PS 38-94
Invasive non-native plant colonization on Marcellus Shale well pads predicted by local road density and drilling intensity

Wednesday, August 13, 2014
Exhibit Hall, Sacramento Convention Center
Kathryn M. Barlow, Intercollege Graduate Degree Program in Ecology, The Pennsylvania State University, Univerisity Park, PA
David A. Mortensen, IGDP Ecology, Department of Plant Science, The Pennsylvania State University, University Park, PA
Patrick J. Drohan, Ecosystem Science and Management, The Pennsylvania State University, University Park, PA
Ashley Hayes, College of Earth and Mineral Sciences, The Pennsylvania State University, Univerisity Park, PA
Kristine M. Averill, Intercollege Graduate Degree Program in Ecology, Pennsylvania State University, University Park, PA
Background/Question/Methods

Disturbed plant communities with newly available resources are ideal habitats for invasive plant colonization. In the last decade, development associated with  Marcellus Shale gas extraction has brought new networks of roads and pipelines, and high-volume traffic, into forests and agricultural landscapes. We hypothesize that within the development footprint in Pennsylvania forests invasive non-native plants (INNP) would be present on the well pads (~3 acres) used to stage gas extraction. In addition, we hypothesize that invasion severity on well pads would depend on the extent of disturbance in the area of the pad prior to construction. To determine the presence and extent of invasion we surveyed 128 well pads and access roads across the PA State Forest system and the Allegheny National Forest for 23 INNP in the summers of 2012 and 2013. Using ArcMap we calculated road density prior to well pad construction (31 acre circular polygon) as a measure of landscape fragmentation and potential invasive plant habitat. We used the number of wells drilled on a pad as an estimate of traffic to the pad.

Results/Conclusions

Invasive non-native plants were present on 70% of well pads. Of those pads invaded, 60% had stem counts between 1-100, 29% had counts from 100-1000, and 11% had counts greater than 1000. Herbaceous non-natives were more abundant than woody species, and the most abundant were Polygonum convolvulus, Centaurea maculosa, Phalaris arundinacea, and Microstegium vimineum. Using a generalized linear mixed model we found that the invasion rate ratio on pads increased by 4% for every 100 stems of INNP on access roads, by 6% for every 100m of road density, and by a factor of 12 for pads with 2-11 wells compared to pads with 0-1 well. Pad age alone was not a significant predictor, but when included as an interaction with the number of wells the invasion rate ratio increased by 11%. As development of the Marcellus and the Utica Shale is projected to increase in intensity and extent of surface disturbance it is clear that such development will accelerate invasive plant invasions. Monitoring and management of these newly disturbed sites will be critical to limit invasive plant success and should be targeted at areas with high density road networks and where well numbers are high.