../_images/mars-peaks-and-valleys2-banner.jpg


The Peaks and Valleys of Mars, Part 2: Searchable Nomenclature

In our first Mars tutorial, you used VTS 3D Geospatial Software Stack to stream an interactive, browseable map of Mars from your own system using public domain data sources. That was fun.

But wait a minute - is your virtual Mars model a real map? A map should have labels, so that aliens (YOU in this case) can use it to get around. After all, Mars is one of the best researched and documented bodies in the solar system so its nomenclature is readily available. And by the long standing standard of digital cartography, a map nomenclature should be searchable.

In following this tutorial, you are going to turn your 3D model of Mars into a real map, complete with labels and search functionality.

Preparation

As a prerequisite, you should complete our first Mars tutorial.

For search functionality, you need to have Node.js 8.x and PM2 process manager installed on your system. Run

$ node --version

to check your Node.js version. If the version you see is bellow 8.x or if you lack a Node.JS environment completely, run the following commands:

$ curl -sL https://deb.nodesource.com/setup_8.x -o nodesource_setup.sh
$ sudo bash nodesource_setup.sh
$ sudo apt-get install nodejs
$ sudo npm install -g pm2

Make sure you use the vts system user for data manipulation commands in the rest of this tutorial:

$ sudo -iu vts

The Labels

In creating your first Mars website you heavily relied on VTS mapproxy and on its ability to stream 2D map tiles and 3D VTS surfaces based on DEMs. Now you will make use of another mapproxy feature: the ability to convert geographic features to VTS geodata free layers.

Your Mars labels will be based on the offcial IAU nomenclature, available as an ESRI shapefile. Download and unzip it to your mapproxy data directory:

$ cd && wget http://planetarynames.wr.usgs.gov/shapefiles/MARS_nomenclature.zip
$ cd mapproxy/datasets/mars-case-study
$ unzip ~/MARS_nomenclature.zip

Create a resource configuration file at /etc/vts/mapproxy/mars-case-study.d/iau-mars-nomenclature.json with the following contents:

[{
    "group": "mars-case-study",
    "id": "iau-mars-nomenclature",
    "comment": "IAU MARS Nomenclature (MDIM 2.1)",
    "driver": "geodata-vector",
    "type": "geodata",
    "definition": {
        "dataset": "mars-case-study/MARS_nomenclature.shp",
        "demDataset": "mars-case-study/mars-mola-dem",
        "displaySize": 2048,
        "mode": "auto",
        "styleUrl": "file:mars-case-study/iau-mars-nomenclature.style",
        "introspection": {
            "surface": {
                "group": "mars-case-study",
                "id": "mars-mola-dem"
            },
            "browserOptions": {}
        }
    },
    "referenceFrames": {
        "mars-qsc": {
            "lodRange": [1,1],
            "tileRange": [[0,0], [1,1]]
        }
    },
    "registry": {
        "credits": {
            "usgs-asc": {
                "id": 8,
                "notice" : "USGS Astrogeology Science Center"
            },
            "iau": {
                "id": 202,
                "notice": "International Astronomical Union (IAU)"
            }
        }
    },
    "credits": ["iau","usgs-asc"]
}]

In VTS terminology, you’ve created a monolithic geodata free layer definition. Among other things, it defines the path to the feature dataset (definition.dataset) and elevation DEM (definition.demDataset).

Geodata free layers are stylable in a manner remotely resembling CSS. The style file is referenced in the above resource definition (definition.styleUrl). The file does not exist yet. Fix this by putting the following into ~/mapproxy/datasets/mars-case-study/iau-mars-nomenclature.style:

{
    "layers": {
        "point-labels": {
        "filter": ["<=","$diameter",2],
        "label": true,
        "label-size": 20,
        "zbuffer-offset": [-1,0,0],
        "culling": 90,
        "visibility-abs": [0,120000]
        },
        "labels-size0": {
            "filter": [">","$diameter",2],
            "label": true,
            "label-size": 20,
            "zbuffer-offset": [-1,0,0],
            "culling": 90,
            "visibility-rel": [{"str2num":"$diameter"}, 1000, 0.08, 0.8]
        }
    }
}

VTS Mapproxy will process this configuration automatically within five minutes. If you do not feel like waiting run

sudo /etc/init.d/vts-backend-mapproxy force-update

All right, time to see this in action. Point your browser to

http://<your server>:8070/mapproxy/mars-qsc/geodata/mars-case-study/iau-mars-nomenclature/

Here is what you should see:

../_images/mars-peaks-and-valleys-labels.jpg

If you’re into the VTS fineprint, here is a bit of styling tricks you’ve just used: the diameter property from the IAU feature dataset (which contains the feature size in kilometers) in conjunction with the “visibility-abs” and “visibility-rel” layer properties makes sure that labels are visible either 1.) when they are bigger than two kilometers and occupy more then 8 percent and less than 80 percent of the current vertical view extent, or 2.) they are smaller than two kilometers and the vertical view extent is less than 120 kilometers. These simple rules ensure that the labels exhibit just the right level of visual density for the map to be informative while avoiding visual clutter.

Another point worth noticing: the streamable map you’ve just created is a fine example of the VTS mapproxy introspection capabilities. It combines four different resources: 1.) the nomenclature, which you’ve just defined, 2.) its style, 3.) the terrain (which the labels refer to in their introspection) and 4.) the orthomosaic (which the terrain refers to in its own introspection). Yes, introspection is quite powerful.

The Search Interface

Search interfaces for maps go by the fancy name of geocoding these days, though if you want to sound even fancier, you might speak of areocoding in this particular case.

Your search interface will make use of the same nomenclature file you’ve downloaded in the first part of this tutorial. To turn this shapefile into a search server VTS clients can talk to, you will use togs, a simple Node.JS script.

Install togs as follows:

$ npm install togs

Togs configuration file is located at $(npm root)/togs/conf/togs.conf. Open it and the following snippet to it:

[interface.mars]
dataset = /var/vts/mapproxy/datasets/mars-case-study/MARS_nomenclature.shp
idProperty = link
searchExpand = true
searchBool = AND
searchExclude = ^center.*|diameter$
iauBoundingBox = true
class = place
type = region
displayName = {name}, {quad_name}
addressRegion = {name}
addressState = {quad_name}

Start togs via pm2 process manager:

$ pm2 start $(npm root)/.bin/togs

and do

$ curl 'http://localhost:8100/mars?q=Chasma&format=json&limit=1'

to test that togs is up and running. The output will be a single-element JSON array, familiar in format to everyone used to working with OSM Nominatim.

You might want to make sure that togs survives the reboot of your server. To do that, perform

$ pm2 startup systemd

and run the last line of the output as superuser.

To make your Mars website aware of your brand new search API, go back to the mapproxy resource you created in the earlier part of this tutorial. Open /etc/vts/mapproxy/mars-case-study.d/iau-mars-nomenclature.json, find the browserOptions object, and modify it as follows:

"browserOptions": {
    "controlSearchSrs": "+proj=longlat +a=3396190 +b=3376200 +no_defs",
    "controlSearchUrl": "http://<your-server>:8100/mars?q={value}&format=json&limit=20",
    "controlSearchFilter": false
}

Do not forget to replace <your-server> above with your server’s hostname or IP address.

To make VTS mapproxy instantly aware of your changes, do

sudo /etc/init.d/vts-backend-mapproxy force-update

Go back to your website at

http://<your server>:8070/mapproxy/mars-qsc/geodata/mars-case-study/iau-mars-nomenclature/

If all went well, your map now includes a search field:

../_images/mars-peaks-and-valleys-search-window.jpg

You can search for the characteristic Martian topographic feature types (chaos, chasma, mons, or crater). Or you can search for the nationalities (Czech, German, Dutch) to find out what nations discovered and named the individual features. There is lots of fun ways to fool around.

That’s it! Your own Mars website sports almost 2000 official place labels, indexed and searchable.

VTS has some more yet uncovered features to offer which can make your Mars website even more interesting. We shall explore these in some of our next tutorials.