search

pySTEPS / pysteps

Python framework for short-term ensemble prediction systems.
https://pysteps.github.io/
BSD 3-Clause "New" or "Revised" License
441 stars 160 forks source link

Can pysteps be applied to prediction of non-radar rainfall data? for example imerg? #359

Closed sunmoumou1 closed 3 months ago

sunmoumou1 commented 3 months ago

I have recently been working on a project that involves using Imerg data to complete rainfall forecasts.

Specifically, I now have a variable, sample, which is a ndarray with a shape of (989811). How can I write code that uses 96968 of the first 8 frames to predict 96963 of the next 3 frames?

Thanks for answering my question. I've looked at the tutorials, but found that they all use radar data, and wonder if the pysteps library can be extended to the situation I'm talking about

aperezhortal commented 3 months ago

Hello @sunmoumou1

Thank you for using PySteps!

PySteps can work with any type of precipitation measurement (radar, satellite, or model) as long as you map the data to a Cartesian Grid. This is required for the optical flow methods to work correctly.

After you have the precipitation data in that grid, you can use PySteps to run the forecasts.

Let us know if you have any other questions.

aperezhortal commented 3 months ago

Hello @sunmoumou1 ,

Did the previous answer help? Can we close the issue?

Thanks!

dnerini commented 3 months ago

closing this for now, see also #339.