We notice a new phrase cropping up a lot lately in the arguments of NIMBY types and other anti-growth advocates – “forest fragmentation.”  It sounds ominous, of course, and, therefore, our friends on the other side of the natural gas debate have seized upon it.  There’s nothing negative that doesn’t make them go wobbly in the knees, after all.  Recently, some of those folks have turned to a USGS report on Bradford and Washington Counties for evidence natural gas development will somehow leave Pennsylvania’s forest devastated and all lovable furry creatures scurrying for protection under ever smaller clumps of trees that remain.

You’ve got to hand it to them.  They’re great marketers, if nothing else, and know the image of a threatened Bambi still resonates with a public that too often knows very little of what’s really happening.  There is some fragmentation, to be sure, but it would seem to be less with the forests than in the minds of those who would look only at statistics supporting their position, while ignoring the bigger patterns contradicting those claims.

The USGS report is entitled Landscape Consequences of Natural Gas Extraction in Bradford and Washington Counties, Pennsylvania, 2004–2010 and assembles some detailed data regarding the impacts of natural gas development (both Marcellus Shale and non-Marcellus activity) on forest and other landscapes in those two counties.  We have no quarrel with the USGS data in so far as it goes, but the extrapolations of that data and misuse by natural gas opponents raises many issues.

A story in the Pittsburgh Tribune-Review entitled “Research maps out gas boom impact in Washington County,” for example, melds together the USGS data with comments from a “soil and forest expert at Penn State studying drilling” who estimated natural gas development “could lead to 278 to 695 square miles of disturbance, comparable to the state’s number of abandoned surface mines.”  The exact source of that quote is unclear, but the report to which it apparently refers is a paper published in the Environmental Management journal in March, 2012.

Entitled “Early Trends in Landcover Change and Forest Fragmentation Due to Shale-Gas Development in Pennsylvania: A Potential Outcome for the Northcentral Appalachians,” the Penn State report is only slightly longer than the title, but that’s a subject for another day.  It covers some of the same territory as the USGS study and provides a bit of useful data, but we can’t help but notice what is NOT discussed in either report.  Consider the following:

Where Is The Context?

Both reports go to great lengths to document problems with forest fragmentation and proceed to suggest natural gas development is a significant contributor to such fragmentation.  Neither report, though, provides any perspective on long-term trends to indicate how development might be effecting land cover over time.  If we are rapidly losing forest cover and natural gas development is adding to the loss, accelerating the pattern of change, that is one thing, but minor conversions of forest land to other uses simultaneous with reversion of brush and pasture land to forest is quite another.  Context, after all, is everything if one doesn’t want to miss the forest for the trees, isn’t it?

Neither the USGS, nor the Penn State, report provide any historical perspective on landscape patterns in areas where Marcellus Shale development is taking place.  They simply assume land cover is static and analyze how natural gas development is or, more often, will affect the patterns without taking into account the dynamic nature of land use and how it results in forest changes independent of this factor.

The U.S. Forest Service gathers statistics on forestation patterns and that data suggests the areas most affected by natural gas development have either gained or held their own in terms of forest cover.  Their data indicates the following for Pennsylvania as a whole, for example:

Pennsylvania Forest Cover (Source: U.S. Forest Service)

There’s not much evidence of decline here, is there?  Forest land has stayed fairly even since 1965, trending down slightly from 1990 and back up since 2004 or so, but notice it is up substantially since 1955 or thereabouts, which is no coincidence.  That is when the nature of farming began to change as the tractor replaced the horse, farms got bigger, smaller farms began to go the way of the buggy whip and less productive crop and pasture land started reverting to forest.  We have, in other words, since seen forest land become a lot less fragmented over the years.  Don’t believe it?  Well, compare some aerial photos from years ago with those on Google Earth.  It’s easy and it’s revealing.

The Wayne County, Pennsylvania, Department of Planning did such a comparison for their entire county, comparing 1959 and 2002 aerials and classifying land by use.  This is what they found, the two maps saying it all:

Wayne County, Pennsylvania Land Use, 1959-2002

This comparison of land use over 43 years plainly indicates forest land has not only grown, but also forest fragmentation has decreased.  The County Planning department noted the following in its reports:

Forestland, in 2002, covered nearly 10% more of Wayne County than it did in 1959.

In 1959 forest covered 55.6% of Wayne County, in 2002 it covered 65.2%.

The most notable change in land use/land cover over the study period was the amount of cropland that converted to pasture/brush use. Over 15,882 acres or 18% of the 1959 cropland was idled and began the process of reforestation. Another 9,593 acres, or 10.9%, was reforested in the 43-year study period.

Get the picture?  Wayne County is typical of many counties throughout the Marcellus Shale region and it has gained large areas of new forest over the last half-century (an update in 2008 revealed the same patterns).  Moreover, as reversion has taken place, new connections between forest patches have clearly developed, as the Wayne County maps demonstrate.  It is against this background any changes today should be measured.  The USGS study indicates 0.41% of the land in Bradford County has been disturbed by Marcellus Shale to date, but only 0.12% of the county’s forest land has been disturbed.  So, how does that compare with the total forest cover?

Well, the Forest Service data, once again, provides some insight.  Their data for both Bradford and Washington Counties from 1989 through 2010 (all the years available) indicates forest land in Bradford County increased by 30,834 acres over the 23 years, a gain of 7.2%.  Washington County lost 15,792 acres or 5.8% but has added 854 acres since 2004.  If there’s been a loss of forest land due to Marcellus Shale development in recent years, it’s so minimal it is, in turn, lost in the much larger gain of forest land from reversion of brush or pasture land into forest.

The Penn State study suggests it isn’t just forest changes that are important, but the type of forest involved.   Their researchers developed a classification entitled “core forest,” which they define as forest being 100 meters (328 feet) or more from the edge of non-forested areas.  It is, in other words, deeper forest where larger trees may be expected to be found.  There is no such category in the Forest Service data but the agency does estimate amounts of land in large, medium and small diameter trees, which provides a useful corollary comparison.  Here’s the trend with respect to land in large diameter trees:

Large Diameter Tree Forest Cover

Notice the acreage in large diameter trees has grown in both counties, but particularly in Bradford County.  Marcellus Shale development no doubt resulted in sacrifice of some forest land in individual circumstances, but can it matter when the acreage in large diameter trees is growing to this degree?  This is the kind of context completely missing from both the USGS and Penn State reports.

What Are The Alternatives?

Likewise, neither report bothers to examine how the forest losses involved with natural gas development might compare to other land uses with similar economic development potential.  It’s a highly relevant question because private land must produce some economic rent.  If that rent cannot be produced from natural gas development, then something else must be done with that land to pay its carrying costs.  This is a reality many are content to ignore but no landowner can ignore it, nor can any local government depending on the property taxes from that land.  One cannot study the impacts of natural gas development without considering the alternatives.

What are those alternatives?  Well, they consist primarily of the following; agriculture, timbering, other mining, commercial development, residential development and industrial development.  We could add public use, but that requires governments with the funds to buy the land and preserve it.  Given that all governments depend on private taxpayers to do so, and most governments are currently cash strapped it, we may comfortably dismiss that option.  Some form of productive use is required.

The most obvious is timbering, which makes use of the renewable resource that a forest constitutes.  There are several considerations with timbering, however.  First, it is fully compatible with natural gas development.  The latter requires very little forest land and some of the road infrastructure can be used for both enterprises.  Secondly, timbering involves significant changes in land cover, far more than natural gas development.  Another USGS study on Land Cover Change in the Eastern United States found “the highest amounts of change were generally in eco-regions with active timber harvesting.”  It indicates 7.6% of forest land changed in some fashion (to or from forest) between 1973 and 2000 and “the primary forest conversion was to mechanically disturbed, clear-cut land and forest land recently harvested.”

Timbering also involves considerable road infrastructure, a subject of Forest Service concern for some time.  Some 132,800,000 acres of the Forest Service’s land is accessed by 522,522 miles of road, which, assuming a modest 50 feet width of disturbance, involves over 3,160,000 acres of land – equivalent to 2.4% of the land base involved.  The USGS study of Marcellus Shale development indicated 0.12% of forest land had been disturbed in Bradford County and 0.42% in Washington County.

The Penn State study doesn’t deal with percentages and uses hectares, which has the unfortunate effect of obscuring the import of the statistics, but it does suggest development of all permits issued as of June 3, 2011 could result in up to 894 hectares (2,209 acres) of forest land conversion.  A review of permits issued by county compared to Forest Service data on coverage indicates there is some some 12,254,215 acres of forest land in those counties, so a conversion of 2,209 acres would represent but 0.02% of the total.  That’s a whole lot less than timbering.

It’s also important to realize timbering doesn’t always pay its way either, meaning it is unrealistic to expect landowners to maintain land in forest without ancillary sources of income such as natural gas development provides.  The estimated use value for timbering, in fact, ranges from a low of $36 to a high of only $297 per acre, according to the Pennsylvania Bureau of Forestry – not exactly something anyone would consider a prime investment.  Holding costs can easily exceed the annual return and obviously do if the opportunity costs are considered.  Therefore, timbering is a both an economically inferior use of land compared to natural gas development and also produces more land disturbance.

Other alternatives, of course, involve much larger conversions of forest land.  Agricultural use and other mining activities require major clearing of land by their very nature.  Residential, commercial and industrial development all typically involve clearing of 25% or more of lots.  Only natural gas development delivers a high economic return with land disturbance percentages in the single digits and below.

It’s The Wildlife Stupid!

There’s still one other factor that must be not be neglected, but gets no attention in either the USGS and Penn State reports and it’s the wildlife itself.  First, while there is no doubt forest fragmentation can be an issue in many cases, it is also undeniably true clearing of pipeline corridors, for example, generates growth of new browse that supports many wildlife species.  A recent study of utility corridors in Michigan also found a positive impact on biological diversity and numerous other studies have demonstrated there can be positive effects as well as negative.

Indeed, this is also recognized by local conservancy groups and the Pennsylvania Fish and Game Commission.  In a 2011 article focusing on gas drilling impacts on forest and wildlife an official with t he Fish and Game Commission noted:

We’ve actually seen deer move back into the area and take advantage of some of the openings that have been created.

This sentiment was also captured by another spokesperson for the PA Game Commission who commented in a Harrisburg Patriot-News article later in 2011.  In that article, Game Commission spokesperson Jerry Feaser stated:

The kind of forest clearing that accompanies shale drilling ultimately benefits many game and non-game species, Feaser said.

While the development can be an inconvenience to hunters now, the areas will ultimately go back to grass, shrubs and early forest.

You take the same acreage and put houses or shopping malls or gas stations on it, and that land is lost forever.

The most important wildlife factor, though, may not be the impact of forest cover changes on wildlife, but the impact of wildlife on the forest.  The single greatest threat to many forests throughout our region today is not man’s activities, but those of the White Tailed Deer, according to Pennsylvania agencies responsible for wildlife management.  A Forest Service article from 2008 details their view of  problem and quotes an earlier report by the agency as follows:

A Pennsylvania Department of Conservation and Natural Resources study found “three quarters of the state’s 2.1-million-acre state forest system have moderate or severe deer damage, with nearly half of surveyed areas showing no new growth at all,” according to a report in the Allentown Morning Call.

Seeing The Forest

So, maybe forest fragmentation from natural gas development isn’t such a big deal after all.  There is no reason to question the data assembled by USGS and Penn State, but what is left out is often more important than what is included when it comes to academic studies.  You can’t see the forest if you’re only looking at a handful of trees.