A Natural History of Millbrook Marsh,
A Wetland In An Urbanizing Setting

Chapter 7

RECOMMENDATIONS

Given the increased attention to protection of the Spring Creek watershed and Millbrook Marsh, outright loss of habitat by filling or drainage is unlikely. There are, however, two major factors that could cause degradation and threaten the continued viability of Millbrook Marsh. The greatest impact upon the marsh is caused by stormwater input and non-native invasive vegetation. The management of these two possibly interrelated problems would improve ecological conditions at Millbrook significantly and increase the integrity of the marsh ecosystem.

As stated throughout the previous chapters, stormwater impacts are numerous. They include scouring of streambeds, streambank erosion, silt deposition on streambeds and possibly corridors beside streams resulting in drier soils, introduction of pollutants that alter the water chemistry. The stormwater input is, therefore, detrimentally affecting macroinvertebrate populations, and reducing benthic habitat for macroinvertebrates and spawning fish.

One reason for stormwater input having such great consequence is the flashy, high velocity and an unusually large amount of water flowing into streams that are too small to accommodate it. For a 10-year storm, the stormwater flow entering at Thompson Run alone is estimated to be between 33 and 46 times that of the normal spring flow, after a fairly wet spring. This is a tremendous flow change with understandably great impact on a small stream and its inhabitants. Peak flows should be estimated for the higher frequency, smaller storm events so that management for them also will be appropriate to prevent streambank erosion and the other problems attributed to stormwater. At the very least, the management practices adopted with respect to stormwater should begin with no increase of stormwater input to Millbrook Marsh. A more proactive approach would include reducing the velocity of incoming stormwater and spreading the input over a greater time period. The stormwater should be introduced into the streams of Millbrook at a slower rate; less quantity over a longer period. The silt being deposited is probably originating from three main areas. Some silt is already suspended in the stormwater along with other contaminants before it reaches the drainage ditch beginning near the Pennsylvania State University Sewage Treatment Plant. A great amount of suspended sediments are added to the stormwater as it travels through the drainage ditch before it joins with waters from Thompson Spring (Figure 7-1). When the water reaches the eroded banks of Thompson Run more silt becomes suspended (Figure 7-2). This continues as the water flows through Slab Cabin Run (Figure 7-3), depending on the strength of the storm.

Sedimentation occurs along Thompson Run, Slab Cabin Run, and Spring Creek, depending upon when and where the water velocity decreases. From the evidence in this report, stormwater has clearly influenced negatively the morphology of Millbrook Marsh streams. The existing aquatic communities in the marsh cannot be maintained if the volume of stormwater increases or the water quality declines (Brooks et al. 1998).

thesis0022.JPG (582151 bytes)
Figure 7-1 Stormwater drainage ditch above Duck Pond. July 1998.

thesis0023.JPG (698959 bytes)
Figure 7-2 Thompson Run streambank erosion in Millbrook Marsh July 1998.

thesis0024.JPG (730509 bytes)
Figure 7-3 Slab Cabin Run eroded streambank in Millbrook Marsh after confluence with Thompson Run. July 1998.

Fish species richness has decreased in Millbrook Marsh. This could be the result of degraded habitat and water quality from stormwater input. Stabilization of the streambanks, including the drainage ditch, is imperative in solving the problem of siltation in Millbrook Marsh. This would improve the habitats of macroinvertebrates and fish, reducing the amount that cobbles, rocks and boulders are embedded. Reducing the velocity of stormwater input will also allow instream and riparian vegetation to reestablish.

Reducing the overall amount of stormwater is the most recommended. Design of new development and redesign of existing development throughout the watershed should begin with the priority of stormwater infiltration and detention on-site. This would reduce stormwater damage in Millbrook Marsh as well as other riparian and wetland areas further down the watershed. To repair existing and prevent further damage, there are various conventional methods of stabilizing streambanks. Riprap is commonly used, especially when the water level fluctuates widely. There is usually a bed of gravel or crushed stone under a layer of larger stone about 30 cm thick. Gabions are wire mesh cages filled with rock but these are used mainly when erosion cannot be controlled with more natural methods. Log cribbing, asphalt, or sodding are other methods of erosion control. There are also bioengineering methods of streambank stabilization which seem to be effective in reducing or eliminating erosion while also restoring native vegetation. The streambanks are returned fairly quickly to a functional and more aesthetically pleasing, natural state. One method enables natural bank stabilization using native hydrophytic vegetation plugs in a coconut matting. It was developed in Germany and has been used along the Little Cedar Creek in Allentown, Pennsylvania (Siegel 1994). After 3 years and several flood events, streambanks of the Little Cedar Creek were fully stabilized with no erosion. This method enhances the natural environment rather than using artificial means of conventional bank stabilization methods. Streambank stabilization within Millbrook Marsh should be implemented using a bioengineering method such as the one mentioned above, to keep the site in the most natural state. If at all possible, similar methods should be used for stabilization of the banks in the drainage ditch leading into Thompson Run, above the Duck Pond.

Water quality monitoring should take place on a regular basis. Additionally, samples should be taken during and directly after storm events to record the specific changes and effects of stormwater upon Millbrook Marsh streams. This monitoring should include stream substrate monitoring and benthic macroinvertebrate inventories. Quality, quantity and velocity of water at normal times should be compared to levels during and immediately following storm events of varying intensities. Research should include determining the reason for the decline in fish species richness. Have the Millbrook Marsh streams decreased in temperature because of the cessation of STP effluent, making them no longer habitable for warm water fish? This scenario is an improvement, and should be confirmed if true. If not, the factors that have caused the species richness decline, possibly a decline in stormwater quality and changes in stream morphology, must be determined and controlled.

Another issue that should be addressed with regard to water monitoring is the discharge amount of Thompson Spring. The data reported here suggests that the discharge has decreased since 1941, although this is not conclusive. Thompson Spring is a major source of water for Millbrook Marsh, via Thompson Run, and additionally, changes which affect Thompson Spring may also affect the many smaller springs within Millbrook Marsh, including the springs of the calcareous fen. The Big Spring in Bellefonte has a discharge today of roughly twice that of Thompson Spring, though it was much more previously. In the past, it underwent a permanent decrease in discharge to about half of its original output. This decrease in discharge occurred quite suddenly and coincided with the opening of a new lime mining operation in the area. This may have been coincidental. No conclusive evidence has linked the two incidents but it is reasonably assumed that they are related. The change in Thompson Spring discharge seems to be gradual rather than sudden, though there are great periods of time between discharge data. Also, other details such as the discharges of other springs in the area at the same time and whether the data was gathered during a time of drought, for example, are not available. The possibility of an overall decrease in Thompson Spring discharge and the reasons for it are issues worthy of address in subsequent studies.

Due to the nature of the wetland ecosystem, maintaining the integrity of the hydrologic functions of Millbrook Marsh is imperative in seeing its success and continuation into the future as a natural wetland. Measures should be taken to prevent any further human-induced changes in the hydrology of the site. Much of the change in hydrology can be attributed to stormwater. Damage to the marsh can and should be discontinued by reducing the flashy high velocity input of urban stormwater. Doing this will stop or slow stream bank erosion and siltation when used in conjunction with streambank restoration methods. The corridors along Millbrook Marsh streams seem to be drier than in the past, evidenced by the fact that they are supporting shrub species that prefer drier sites. This is probably at least partially due to the silt deposition of overflowing streams during storm events. This issue should be investigated further with soil testing to determine whether that hypothesis is true. If so, or if the streambed scouring is causing a water table draw-down, stormwater control and reduction will decrease or prevent further drying out of the corridors beside streams.

The second major item of concern is that of the change in vegetation communities over time, a combination of cover type change and native to non-native flora. A baseline inventory and mapping of the extent of non-native species on site will be an important tool in management. There are 2 shrub species that have invaded large sections of Millbrook and are the most detrimental of the non-natives at this time that exist on site. The great increase in shrub cover, mainly Tartarian honeysuckle and Multiflora rose, warrant their removal, but this should be undertaken with care. Where vegetation is removed, the disturbed area must be replanted with native vegetation appropriate to the particular area within Millbrook Marsh, otherwise non-native vegetation will most likely return due to the seedbank in the soils. The appropriate replacement vegetation should be determined based upon the historical photographs, surrounding vegetation and the soil moisture level. Extreme care should be taken if herbicides must be used, as in the case of a third invasive shrub, Autumn olive. It is nearly impossible to remove all the roots of the Autumn Olive, which will resprout if left behind. There are only several individuals at Millbrook Marsh presently, so removal of them at this time will prevent a greater problem in the future.

Two herbaceous plant species that must be constantly monitored for are Japanese knotweed and Purple loosestrife. These two wetland invaders are located not far from Millbrook Marsh and if allowed to become established, would be practically impossible to eradicate. The most likely to arrive first is Japanese knotweed, which must be completely removed immediately upon detection. Purple loosestrife must also be removed immediately. With both plant species, it is important to remove all roots and plant parts (Blossey 1997).

After the primary invasive non-natives have been removed and native vegetation has become established, the secondary invasives should be replaced with the appropriate native vegetation. When this has been accomplished, a monitoring schedule can be implemented and non-natives removed before they become fully established.

For a more complete account of the natural history of Millbrook Marsh, continued study is necessary for some aspects of the ecosystem. Several inventories are nonexistent. The initial baseline inventories needed in Millbrook Marsh include those for mammals, reptiles and amphibians. This could be undertaken by students in wetland and wildlife courses at Penn State, members of the Nature Center under supervision, or in subsequent research projects at Millbrook Marsh. There is a substantial list of vascular plants for Millbrook, but certain areas of the Marsh have not been inventoried in quite some time. The calcareous fen, in particular, should be studied in detail, as well as other parts of the emergent wetland areas. Close attention should be given to rare, threatened and endangered plant species and their location, especially to their vulnerability to development along the edges of Millbrook Marsh, stormwater, and the increase in visitors because of the new Millbrook Marsh Nature Center. The bird list for Millbrook Marsh is a fairly good baseline inventory. More information should be gained regarding the birds that nest in the marsh, those that are year-round residents, and those that are summer residents. The present bird inventory includes 7 facultative, 2 facultative wetland and 7 obligate wetland bird species. Another bird inventory should be undertaken, especially looking for the secretive species that are expected but have not yet been listed. It would not be unusual to find other wetland specialists within the marsh.

All aspects of the calcareous fen should be studied in more detail. It is possible that the fen and the vegetation it supports would benefit from removal of the urban fill. Soil drilling and testing should be performed to delineate the extent of the fill, and determine if removal is feasible. Since this area is privately owned, approval would be necessary.

Another area that should be studied in detail is the Bathgate Spring Run section. This portion of the marsh is in danger of stormwater damage, so any flora or fauna there that might be damaged should be moved before the stormwater is diverted into the stream. The vegetation can be used in other areas of the marsh where non-natives are removed. In the unfortunate case that the stormwater is directed into Bathgate Spring Run, the opportunity exists for a detailed stormwater impact study. If the best interests of the Millbrook Marsh ecosystem are considered, the stormwater will be diverted away from Bathgate Spring Run. Already, 19% of Pennsylvania’s assessed streams are degraded (Thorne et al. 1995) and rather than adding to that figure, it should be reduced. Stormwater addition to Bathgate Spring Run will most likely cause scouring of the healthiest streambed in Millbrook Marsh (Figure 7-4) and result in additional sedimentation further downstream. On a grander scale, it will also add to the 2.6 billion pounds (1.2 billion kg) of sediment that is washed into the Chesapeake Bay from the Susquehanna River Basin alone (Thorne et al. 1995). Degradation of the stream morphology, instream and riparian vegetation, invertebrate and fish communities, and higher order wildlife will occur as has happened in Thompson Run and Slab Cabin Run.

Monitoring, additional investigations, and baseline inventories are necessary to determine changes in Millbrook Marsh and to ascertain what must be done to prevent further degradation now and in the future. Furthermore, the increase in use as a recreation area because of the Nature Center will result in potential harm by surpassing the human carrying capacity of the marsh. This can be counteracted by proper management. To ensure the future of Millbrook Marsh, the ecosystem, and the Millbrook Marsh Nature Center, several changes put forth here and in the Protection and Management Plan (Brooks et al. 1998) must occur. Continued degradation will occur if management practices, or the lack thereof, remain static. With this initial summary of the natural history of Millbrook Marsh and the recommendations from the Millbrook Marsh Protection and Management Plan, the ecosystem can thrive and become a place of discovery, learning, inspiration, and preservation.

thesis0025.JPG (683832 bytes)
Figure 7-4 Confluence of Bathgate Spring Run (right) and Thompson Run (left) in Millbrook Marsh July 1998.

On to References