameters, point to polygons, (polygon to points). Read this.

See /home/simon/Documents/Si Work/PostDoc Work/Kroetz & Dedman Sawfish BRT/Substrate/NOAA-NOA-USGS Seabed descriptions/

As a different option to solar irradiance sometimes you want a categorical variable for day/night/or crepuscular periods. The function I currently use pulls down sunrise/sunset times using package suncalc (for the local time) and then adds a column to the data "TimeofDay" with either a "Day" or "night" distinction. It also has an option to add another category of crepuscular. The user can choose if crepuscular means 1 hour before/after sunrise and sunset or can change it to be 2 hours etc and it will include a "dawn" or "dusk" category.

I currently pull the data down from ERRDAP often for multiple locations across years. Since it doesn't let you do this all in one go (I think it only downloads one month at a time) I adapted code from Luke Miller https://github.com/millerlp/Misc_R_scripts/blob/master/coops_water_temp_retrieval_monthly_v3.R
Currently I use it to get hourly average wind speed
(I don't know if any of my notes are useful but wanted to let you know how I currently do it to try to help! Definitely interested in a more streamlined process).