Making the Benefits Visible

The Lancaster Conservancy is a non-profit whose mission is to find and preserve natural lands, primarily in Lancaster and York counties, to create habitat for animals, protect plants and trees, and provide public access to nature. This land preservation is accomplished through a combination of land purchase and easements, which are legal agreements restricting land use into the future.

This project is a collaboration with the York and Lancaster County Planning Commissions, the Lancaster County GIS Department, and the Lancaster Conservancy. The project will illustrate the use of geospatial technology to assist in selecting land for preservation to accomplish the multiple goals of these agencies. As part of this process, we will use 'viewshed analysis', described further below, to allow us to consider public visibility as a factor when selecting parcels of land for preservation.

While the stated goal of this project is preservation of scenic areas, it is important to step back and consider the context and stakeholders involved, and in particular how we define 'scenic'.

Since we are working with public agencies, we are ultimately serving residents of York and Lancaster counties, including those living within the study area as well as those who may travel there. For many of us, particularly city dwellers, it is restorative to be somewhere where we can see a lot of trees and water. Others may be seeking more stimulation; looking to be energized by a view of nature's power in the form of waterfalls or rapids, or a view from a hill top.

If our goal is a feeling of well being for our fellow county residents, let's realize this is the result of a process over time, not just the mere existence of a scenic area. It is about how we get there, the variety of terrain and other factors. Some people may want to be alone or in small groups, others may want to share with many people. Many people enjoy a natural scene with few obvious signs of human activity. Others like to feel less isolated, perhaps with townscapes as part of their view.

Beyond how a person feels about what they see, we should also have some awareness of whether they feel welcome in a given place. This is, of course, a huge influence on whether they feel 'restored' after spending time somewhere. Also consider if a given scenic area or trail offers options for both larger and smaller groups of people to choose their preferred path for a secluded experience or a more social one.

While we think about things for people to see, it makes sense to consider opportunities to raise awareness and support for environmental initiatives. Areas where trees are planted for riparian buffers to protect water quality can double as scenic areas, which in turn can get visitors thinking about how to support such efforts.

Given all of the above, a project like this one can only touch on some of the issues. We will now narrow our focus as we describe our methodology.

As mentioned above, there is value in making good environmental practice visible to the public. This allows people to enjoy nature in the present, but also can inspire them to become involved in preserving the environment into the future. We'll now look at a method for selecting parcels for preservation which can help this to happen. Let's consider two environmental goals: improving water quality, and sustaining biodiversity. A common strategy targeting both of these goals is the creation of riparian buffers.

•  Riparian buffer : A riparian buffer or stream buffer is a vegetated area (a "buffer strip") near a stream, usually forested, which helps shade and partially protect the stream from the impact of adjacent land uses.

This project will consider visibility from three sets of observation points; trails, the Susquehanna river centerline, and singular scenic areas. In short, our methodology will seek to determine land parcels conducive to the preservation and extension of riparian buffers which are within a viewshed of one of the above observation points. In addition, by request, we will be broadly categorizing other potential scenic areas. For now, we will consider three categories: 'natural' (in this case, forest), 'pastoral' (pasture land cover) and 'water'.

Though not the only choice, ArcGIS Pro is a software tool well suited to the analysis needed for this project, and we will be using it.

Required data for geospatial analysis

To define the study area:

•  municipality boundary vector  (shapefile) data

And parcel boundaries:

•  York County parcel data 
•  Lancaster County parcel data 

To calculate viewsheds:

• our  observation point features  (including on trails & river center line)
•  elevation raster(s)  covering the study area (for 10m resolution DEMs used here, elevation data residing on ESRI managed ArcGIS server was used)

To locate existing forest and otherwise categorize potential scenic areas:

• land use raster data for  Lancaster  and  York  counties
•  text description  for land use classes

To locate smaller streams feeding the Susquehanna (often not visible in land use layer)

•  flowline for Susquehanna river  from the National Hydrography Dataset (this data is from 2004 and land development may have altered some watercourses since then)

To locate previously preserved parcels:

• preserved parcel data provided by York and Lancaster counties
•  additional nature preserves  owned or managed by Lancaster Conservancy

Workflow

Now we will describe the workflow used to transform the input data listed above into a map showing parcels suggested for preservation, color coded according to the length of watercourses within each parcel which flow into the Susquehanna river.

Though results will be shown further below, it may help understanding of the workflow to preview the map we will produce. You may do that  here .

A powerful feature of ArcGIS is a tool called 'Model Builder' which uses a graphical interface to create a geospatial analysis workflow. The diagram of our model below describes most of the workflow we are using.

The model in the above diagram does not show the entire workflow; some transformation of the raw input data preceded its creation. We'll reference the red letters in the model diagram where relevant.

• the study area was created from the PA municipalities shapefile by selecting Lancaster and York County municipalities adjacent to the Susquehanna river and saving this as a feature class; all other input data was clipped to this study area
• the currently preserved parcel data provided by the counties (see A) was merged with additional parcels under management by the Lancaster Conservancy (see B) as nature preserves
• the merged preserved parcels above were 'erased' from parcel data supplied by the counties (see C), resulting in an 'unpreserved parcels' feature class
• the observation point features referenced in 'Required data' were used as input to an ArcGIS server/cloud based viewshed calculation, using elevation data resident on the server and the separate viewsheds were merged and dissolved
• the viewshed feature class resulting from above (see D) was used to 'clip' from 'unpreserved parcels' only those parcels within the viewshed
• land use raster data for York and Lancaster county, obtained from the Chesapeake Conservancy, was merged and converted to vector data, and selection was used to create a feature representing 'forest' land use (see E)
• though not used in the parcel selection model here, separate features were created via land use selection for 'pasture' land use, and multiple wetland land use codes were combined into a 'wetlands' category
• by 'summarizing' the amount of forest within each parcel, and then 'selecting', the model can select only those parcels with a given minimum area and percentage of forest -- these minimum values are specified by the parameters (in blue -- see G)
• a select by location tool was used to further restrict parcel selection to those parcels adjacent to already preserved parcels -- this represents our final selection of parcels for preservation
• a 'flow line' feature class (see F), representing tributaries of the Susquehanna river, was used with the above parcel preservation candidates as input to a 'summarize within' tool, which replicates the input parcel data and adds a field with the length of streams within the parcel boundary
• the 'summarize within' tool can also be used to measure how much of any land use category is within any parcel
• parcels selected for preservation were then classed in one of three color coded categories based on the contained stream length for the final resulting map

Directly below is an image of the combined viewshed which was created by merging viewsheds for our three categories of observation points -- in the middle of the Susquehanna River, on select trails, and at singular scenic points.

The above model allows us to target parcels to facilitate a goal of acquiring contiguous blocks of forest with streams. This could allow us to preserve and extend riparian buffers, as well as habitats for compatible species. By focusing on parcels that are within view of recreational areas, we are trying to enhance visitor's experience, and perhaps inspiring environmental volunteerism and advocacy.

There are numerous ways the model could be improved. The National Hydrography Dataset tributary data is from 2004 and an update may be warranted. Beyond that, simple stream length within a parcel is a crude metric for the potential impact of measures that preserving that parcel could facilitate. Further work might include flow analysis that considers upstream land uses.