Find elevation for series of (lat,lon) combinations to make 3d model of earth with relief

Question

I would like to say I am a complete novice to the world of GIS, so if I ask stupid questions or make strange assumptions, please point it out and I will try to clarify myself.

I would like to make a 3D-model of the earth with relief. To do this, I was thinking to divide the earths surface in a matrix of about 10000x10000 (dividing the earth horizontally and vertically) points and then find the elevation for all these points. So I need to find an excel file somewhere with a lot of (lat,lon) points on this earth and their elevation. No idea where to start though. First of all I looked into the google elevation api, but there is a limit to request max 2,500 elevations per day, so this does not seem like a good option. Another alternative I have looked into is the gmted2010 data on earthexplorer.usgs.gov, but I only seem to be able to download the data as tiff files, while I just would like to download an excel file. Can someone advise me the best place to find (lat,lon) combinations of the entire earth's surface with their elevation? Furthermore I also want to know how to find coordinates of the points that make up the borders of the countries in our world?

---

I will first try to get an answer to the first part here and then later post my other question about the boundaries. It was more intended to guide people in the right direction when choosing between different solutions to offer me (i.e. which one is most compatible with the boundary task). Radouxju, I tried your solution, but am still confused on some parts (I repeat: I am a complete newbie to all this). So it would be great if you could clarify some things to me:) First of all I downloaded one of the NOAA tiles to try it out and opened it in Photoshop, the only software of the possibilities they list, that I posses. I did everything as they explained here (http://www.ngdc.noaa.gov/mgg/topo/report/s11/s11Gx.html) and now have a really good looking graphic on my screen, coloured according to the elevation. I looked everywhere for solutions to export this to ASCII, but without succes. The only thing that mentioned ASCII was saving as EPS and choosing ASCII as encoding. But this still saves it as a graphic. So no idea how to get from the file I downloaded to ASCII.

Answer 1

I originally wrote this as a comment, but I think it addresses your first question.

What do you want to ultimately achieve with your elevation model? I could explain why if you'd like, but I suspect you really don't need an Excel spreadsheet to achieve this, and in fact that is really not a good way to go about it (for one, 10,000 x 10,000 is already more rows than an .xlsx can store). The raster dataset of elevation (the .tiff or .geotiff) is already what you want: it represents a regular grid of cells, each containing a representation of the elevation of the cell, and the position of the cell on the earth's surface is known. What you probably need is a desktop GIS capable of interpreting such a dataset and displaying it to you visually. (QGIS is a great, free, open-source option.)

To obtain "coordinates of the points that make up the borders of the countries in our world", what you probably want is a vector dataset of current landmasses and/or political boundaries. There are several websites that offer such a dataset, just google "countries shapefile" or something similar. A desktop GIS will be able to read a variety of data formats used to represent such information. A country is likely to be represented as a polygon (i.e. a shape described by a series of coordinate pairs). As each coordinate describing the shape by definition has a location, a desktop GIS can be used to look at each coordinate ("vertex"), and then look at the underling cell of the elevation dataset (.tiff) you've already acquired. Then you can associate each vertex with an elevation (this is commonly referred to as a z-coordinate, as in the tuple (x,y,z) used to describe a location in three dimensions).

Answer 2

I suggest that you start from an existing raster DEM (such as NOAA 1 km DEM). Then you can resample it at a coarser spatial resolution (if it is too heavy, on my example I took 0.2 degrees) and convert it to ASCII (easy to import in an excel sheet), but maybe you can already use "as is". (For resampling, you can use gdal_translate).

gdal_translate -of XYZ -tr 0.2 0.2 inputdem outputdem.txt

note that the Earth is not a square in lat/long, so I do not recommend using 1000*1000 matrix