Replies: 1 comment 3 replies
-
|
Hi, Furthermore, when zooming to the philipines on the online explorer, I do not find any major issue (smaller mismatch always expected owing to differences in grain). Here shown exemplary for level 1 with half-transparency:
|
Beta Was this translation helpful? Give feedback.
-
|
Hi Martin, thank you for your prompt response, I really appreciate it! The shapefile I'm using is the official adm level 0 shapefile (country boundary) from the Philippine gov GIS database: https://www.geoportal.gov.ph/ which is originally projected in WGS84. I followed your advice and uploaded the raw v004 level 1 data uncropped / not resampled (and also in WGS84). I still see the same misalignment issue. I've uploaded some additional examples, both using my shapefile in WGS84 but also for an additional check with Google Maps layer (WGS84 too). I've added the arrows to try to show the area of misalignment - I hope that's clear enough (it's basically the coastline). When I'm looking at your map, I also see the same problem (e.g. for the island in the left bottom corner - the pixel area (i.e. habitat data) is shifted slightly to the left. To give you more context, I'd really like to use your maps for mangrove modelling where it's crucial for the coastal area to be alligned correctly. At the moment, when I'm overlaying the data from the Global Mangrove Watch it is perfectly alligned with my country shapefile/Google maps etc. but a high proportion of the mangrove extent is outside the terrestrial habitat map layer (again, including when I'm using raw unprojected WGS84 data for mangroves). I appreciate that the data you developed is done at a global scale so it's completely understandable that there might be some mismatch but I just thought I'd flag this issue I'm experiencing. It would also be useful to have a confirmation that this misalignment is within the expected margin rather than something that could maybe be resolved somehow (sorry I haven't looked at your GEE code yet). Thanks so much again for your help.
|
Beta Was this translation helpful? Give feedback.
-
|
Heya, i think you hoping of too much from a 100m land cover map. If you are interested specifically in fine-scale coastal areas/mangroves, then I would recommend using directly the best available data there such as from Mangrove Watch (which is not integrated in the habitat map yet either). Coastal areas are generally very dynamic and as cross-realm habitats i am not surprised if they don't match perfectly. There are entire datasets created just for perfect coast alignment, which however does not help for the habitat mapping as it is based on integration of several different datasets can not create values for areas. |
Beta Was this translation helpful? Give feedback.
-
|
FYI: Here is how it looks for me. Administrative map from here. |
Beta Was this translation helpful? Give feedback.
-
Hi Martin,
I was wondering if you could please help me with one question - I am trying to use level 1 habitat maps and crop them to cover just the Philippines. When I do so and generate a map, I can see that there is some misalignment of the original raster and the land area. I think that I have checked everything including potential CRS issues but it seems that it is the original raster data that is slighlty misalligned. I've attached a photo to illustrate the problem - the orange area represents the country border, all the grey cells show level 1 habitat types - you should be able to see that the habitat type data is almost shifted slightly upwards.
Is this something to be expected? If not, could you please advise if there is any possible fix other than manual georeferencing to allign the data?
Thank you for your help in advance!
Beta Was this translation helpful? Give feedback.
All reactions