Pollution UpDate

For the past year, the Raymond Proffitt Foundation has had a biologist investigating streams affected by longwall mining in southwest Pennsylvania. According to letters they sent us in March, about Enlow Fork, the Office of Surface Mining claims that their studies show that longwall mining has not caused "significant environmental harm" to streams. That phrase has a regulatory meaning that we have not been able to decipher–and it doesn’t seem that OSM knows, either. Why not? OSM has never found, in the entire history of longwall mining in Pennsylvania, that any longwall mining has caused "significant environmental harm."

We received another letter from OSM this week regarding Laurel Run. This one says that, after they conducted a "thorough review of the statutory and regulatory provisions of the approved state program," OSM concluded that "stream dewatering is an anticipated and allowable impact of underground mining." Seeing how Act 54 says that DEP must follow PA’s water quality standards, we are puzzled how dewatering our waters is "anticipated and allowable."

Keep in mind that longwall mining has dried up hundreds of springs that provided clean, cold water to landowners and larger streams; dried up sections of larger streams; and subsided thousands of feet of streams, such as Enlow Fork, that then fill with sediment. Sediment is the number one pollutant in our streams. It smothers the aquatic insects that fish depend on for food, it eliminates the habitat for fish that required clean, stony bottom streams, and changes the fish community to one that thrives in this polluted system. According to OSM, none of these major impacts to our streams rises to the level of "significant environmental harm;" in fact, they are "anticipated and allowable."

It has been clear for some time that DEP is not doing their job. The letters from OSM indicate they are not doing theirs, either. Our reply to OSM’s March letter is below. We are preparing a reply to their letter on Laurel Run.

Existing data

The following is a review of existing data and OSM’s response to RPF. Our analysis shows that significant, imminent environmental harm has occurred on Enlow Fork

Consol data

Consol’s data from Enlow Fork was in the DEP files and could easily have been reviewed by OSM. Data collected by CEC in 1998 and 1999 indicate that longwall mining has affected Enlow Fork. CEC collected pre-mining data on panels 8C, 9C, and 10C, and post-mining data on the 6C panel (Mining had already occurred in panel 6C when they conducted their study). When CEC re-sampled in 1999, only site E-05, situated in panel 8C, had been undermined. We do not have access to their latest data to evaluate impacts from mining panels 9C and 10C.

CEC’s report shows aquatic habitat has changed as a result of longwall mining. For example, prior to mining, Station E-05 consisted of 360 feet of pool habitat, 180 feet of riffle habitat, and 60 feet of run habitat. CEC sampled this 600 ft. reach for fish in October 1998. After mining, Station E-05 (sampled by CEC in May 1999) consisted entirely of pool habitat. Apparently, the station limits were changed somewhat to accommodate the new habitat configuration, as Station E-05 is shorter and wider in May 1999 than it was in October 1998.

This change in habitat has also changed the fish assemblage. Before mining, CEC collected 189 fish (0.1 fish per second of electrofishing time), compared with 26 fish after subsidence (.01 fish per second of electrofishing time) (Table 1). Not only is there a tenfold decrease in the catch per unit effort of fish at this station, but riffle dwellers such as darters and stonerollers have been greatly reduced, and smallmouth bass have been eliminated. There is not enough data to ascertain if this change in the assemblage is permanent. Still, at this point in our evaluation of longwall impacts to stream habitat and the fish assemblage, we must conclude that longwall mining subsidence is deleterious to fish and their habitat on Enlow Fork.

Table 1. Fish captured by CEC in panel 8C, Station E05, on Enlow Fork, before and after longwall mining.

	Date	Oct-98	May-99
	Fishing Time (minutes)	31.0	38.7
	Length	600	500
	Width	48	65
	Area	28800	32500
Common Name	Scientific Name	Count	Count
central stoneroller	C. anomalum	89	3
hornyhead chub	N. biguttatus	1	0
sand shiner	N. stramineus	2	0
bluntnose minnow	P. notatus	6	1
white sucker	C. commersoni	10	0
northern hogsucker	H. nigricans	17	1
golden redhorse	M. erythrurum	0	6
stonecat	N. flavus	3	5
rock bass	A. rupestris	15	3
smallmouth bass	M. dolomieui	10	0
greenside darter	E. blennioides	21	2
rainbow darter	E. caeruleum	15	3
fantail darter	E. flabellare	0	2
	Total	189	26

Although data are not available for pre-mining conditions at Station E-01 (panel 6C), CEC’s fish sampling indicates a similar fish assemblage to post-mining conditions at Station E-05. In October 1998, CEC collected thirty-six fish in a 600-foot reach at Station E-01. They did not collect any darters, and only 2 stonerollers. In May 1999, results were even more dismal; only 26 fish were captured (to bolster efficiency, CEC used both a boat and backpack electrofishing on this survey). Although there are no pre-mining data for Station E-01, the data are suspiciously similar to the post-mining sample of E-05. The pattern is clear: when longwall mining subsides a reach of stream, and it fills with sediment, we find damage to the fish assemblage.

USEPA data

In response to a request from FWS, USEPA Region III Aquatic Biology Group conducted a field study of the effects of longwall mining subsidence on stream habitat with FWS. USEPA sampled two reference sites un-impacted by longwall mining: Dunk-1 (Dunkard Fork – unaffected by mining) and Enlow –1 (downstream of longwall mining activity on Enlow Fork), and two sites within longwall panels: Enlow-2 and Enlow –3 (panels 6C and 7C, respectively). USEPA used USEPA’s Environmental Monitoring and Assessment Program (EMAP) protocol to sample the physical habitat within these reaches.

USEPA found longwall mine subsidence zones had significantly more pool habitat than the reference reaches. They also found a significantly higher percentage of fines and sand in the subsided reaches. For example, the two subsided pools (Enlow 2 and 3) had 43% and 32%, respectively, of the bottom substrate composed of fine grain material (less than 2mm in diameter). In contrast, the reference sites (Dunkard-1 and Enlow-1) had only 3% and 18%, respectively, of the substrate composed of fine grain material. Channels within subsided reaches were also less stable, based on the Relative Bed Stability Index (an index developed by USEPA to measure bed stability). Based on these findings, and sampling by CEC, PFBC, and FWS, it is likely the aquatic macroinvertebrate fauna, and fish community structure within the subsided reaches has been adversely affected.

OSM investigation:

On Site Review for NOI

Page 2 of the report prepared by Inspector #50 states that “[n]o landslides in the stream were noted.” There has been a significant hillside slump along Enlow Fork that has filled the channel with fallen trees and caused a major erosion problem downstream from the 9C panel. This slumping began in 2000, and has gotten substantially worse. About 250 feet (laterally) of the hillside has fallen into the stream, causing trees to divert the flow into the opposite bank, causing more erosion. According to DEP, this slide began after the 10C panel caused a spring to migrate laterally.

Assessment of the Hydrologic and Biologic (sic) Impacts of Longwall Mining…

It is difficult to assess OSM’s data interpretation since the report does not actually present any data. The report indicates that leaf-packs and substrate were sampled using D-frame nets. There is no information as to what proportion of the sample is leaf-pack and what is actual substrate. We would not expect the macroinvertebrate assemblage of leaf packs to exhibit changes due to habitat changes. We hypothesize that there would be differences in the macroinvertebrate fauna in cobble/gravel substrate of unsubsided pools compared to the fauna of silty subsided pools. No effort was made by OSM personnel to analytically block their sampling effort to test any such hypothesis. OSM should have compared the benthic community in pools, in unsubsided reaches and unsubsided reaches.

The observation by OSM that “[p]anels 7C through 10C did not exhibit any observable pooling that appeared chiefly attributable to subsidence.” is an indication that OSM did not even analyze data collected by CEC on Enlow Fork that could have been found in DEP’s files, that OSM supposedly reviewed. Even the mining company’s consultants admit mining has caused pooling in those reaches. There has been significant pooling in panels 7C through 10C, and data collected by CEC and others documents this. OSM’s cursory analysis of the habitat and failure to examine studies conducted on this stream has caused you to mischaracterize the damage to this stream

OSM also contends that accruing sediment in the “[f]ew areas of enhanced pooling” is “[s]imilar to what occurs in naturally-pooled sections”. This can only be a misunderstanding of stream geomorphology. Pools are erosional features and do not naturally fill with the amount of silt that has filled most of the subsided reaches. That is why they are pools, - the erosive power of the stream excavates these areas. Although silt accumulates behind beaver dams, it does so because flows are blocked, similar to the function that the gates perform. Moreover, they are usually eliminated naturally by high flows on large streams, such as Enlow Fork. High flows are not likely to eliminate the gates that are acting as dams on Enlow Fork because the gradient on Enlow Fork is fairly flat, and much of the stream bottom is controlled by bedrock. Although we would agree that some fine sediment settles out in pools, the sediment deposition that is occurring on Enlow Fork, Rocky Run, Templeton Fork, Robinson Fork, Hoovers Run, Roberts Run, and other subsided streams can be measured in feet, not millimeters.

Your comment that “[s]treams evolve over geologic time” seems to imply that the habitat on Enlow Fork will eventually be restored through natural processes. This is simply not true. When a stream subsides as a result of longwall mining, the gravel/cobble/boulder substrate is covered with a thick layer of fine grain sediment. Our consultant has observed this same phenomenon in all the other streams that he has studied. Although the subsided reach will eventually fill with sediment and restore the channel to some semblance of its original width and depth, the substrate will be permanently lost. So will the aquatic community that was intolerant of silt-laden substrates. Your claim that streams evolve is similar to the arguments made in attacks on the Endangered Species Act. Since species have gone extinct naturally, what’s the big deal if we knock off a few more? Like endangered species, the streams that longwall mining is adversely affecting through subsidence are being changed by a process that may occur over geologic time – not in the blink of an eye.

We would not expect water chemistry in general to be affected by longwall mining. However, conductivity has substantially increased due to the permitted mine discharge on Enlow Fork. Conductivity has increased from 2 to 4.5 times since measured by Dr. Cooper in the 1970’s (Table 3). OSM found a trend of decreasing specific conductance on Enlow Fork, and attributed it to “[i]ncreased base flow and inflow from tributaries.” We have not consistently seen this trend on Enlow Fork. Data collected by CEC in July of 1999 indicated a slight drop in conductivity (22%) but nothing like the drop in conductivity reported by OSM.

class=Section2>

Table 3. Stream conductivity measured on Enlow Fork by Cooper in 1976 and CEC in recent studies of Enlow Fork. * indicates an average of riffle and pool measurements for that station.

Collector

Cooper

CEC

Station

E01

E02

E04

E05

E06

E01

E02

E04

E05

E06

E01

E02

E02B

E04

E05

E05B

E06

Panel

10C*

9C*

8C*

10C*

9C*

10C

Date

Jul-76

Oct-98

May-99

Jul-99

302

300

1008

896

969

989

1119

728

690

697

738

717

1074

1174

1157

1165

1217

1278

1380

class=Section4>

Page 2 of cover letter– You state that “Although very little change was observed above the undermined sections of Enlow Fork…” We do not understand how you can make any statement about whether there were any changes since you did not conduct any studies, and most of the stream was ice covered. The ice was thick enough that one could walk on ice the entire length of the subsided reach at the 6C panel. This is your Way Point 12, where you supposedly made your observations. This is the same reach that EPA studied and found that fines (sand and silt) comprised 42% of the bottom composition. This is an abnormal situation, but typical of the impacts associated with stream subsidence.

Report by Inspector #50, page 2 – We are not sure why you would be reviewing “[t]he surface area adjacent (our emphasis) to the Enlow Fork, Templeton Fork, and Rocky Run streams…” as a way of determining if “[s]ignificant environmental harm had occurred or was now occurring to these stream areas.” In order to determine if impacts to the subsided reaches had occurred, OSM should have either sampled the fish, macroinvertebrates, or substrate in the subsided reaches and compared them to unsubsided reaches on the same stream, or a reference stream, similar to the work that EPA and FWS has done. It is your unscientific procedure that has caused you to erroneously conclude that environmental harm has not occurred.

Report by Inspector #50, page 2 – As previously pointed out, the hillside downstream from the 9C panel began slumping into Enlow Fork in 2000, and continued into 2001. Perhaps your staff did not walk downstream of the 9C panel.

Page 5, third paragraph – The report states, “Two samples were taken in riffles, two in pools, and one in a glide.” We would appreciate your informing us exactly where these samples were taken (way points would be sufficient). The samples taken in the riffles were likely a waste of time since these areas are in the gates, an area unaffected by mining. The glide area you sampled was also a gate, and was a shallow riffle before Consol was permitted to excavate down into the riffle, and then dump the material off to the side, thereby eliminating at least 50% of the riffle (You state that Consol “[g]raded the stream bed between the gates…” when, in fact, Consol excavated the gate between the 9C and 10C panel). So, three of your five samples appear to be taken from areas that we would not have sampled if we were trying to determine if subsidence affects the benthic community. Information from the pool samples would be useful if you compared the benthic community in pools located in subsided and unsubsided reaches.

Page 6, third line – Your statement that “[t]here appears to be no ‘significant environmental harm’ to the benthic zone of the reaches of stream affected by longwall mining” is not supported by your “studies.” Three of the five sites were in reaches unaffected by subsidence. The issue is not whether there is an “adequate benthic community,” whatever that means, but whether significant environmental harm has occurred.

Page 7, first paragraph – Your statement that “[s]ections for panels 7C through 10C did not exhibit any observable pooling that appeared chiefly attributable to subsidence” is especially unbelievable. Even the coal company acknowledged that longwall mining has caused the stream to subside in those reaches. In fact, they cut through the gate at the downstream end of the 9C and 10C panel in an effort to drain the pool. Everyone, including DEP, EPA, FWS, PFBC, and the coal company acknowledges that the subsidence is due to longwall mining. You are the only agency that has made that insupportable statement.

Page 7, last paragraph – We do not agree with your assertion that “There is no mechanism by which pooling will cause increased bank erosion.” You are correct in stating that “[d]eeper water and larger cross-sectional area of pools will reduce the stream velocity for that area.” Therein lies the problem. At normal stream widths and depths, Enlow Fork was able to transport sediment. Very little settled out in the channel. EPA’s study that compared subsided and unsubsided reaches on Dunkard and Enlow Forks proved this. Under the new conditions, as high water, carrying large amounts of sediment enters the deeper, wider channel, the velocity slows, causing sediment to settle out. In a straight reach, such as that on the 6C panel, the velocity is greatest in the center of the channel, which is where the largest size particles are carried. As the velocity slows, sediment, leaves, and branches settle out in the middle of the channel.

Eventually, a mid-channel bar will form, which will split the flow. The stream’s erosive power will now be directed to the adjacent banks, thus exacerbating bank erosion.

Cooper, E.L., C.C. Wagner, and W.G. Kimmel. 1976. Biological survey of selected locations in the Wheeling Creek watershed Greene and Washington counties, Pennsylvania and Marshall County, West Virginia. USDA Soil Conservation Service Contract Number AG42scs-00423, Harrisburg, PA.

Civil and Environmental Consultants, Inc. (CEC). 1999a. Enlow Fork Fall 1998 biomonitoring report for Consol Pennsylvania Coal Company. Bailey Mine Expansion Area. Civil and Environmental Consultants, Inc. Pittsburgh, PA.

Civil and Environmental Consultants, Inc. (CEC). 1999b. Preliminary assessment of water quality and fish monitoring data for Enlow Fork – Spring 1999, for Consol Pennsylvania Coal Company. Bailey Mine Expansion Area. Civil and Environmental Consultants, Inc. Pittsburgh, PA.

Civil and Environmental Consultants, Inc. (CEC). 1999c. Comparison of fish use of unmined pools and mined pools in Enlow Fork during the Spring 1998 drought, for Consol Pennsylvania Coal Company. Bailey Mine Expansion Area. Civil and Environmental Consultants, Inc. Pittsburgh, PA.