How can I select data points based on similar coordinates and year with Python?

Question

I have two datasets (shp files (converted into csv)) in which one contains fire history and the other lightning strikes. Both data sets have latitude and longitude coordinates (same projection) and year of fire/lightning strike. I want to be able to select the lightning strike that caused the fire a certain year. Thus, I need to... 1) select the lightning strikes within a certain radius of the fire 2) from these lightning strikes select those within the same year.

Additional: I need it for every fire area (1300 areas) so yes automated as manual selection would take way too long. Further, I have included all lightning strikes of a fire area within a buffer of 1 km from the outer parts of the fire area. So in terms of coordinates, we are not talking about changes of more than .1 or .2 The temporal uncertainty can be bigger, as the lightning dataset varies largely within itself. Let's say 3 to 4 weeks of uncertainty. Oh, and output is less important, preferably a shape file. Thank you for your interest. –

I have tried with gpd.sjoin, however, this only joins the two data sets.

Wouldn't it make sense to check the year first, and then get the distance? The distance checking is going to be much more resource intensive than the year checking. — user1558604, Dec 16 '19 at 16:59
I think you'd be better off asking this at gis.stackexchange.com — Paul H, Dec 16 '19 at 18:17

Thomas · Answer 1 · 2019-12-17T11:38:58.357

Thank you for your reply.

Well, these datasets are CSV files that are imported in Python. Running this in GIS will take to much computing power as it is at the moment. I will have to conduct the task in Python and create new data files from my findings.

E.g. Within or near-by a fire area (burned: 17/06/2014) (coordinates: 67.2, 129.9) there are multiple lightning strikes (from different years). In total, I have more than 1300 fire areas and 27,000 lightning strikes within a given period. Therefore, I do not believe I can start off with comparing dates of fire and lightning as this will result in joining lightning and fire areas with different coordinates.

Additional: I need it for every fire area (1300 areas) so yes automated as manual selection would take way too long. Further, I have included all lightning strikes of a fire area within a buffer of 1 km from the outer parts of the fire area. So in terms of coordinates, we are not talking about changes of more than .1 or .2 The temporal uncertainty can be bigger, as the lightning dataset varies largely within itself. Let's say 3 to 4 weeks of uncertainty. Oh, and output is less important, preferably a shape file. Thank you for your interest. –

How can I select data points based on similar coordinates and year with Python?

1 Answers1