Any GIS gurus out there willing to help out a newbie?

I’m not too bad when it comes to computers and technology. I’m the guy my friends come to for advice, or to help them out when their computers go haywire.

But i’m buggered if i can get off the ground with (what i think is) some very basic GIS stuff, and i’m hoping i can get some pointers here.

My wife is a historian, who works on New York City in the 19th century, and would like to be able to take an old map of the city (as it was in the mid-19th century) and put information on it, like the location of particular houses, brothels, police stations, etc. This 1836 Colton Map would be perfect.

What i would like to be able to do is overlay this into a GIS program so that she could start inserting her markers.

I’ve downloaded and installed some free, open source GIS tools which, if my reading is correct, should be plenty powerful for the job. I have Quantum GIS, GRASS, uDig, and OpenEV on my computer. The computer itself is a Core 2 Quad with 6 GB of memory, so it’s got enough power for the job.

But, while i have the programs, i don’t really know how to start to get them to do what i need. I know i need to get the maps, but it’s not clear to me what type of files i need to start with - raster, vector, shapefiles? It’s all sort of confusing. There’s a lot of GIS stuff available online, but i don’t know what to download, in what order, to get what i need.

For those willing to help, there’s actually an online version of what i’d like to do, so you can see it for yourself.

Go here, scroll down a bit, and then click on the button “Launch GIS Professional Browser.” (Java required)

When it launches, activate the layer “NYC, 1836, Colton” and hit the Refresh button.

What you see there is exactly what i want to be able to load into my GIS software, so that my wife can add markers and stuff to it.

Is this a task that only a GIS expert can accomplish, or is this something that i should be able to do with a bit of help? I know so little about this, that i don’t even know what i don’t know. :slight_smile:

Any advice most welcome.

MrSid is a raster format however it’s not usually supported by freeware programs due to the cost of licencing, if the viewer can’t save it in a more common raster image I’d suggest using IrfanView and assuming the map is accurately drawn you’ll need to manually calibrate the image using your GIS application.

I use ESRI products myself and don’t know anything about Quantum GIS. I’ll take a look at it if I get a chance.

In any case, I just exported and downloaded the image (there is an option on the page you linked to the map). It comes in as a .jpg which should work fine if I understand what you are doing.

A shapefile will contain lines, points or polygons. Depending on what Quantum can do, you could create a seperate shapefile and insert your markers (points) in there. Using the map as simply another layer. In that way, you could attribute the points, label them, and use different symbology based on attributes.

Hope that helps.

Shapefiles are vector files. But that NY map is a raster one. Generally you fit it by marking several known co-ordinates on the layer and distorting to known spatial data, like an OGC or USGS layer.

ETA: In QGIS, use the georeferencer plugin for this.

Let me get this straight - you want to have that map as a background layer, and then fit co-ordinate geospatial data to it? Or do you just want to take the map and mark known locations on it with icons? Because if it’s the latter, you’re better off with layers in Gimp or Inkscape, frankly. GISes are for when you actually want to do something with the data like statistics or other computational stuff. Say, you want to find all the police stations that are within 500m of a brothel but <100m from the river. If all you’re after is scale maps, then GIS is more than you need.

Thanks for the responses.

The MrSID format of the downloadable map is not a problem; i’ve used MrSID images before, and can convert them to other formats like JPG without a problem, if that’s what’s necessary.

Your explanation does help, enipla. I’ll give that a go.

Although, after reading MrDibble’s post, it could be that i’m trying to make things too complicated here. The main purpose of this exercise, at least right now, is simlpy to “to take the map and mark known locations on it with icons.” There is no need for the sort of computational stuff that GIS is best known for.

I think i was just under the impression that, because GIS is so powerful, it would be relatively easy to use for something simple like this.

Anyway, i’ll head off and try a few things, and report back.

I am not a GIS expert but I could probably be called an expert of sorts on geographic databases.

It sounds like you want to georeference some old maps and then overlay newer data on top, then mark a copy of the old data with new? What do you want at the end of the process? An image of the older map with whatever marks you’ve made?

If I’ve guessed correctly then this is pretty standard fare for something like arcGIS. Import the image of the old map (as a .jpg or .gif maybe?). Open whatever other data you have and display it on top. The example you give is some raster data with some vector data on top. The trick is getting them lined up.

Your GIS should have tools that allow you to georeference the two datasets. That means picking some common points in each set so that the software can line them up. If you have two raster images (i.e. jpgs or gifs) you have a special problem with the display but I bet your GIS has tools for that too.

Then you should be able to annotate your original image and save it with the changes (keep a copy of the original!)

What the GIS is giving you is the ability to line up and overlay geo data that may not be using the same projection and then edit that data in a simple way. If you just want to take an image and put marks on it, you could use Paint or something similar. Except the file format of the 1836, Colton map is not so easily dealt with as bluezooky explained.

The thing is, i don’t have “data” to open up. At least not in the sense of a dataset with co-ordinates.

I’ve converted the NYC map image from MrSID to a JPG, and have opened it as a raster layer in my GIS program.

What i want to be able to do now is something like what enipla suggested earlier: “attribute the points, label them, and use different symbology based on attributes.”

So, for example, i’d like to be able to zoom in to the intersection of Mott and Spring Streets, where i know there was a brothel, and put a little marker there (say, a red triangle) that indicates a brothel. Then i want to move to Bowery, and place, for example, a little blue triangle on the location of a police station.

But i can’t work out how to do that. I can add a new vector layer to my project, so i created a vector layer (polygon) called “brothels.” I assume this is what i need to do? But i can’t work out how or where to start adding specific points and symbols to the map.

Sorry to be such an amateur about all this; usually i’m pretty good at figuring this sort of stuff out, but this has me stumped.

I would say that you need to add a point layer if you can do that. I’ll download Quantum GIS an poke around.

You are correct, and it seems things are going well now.

Just after i posted my last post, i played around some more and finally worked out how to add a point layer and start allocating points.

So, i now have a Brothels point layer, with little red triangles representing the brothels, and a Police Station point layer, with little blue triangles as the symbols.

At the moment i’m trying to work out how to add specific information labels to each point (i.e., to each individual brothel or police station), but that’s not a huge concern. Just being able to make the point layers is an excellent start.

Thanks for all the advice. I’m keep playing around and hopefully increase my understanding of how this all works.

Good going. You should be able to label your features from info in the attribute table from the points layer. You may have to add the fields you want to that table.

Yep, just been working on that.

It seems, however, that you have to be careful to add all of the fields to the attribute table at the moment when you create the new layer. I can’t find any way to add fields to the attribute table after the layer has been created. This isn’t a huge deal, but it would be nice to be able to modify the attribute table.

I have trouble believing that there’s no way to do this, so i’ll keep digging around.

I looked too. Don’t see a way to add a field either.

However, the table behind a shapefile is just a .dbf (dbase file). Find the shape file with explorer, and you will see a .dbf with the same name as your point layer.

You could bring this into Access, add a field and then export it back out. I highly recommend doing a backup/copy first, but it should work no prob.

Also, if you right click the point layer in the table of contents (left pane) choose properties you will get a labels tab.

eta. Oh, looks like you got labels figured out.

Thanks. If it becomes essential to be able to change the attribute table, i’ll give that a go.

Thanks for all your help. I think we’ve got it running well enough that it will serve her purposes for now, and we can learn new stuff as we go along. I think i’m going to invest in a GIS book, or borrow one from the library, and do a bit of reading.

Most of us who do maps for publications move them out of GIS (and into Illustrator) as soon as possible because GIS is a very nonintuitive way to make what is, in the end, a drawing. It’s hard to directly position things in GIS.

A much better solution for you would be to simply put the marker icons on another layer in Photoshop, Illustrator, InDesign, or whatever you’re comfortable using. You don’t have hundreds, or even dozens, of markers to place, and you have them, in many cases, as intersections rather than modern street addresses. GIS is just the wrong tool for the job. It’s using a programmable robot to butter a slice of toast.