![\"\"](\"https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F2332154%2F0f221e50da59e0e2b87671660b841ca5%2FCapture%20decran%202020-02-14%20a%2020.14.18.png?generation=1581707698877212&alt=media\")
Hi,\n Any one has the idea why Area of extent of GLDAS image is smaller than other area of extent of data source available in folder s5p_no2 and gfs? Do you think data been missing here?
\n\nThe top layer in attached image is GLDAS and bottom layer is gfs.
\n\nIf data is missing, any idea how to fetch the data for the missing tiles?
\n\nAny thoughts, please share.
", "messages": [ { "id": "745787", "postDate": "02/14/2020 07:44:53", "content": "Hi,\n Any one has the idea why Area of extent of GLDAS image is smaller than other area of extent of data source available in folder s5p_no2 and gfs? Do you think data been missing here?
\n\nThe top layer in attached image is GLDAS and bottom layer is gfs.
\n\nIf data is missing, any idea how to fetch the data for the missing tiles?
\n\nAny thoughts, please share.
", "rawMarkdown": "Hi,\n Any one has the idea why Area of extent of GLDAS image is smaller than other area of extent of data source available in folder s5p_no2 and gfs? Do you think data been missing here?\n\nThe top layer in attached image is GLDAS and bottom layer is gfs.\n\nIf data is missing, any idea how to fetch the data for the missing tiles?\n\nAny thoughts, please share.", "votes": null }, { "id": "745795", "postDate": "02/14/2020 07:59:57", "content": "I also see certain geo location provided in gppd file falling outside the provided GLDAS image. See attached image
", "rawMarkdown": "I also see certain geo location provided in gppd file falling outside the provided GLDAS image. See attached image", "votes": null }, { "id": "746234", "postDate": "02/14/2020 19:16:31", "content": "Hi,
\n\nGLDAS data are simulated data and cover only emerged lands. Due to the resolution, islands are sometimes not covered by GLDAS.
\n\nSee, a higher view of GLDAS over Puerto Rico to vizualize the resolution:\n
Thank you, this screenshot is very helpful for illustrating that the GLDAS data only covers land and does not extend into the ocean.
", "rawMarkdown": "Thank you, this screenshot is very helpful for illustrating that the GLDAS data only covers land and does not extend into the ocean.", "votes": null }, { "id": "746448", "postDate": "02/15/2020 02:29:12", "content": "Thanks !! It answers my question.
", "rawMarkdown": "Thanks !! It answers my question.", "votes": null }, { "id": "750412", "postDate": "02/19/2020 10:56:26", "content": "I wanted to touch upon this on my topic but I still can’t make it fit in somehow.
\n\nAnyways, there are multiple ways to tackle this. One was is to change the the resolution of one of the data sources and warp it to align with the image Sentinel 5 imagery. Obviously you have to be aware of the transformation you’re doing, and that you are basically changing the location of the data you have originally. However this can significantly simplify your workflow and is an interesting step when starting out.
\n\nThe way I would do this (and like a lot of things in remote sensing, it’s not the only way to do it) would be via GDAL (with gdalwarp, see this link on how to do so, either in a script or in the command line as the link shows.
\n\nI don’t know if this helps for what you wanted to do, but I think it’s an interesting approach to a common problem in remote sensing.
", "rawMarkdown": "I wanted to touch upon this on my [topic](https://www.kaggle.com/c/ds4g-environmental-insights-explorer/discussion/130221) but I still can’t make it fit in somehow.\n\nAnyways, there are multiple ways to tackle this. One was is to change the the resolution of one of the data sources and warp it to align with the image Sentinel 5 imagery. Obviously you have to be aware of the transformation you’re doing, and that you are basically changing the location of the data you have originally. However this can significantly simplify your workflow and is an interesting step when starting out.\n\nThe way I would do this (and like a lot of things in remote sensing, it’s not the only way to do it) would be via GDAL (with gdalwarp, see [this link on how to do so](https://stackoverflow.com/questions/10454316/how-to-project-and-resample-a-grid-to-match-another-grid-with-gdal-python), either in a script or in the command line as the link shows.\n\nI don’t know if this helps for what you wanted to do, but I think it’s an interesting approach to a common problem in remote sensing.", "votes": null }, { "id": "751039", "postDate": "02/20/2020 00:25:29", "content": "You cannot do GDAL Wrap here, which is used to reshape the raster. Here, all image sizes are same 148 * 475, still GLDAS not properly overlapping with other rasters. One thing I observed is GLDAS image is having NAN values where ever the tiles are missing. Hence we need to use with some nearest neighbor techniques to come out of this issue.
", "rawMarkdown": "You cannot do GDAL Wrap here, which is used to reshape the raster. Here, all image sizes are same 148 * 475, still GLDAS not properly overlapping with other rasters. One thing I observed is GLDAS image is having NAN values where ever the tiles are missing. Hence we need to use with some nearest neighbor techniques to come out of this issue.", "votes": null }, { "id": "751555", "postDate": "02/20/2020 09:19:01", "content": "Why couldn't we use gdalwarp?
\n\nI haven't spend enough time looking at the data yet, that's one thing I want to spend more time on, so you might know better than I do! I was just pointing out the original ideas I've had. I'm really curious to see how people are going to tackle this competition.
", "rawMarkdown": "Why couldn't we use gdalwarp?\n\nI haven't spend enough time looking at the data yet, that's one thing I want to spend more time on, so you might know better than I do! I was just pointing out the original ideas I've had. I'm really curious to see how people are going to tackle this competition.", "votes": null }, { "id": "752447", "postDate": "02/21/2020 03:04:46", "content": "Since size of all raster provided was same (148 * 475). If the size of raster is uneven then as you said we should use GDAL Wrap. Hope this information helps
", "rawMarkdown": "Since size of all raster provided was same (148 * 475). If the size of raster is uneven then as you said we should use GDAL Wrap. Hope this information helps", "votes": null } ], "comments": [ { "id": 745795, "author_name": "nagabilwanth", "author_url": "", "post_date": "02/14/2020 07:59:57", "content": "I also see certain geo location provided in gppd file falling outside the provided GLDAS image. See attached image
", "votes": null, "replies": [] }, { "id": 746234, "author_name": "gpoulain", "author_url": "", "post_date": "02/14/2020 19:16:31", "content": "Hi,
\n\nGLDAS data are simulated data and cover only emerged lands. Due to the resolution, islands are sometimes not covered by GLDAS.
\n\nSee, a higher view of GLDAS over Puerto Rico to vizualize the resolution:\n
Thank you, this screenshot is very helpful for illustrating that the GLDAS data only covers land and does not extend into the ocean.
", "votes": null, "replies": [] }, { "id": 746448, "author_name": "nagabilwanth", "author_url": "", "post_date": "02/15/2020 02:29:12", "content": "Thanks !! It answers my question.
", "votes": null, "replies": [] } ] }, { "id": 750412, "author_name": "maxlenormand", "author_url": "", "post_date": "02/19/2020 10:56:26", "content": "I wanted to touch upon this on my topic but I still can’t make it fit in somehow.
\n\nAnyways, there are multiple ways to tackle this. One was is to change the the resolution of one of the data sources and warp it to align with the image Sentinel 5 imagery. Obviously you have to be aware of the transformation you’re doing, and that you are basically changing the location of the data you have originally. However this can significantly simplify your workflow and is an interesting step when starting out.
\n\nThe way I would do this (and like a lot of things in remote sensing, it’s not the only way to do it) would be via GDAL (with gdalwarp, see this link on how to do so, either in a script or in the command line as the link shows.
\n\nI don’t know if this helps for what you wanted to do, but I think it’s an interesting approach to a common problem in remote sensing.
", "votes": null, "replies": [ { "id": 751039, "author_name": "nagabilwanth", "author_url": "", "post_date": "02/20/2020 00:25:29", "content": "You cannot do GDAL Wrap here, which is used to reshape the raster. Here, all image sizes are same 148 * 475, still GLDAS not properly overlapping with other rasters. One thing I observed is GLDAS image is having NAN values where ever the tiles are missing. Hence we need to use with some nearest neighbor techniques to come out of this issue.
", "votes": null, "replies": [] }, { "id": 751555, "author_name": "maxlenormand", "author_url": "", "post_date": "02/20/2020 09:19:01", "content": "Why couldn't we use gdalwarp?
\n\nI haven't spend enough time looking at the data yet, that's one thing I want to spend more time on, so you might know better than I do! I was just pointing out the original ideas I've had. I'm really curious to see how people are going to tackle this competition.
", "votes": null, "replies": [] }, { "id": 752447, "author_name": "nagabilwanth", "author_url": "", "post_date": "02/21/2020 03:04:46", "content": "Since size of all raster provided was same (148 * 475). If the size of raster is uneven then as you said we should use GDAL Wrap. Hope this information helps
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "745787": "Hi,\n Any one has the idea why Area of extent of GLDAS image is smaller than other area of extent of data source available in folder s5p_no2 and gfs? Do you think data been missing here?\n\nThe top layer in attached image is GLDAS and bottom layer is gfs.\n\nIf data is missing, any idea how to fetch the data for the missing tiles?\n\nAny thoughts, please share.", "745795": "I also see certain geo location provided in gppd file falling outside the provided GLDAS image. See attached image", "746234": "Hi,\n\nGLDAS data are simulated data and cover only emerged lands. Due to the resolution, islands are sometimes not covered by GLDAS.\n\nSee, a higher view of GLDAS over Puerto Rico to vizualize the resolution:\n", "746301": "Thank you, this screenshot is very helpful for illustrating that the GLDAS data only covers land and does not extend into the ocean.", "746448": "Thanks !! It answers my question.", "750412": "I wanted to touch upon this on my [topic](https://www.kaggle.com/c/ds4g-environmental-insights-explorer/discussion/130221) but I still can’t make it fit in somehow.\n\nAnyways, there are multiple ways to tackle this. One was is to change the the resolution of one of the data sources and warp it to align with the image Sentinel 5 imagery. Obviously you have to be aware of the transformation you’re doing, and that you are basically changing the location of the data you have originally. However this can significantly simplify your workflow and is an interesting step when starting out.\n\nThe way I would do this (and like a lot of things in remote sensing, it’s not the only way to do it) would be via GDAL (with gdalwarp, see [this link on how to do so](https://stackoverflow.com/questions/10454316/how-to-project-and-resample-a-grid-to-match-another-grid-with-gdal-python), either in a script or in the command line as the link shows.\n\nI don’t know if this helps for what you wanted to do, but I think it’s an interesting approach to a common problem in remote sensing.", "751039": "You cannot do GDAL Wrap here, which is used to reshape the raster. Here, all image sizes are same 148 * 475, still GLDAS not properly overlapping with other rasters. One thing I observed is GLDAS image is having NAN values where ever the tiles are missing. Hence we need to use with some nearest neighbor techniques to come out of this issue.", "751555": "Why couldn't we use gdalwarp?\n\nI haven't spend enough time looking at the data yet, that's one thing I want to spend more time on, so you might know better than I do! I was just pointing out the original ideas I've had. I'm really curious to see how people are going to tackle this competition.", "752447": "Since size of all raster provided was same (148 * 475). If the size of raster is uneven then as you said we should use GDAL Wrap. Hope this information helps" }, "source": "meta" }