HYDROLOGY AND AQUATIC HABITATS
 As with all wetlands, the hydrologic regime of Millbrook Marsh plays the primary role in controlling function for the site and in the surrounding watershed and subwatershed.  Millbrook Marsh serves in a flood control capacity by storing surges of stormwater or meltwater from further up the watershed, thereby reducing peak flows and crest levels further downstream.  During these flooding events the marsh vegetation helps filter out sediments and potential contaminants in the stormwater.  The marsh also functions as a groundwater discharge area for many springs.

 The hydric soils of Millbrook Marsh occur due to the periodic flooding of Slab Cabin Run and Thompson Run, the discharge of numerous continuously flowing springs, and to the high water table in some areas.  These various hydrological conditions contribute to the high plant species diversity at Millbrook.

Springs

 The great number of springs on the site (Map 7) is due to the presence of the very high water table and to the geology of the area.  Because of the large size of the interbedded limestones and dolomites (Chapter 4), groundwater usually occurs along joints and fractures caused by structural deformation, bedding planes, and in solution cavities caused by water.  Joints in the bedrock are influential in determining the occurrence and movement of groundwater in the rock types of this area.  Fracture traces are also important in the movement of the groundwater here (Clark 1965).

 Several major springs exist outside of the site in the Bathgate neighborhood, on the other side of Puddintown Road.  Bathgate Spring, at an elevation of 296 m (970 ft) is on the northeast side of Orchard Road and is contained in a springhouse.  Another, also called Bathgate Spring, flows out of the ground on the farm at the end of Bathgate Drive.  This spring is at an elevation of 305 m (1000 ft).  The discharge reported in 1904 was 0.033 cms and the 1971 measurement was 0.034 cms with a water temperature of 11.0° C.  Both have been used by the Lemont Water Company as a source of public water (Wood 1980).  These two springs contribute the continuous streams of water which enter the site along the bend in Puddintown Road and near the tractor path.  At the confluence of the two springs is an area where the water table is very high, resulting in persistent saturation.  This section contains emergent and shrub wetlands located on the Farm 12 acreage.   At this time, Bathgate Spring Run is one of the healthiest of the riparian-wetland areas on the site, with little or no undercutting of the streambank, and no scouring of the stream bottom.  It is the only stream on site that is not affected significantly by urban stormwater surges (Figure 3-1).  By far the largest spring outside of Millbrook Marsh is Thompson Spring, the source of Thompson Run.
 
 

 In April 1941, B. F. Donley reported 4,192 gallons per minute, or 6.036 million gallons per day (0.247 cms) (PA Fish Commission 1941).  A 1944 measurement showed a 0.237 cms discharge (Woods 1980).  Historically, the output of Thompson Spring has been approximately 5 million gallons per day (0.219 cms) (Clark 1965).  In 1971 a discharge of 0.170 cms was measured with the water temperature at 10.0 to 10.3° C (Woods 1980).

 Another major groundwater discharge area lies within the marsh, but outside the boundary of the Millbrook Marsh Nature Center.  This concentration of springs is found at the south edge of the marsh, near the west and north ends of Watkins Road and Shoferd Lane.  The springs discharge through the limestone bedrock causing the ground to be consistently saturated by water with an alkaline pH level.  Water testing done for the Study of Calcareous Fen Communities in Pennsylvania  during the summer of 1995 showed pH levels of the water at this site in Millbrook Marsh to be 7.6, 7.5, and 7.2 (Western Pennsylvania Conservancy 1995).  This alkaline groundwater supports the calcareous fen, with a unique and rare community of plant species.  The PennDOT study (1980) reported a pH of 8 at 3 locations in the marsh.  Other springs found throughout Millbrook include several along Thompson Run, Slab Cabin Run, and one at the east and north end of Shoferd Lane.

 The percentage of the total discharge that the springs contribute to the marsh varies seasonally.  In June of 1997 the flow leaving Millbrook Marsh at Slab Cabin Run equaled a total discharge of 0.608 cms (Table 3-1).  About 7 percent of that, or 0.045 cms, came from the Bathgate Springs.  About 0.028 cms (5 percent) originated from smaller springs throughout the site.  The total from Bathgate and the smaller springs was 0.073 cms.  There was a proportional shift by October.  After the dry summer, the total discharge was 0.274 cms, and 21 percent of the water, 0.059 cms, was contributed by Bathgate Springs and the other unmeasured springs (Brooks et al. 1998).  The amount of water contributed by springs was 80 percent of the June flow, reduced only 0.014 cms, while the total flow was only 45 percent of the June total flow, a reduction of 0.334 cms.
 
 
 
 

Stormwater

 Wetlands associated with streams and rivers often provide a flood control function by dissipating and temporarily storing natural and stormwater flows.  This stormwater, if not managed carefully however, can also cause significant damage to wetlands and riparian corridors.  This results in a reduced flood control function.  Substantial disturbance has occurred in Millbrook Marsh, as evidenced by the observations of adjacent property owners and long-time residents (N. Deno, B. Niebel, pers. comm.) and by examining the stream characteristics.  The stream surveys conducted in reaches of Slab Cabin Run and Thompson Run within Millbrook Marsh during 1997 yielded evidence that high stream flows attributable to stormwater runoff have negatively influenced the morphology of both streams (Brooks et al. 1998).

 In a more natural or less degraded system there would be few or no point source entries of stormwater.  Instead there would be overland sheet flow and an overall, and in most cases, relatively gradual rise and fall in stream flow.  This is not the situation at Millbrook Marsh where there are multiple stormwater outlets discharging throughout the site and flowing into the streams offsite (Map 7).  There are stormwater outfall pipes located outside the bounds of the Millbrook Marsh Nature Center, but still within Millbrook Marsh.  Others are outside the marsh but still affect the site.  A major input of stormwater flows into Thompson Run before it crosses to the north side of East College Avenue.  This has been the primary cause of scouring and bank erosion observed in Thompson Run.  One of the detrimental aspects of stream bed scouring is the lowering of the water table.  Further downstream the result is an increase in silt deposited on the stream beds of Thompson Run and Slab Cabin Run.  The majority of stormwater originates in both the Borough of State College and the Pennsylvania State University campus.  It flows first into a ditch on the east side of the University Waste Water Treatment Plant (Figure 3-2) and then joins with part of Thompson Run above the Duck Pond.  The Penn State campus drainage area is approximately 172 ha (426 ac).

The estimated discharge in 1994 for a 10-year storm was 14.2 cms to 19.8 cms (500 to 700 cfs).  The portion of stormwater contributed by drainage areas outside of the University was estimated to be between 5.7 cms and 8.5 cms (200 and 300 cfs) (Gaudlip, pers. comm).  The Borough has two main drainage areas totaling about 99 ha (245 ac) that drain (with the University stormwater) into the ditch above the Duck Pond.  One, the Calder Way system, drains approximately 89 ha (220 ac), and the other, the Holly Alley system, drains approximately 51 ha (125 ac).  Estimates of discharge were made in 1988 to 1990.  The expected discharge of the Holly Alley system for a 10-year storm was 4.8 cms (170 cfs) (Wagner, pers. comm.).  The combined Borough and University area is 312 ha and the combined 10-year storm discharge estimate is 27 to 33 cms.  In addition to the stormwater entering Thompson Run from above the Duck Pond, there is also stormwater draining into Walnut Spring Run, which then enters Thompson Run just before it crosses under East College Avenue and into Millbrook Marsh.  Thompson Run also has two stormwater outlets just after East College Avenue, the point at which it enters the south western corner of the Marsh.
 
 

 Two more stormwater entry points are located at the southeastern corner of Millbrook Marsh.  One is located at the entrance to the upper parking lot of the State College Township Building, and one is at the lower level parking area, and includes the sheet drainage from the parking lot (Figure 3-3 and 3-4).  This flows through a swale into the floodplain.  These two stormwater outlets drain into the marsh itself, not into the associated streams.

 At the south eastern corner of Millbrook Marsh one stormwater outlet drains into Slab Cabin Run at the crossing of East College Avenue.  Approximately 60 m from East College Avenue is another stormwater outlet flowing into Slab Cabin Run.

 Another stormwater outfall is located along the boundary adjacent to the Mt. Nittany Expressway about half way between Puddintown Road and East College Avenue.  It is on the west side of the paved footpath, and drains into a short swale.
 
 

Water Quality

 Water quality in Millbrook Marsh and the associated streams has improved in some ways, yet decreased in other ways.  The sewage treatment plant at University Drive and East College Avenue, in operation since about 1913, ceased input of treated sewage effluent into Thompson Run in 1983 (J. Gaudlip, pers. comm.).  The effluent is now piped out and spray irrigated elsewhere.  The input of stormwater, though, is increasing as more of the drainage area acreage is developed with impervious surfaces.  The result is an increase in stormwater contaminants and an increase in discharge amount and velocity during storm events, causing more scouring, erosion, sedimentation and possibly water table drawdown.  The integrity of the water in Millbrook Marsh is affected by the actions and pollutants mainly associated with stormwater at this time.  Those things generally impacting water quality include: dust accumulated during dry periods such as atmospheric pollutants on roofs, traffic emissions from roads and parking areas, solids from road deterioration, vehicle and tire wear including vehicle fluids, road salts, leaking septic tanks, detergents from car washing, leaking or illegal dumping of oil and other fluids, herbicides and pesticides from lawn and garden care, erosion of drainage channels, buildup of solids in sewers, and litter (Novotny 1995).  The stormwater received by Thompson Run from the University lands includes coal pile runoff from the PSU-West Power Plant during rain events (J. Gaudlip, pers. comm.).  The power plant was built some time in the 1930’s, and until about 1987 or 1988, the stormwater going into Thompson Run also included continuous input of boiler blowdown, condenser cooling water, intermittent input of softener recycle water and cooling tower blowdown as well as the coal pile runoff (Metcalf and Eddy 1961, Bureau of Water Quality Management 1989).

 The present condition of Millbrook Marsh is shown by data gathered in the summer and autumn of 1997. As well as the typical heavy metals and petroleum derivatives in the stormwater entering Thompson Run, there is substantial sediment load originating in the ditch that conveys stormwater from the discharge pipes to the Duck Pond.  This stormwater ditch consists of vertical unvegetated banks of approximately 2.5 m which continue to erode and add sediments to the stormwater.  Impaired water quality from the input of urban stormwater runoff is the most likely cause for the moderate degradation in the macroinvertebrate communities at 5 of 6 stations throughout the marsh (Chapter 7).  The U. S. Environmental Protection Agency’s Rapid Bioassessment Protocol III (RBP III) was used to determine the condition of the macroinvertebrate community of Millbrook Marsh and is a good indication of aquatic ecosystem health.  RPBIII measures the amount of degradation of stream communities due to pollution, compared to an unaffected reference site, and results in one of 4 Bioassessment categories ranging from "Non Impaired" to "Severely Impaired".  Overall, RBP III suggests a moderate departure from good water quality for streams within Millbrook Marsh (Table 3-2) (Brooks et al. 1998).
 
 
 
 
 

 Historical water quality data for Millbrook Marsh includes sporadic testing of Slab Cabin Run, Thompson Run and Bathgate Spring Run from surveys conducted from 1941 through 1990.  Data for Thompson Run and Slab Cabin Run below the confluence with Thompson Run reflect the sewage effluent input from State College Borough and the Pennsylvania State University until cessation in 1983.  The historical data for Millbrook Marsh is from various reports from the Pennsylvania Department of Environmental Protection, the Pennsylvania Fish and Boat Commission, Pennsylvania State University theses, and the Pennsylvania Department of Transportation (PennDOT).  Two stations along Slab Cabin Run in Millbrook Marsh were among the sampling locations for the water collection of May 1980 (PennDOT 1981).  Historical data for Slab Cabin Run from 1972 to 1980 was also collected by the Pennsylvania Department of Transportation and compared to the 1980 data.  Water quality data collected in May 1980 and the data from 1972 to 1980 included dissolved oxygen concentration, turbidity, nitrate concentration, total nitrogen concentration, total and ortho-phosphorus, total suspended solids, chloride, sodium, fecal coliform bacteria, pH, and oil and grease concentrations (PennDOT 1981).

 Spring Creek and its tributaries, including Slab Cabin Run, Thompson Run, and Bathgate Spring Run, are designated at least Cold Water Fish (CWF) streams (Bureau of National Affairs 1994).  Slab Cabin Run is designated a High Quality Cold Water Fish stream from the source until its first crossing with PA Route 26 in the Pine Grove Mills area.  The critical use for Slab Cabin Run, Thompson Run and Bathgate Spring Run is considered to be aquatic life.  The criteria designated to protect aquatic life is given in the following water quality data (Bureau of National Affairs 1994).

Dissolved Oxygen

 The criteria for the level of dissolved oxygen in Cold Water Fish streams to protect aquatic life is a minimum daily average of 6.0 mg/l.

Dissolved oxygen levels on January 15, 1959 at 1:00 pm were 6.80 mg/l at Slab Cabin Run and Route 545 (now Route 26, East College Avenue), and 8.72 mg/l at Thompson Run and Rte 545 (Spring Creek Surveys 1959) (Table 3-3).  The level in 1959 in Slab Cabin Run was between 7.3 and 8.5 mg/l in October and November (McDonnell 1960).  Levels of dissolved oxygen in 1960 for Thompson Run and Slab Cabin Run were sampled around the clock on October 25 and 26.  Samples were taken between 4:10 am and 11:40 pm.  The dissolved oxygen levels in Thompson Run below the Duck Pond ranged from 6.20 to 5.40 mg/l.  The average of the 4:00 am to 6:00 am samples taken between October 5 and November 11, 1960 was 5.10 mg/l.  Slab Cabin Run, at Route 545 average was 8.67 mg/l with a range from 12.45 to 10.70 on October 25 and 26  (Metcalf and Eddy 1961).  June 8, 1962 sampling occurred at about 12:15 pm and showed dissolved oxygen levels in Thompson Run at Route 545 to be 8.2 mg/l and showed Slab Cabin Run to be 2.6 mg/l (Vrenna 1962).  Levels in May 1980 were at or near saturation, more than 2 mg/l higher than the state minimum level.  Slab Cabin Run, about 90 m above the Thompson Run confluence (PennDOT Station 3), was 10.8 mg/l and  was 10.2 mg/l about 230 m from East College Avenue (PennDOT Station 4) (PennDOT 1981).  An elementary school class surveyed Slab Cabin Run in October 1986, and found levels of 14mg/l, 12.3mg/l, and 13.6mg/l at three locations within the marsh.  July 1989 data showed an improvement in the dissolved oxygen levels of Thompson Run after the cessation of the treated sewage effluent input.  Approximately 300 m below Route 26, the dissolved oxygen level was 9.8 mg/l.  Slab Cabin Run at about 800 m before the confluence with Thompson Run was 10.80 mg/l (Hughey, 1990).
 
 

Turbidity
 The standard method of determining turbidity in the past has been the Jackson candle turbidimeter, resulting in units of JCU, or Jackson Candle Units.  These are also labeled JTU, or Jackson Turbidity Units, and no conversion is needed between the two (S. McGonigal and K. Confer, pers. comm.).  One JTU represents the amount of light intercepted by 1 mg/l of a standard sand.  The lowest turbidity value that can be measured, though, is 25 JTU.  Because of this, the standard method has become based upon the electronic nephelometer, resulting in NTU, or Nephelometric Turbidity Units.  JTU’s approximate, but are not truly equal to NTU’s (Brown 1980).  The maximum contamination level of turbidity for CWF aquatic life is not more than 100 NTU from September 16 to May 14 and not more than 40 NTU between May 15 and September 15 of any year (Bureau of National Affairs 1994).
 Water chemistry data for July 11 and 13, 1978 showed turbidity of Thompson Run at 6 JCU and Slab Cabin Run less than 5 JCU (Miller 1979) (Table 3-4).  On July 5-7, 1989, turbidity was 6.80 NTU in Slab Cabin Run and 4.00 NTU in Thompson Run (Hughey 1990).  Turbidity levels were moderate for the historical and 1980 data, which was 25 JTU at PennDOT Station 3 and 27 JTU at PennDOT Station 4.  At all other Spring Creek and Slab Cabin Run stations during the 1980 testing the turbidity levels were between 4.8 and 9.3 JTU (PennDOT 1981).
Table 3-4
Turbidity Levels in Millbrook Marsh Streams, 1978-1989
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run
July 1978 (Miller 1979)   5 JCU 6 JCU
1980 (PennDOT 1981) 25 JTU27 JTU
July 1989 (Hughey 1990)   6.8 NTU 4 NTU

Nitrogen
 The allowable level of ammonia nitrogen for CWF streams and aquatic life is calculated for each water sample.  The criteria is dependent upon temperature and pH, and must be calculated using values taken between July and September.  The maximum level of nitrate plus nitrite is 10 mg/l as nitrogen (Bureau of National Affairs 1994).
 In January 1959, ammonia was 16.0 (Table 3-5) and Kjeldahl nitrogen was 41.7 (Table 3-6) in Thompson Run.  The assumed units for this data, since it was not given, is lb/d, pounds per day.  Kjeldahl nitrogen is the measure of both organic and ammonia nitrogen, and so at least equal to the amount of organic nitrogen.  In Slab Cabin Run, ammonia was 3.6 and Kjeldahl nitrogen was 50.0 (Spring Creek Survey 1959).  The assumed units for this data also is lb/d since in October 1960 at Thompson Spring the total amount of nitrogen, organic and inorganic, was 4.38 mg/l or 136.5 lb/d and at Bathgate Springs it was 2.25 mg/l or 12.1 lb/d (Metcalf and Eddy 1961).  Ammonia in 1960 in Thompson Run was at 5.4 mg/l and 8.1 mg/l, Kjeldahl nitrogen at 11.0 mg/l and the nitrate level was 0.27 mg/l (Table 3-7).  For Slab Cabin Run, ammonia was 0.4 mg/l above the confluence with Thompson Run and 6.2 mg/l below it.  Kjeldahl nitrogen was 0.8 mg/l above and 7.4 mg/l below the confluence.  Nitrate was 0.08 mg/l above and 0.29 mg/l below Thompson Run.  Bathgate Spring Run in the northwest corner of Millbrook Marsh near the bend in Puddintown Road, had an ammonia level of 0.21 mg/l, Kjeldahl nitrogen was 0.9 mg/l, and nitrate was 0.46 mg/l (Metcalf and Eddy 1961).  Thompson Run was tested in 1963 between 12:00 pm January 15 and 9:00 am January 16, 1963.  Nitrate levels were between 1.50 and 2.75 mg/l (Goodwin 1963).  In July 1978 Thompson Run levels of nitrate were 4.84 and 5.06.  Ammonia was 2.2 and 1.1.  In Slab Cabin Run, nitrate was 3.74 and ammonia was 0.07 (Miller 1979).  Units for the 1978 nitrogen data were not given.  Assumed units are mg/l.  In 1980, nitrate levels were moderate at 1.80 mg/l for PennDOT Stations 3 and 4.  Total nitrogen was 0.5 and 0.6 mg/l for Stations 3 and 4 respectively (PennDOT 1981). The October 1986 data showed nitrate levels at 0.1mg/l, 0.1 mg/l and 0.4 mg/l.  The most recent nitrogen data found was recorded in July 1989.  Nitrate in Slab Cabin Run was 1.92 mg/l and ammonia was 0.04 mg/l.  In Thompson Run, nitrate was 2.88 mg/l and ammonia 0.14 mg/l (Hughey 1990).

Table 3-5
Ammonia Levels in Millbrook Marsh Streams, 1959-1989
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run
January 1959 (Spring Creek Survey 1959) 3.6 lb/day   16 lb/day
October 1960 (Metcalf and Eddy 1961) 0.4 mg/l 6.2 mg/l  5.4 mg/l8.1 mg/l 0.21 mg/l
July 1978 (Miller 1979)   0.07 mg/l 2.2  mg/l1.1 mg/l
July 1989 (Hughey 1990)   0.04 mg/l 0.14 mg/l
 

 Table 3-6
Kjeldahl Nitrogen in Millbrook Marsh Streams and Springs, 1959-1980
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thomp-son Run Bathgate Spring Run Thompson Spring Bathgate Springs
January 1959 (Spring Creek Survey 1959) 50.0 lb/day   41.7 lb/day
October 1960 (Metcalf and Eddy 1961) 0.8 mg/l 7.4 mg/l  11.0 mg/l 0.9 mg/l 4.38 mg/l136.5 lb/day 2.25 mg/l12.1 lb/day
1980 (PennDOT 1981) 0.5 mg/l0.6 mg/l

Table 3-7
Nitrate Nitrogen in Millbrook Marsh Streams, 1960-1989
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run

October 1960 (Metcalf and Eddy 1961) 0.08 mg/l 0.29 mg/l  0.27 mg/l 0.46 mg/l
January 1963 (Goodwin 1963)    1.50 - 2.75 mg/l
July 1978 (Miller 1979)   3.74 mg/l 4.84 mg/l5.06 mg/l
1980 (PennDOT 1981) 1.80 mg/l
October 1986 (unpublished school report)   0.1 mg/l0.1 mg/l0.4 mg/l
July 1989 (Hughey 1990)   1.92 mg/l 2.88 mg/l
 

Phosphorus
 There was no maximum contamination level criteria regarding phosphorus for Pennsylvania CWF streams, nor were there maximum levels of phosphorus for Pennsylvania drinking water.
The 1960 surveys showed the level of total phosphorus in Thompson Run was 4.0 mg/l, Upper Slab Cabin Run was 0.4 mg/l, Lower Slab Cabin Run was 3.9 mg/l and Bathgate Spring Run was 0.06 mg/l (Table 3-8).  Ortho-phosphorus was 2.3 mg/l in Thompson Run, 0.09 mg/l in Upper Slab Cabin Run, 1.9 mg/l in Lower Slab Cabin Run, and 0.6 mg/l in Bathgate Spring Run (Metcalf and Eddy 1961) (Table 3-9).  In 1962, Thompson Run phosphate level was 3.1 mg/l, and Slab Cabin Run, below the confluence with Thompson Run, was 4.0 mg/l (Vrenna 1962) (Table 3-10).  The January 1963 level of phosphate in Thompson Run ranged from 4.4 to 9.7 mg/l (Goodwin 1963).    In 1980 ortho-phosphorus levels were below detectable limits (<0.01 mg/l) and total phosphorus was 0.8 mg/l at Station 3 and 0.10 mg/l at Station 4 (PennDOT 1981).  In 1986, the phosphate level of Slab Cabin Run directly after the Rte. 26 bridge was 0.4 mg/l, and undetected at the other two stations (unpublished school report).  On July 5-7, 1989, total phosphorus was 0.07 mg/l in Slab Cabin Run, and 0.06 mg/l in Thompson Run (Hughey 1990).
 Table 3-8
Total Phosphorus Levels in Millbrook Marsh Streams, 1960-1989
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run
October 1960 (Metcalf and Eddy 1961) 0.4 mg/l 3.9 mg/l  4.0 mg/l 0.06 mg/l
1980 (PennDOT 1981) 0.8 mg/l0.10 mg/l
July 1989 (Hughey 1990)   0.07 mg/l 0.06 mg/l

Table 3-9
Ortho-phosphorus Levels in Millbrook Marsh Streams, 1960-1980
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run
October 1960 (Metcalf and Eddy 1961) 0.09 mg/l 1.9 mg/l  2.3 mg/l 0.6 mg/l
1980 (PennDOT 1981) < 0.01 mg/l

Table 3-10
Phosphate Levels in Millbrook Marsh Streams, 1962-1986
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run
1962 (Vrenna 1962)  4.0 mg/l  3.1 mg/l
January 1963 (Goodwin 1963)    4.4 - 9.7 mg/l
October 1986 (unpublished school report) 0.4 mg/l
 

 Chloride
 The Pennsylvania water quality standards criteria for chloride was listed for 4 uses only, not including the critical use "Aquatic Life".  For the critical use of "Special Protection", a maximum of 150 mg/l chloride is given.  For the critical use "Water Supply, the maximum is 250 mg/l (Bureau of National Affairs 1994).  The other 2 uses were those that apply only to selected portions of the Delaware River Basin.
 The chemical testing of January 1963 showed a range of 15 to 22 mg/l chloride in Thompson Run (Goodwin 1963) (Table 3-11).  In July 1978, the chloride level in Thompson Run was 33 mg/l, and in Slab Cabin Run, chloride was 18 mg/l.  The 1980 chloride levels in Slab Cabin Run, were 5.8 mg/l at PennDOT Station 3 and 6.4 mg/l at PennDOT Station 4 (PennDOT 1981).  In July 1989, chloride in Thompson Run was 31 mg/l and in Slab Cabin Run it was 14 mg/l (Hughey 1990).
Table 3-11
Chloride Levels in Millbrook Marsh Streams, 1963-1989
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run
January 1963 (Goodwin 1963)    15 - 22 mg/l
July 1978 (Miller 1979)   18 mg/l 33 mg/l
1980 (PennDOT 1981) 5.8 mg/l6.4 mg/l
July 1989 (Hughey 1990)   14 mg/l 31 mg/l

Fecal Coliform
 As with chloride, there is no criteria for fecal coliform levels regarding aquatic life in Cold Water Fish streams.  For "Recreation (including aesthetics)", from May 1 through September 30, the maximum fecal coliform level  is a geometric mean of 200/100ml based on 5 consecutive samples from different days.  For the remainder of the year the level is the geometric mean of 2,000/100ml.  The criteria for use as a water supply is a maximum of 5,000/100ml as a monthly average value (Bureau of National Affairs 1994).
 In January 1959 fecal coliform count was 24,000 in Thompson Run at the Duck Pond Dam, 2,400 just below Route 545, and 240,000 in Slab Cabin Run (Spring Creek Surveys 1959) (Table 3-12).  The units, not given for the 1959 data, are assumed to be cts/100ml.  The 1980 levels of fecal coliform bacteria were very high, at 583 cts/100ml at PennDOT Station 3 and 572 cts/100ml at PennDOT Station 4 (PennDOT 1981).
Table 3-12
Fecal Coliform Levels in Millbrook Marsh Streams, 1959-1980
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run
January 1959 (Spring Creek Survey 1959)   240,000 cts/100ml 24,00 cts/100ml
1980 (PennDOT 1981) 583 cts/100ml572 cts/100ml

pH
 There are 2 criteria given for pH for CWF streams for the protection of aquatic life.  The pH should be from 6.0 to 9.0 inclusive or from 7.0 to 9.0 inclusive (Bureau of National Affairs 1994).
 A 1941 stream survey recorded the pH at Thompson Run at 4 different locations with regard to the sewage effluent input pipe (Donley 1941).  The pH above the effluent pipe was 7.9, at the mouth of the pipe it was 7.15, below the pipe and also at the mouth of the stream it was 7.8 (Table 3-13).  In January 1959, the field determined pH of Thompson Run was 8.0 below Route 545 and Slab Cabin Run was 7.4 (Spring Creek Surveys 1959).  In 1960, Thompson Spring was 7.7, Thompson Run after the Duck Pond was 8.2, and below Route 545 the pH was 7.8.  The pH of Slab Cabin Run below Route 545 was 8.3 and Bathgate Spring Run was 7.7 (Metcalf and Eddy 1961).  In July 1962, Thompson Run pH was 7.7 and Slab Cabin Run was 7.6 (Vrenna 1962).  July 1978 pH of Thompson Run just after Route 545 was 8.0, and was 7.8 just before Slab Cabin Run.  Slab Cabin Run pH was 8.4 (Miller 1979).  In 1980 the pH at PennDOT Station 3 was 5.8 and at PennDOT Station 4 it was 6.0 (PennDOT 1981).  The pH of the first 19 m of Thompson Run was 7.2 in 1983 and 1984 (Miller 1985).  In an unpublished elementary school project from the files of the Pennsylvania Fish and Boat Commission, the 1986 pH of Slab Cabin Run measured 7.5 at the station closest to Puddintown Road, and 8 at the other two stations.  In November and December of 1987, the pH of Thompson Run was 7.9 and 8.3, and Slab Cabin Run was 8.3 and 8.7 (Bureau of Water Quality Management 1989).  Most recently, July 1989 pH of Thompson Run was 7.7 and Slab Cabin Run was 7.4 with 7.6 determined in the field (Hughey 1990). In the summer of 1995, the pH of the calcareous fen water was 7.6, 7.5, and 7.2 at 3 locations within the fen area (Western Pennsylvania Conservancy 1995).
Table 3-13
pH of  Millbrook Marsh Streams and Springs, 1941-1989
 Upper Slab Cabin Run Lower Slab Cabin Run Slab Cabin Run - reach not specified Thompson Run Bathgate Spring Run Thompson Spring Fen Springs
April 1941 (Donley 1941)    7.97.157.8
January 1959 (Spring Creek Survey 1959)   7.4 8.0
October 1960 (Metcalf and Eddy 1961) 8.3 7.8  8.27.8 7.7 7.7
1962 (Vrenna 1962)   7.6 7.7
July 1978 (Miller 1979)   8.4 8.07.8
1980 (PennDOT 1981) 5.86.0
October 1986 (unpublished school report) 8.0 7.5
November and December 1987 (Bureau of Water Quality Management 1989)   8.38.7 7.98.3
July 1989 (Hughey 1990)   7.47.6 7.7
Summer 1995 (Western Pennsylvania Conservancy       7.67.57.2

Other Contaminants
 The only available sodium data is from 1980 at which time the level was 5.2mg/l at PennDOT Station 3 and 5.3 mg/l at PennDOT Station 4.  Petroleum products, oil and grease, were very high at 4.2 mg/l and 3.9 mg/l (PennDOT 1981).  In 1987, abnormally high levels of total dissolved solids, metals, and phenolics were found in discharges into Thompson Run (Bureau of Water Quality Management 1989).

 Water quality in Thompson Run and Slab Cabin Run was greatly compromised by the input of the Pennsylvania State University Sewage Treatment Plant effluent and stormwater from about 1913 until the cessation of the sewage effluent input in 1983.  By 1980, Thompson Run was receiving four-fifths of State College’s treated sewage effluent (PennDOT 1981).  The 1961 report to the Spring Creek Committee by Metcalf and Eddy Engineers confirmed the fact that the effluent discharge into Thompson Run caused general deterioration of Spring Creek.  Thompson Run and Slab Cabin Run showed significant quantities of BOD, nitrogen, and phosphorus, and had profuse aquatic plant growths in some reaches.  This stemmed from the sewage effluent input (Metcalf and Eddy 1961).  The water quality of Thompson Run, Slab Cabin Run, Spring Creek, the Pennsylvania State University Sewage Treatment Plant and the University Area Joint Authority Sewage Treatment Plant was investigated again in 1978.  The report concluded that in general, the 3 streams had improved chemically since the 1960’s, reflected by lower levels of BOD, ammonia-nitrogen and phosphate, probably because of improved treatment at the Pennsylvania State University Sewage Treatment Plant, but that specifically, Thompson Run had not improved much.  Biologically the condition was similar to that of 1959 (Miller 1979).
 The PennDOT report concluded that concentrations such as those of oil and grease, phosphorus, and low pH could affect survival of some aquatic species if the levels remained for extended time periods (PennDOT 1981).  Thompson and Slab Cabin Runs continue to receive increasing amounts of stormwater runoff from roads and the surrounding urban areas and neighborhoods.
Numerous fish kills and other contamination events have been reported to the Pennsylvania Fish Commission and the Pennsylvania Department of Environmental Resources.  On April 14, 1972, an investigation was initiated on a fish kill in Thompson Run and Slab Cabin Run to their confluence with Spring Creek.  A Centre County Waterways Patrolman, Paul Antolosky, reported that 200-300 minnows, 3 trout, a few goldfish, and 140 suckers were killed.  "The pollution probably was caused by some substance in the storm sewers (rather than PSU STP effluent)", but though water samples were taken and analyses made of pH and metals, the cause of the kill was undiscovered (Hemmerly 1972, p. 1).  Another report of an acute pollution incident occurred on August 31, 1990, which resulted in a fish kill in Thompson Run immediately downstream of the Duck Pond to the area adjacent to the Route 26 bridge.  The source of the pollution was determined to be the White Building Swimming Pool of the University (Burger 1990).  There was some residual chlorine odor in the affected area of Thompson Run, and evidently the contamination was carried into Thompson Run via the stormwater outlets.  These are examples of reported contaminations; undoubtedly other incidents occur without investigation.

Stream Morphology, Habitat, and Discharge
 In 1997, Slab Cabin Run  averaged 5.5 m (18.1 ft) wide and 0.3 m (1.0 ft) deep.  It consisted of approximately equal proportions of riffle, glide, and pool habitats (Table 3-14).  Only 36% of the bank was rated as stable.  Gravel was the predominant substrate and silt ranked second (Brooks et al. 1998).  The earliest data available for Slab Cabin Run is the "Commonwealth of Pennsylvania Board of Fish Commissioners Stream Survey Report" of April 7, 1941.  The "lower 2/3 mile" (1073 m) from the Thompson Run confluence to Spring Creek had a sandy bottom and scarce aquatic vegetation (Donley 1941).  Slab Cabin Run was included in the study area for a thesis on the ecology of the Muskrat on Spring Creek (Smith 1954).  In 1953, it was a narrow, shallow stream with a comparatively silt-free bottom until after the junction with "Willow Run", at which point it became much wider, silt laden, and 0.6 - 1.2 m deep.  Other statements within the report, such as the description of the cattail-sedge marsh just above the confluence of Willow and Slab Cabin Runs, suggest that "Willow Run" is actually Thompson Run.  The PennDOT study of 1980 describes Slab Cabin Run as 4.8 m (15.75 ft) wide and 0.76 m (2.49 ft) deep.  The stream’s 0.6 to 1.2 m (1.96 to 3.94 ft) banks were vertical in 1980 and the substrate was a rock and gravel dominated stream bed (PennDOT 1981).  In 1987, the substrate of Slab Cabin Run in Millbrook Marsh downstream of the confluence with Thompson Run was found to be "quite silted" (Bureau of Water Quality Management 1989).  By 1989, the riffle substrate in Slab Cabin Run about 800 m above Thompson Run was mostly gravel mixed with sand and boulders and the rocks were about 30% embedded (Hughey 1990) (Figure 2-4, 3-5, 3-6).
 In 1997, Thompson Run averaged 4.8 m wide and 0.3 m deep.  Riffles made up 55% of the instream habitat and 88% of the bank was rated as stable.  Gravel was the predominant substrate and cobble ranked second (Table 3-14) (Brooks et al. 1998).  The earliest data for Thompson Run is also a Stream Survey Report, from April 8, 1941.  The stream bottom had a fine gravel to sandy substrate and abundant aquatic vegetation (Donley 1941).  In 1989, Hollender described Thompson Run as 300 m long, 4.4 m (14.4) wide, with a rubble and gravel substrate (Hollender 1989).  Also in 1989, Hughey (1990) described the riffle substrate as mostly sand and fine gravel that "looked as if it had washed into the stream from the shoulder of PA 26" with the large rocks being about 40% embedded.
Table 3-14.
Summary of habitat composition in reaches of Slab Cabin Run and Thompson Run within Millbrook Marsh (Brooks et al. 1998)
       % substrate composition
 %riffle %glide %pool % stablebank %bankfishcover %instreamfishcover boulder cobble gravel sand silt
Slab Cabin Run :
 27 38 35 36 32 21 6 11 42 10 31
Thompson Run
 55 35 10 88 43 17 4 26 63 2 5
 

 The high percent of  unstable banks in Slab Cabin Run found in 1997 are evidently due to high stream flows attributable to stormwater runoff (Brooks et al. 1998).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Figure 3-5  Gravel bar formation in Thompson Run.  July 1998.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Figure 3-6  Gravel bar formation in Thompson Run.  July 1998.
 One reach of Slab Cabin Run, sampled in 1987, had an inordinate amount of silt as a result of extensive erosion from the mid-reaches of Slab Cabin Run.  The heavily silted portion is also below the confluence with Thompson Run (Bureau of Water Quality Management 1989) which most likely contributed a substantial amount of sediments.  Slab Cabin Run has a higher proportion of silt than Thompson Run.  This is possibly due to a combination of the high sediment load from upstream areas in Slab Cabin Run and the higher water velocity in Thompson Run. With the velocity decreasing after entering the wider Slab Cabin Run, sediments can then settle out of the water.  Bar formation is evident below the confluence (Figure 3-7, 3-8).
 In June 1997 the discharge of Slab Cabin Run leaving the marsh was 0.608 cms and about 40% of this flow, 0.243 cms was from upper Slab Cabin Run (Table 3-15).  Thompson Run discharge was 0.292 cms.  Most of the flow in Thompson Run, which contributed about 48% of the flow leaving the marsh, originates from Thompson Spring.  The remaining 7% of flow, 0.045 cms, came from Bathgate Spring and 5%, 0.028 cms, originated from smaller springs that were not measured.  In October 1997, following a dry summer, the discharge of Slab Cabin Run leaving the marsh was 0.274 cms and only 13% of this water, 0.035 cms was contributed by upper Slab Cabin Run.  Thompson Run contributed 66% of the flow (0.180 cms) and the remaining .059 cms, 21%, was provided by Bathgate Spring and other unmeasured springs (Brooks et al. 1998).
 In July 1959, McDonnell reported the discharge of Thompson Run as 0.246 cms.  Stream discharge data from October 1959 showed that the discharge of Slab Cabin Run leaving the marsh was 0.303 cms.  According to McDonnell’s data, Upper Slab Cabin
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Figure 3-7  Confluence of Slab Cabin Run (left) and Thompson Run (right), looking upstream.  July 1998.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Figure 3-8  Slab Cabin Run just after Thompson Run confluence.  Degraded streambank is opposite the confluence.  July 1998.
 Run contributed 0.082 cms and Thompson Run, 0.221cms (McDonnell 1960).  Stream discharge was measured in some areas and calculated in others.  The resulting data does not include Bathgate Springs or other smaller springs within Millbrook Marsh.
 The discharge data illustrates the importance of ground water in maintaining the aquatic communities in the marsh.  The Bathgate and smaller springs in Millbrook Marsh seemed to discharge fairly consistent amounts over the course of the 1997 growing season, for the most part independent of precipitation.  In October after the dry summer, Bathgate Springs were emitting 73% and smaller springs were still emitting 92% of what they did in June (Table 3-15).  Thompson Spring also remained fairly constant, especially compared to Slab Cabin Run.  The flow in October of Thompson Run was 61% of the June flow, where Slab Cabin Run flow was only 14% of the June flow.  Slab Cabin Run appeared to J. H. Clark (1965) as an effluent, or gaining stream only from the south edge of Lemont to the confluence with Spring Creek which includes the reaches within Millbrook Marsh.  An effluent stream is one sustained by the consistent yield of springs.  Slab Cabin Run has groundwater discharging into it along that length that it is an effluent stream.  Above that point in Lemont, though, it is both a discharge and recharge of groundwater.  At some points it disappears into sinkholes and rises again at other points.  Along those lengths it alternates between a water-table effluent stream and an underdrained influent stream, depending on the season.  Slab Cabin Run in those reaches above Lemont can go dry during times of drought.  The flow is greatly decreased during drought times because of the lack of flow upstream.  Within Millbrook Marsh it will not stop completely since it is an effluent stream in that area.
 
 
 
 

Table 3-15
1997 Millbrook Marsh Streams and Springs Discharge

June

October

	cms	% of total discharge	cms	% of total discharge	% of June discharge
ThompsonRun,Route26	0.292	48	0.180	66	61
SlabCabinRun,Route26	0.243	40	0.035	13	14
ConfluenceofBathgateSprings	0.045	7	0.033	12	73
OtherSprings	0.028	5	0.026	9	92
SlabCabinRun,PuddintownRoad(TotalMMdischarge)	0.608	100	0.274	100	45

 Thompson Run, on the other hand, is an effluent stream along the entire length.  It is fed by the sustained yield of springs, mainly Thompson Spring, and therefore, does not decrease as Slab Cabin Run does.  Though the discharge from Thompson Spring is great, about half of Bellefonte’s famous Big Spring, the historical data shows a possibility of a gradually decreasing discharge.  Discharge in 1941 was measured at 0.247 cms, in 1944 at 0.237 cms, in 1965 at 0.219 cms, in 1971 at 0.170 cms, and then in 1985 at 0.130 to 0.192 cms (PA Fish Commission 1941, Clark 1965, Wood 1980, Miller 1985)).  The historical data is limited, though, and more in-depth research and consistent monitoring is necessary for any conclusive statement regarding Thompson Spring.

Discussion
 Basically, the morphology of Millbrook Marsh streams, with the exception of the two branches of Bathgate Spring Run, has been and is still being negatively affected by stormwater input.  High velocity, excessive amounts and erratic input of stormwater have caused streambank erosion and high sedimentation rates on the stream bottoms.  The results of the excessive stormwater input are wide reaching.  High siltation causes habitat degradation for benthic macroinvertebrates, resulting in lowered species diversity and species richness (Chapter 6).  Benthic macroinvertebrates are part of the food supply for fish, amphibians, reptiles, some birds and mammals, so the stormwater input indirectly affects the populations of these species as well.  In addition, herbaceous riparian vegetation has decreased or been eradicated, as seen in some reaches of Slab Cabin Run and Thompson Run.  Another possible effect is the lowering of the water table, which affects vegetation types in the marsh as a whole.  Essentially, the stormwater entering Millbrook Marsh has most likely been the initial cause of many of the changes within the system.
 The parameters of water quality that are generally affected by sewage treatment plant (STP) effluent have improved since 1983.  The sporadic fish kills that occurred before 1983 were due to either STP effluent or stormwater.  Fish kills have occurred after 1983 due to stormwater alone.  The stormwater input is still affecting water quality.  Water quality testing must be done during and after storm events to gather enough data for comparison to historical data, and to determine whether levels of contaminants are below the acceptable limits for the protection of aquatic life.  Historically, dissolved oxygen levels have varied.  Only in June 1962 was the level below the minimum 5.0 mg/l in Slab Cabin Run, but Thompson Run was close to that minimum in October and November 1960.  There is no data available for Thompson Run or Lower Slab Cabin Run after the STP effluent ceased in Thompson Run.  Water samples that include stormwater from Thompson Run should be tested for dissolved oxygen levels to enable comparison.  Turbidity data is sparse, but the high level of turbidity in 1980 in Upper Slab Cabin Run reflects stormwater input.  Another point is that JCU’s are inaccurate under a turbidity level of 25 though the 1978 report shows 5 JCU and 6 JCU.  It was not above the acceptable limits for CWF aquatic life.  There is not enough Kjeldahl nitrogen data for comparison.  The ammonia nitrogen data is sparse, but levels seem to have decreased from 1959 to 1989 in both Thompson and Slab Cabin Runs.  Nitrate levels, on the other hand, have increased in Thompson Run from 1960 to 1989.  The lower pH levels in Millbrook Marsh streams could be reflective of rain before the water sample was taken, but this is not indicated except in the case of the 1980 PennDOT data.  The levels of pH were very low, due to the high amount of rainwater in Slab Cabin Run.  The pH at that time was more acidic than allowed for a CWF stream with the designated use of protection of aquatic life.  The level of phosphorus was much lower in 1989 than in 1960, and the levels of phosphate and orthophosphorus were very high in the 1960’s as well.  Phosphorous levels were most likely particularly high at that time because of much wider use then than now.  The high phosphorus levels in Thompson Run and Lower Slab Cabin Run were probably due mainly to the STP effluent input.  The only phosphorus data available for Thompson or Lower Slab Cabin Run is from July 1989, and reflects a significantly lower total phosphorus level.  Chloride levels were not exceeded for water bodies designated for recreational use or use as a water supply.  More testing should result in a better data set for comparisons of STP effluent influence and stormwater influence on water quality in Millbrook streams, though it is evident in the water quality data gathered after 1983 that the streams have improved since the cessation of treated sewage effluent input.  Continued and increasing stormwater input is also very clearly affecting the water quality and stream conditions at Millbrook, and therefore, affecting the flora and fauna of the marsh.  This is apparent when looking at the 1980 PennDOT report, since the sampling stations were on Slab Cabin Run above the confluence with Thompson Run, and so did not include sewage effluent.  The data was gathered after 2 days of rain in State College. The stormwater input had significant impact upon the results.  The high turbidity levels were due to being sampled immediately after the rain, before the stream had time to recover from the sediment load from runoff.  Chloride levels were probably influenced by proximity to State College.  Sodium levels were affected by the use of road salts in winter, transported into Slab Cabin Run with stormwater runoff.  The very high fecal coliform counts, substantially higher than the state and Federal standards for bathing water at 200 cts/100ml, were attributed to runoff from parts of State College, Dalevue and Lemont neighborhoods, septic tank leachates, and stormwater runoff from the nearby streets and pastures during the previous rain.  The stormwater runoff also carried petroleum products from the roads and caused the very high levels of oil and grease in the stream.  The unusually low pH of the streams, 5.8 and 6.0 instead of 7.6 to 8.7 in Slab Cabin Run, was most likely caused by large amounts of acid rainwater added to the streams.  Pennsylvania receives the largest amount of acidic precipitation in the United States, with an average pH of almost 4 (Thorne et al. 1995).  In 1980, acid rain in State College was usually between 3.5 and 5.0 (PennDOT 1981).  Ten years ago, the pH of State College area rain was usually about 4.5.  This has declined to a pH that is commonly now about 3.5 (S. McGonigal, pers. comm.).
Phosphorus is contributed not only by the sewage effluent, but also by the stormwater from agricultural and residential lands and the use of fertilizers and other chemicals, animal wastes, litter, and decaying vegetation.  The data from 1980 included only stations located on Slab Cabin Run above its confluence with Thompson Run, and therefore, the high levels of contaminants resulted from field and street runoff from Dalevue, Lemont, parts of State College, and the associated on-lot sewage facilities (septic systems) there.  Consequently, the levels in the 1980 data are representative of stormwater runoff, rather than the sewage treatment plant effluent.
Levels of contaminants in Thompson Run and Slab Cabin Run below the confluence were not tested after a period of rain, but it would be reasonable to believe that levels would be higher than those of Slab Cabin Run above the confluence due to the greater amount of urban drainage area stormwater coming from State College borough and the Penn State campus.  Discontinuation of sewage effluent input lowered the phosphorus levels of Thompson Run and downstream waters, but some phosphorus loading is most likely occurring from the stormwater input, as seen in the 1980 phosphorus data for Slab Cabin Run from the PennDOT study.
.  The number of stormwater inputs has increased since 1980, as has the amount of developed land, roads, other impervious surfaces, and hence, the total stormwater amount has increased as well.  Levels of turbidity, total suspended and dissolved solids, fecal coliform bacteria, contaminants associated with fertilizers and pesticides, oil and grease and other road surface materials and low pH will continue to increase with increases in stormwater input, and possibly resulting periodic fish kills and water quality impairment from accidental spills or other sporadic contamination from stormwater, as has happened in the past.