That's great 👍 thank you so much. I've got many more in the pipeline, just had to have a bit of a break due to many work commitments and other things going on. But more on the way soon!
Thanks so much. All these nice feedbacks mean a lot, I'm really happy i can help. More videos are planned, just trying to cope with a lot of things in life at the moment, but promise that another 6 or 7 will come out over the May to July period. Thanks for subscribing, that way you will hopefully get news when the new videos are ready.
Very well explained Prof. Tompkins. I found a lot of other ways of creating land-sea mask on Stackoverflow but nothing of this quality and ease. Please please make many such videos. Model bias evaluation is something that I am really looking forward. Thank you very much.
Thanks for your kind comments. If you check out my latest video "combining gridded fields and broadcasting" it shows you exactly how you can calculate biases between models and observations... I hope it useful for you.
I'll add it to my to do list. I better do one on model drift too, just to avoid any bias 🤣 seriously though, I'll try to get to it asap, recently I've been a bit buried with teaching and paper writing... Not enough time in the day!
thanks professor for such a perfect video, May I ask which climate data have the highest resolution ? Cordex core as I know have 25 km ? what about other data on copernicus for CMIP6 climate projections ? Can I find data of 10 km or less resolution for Europe in particular for Italy ? thanks in advance
Thanks a lot for such a great video. It is very useful for me to make a mask. But I have a question about the mask of Antarctica. I found it is incomplete, and there are some missing grids. Could you please kindly tell me how to figure it out? Another question is that I have two global NC files with different resolutions. If I mask each file first, exclude the ocean areas to improve efficiency, then calculate indices for these two files, remap the results into 1*1 degree, and average them. I'm afraid that there will be some problem with the junction of land and sea areas because of the different original grid sizes due to the resolution. So I would like to make a mask which is bigger than my land area and also exclude most ocean areas, can I ask is there any way to do this?
Regarding the second part, excellent question. I would perhaps map both fields to the coarse grid. Average. And then apply the mask on the coarse grid. I'll think about it some more because you may also want to map the fine scale mask to the coarse, setting any cell to missing if the fine scale cell is missing, let me mull over how to do that best. Don't forget that the built in topography is only 0.5 deg. So for fine resolutions it is quite approximate. If you want anything better then it is preferable to download an appropriate topography or land sea mask file. These longer more detailed questions are easier to answer on stack overflow by the way, if you post there using the tag netcdf and cdo-climate it is easier to write a nice formatted answer. Thanks for your interest in my channel, i hope you find it helpful.
thanks Prof. i just wanna make sure about the data. If i didnt have the golbal data, example that only tropics data, and then to create mask data in my data, can be running? or we have to download the global data?
Unfortunately my country, the Netherlands, is mostly below sea level. It would simple dissappear from any land map this way. Any alternative to deal with polders?
Yes you are right, this method is inaccurate for pixels below sea levels (some areas near your coast, and also near the African rift for example) and has the opposite problem in that it classes elevated lakes as land. This is meant to be a quick soln that you can do on the fly without downloading extra data and works pretty well all said and done. There is a land-sea mask available in ERA5 on the native ERA grid, (var-lsm) - You can download that and compare the two... (use gec,0.5 to convert it to a binary field) - Overall they are not far off each other, just the issues above, and some noise along the coasts due to the different native grids (0.5 and 0.25 deg)
I followed similar process for masking my ocean data for precipitation from 1979 to 2019 but I got the masked data only for the first time level...next time steps were not done when masking was done.What shd be done.....????
@@gokulvishwanathan3993 I'd like to help but to do it I need to see the data. Can you post a question on stack overflow with it and tag it with cdo-climate?
Hi, I tried this method but I got the result like Obesrvation_data.nc file and Model.nc file after regrid the vale of the output is same as observation data. this method was copied all the values of the observation file into the model file. Please help me how i regrid my model file without ocean values after remap with the base of observation data.
Saving lives as usual Prof.Thompkins!!. Found out this video one day before a deadline lol
Hi Adrian, thank you for all these amazing videos. Your videos are always my "go to" whenever I encountered problems in my research.
Very nice and entertaining videos Adrian! Cheers, Csaba :)
Thanks for the nice feedback Csaba. Hope all is well with you!
Thanks a lot!! What an amazing video. Very helpful!!! Gonna share it with my colleagues. Awesome channel btw.
That's great 👍 thank you so much. I've got many more in the pipeline, just had to have a bit of a break due to many work commitments and other things going on. But more on the way soon!
Subscribing because you made it very easy! Thanks a lot.
Thanks so much. All these nice feedbacks mean a lot, I'm really happy i can help. More videos are planned, just trying to cope with a lot of things in life at the moment, but promise that another 6 or 7 will come out over the May to July period. Thanks for subscribing, that way you will hopefully get news when the new videos are ready.
thank you
You're welcome 😁
This video just made my day, thanks a lot
Glad you enjoyed it!
Very well explained Prof. Tompkins. I found a lot of other ways of creating land-sea mask on Stackoverflow but nothing of this quality and ease. Please please make many such videos. Model bias evaluation is something that I am really looking forward. Thank you very much.
Thanks for your kind comments. If you check out my latest video "combining gridded fields and broadcasting" it shows you exactly how you can calculate biases between models and observations... I hope it useful for you.
Thank you so much for a great video! How would you do if you are trying to mask out the land and retain the ocean data?
Just use the less than operator instead of greater than 😉
Please make a Video on Bias correction
Thanks
I'll add it to my to do list. I better do one on model drift too, just to avoid any bias 🤣 seriously though, I'll try to get to it asap, recently I've been a bit buried with teaching and paper writing... Not enough time in the day!
thanks professor for such a perfect video, May I ask which climate data have the highest resolution ? Cordex core as I know have 25 km ? what about other data on copernicus for CMIP6 climate projections ? Can I find data of 10 km or less resolution for Europe in particular for Italy ? thanks in advance
amazing
Thank you
Thanks a lot for such a great video. It is very useful for me to make a mask. But I have a question about the mask of Antarctica. I found it is incomplete, and there are some missing grids. Could you please kindly tell me how to figure it out? Another question is that I have two global NC files with different resolutions. If I mask each file first, exclude the ocean areas to improve efficiency, then calculate indices for these two files, remap the results into 1*1 degree, and average them. I'm afraid that there will be some problem with the junction of land and sea areas because of the different original grid sizes due to the resolution. So I would like to make a mask which is bigger than my land area and also exclude most ocean areas, can I ask is there any way to do this?
Regarding the second part, excellent question. I would perhaps map both fields to the coarse grid. Average. And then apply the mask on the coarse grid. I'll think about it some more because you may also want to map the fine scale mask to the coarse, setting any cell to missing if the fine scale cell is missing, let me mull over how to do that best. Don't forget that the built in topography is only 0.5 deg. So for fine resolutions it is quite approximate. If you want anything better then it is preferable to download an appropriate topography or land sea mask file. These longer more detailed questions are easier to answer on stack overflow by the way, if you post there using the tag netcdf and cdo-climate it is easier to write a nice formatted answer. Thanks for your interest in my channel, i hope you find it helpful.
thanks Prof. i just wanna make sure about the data. If i didnt have the golbal data, example that only tropics data, and then to create mask data in my data, can be running? or we have to download the global data?
Yes that's also fine. You don't need to have global gridded data to be able to use cdo
Unfortunately my country, the Netherlands, is mostly below sea level. It would simple dissappear from any land map this way. Any alternative to deal with polders?
Yes you are right, this method is inaccurate for pixels below sea levels (some areas near your coast, and also near the African rift for example) and has the opposite problem in that it classes elevated lakes as land. This is meant to be a quick soln that you can do on the fly without downloading extra data and works pretty well all said and done. There is a land-sea mask available in ERA5 on the native ERA grid, (var-lsm) - You can download that and compare the two... (use gec,0.5 to convert it to a binary field) - Overall they are not far off each other, just the issues above, and some noise along the coasts due to the different native grids (0.5 and 0.25 deg)
I followed similar process for masking my ocean data for precipitation from 1979 to 2019 but I got the masked data only for the first time level...next time steps were not done when masking was done.What shd be done.....????
Very strange. I think to answer this I'll need to see details. Can you upload the question on stack overflow?
@@climateunboxed Similar situation for me
@@gokulvishwanathan3993 I'd like to help but to do it I need to see the data. Can you post a question on stack overflow with it and tag it with cdo-climate?
Hi,
I tried this method but I got the result like Obesrvation_data.nc file and Model.nc file after regrid the vale of the output is same as observation data. this method was copied all the values of the observation file into the model file.
Please help me how i regrid my model file without ocean values after remap with the base of observation data.
Hi. Not sure I understood the problem, perhaps post to stack overflow with output and tag cdo-climate and I'll try to help there