Some questions about grid discontinuities
22 views
Skip to first unread message
Wang
Nov 10, 2023, 3:01:27 AM11/10/23
to Biodiverse Users
Hello,Shawn
My grid locations were identified based on latitude and longitude (in decimal degrees), which identified the geometric center of the grid, but I eventually found gaps and discontinuities between the grids. So I want to ask you how to solve this problem
Shawn Laffan
Nov 10, 2023, 4:18:06 PM11/10/23
to biodiver...@googlegroups.com
Hello Wang,
It is difficult to be sure how to diagnose this issue without seeing the input data.
Are the input data points on a regularly spaced grid? This can be checked by displaying the coordinates in a GIS.
If they are on a regular grid then you can use the grid spacing as the cell size. You also need to ensure the cell origins are set to match the point grid, i.e. to make the centre of each Biodiverse group (cell) match the
input points.
As an example, consider a data set in decimal degrees where the x-coordinates all end in .176 and the y-coordinates all end in .778. To have cell sizes of [1,1] (one degree squares) that align with the input point data, the cell origins can be set to [0.676, 0.278]. These values are the point origins offset by half the cell size.
This is the approach we used in Zhang et al. (2022) https://doi.org/10.1111/jipb.13189
If the data are on a regular grid, but the cell size still results in gaps after accounting for the origins, then you could use a larger cell size. Be sure to use an integer multiple of the point spacing (i.e. 2x, 3x etc) so each larger cell size has an equal number of possible points. The cell origins still need to be specified but the values can be the same as before.
Another option is to import the data into Biodiverse using cell sizes set to zero. The display will then use a cell size calculated from the distances between all points so it will appear to be a regular grid. However, the underlying data will still be points so one cannot export results to the raster formats. One also cannot use these data to aggregate environmental layers to match the cell sizes (see https://biodiverse-analysis-software.blogspot.com/2022/05/importing-group-properties-directly.html ).
Regards,
Shawn.
It is difficult to be sure how to diagnose this issue without seeing the input data.
Are the input data points on a regularly spaced grid? This can be checked by displaying the coordinates in a GIS.
If they are on a regular grid then you can use the grid spacing as the cell size. You also need to ensure the cell origins are set to match the point grid, i.e. to make the centre of each Biodiverse group (cell) match the
input points.
As an example, consider a data set in decimal degrees where the x-coordinates all end in .176 and the y-coordinates all end in .778. To have cell sizes of [1,1] (one degree squares) that align with the input point data, the cell origins can be set to [0.676, 0.278]. These values are the point origins offset by half the cell size.
This is the approach we used in Zhang et al. (2022) https://doi.org/10.1111/jipb.13189
If the data are on a regular grid, but the cell size still results in gaps after accounting for the origins, then you could use a larger cell size. Be sure to use an integer multiple of the point spacing (i.e. 2x, 3x etc) so each larger cell size has an equal number of possible points. The cell origins still need to be specified but the values can be the same as before.
Another option is to import the data into Biodiverse using cell sizes set to zero. The display will then use a cell size calculated from the distances between all points so it will appear to be a regular grid. However, the underlying data will still be points so one cannot export results to the raster formats. One also cannot use these data to aggregate environmental layers to match the cell sizes (see https://biodiverse-analysis-software.blogspot.com/2022/05/importing-group-properties-directly.html ).
Regards,
Shawn.
On 10/11/2023 19:01, Wang wrote:
Hello,Shawn
My grid locations were identified based on latitude and longitude (in decimal degrees), which identified the geometric center of the grid, but I eventually found gaps and discontinuities between the grids. So I want to ask you how to solve this problem
--
You received this message because you are subscribed to the Google Groups "Biodiverse Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to biodiverse-use...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/biodiverse-users/c6c7394a-1c34-4c6a-a406-317da2eb8c47n%40googlegroups.com.
Wang
Nov 11, 2023, 9:12:19 PM11/11/23
to Biodiverse Users
Hello,Shawn
I am still very confused about this question. Please help me look at my data and I will send it to your email.Reply all
Reply to author
Forward
0 new messages