Recover historical shoreline extract from Google Earth Tool and .mdb using Ubuntu OS, NOAA T-00267

Question

For a birthday present for my father-in-law, I want to get historical shoreline data viewable in Google Earth into a format I can read into R and use with the sf package, specifically table T-00267, that I will incorporate into a map of his neighborhood.

The KMZ file can be downloaded here, https://shoreline.noaa.gov/data/datasheets/t-sheets.html

I download shoreline extract for table T-00267, and get a Microsoft database file called ec10a01.mdb

I'm using Ubuntu, so I downloaded mdbtools. I run in the location the file is saved...

mdb-tables ec10a01.mdb

and see there are files t00267line_SHAPE_Index t00267line in the database. I then try to save the file like

mdb-export  ec10a01.mdb t00267line > t00267line.csv

And I get a corrupted file that seems like it has some shape data in it that I'm not accessing right. If there are other ways to get this data besides through Google Earth, I'm open to alternative solutions.

---

The historical data for MD was also used in the creation of these shoreline erosion datasets found here http://www.mgs.md.gov/coastal_geology/shoreline%20change.html

---

Answer 1

As mentioned in comments, the .mdb file is an ESRI Personal GeoDatabase file. There is a driver for GDAL/OGR so it can be read by open source tools, but there is some configuration required. You need to install ODBC tools and drivers.

The process is detailed here:

https://gdal.org/drivers/vector/pgeo.html

With that complete I can for example query the .mdb for layers:

$ ogrinfo PGeo:ec10a01
INFO: Open of `PGeo:ec10a01'
      using driver `PGeo' successful.
1: t00250bis1line (Multi Line String)
2: t00250bis2line (Multi Line String)
3: t00255line (Multi Line String)
4: t00266line (Multi Line String)
5: t00265line (Multi Line String)
[etc - it contains 123 layers]

and I could use ogr2ogr to convert these to something less troublesome, like a GeoPackage (which could be read simply into R using the sf package).

I can read the PGDB into R with all this config and setup, eg:

 library(sf)
 s = st_read("PGeo:ec10a01","ec10a01polys_83")
 plot(s$Shape)

(note if this is the only .mdb file you want then I don't mind doing the conversion for you and putting the output somewhere accessible, but if you have a lot to do I think its better you configure your system so you can do it...)