CBP/TRS 129/94 Contaminant Trace Element Loads at the Susquehanna River Fall Line during the Spring, 1993 High Flow Event Addendum to the Fall Line Toxics 1992 Final Report Chesapeake Bay Program Of recycltd paptr ------- Contaminant Trace Element Loads at the Susquehanna River Fall Line during the Spring, 1993 High Flow Event Addendum to the Fall Line Toxics 1992 Final Report Printed by the U.S. Environmental Protection Agency for the Chesapeake Bay Program ------- Contaminant Trace Element Loads at the Susquehanna River Fall Line during the Spring, 1993, High Flow Event Addendum to the Fall Line Toxics 1992 Final Report prepared by: the Maryland Department of Environment and: the U.S. Geological Survey, Department of the Interior for the: IJSEPA Chesapeake Bay Program Office INTRODUCTION The largest freshwater discharge event on record to the Chesapeake Bay occurred in the spring of 1993 as a result of several climatological factors. In 1993, as much as 92 inches of snow was recorded for the winter in some parts of the upper Susquehanna River watershed. This snowpack was washed out by a single strong rainstorm that lasted approximately nine days. The average precipitation for this storm in the basin above Harrisburg was 7.5 inches. The total storinflaw for the Susquehanna River from this storm exceeded that of the flow from Hurricane Agnes, a devastating storm that occurred in 1973. Peak flow measured at Conowingo Dam during the spring storm event of 1993 was 500,000 cubic feet per second (cfs), approximately half of peak flow measured for Hurricane Agnes (1,130,000 cfs). However, because of its longer duration and the larger residual of stored water in the watershed, the 1993 storm transported a total of 816 billion cubic feet of water through the Conowingo Dam, compared to Hurricane Agnes total discharge of 521 billion cubic feet. During both storm events, a significant quantity of sediment material, including the associated trace element contaminants, was transported into the Bay from the watershed. In order to study the dynamics of contaminant transport and to calculate accurate loads during this period of high flow in 1993, the United States Geological Survey (USGS) conducted a short-term intensive water quality sampling study on the Susquehanna River at the Conowingo Dam in Maryland. The Susquehanna River is the largest tributary to the Chesapeake Bay, contributing approximately 50% of the freshwater inflow to the Bay. It drains an area that is impacted heavily by agriculture, and coal and mineral mining industries. Additionally there is a significant number of municipalities that discharge effluents into the Susquehanna River. There are therefore significant sources of trace element contaminants from the Susquehanna watershed and mobilization of these to the Chesapeake Bay can be strongly enhanced during a large storm event. A second and also potentially important reservoir of trace element contaminants is stored in the sediments behind each of the three dams in the lower Susquehanna River. These are, in order of upstream to downstream, Safe Harbor, Hoitwood, and Conowingo. During high discharge, scouring of stored sediments behind each dam can occur. According to Lloyd Reed of the U.S. Geological Survey, (personal communication, 1993), scouring occurs at the Conowingo Dam when discharge exceeds 200,000 cfs. This discharge level was exceeded during most of the storm event in March/April, 1993. METHODS Water samples were collected at the Conowingo Dam site during March 25 through May 4, 1993. Samples were collected using ultra-clean techniques two to three times per day for the nine day period of highest flow (March 25 - April 3, 1993), and an 1 ------- additional 11 times for the remaining 32 days of high flow. Samples were analyzed for suspended sediment, dissolved aluminum (Al), and total-recoverable and dissolved fractions of arsenic (As), cadmium (Cd), chromium (Cr), copper (Cu), iron (Fe), mercury (Hg), nickel (Ni), lead (Pb), strontium (Sr), and zinc (Zn). Data for this study are in the Appendix to this Addendum, entitled 1993 Trace Element Data for the Susquchanna River at Conowingo Dam, MD. Loads of trace elements are calculated with a log-linear regression model that fits parameters for discharge, time, and seasonality to the concentration data. When there is not a significant relationship between the parameters of the regression model, loads are calculated by the Interpolation/Integration (II) model that interpolates concentration data, multiplies these by daily mean discharge measurements, and then integrates over the year. Methodologies for sampling and analysis are described in detail in the Fall Line Toxics Program 1992 Final Report. RESULTS Figure 1 presents a time series plot for discharge and suspended sediment measurements at Conowingo Dam, MD, for the period covering 1992 and 1993. The maximum discharge occurred in the spring of each year. However the plot also emphasizes that discharge was much higher in the spring 1993 than in 1992, and that storms which generate high discharges are an important mechanism for the mobilization of sediments. Concentration data for total-recoverable and dissolved Pb and Zn during this time period are presented in similar format in Figures 2 and 3, respectively. These two trace elements are normally associated with the particulate phase of surface water and therefore concentrations of total-recoverable Pb and Zn are correlated to discharge; particularly during the spring storm events. During base flow, total-recoverable concentrations of these 2 elements vary within a small range around the reporting limits of this study. Dissolved concentrations of Pb and Zn do not show a correlation to discharge or seasonality. Figure 4 presents the time series plot for discharge and the concentrations of dissolved and total-recoverable Cr at Conowingo Dam in 1993. There appears to be no relation between discharge or seasonality with either phase of Cr. Interestingly enough, the concentration of total-recoverable Cu exhibits a dual behavior at the Conowingo site. Figure 5(a) presents the time series plot for dissolved and total-recoverable Cu. During most of this time period, there was no more than a minimal correlation between discharge and total-recoverable Cu concentration (R=O.79 for the entire two year period). In fact there are several base flow concentration values that fail within the mid- to upper-range of stormflow data. However, during the storm event in late March and April, 1993, the behavior of Cu concentration changes and follows the discharge profile very closely. Figure 5(b), which is an expanded version of the 1993 storm event portion of the time series, emphasizes this point. The average total- recoverable Cu concentration was also slightly higher during this storm event than for base flow in 1993; 4.09 ± 1.97 (1 S.D.) as compared to 2.53 ± 1.51. The data suggest that total-recoverable Cu concentration is not related to discharge rates, except during abnormally high stormflows. There were no discernible relationships or trends for dissolved Cu during this study period. Quality control data for all the trace elements presented in this addendum are given in the Fall Line Toxics Program 1992 Final Report. Briefly, all blank data for total-recoverable Cu, Pb, and Zn were less than the reporting limit. Dissolved blank data for these trace elements had measurable concentrations, but with values significantly less 2 ------- Discharge and at Conowingo Suspended Sediment Dam, Maryland 400000 D C l ) C) 300000 CD 200000 100000 0 Dec Mar Jun 1992 Sep Dec Mar Figure 1. Time series of suspended sediment concentration Conowingo Dam, MD for the time period from 1992 to 1993. Jun Sep Dec 1993 and instantaneous discharge for the Susquehanna River at 250 -J a) E C C 0 CD I -. C C) 0 C 0 C-) 500000 200 150 1 00 50 0 3 ------- Total Recover.able and Dissolved Lead at Conowingo Dam, Maryland Dec Mar Jun Sep 1992 Figure 2. Time series of total-recoverable (tot.-rec.) and dissolved Pb Susquehanna River at Conowingo Dam, MD for the time period from dissolved Pb are 1.0 and 0.06 tg/L, respectively. Jun Sep 1993 Dec 500000 400000 D CR C) 300000 ca CD 200000 100000 0 (aq.) concentrations and instantaneous discharge for the 1992 to 1993. Reporting limits for total-recoverable and -J a) a C C 0 CO I -. C w 0 C 0 C-) 15 10 5 0 Dec Mar 4 ------- Total-Recoverable and at Conowingo Dissolved Dam, Maryland Zinc 400000 300000 200000 100000 0 Dec Mar Jun Sep 1992 Jun Sep 1993 Figure 3. Time series of total-recoverable (tot.-rec.) Susquehanna River at Conowingo Dam, MD. respectively. and dissolved (aq.) Zn concentrations and instantaneous discharge for the Reporting limits for total-recoverable and dissolved Zn are 10 and 0.08 tg/L, 100 C) a 500000 0 Cu 1 C U) U C 0 () 80 60 40 20 0 Dec Mar Dec 5 ------- Total-Recoverable at Conowingo and Dissolved Chromium Dam, Maryland Aug Oct Dec 400000 9 (I ) C, a) 300000 200000 2 U) 100000 0 Figure 4. Time series of total-recoverable (tot.-rec.) and Susquehanna River at Conowingo Dam, MD for 1993. .tgfL, respectively. dissolved (aq.) Cr concentrations and instantaneous discharge for the Reporting limits for total-recoverable and dissolved Cr are 1 and 0.2 6 -j 0) a C C 0 I- C 0 0 C 0 U 500000 5 4 3 2 1 0 Dec Feb Apr Jun 1993 6 ------- (a) Total-Recoverable and Dissolved Cu for 1992 to 1993 Mar Jun Aug 1992 (b) Total-Recoverable during the 1993 March Sep Dec Cu and Discharge Storm Event May Figure 5. Time series of total-recoverable and dissolved Cu and instantaneous discharge for the Susquehanna River at Conowingo Dam for (a) 1992 to 1993 and (b) during the spring storm event of 1993. Reporting limits for total-recoverable and dissolved Cu are 1 and 0.02 p.gfL, respectively. -J a C 0 C 0 U C 0 C.) -I a C C 0 C 0 C.) C 0 C.) 10 8 6 4 2 0 .10 8 6 4 2 0 Nov Jan Apr Jul 1993 500000 400000 D C) 300000 c 0 200000 C) 100000 0 500000 400000 C U) C) U) 300000 0 200000 C) U) 100000 0 April 7 ------- than the environmental data. Quality control data for Cr indicated that contamination by this element may have occurred in 1992 and in part of 1993. DISCUSSION Trace Element Concentrations Much of the trace element concentration data collected during the storm event of 1993 exhibited predictable behavior related to the elevated suspended sediment loads. The elements Pb and Zn are examples of soft-metal trace-element contaminants that are easily scavenged onto particulate surfaces. Sources of the contaminated suspended sediments during a storm event include runoff from the watershed as well as scouring of stored sediments from behind the dams of the lower Susquehanna River. During high flow, scouring is probably a significant source of sediment with trace elements sorbed to it. Data for Al and Fe concentrations, presented in the Appendix, also correlate well to suspended sediment loads. The environmental geochemistry of Cr in this system is not obvious from the results of this study. Sources of Cr to the Susquehanna River are probably primarily industrial or atmospheric deposition. It is not obvious from the data that Cr is being stored in sediments behind the Conowingo Dam, or that storms in general have a major influence on the behavior of this trace element. More study is required for better understanding of the geochemical behavior of Cr. The dual behavior of total-recoverable Cu in 1993 at Conowingo Dam suggests that there are multiple sources of this trace element in the Susquehanna watershed. During base flow, Cu sources may include drainage from active or abandoned coal and mineral mining operations, and industrial effluents such as from the steel and iron industries. During stormflow, it is possible that the Cu concentration is being augmented by sediments scoured from behind the darn, but if this were a significant source of Cu, one would expect to see elevated Cu concentrations whenever discharge was high enough to induce sediment scouring. While scouring occurred during almost the entire high stormflow period, from March 25 through April 28, most of the Cu concentrations for this period were within the range of base flow concentrations. There were five points that exceeded this range and these occurred only during the first three days, when peak flow occurred in 1993 storm event. The induction of a different source of Cu, other than sediment scouring, during abnormally high stormflow provides a better explanation for elevated concentrations and the unusual relationship to discharge during this storm event. The most likely alternate source is from municipal effluents. Municipal effluents, contaminated from Cu plumbing used for water supplies, have been suggested as a general source of this trace element into river systems. However, since many of the sewage treatment plants in the Susquehanna River watershed have recently installed secondary treatment, municipal effluents are probably not a primary source of Cu during base flow. This is not necessarily the case for stormflow. Many of the municipal sewage treatment plants in the Susquehanna watershed have combined sewage treatment, in other words they combine storm drain effluents with municipal sewage. Treatment plants can become overwhelmed during unusually high flow, such as occurred in the spring of 1993, and be forced to discharge raw sewage directly into the Susquehanna River or its tributaries. Under very high flow regimes, Cu becomes correlated to discharge, and this may be related to the municipal overflows during storm surges. This would also explain the slightly higher Cu measurements that occurred at the peak flow of the storm event. 8 ------- Concentration data for the other trace elements, As and Cd, were consistently below detection limit, so interpretation of theft geochemical behavior during the 1993 storm event is not possible. Load Estimates Concentration data collected and mean daily discharges for river flow throughout 1993 were used to estimate annual loads of contaminant trace elements for that year. These load data are presented in Table 1. Table 1. Annual loads of total-recoverable and dissolved trace elements for the Susquehanna River at Conowingo Dam, MD. Units are in metric tons per year. Lead values in italics were calculated with the Integration/Interpolation model (11). For constituents using the H model with censored data (measured concentrations less than the reporting limit), the load estimates were calculated twic&to determine a range in values. Censored data were assigned a value of zero for the calculation of a lower boundary or minimum load, and a value of the analytical reporting limit for the calculation of an upper boundary, or maximum load. All other load estimates (in normal print) are calculated with the log linear regression model (AMLE, Cohn) and ranges in loads are statistical estimates of variance (± 1 standard deviation) made by the model. Both models are described in detail in the Fall Line Toxics Program, 1992 Final Report. The upper ranges of annual load estimates for total-recoverable Cr, Cu, Pb, and Zn are presented in bar graph format in Figure 6. For comparison, load estimates for 1992 are included hi this graph. The spring portion (March, April, and May) of the annual load for each trace element is indicated as the stippled portion of each bar to evaluate the relative contribution of the large storm event to the annual load. Total loads for Cr, Cu, Pb, and Zn in the Susquehanna River at Conowingo Dam were consistently higher in 1993 than in 1992. The differences between the spring contributions of loads for each year are, however, more significant. For example, the spring contribution of the total annual Zn load delivered to the Chesapeake Bay in 1992 was 37% of the total annual load, while in 1993, a more significant 70% of the annual Zn load was transported in the spring. This same phenomenon was demonstrated for all of the other total- recoverable loads, and demonstrates the importance of stormilow to annual contaminant loads. The increased contaminant loads from the spring storm event of 1993 can be partially attributed to elevated concentrations of the contaminants, but more importantly to the large volume of water that was discharged from the watershed. 9 ------- Annual Loads of Contaminant Trace Elements at Conowingo Dam, Maryland Pb 0 1992 1993 1992 1993 1992 1993 1992 1993 Figure 6. Annual load estimates of total-recoverable Cr, Cu, Pb, and Zn for the Susquehanna River at Conowingo Dam, MD. The spring contribution of each load (March, April, and May) is indicated as the stippled portion of each bar. U) C 0 C) I- C) E 1200 1000 800 600 400 200 0 C C Cr Cu E.:...:.::.,.::.: I _____ _____ 10 ------- CONCLUSIONS The results of this swdy suggest the following conclusions: (1) Contaminant concentrations of trace elements that are associated with particulate phases, for example Pb and Zn, are correlated with discharge, particularly during very large storm events such as occurred in the spring of 1993. (2) The concentration behavior of Cu is more complex than for some of the other trace elements, having multiple sources that are apparently similar in importance. (3) Scouring of stored sediments from behind the Conowingo Dam may contribute significantly to the suspended sediment load, and hence contaminant trace element loads of Pb and Zn. (4) All annual load estimates for 1993 were higher than had been observed in 1992. This was at least partially due to the large volume of water that was transported during the spring stormilow of 1993, and a disproportionately large fraction of the annual load was contributed during this storm event. 11 ------- APPENDIX 1993 TRACE ELEMENT DATA FOR THE SUSQUEHANNA RiVER AT CONOWINGO DAM, MD 12 ------- SUSQUEHANNA R AT CONOWINGO. MD SEDI- SED CHRO- 1IENT SUSP. ALUM- CADMIUM CHRO- MIUM COPPER DIS- SIEVE INUM. ARSENIC CADMIUM TOTAL MIUM, TOTAL COPIER, T iAL CHARGE, DIAN. DIS- DIS- ARSENIC DIS- R!COV- DIS- RECOV- DIS- RZCOV- SUS- I FINER SOLVED SOLVED TOTAL SOLVED ERABLE SOLVED ERABLE SOLVED ERABLE DATE PENDED THAN (UG/L (UG/L (UG/L CUG/L (UG/L (UG/L (UG/L (UG/L (UG/L CT/DAY) .062 t41 AS AL) AS AS) AS AS) AS CD) AS CD) AS CR) AS CR) AS CU) AS CU) (80155) (70331) (01106) (01000) (01002) (01025) (01027) (01030) (01034) (01040) (01042) WATER-QUALITY DATA. CAI.ENDAR YEAR JANUARY 1993 TO DECEMBER 1993 DIS- BARO- SPE- PU ALKA- CHARGE. METRIC SPE- CIFIC WATER UNITY INST. PRES- CIFIC CON- WHOLE WAT DIS SEDI- CUBIC TEMPER- TEMPER- SURE CON- DUCT- OXYGEN, FIELD TOT IT MENT, FEET ATURE ATURE ( tl DUCT- A liCE DIS- (STAND- FIELD SUS- DATE TIME PER WATER AIR OF AliCE LAB SOLVED AND /L AS FENDED SECOND (DEG C) (DEG C) HG) (US/CM) (US/CM) (MG/L) UNITS) CACO3 ( /L) (00061) (00010) (00020) (00025) (00095) (90095) (00300) (00400) (39086) (80154) JAN 05... 07... 08... 20... MAR 11. . 25... 27... 28... 30... 31... 31... 1430 1330 1300 1130 1400 1130 1245 1330 1645 0145 1400 95900 101000 114000 69200 70900 145000 162000 184000 314000 321000 415000 5.0 5.0 6.0 5.5 4.0 5.0 6.0 7.0 7.0 7.0 8.0 18.0 13.0 9.0 7.0 10.0 9.0 11.0 15.0 16.0 8.0 15.0 742 769 766 774 763 771 766 759 757 761 759 138 171 163 219 290 270 225 220 160 165 146 155 150 154 201 255 258 207 202 154 149 132 13.0 13.3 12.9 13.9 12.5 13.0 13.2 12.7 12.5 12.3 7.4 7.9 7.8 7.3 7.3 7.5 7.3 7.3 7.2 7.2 26 27 28 31 45 25 31 17 25 20 25 14 18 7 14 21 32 28 79 67 97 APR 01... 01. . 02 . 03 . . 03... 03... 03. 04 . 04.. 05 . 08 12. . 13 . 15... 19 . . 22... 28 . 1230 2300 1315 1030 1715 1900 2230 1230 1945 1230 1500 1545 1415 1045 1515 1515 1115 448000 431000 480000 477000 461000 460000 430000 404000 378000 326000 190000 337000 295000 207000 300000 193000 244000 6.0 11.0 7.0 14.0 7.0 12.0 7.0 12.0 6.0 6.0 - - 7.0 10.0 7.0 8.0 7.0 15.0 9.0 16.0 8.0 14.0 10.0 19.0 9 0 16.0 12.0 22.0 12.0 8.0 11.0 20.0 749 749 752 762 763 765 764 766 763 753 759 760 759 747 771 128 120 114 119 116 220 114 124 172 196 128 133 148 174 118 116 115 109 111 111 111 116 151 147 118 129 134 150 156 12.7 12.5 12.0 12.6 11.7 13.2 12.8 13.3 12.9 12.5 12.6 10.8 11.9 11.1 11.9 7.3 7 2 7.0 7.0 7.1 7.2 7.3 7.4 7.5 7.3 7 4 7.4 7.5 7.5 - - 20 12 20 17 19 19 23 21 26 29 22 23 28 32 91 338 214 251 230 234 167 175 177 131 28 95 118 99 92 46 50 MAY 03... JUN 09 . . 23 1215 1130 1330 121000 6700 46900 16.0 23.0 28.0 23.0 28.0 37.0 770 760 766 178 337 385 167 318 10.3 6.1 5.4 7 3 7.5 7.9 - 31 66 15 8 JUL 07... 28 0915 0945 6400 5960 29.0 30.0 35.0 34.0 767 762 395 366 359 350 5.5 4.5 7.7 7.4 68 78 7 JAIl 05... 6470 99 30 <0.60 <1 0.16 <1 1.90 5 1.47 2 07 08 .. 3820 5540 100 96 30 20 <0.60 - <1 0.10 <1 0.63 - 0 69 5 20 .. 1310 94 30 - - MAR 11 . 25 . 27... 28 .. 30... 31... 31 2680 8220 14000 13900 67000 58100 109000 99 96 96 98 99 99 98 30 30 <10 20 <10 <10 <10 - <0.60 <0.60 <0.60 <0.60 0.65 <0.60 <1 <1 <1 <1 <1 1 <1 - 0.10 0.10 0.10 0.10 0.10 0.10 <1 <1 <1 <1 <1 <1 <1 - <0.20 0.36 0 59 <0.20 <0.20 <0.20 <1 <1 1 1 3 <1 5 -- 0 74 1.49 0.79 1.24 0 36 0.39 2 <1 2 2 4 4 3 APR 01 . . 01 02 . . 03 03... 03. 03 . 04.. 04 . 05 . 08 12 13 . 15. 19 . 22 . 28... 110000 393000 277000 323000 286000 291000 194000 191000 181000 115000 14400 86400 94000 55300 74500 24000 32900 98 97 97 97 97 97 98 95 94 92 97 92 96 99 97 99 96 <10 <10 <10 <10 20 - 40 20 10 20 30 10 40 30 20 20 <0.60 <0.60 <0.60 0.88 1.03 - <0.60 <0.60 <0.60 <0.60 <0.60 <0.60 <0.60 <0 60 <0.60 <0.60 1 2 2 2 2 - 1 2 1 <1 <1 1 1 <1 <1 <1 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 0.10 <1 <1 <1 <1 Cl Cl Cl Cl <1 <1 <1 <1 <1 <1 <1 <0.20 <0.20 0.65 <0.20 <0.20 <0.20 0 35 <0.20 <0.20 <0.20 <0.20 <0.20 <0.20 0.44 0.46 1 6 4 3 5 - 4 2 1 <1 2 1 4 1 <1 0 30 0.66 0 90 0.42 0 50 0 41 0 64 0.44 0.70 0.81 0.90 0.59 1.08 0.66 0 76 4 9 6 7 6 6 5 4 2 5 5 4 4 3 3 MAY 03 4900 94 20 <0.60 <1 0.10 <1 0.46 <1 0.53 <1 JUN 09 23 - 1010 99 - <0.60 <0.60 <1 - 0 10 0.10 <1 1.13 0.38 2 0.68 0.86 4 JUL 07 28. - 113 - 88 20 20 <0 60 1.26 <1 <1 0 10 0.15 <1 <1 0.83 0.59 <1 <1 0 81 1 03 2 . 1 ------- r.,r.as ft AT 10 ISiTU SALftT MTA. W A lRfl If) TO ia - . MU ktkciiiiI*ci a a a a 12 - O. 2 2 ci.3.o -- -- -. as :11 1 ii :: ::! c It 11111 Sd U I It II HI U 1::: iu i *1 II 10I! l i i jsfl in S Dl i i i f *11 1 1 i i U I u ------- PVur I At IS . t - t DI7A, air I i n s to ins Mn 4nttLt i ]I::: 31::: J1... It:: 1111 WI 1111 111 . 201100 111111311 11:1 11:1 0.0 U Itt I tt LI iH m 311 774 31 1 111 134 II i i i 111 141 I I - ii 1:1 12.0 Itt 3:1 3:1 7.0 3:1 Il II I l 1 J 127 U Mn cn )::: ajj... I... It:: 1 0 L j 74100 13111 - - )I 10 11 II C , 10 1 I I I ft CAl Cl t1I I 1:11 *11 0.10 *11 1 c i Cl t t11 114 0.20 11 1 41 2 1 1:11 111 0.07 111 1 ci 5 1 Mn Ci -- 41.1 1 Ci - - *11 4 0.10 3 *11 I 1111 3 13 . 70 11921 I:: 3 4 10, 25 11 Ut Is o *1 3300 40.0 . 9:11 1.23 lit C C C: 4 1 I: ------- |