Historic Wetland Mapping Learning Project

Project Summary

This Wikiversity learning project is part of a study undertaken by Regional Plan Association and funded by the Doris Duke Foundation, with additional support from the Leon Lowenstein Foundation.

In partnership with OpenHistoricalMap, TeachOSM, and others, this learning project will help map out the historic wetlands of the New York metropolitan area. This learning project will support a larger study and toolkit on climate adaptation and habitat restoration.

This project expands upon previous research undertaken by Regional Plan Association to map the historic wetlands of the New York-New Jersey Harbor, seen in the image on the right.

If you have questions about the project, you can contact Sanjay Seth (sanjay@rpa.org), who is a Research Analyst at Regional Plan Association.

Learning Goals

In the following tutorial, participants will learn how to use Wikimedia Warper and TeachOSM Task Manager to undertake a historic mapping exercise.

In order to support a regional study on climate adaptation and habitat restoration, project participants will learn skills related to historic mapping, like georeferencing historic maps and tracing historic map features. (What is georeferencing? Click this link for more information: w:georeference) Additionally, participants will learn how to collaborate virtually to produce important research.

Getting Started with Wikimedia Warper

The United States National Oceanic and Atmospheric Administration (NOAA) has published thousands of historic Coastal Survey maps.

While many of the more recent historic maps have been georeferenced, most maps prior to 1900 have not been georeferenced. Pre-1900 maps are the target of this project, as they more clearly depict habitat conditions prior to modern coastal impacts.

About 175 of these maps have been uploaded to the Wikimedia Warper. These maps need to be georeferenced.

Once these maps are georeferenced, project participants will be able to use the TeachOSM Task Manager and OpenHistoricalMap to trace the historic wetlands of the New York metropolitan area. This information will inform strategies to adapt to the impacts of climate change over the next several decades.

Creating an Account

Navigate to the Wikimedia Warper and click on 'Create Account' in the top-right corner of your screen. You should see the following screen. Create an account and login.

Locating the Historic Maps

Once you have created an account, navigate to the historic maps for this learning project: http://warper.wmflabs.org/layers/3

These maps are in the 'Mosaic' titled 'Category:RPA Historic Wetlands'. Here, you will be able to select a map to georeference. Click on the title 'Category:RPA Historic Wetlands' to open the Mosaic and display the maps it contains.

Selecting a Map to Georeference

Once you navigate to the historic maps, scroll down on the page to find a list of maps included in the 'mosaic'. Here, you can find a map that is 'unrectified' – which means that it is not georeferenced – and open that map.

Georeferencing a Map with Wikimedia Warper

Load the Map

Once you select an 'unrectified' map, the warper will load the map. This can take several minutes, depending on the size of the map.

Click on the 'Rectify' tab to begin georeferencing. You should see a screen that looks like this:

Determine the Map's Coordinates

If you are unfamiliar with latitude and longitude, read this before proceeding further with this learning project.

This step often takes some sleuthing and some knowledge of maps. Be patient and focus on details in the map to interpret the latitude and longitude of lines displayed on the map.

In most cases, you will only see the 'minute' value on a latitude or longitude line displayed on these maps. You are searching for the place where the cartographer has noted both the degree and minutes value for a given coordinate.

Once you find this, use this as your key coordinate to determine the values of all the other coordinates. The distance between each coordinate is always constant on each map.

Note: the coordinate nearest to your key coordinate is almost always a distance of one 'minute' of latitude and/or longitude away. If so, this holds true for the entire map and can help you quickly deploy control points.

(Tip: In rare occasions, you will see two coordinate systems, one darker and one lighter. The coordinate systems of older maps were corrected. You may need to use some trial and error to determine which coordinate system is correct, but it's generally straightforward to tell. Look for the darker lat/long lines and coordinates.)

Access the Control Points Panel

Once you are in the 'Rectify' tab and have a better sense of the coordinates of your map, hover your mouse over the 'Control Points' text in the 'Control Panel' at the bottom of the page. Click on the 'Control Points' text. This will allow you to add control points manually.

Once you click the 'Control Points' text, the Control Panel should expand to show you this:

Add Control Points Manually

These maps do not have features that allow you georeference the map based on current conditions. You will have to add control points manually.

To add points manually, click the 'Add Control Point Manually' button. On your historic map, you should see points in the top-left side of the map. On the modern map (right), you should see points in the center of the map, near the Gulf of Guinea (off the coast of Ghana).

Your page should look like this:

(Tip: If you increase the zoom of your browser, you can make your map larger.)

Position Control Points

Once you have some control points, you will need to position them on your map. Each point corresponds to a coordinate both on the historic map (Image X, Image Y) and on the modern map (Long, Lat). The relationship between these coordinates allows us to georeference (rectify) the image.

When placing a control point onto a coordinate of the historic map, it is important to zoom in as much as possible. You want to place the control point in the exact center of the coordinate.

Once you place the coordinate, note the latitude and longitude. (In all cases, the longitude for these maps will be negative. However, the maps will not display a negative sign.)

Convert to Decimal Degrees

You must convert this coordinate from degrees, minutes, seconds (DMS) to decimal degrees. (See this Wikipedia page for more information on w:decimal degrees.)

This conversion can be done in a spreadsheet. However, there are also websites that do the conversion for you. This United States Federal Communications Commission website offers a tool to convert from DMS to decimal degrees.

Input Decimal Degrees into Control Point

Once you have converted to decimal degrees, input that value into the control point. Repeat the process for the other coordinates on the map.

You should see something like this:

While only three control points are necessary to run the warper, it is desirable to have at least 10. In many maps, it is possible to do more than 20. The more control points, the more accurate the warper will be.

It is also possible to import control points from a CSV file. However, you will need the 'Image X' and 'Image Y' values, which will require manual placement of the control points upon the historic map.

Warp and Mask the Map

Once you have set the control points for the map, you must mask or crop the map. Switch to the 'Crop' tab and draw a polygon around your map area.

As these maps will be pulled together into a 'Mosaic' to form a single historic map, it is critical that any excess information be cropped away.

Using the tools in the top-right corner of the mapping area, draw a polygon around the map. Zoom in close to avoid cutting off important information. Once you have finished, click 'Mask Map!' This function will also warp the map.

Preview the Map

Once you have masked and warped your historic map, preview the map to ensure that its position and warp is relatively accurate.

As more maps are georeferenced, it is critical that you work to mask your map so that it does not overlap significantly with any other map.

If you see a strong rightward or leftward shift, you may need to check your control points. Although it is historical, it will still look like a good fit, if the previous steps were performed correctly.

There may be some rare cases where, even if all of the steps were performed correctly, your map will not render appropriately. In those cases, please contact Sanjay Seth (sanjay@rpa.org) with a link to your map.

Getting Started with TeachOSM Task Manager

Tracing Wetlands with TeachOSM Task Manager

Once all of the maps are georeferenced, the learning project will move on to tracing the wetlands within a Task Manager.

More information here soon.