A different method for controlling erosion and sediment transport during highway construction was used in each of four adjacent drainage basins in central Pennsylvania. The basins ranged in size from 240 to 490 acres (97 to 198 hectares), and the area disturbed by highway construction in each basin ranged from 20 to 48 acres (8 to 19 hectares). Sediment discharge was measured from each basin for 3 years before construction began and for 2 years during construction. In one of the basins affected by the construction, three offstream ponds were constructed to intercept runoff from the construction area before it reached the stream. In another basin, a large onstream pond was constructed to trap runoff from the construction area after it reached the stream. In a third area, seeding, mulching, and rock dams were used to limit erosion. In the fourth area, no sediment controls were used. The effectiveness of the various sediment-control measures were determined by comparing the sediment loads transported from the basins with sediment controls to those without controls. For most storms the offstream ponds trapped about 60 percent of the sediment that reached them. The large onstream pond had a trap efficiency of about 80 percent, however, it remained turbid and kept the stream flow turbid for long periods following storm periods. Samples of runoff water from the construction area were collected above and below rock dams to determine the reduction in sediment as the flow passed through the device. Rock dams in streams had a trap efficiency of about 5 percent. Seeding and mulching may reduce sediment discharge by 20 percent during construction, and straw bales placed to trap runoff water may reduce sediment loads downstream by 5 percent.
The Susquehanna River presently transports about 3.0 million tons of sediment annually (110 tons per square mile). Only about 1.8 million tons of sediment enters the head of Chesapeake Bay annually because some sediment is trapped behind the power dams on the lower Susquehanna. Measured annual sediment yields from subbasins in the Susquehanna range from 40 to 440 tons per square mile. The highest yields are from parts of the glaciated section of the basin, in the anthracite coal region, and the Piedmont province. The lowest yields are from parts of the glaciated section of the basin and the Appalachian high plateau. Available data indicate that there has been a downward trend of sediment discharge in recent years. In the future, the high sediment yields associated with urbanization may offset this present downward trend.
Regression equations for estimating the magnitude and frequency of floods on ungaged streams in Pennsylvania with drainage areas less that 2,000 square miles were developed on the basis of peak-flow data collected at 313 streamflow-gaging stations. All streamflow-gaging stations used in the development of the equations had 10 or more years of record and include active and discontinued continuous-record and crest-stage partial-record streamflow-gaging stations. Regional regression equations were developed for flood flows expected every 10, 25, 50, 100, and 500 years by the use of a weighted multiple linear regression model. The State was divided into two regions. The largest region, Region A, encompasses about 78 percent of Pennsylvania. The smaller region, Region B, includes only the northwestern part of the State. Basin characteristics used in the regression equations for Region A are drainage area, percentage of forest cover, percentage of urban development, percentage of basin underlain by carbonate bedrock, and percentage of basin controlled by lakes, swamps, and reservoirs. Basin characteristics used in the regression equations for Region B are drainage area and percentage of basin controlled by lakes, swamps, and reservoirs. The coefficient of determination (R 2) values for the five flood-frequency equations for Region A range from 0.93 to 0.82, and for Region B, the range is from 0.96 to 0.89. While the regression equations can be used to predict the magnitude and frequency of peak flows for most streams in the State, they should not be used for streams with drainage areas greater than 2,000 square miles or less than 1.5 square miles, for streams that drain extensively mined areas, or for stream reaches immediately below flood-control reservoirs. In addition, the equations presented for Region B should not be used if the stream drains a basin with more than 5 percent urban development.
Methods for computing streamflow statistics intended for use on ungaged locations on Pennsylvania streams are presented and compared to frequency distributions of gaged streamflow data. The streamflow statistics used in the comparisons include the 7-day 10-year low flow, 50-year flood flow, and the 100-year flood flow; additional statistics are presented. Streamflow statistics for gaged locations on streams in Pennsylvania were computed using three methods for the comparisons: 1) Log-Pearson type III frequency distribution (Log-Pearson) of continuous-record streamflow data, 2) regional regression equations developed by the U.S. Geological Survey in 1982 (WRI 82-21), and 3) regional regression equations developed by the Pennsylvania State University in 1981 (PSU-IV). Log-Pearson distribution was considered the reference method for evaluation of the regional regression equations. Low-flow statistics were computed using the Log-Pearson distribution and WRI 82-21, whereas flood-flow statistics were computed using all three methods. The urban adjustment for PSU-IV was modified from the recommended computation to exclude Philadelphia and the surrounding areas (region 1) from the adjustment. Adjustments for storage area for PSU-IV were also slightly modified. A comparison of the 7-day 10-year low flow computed from Log-Pearson distribution and WRI-82-21 showed that the methods produced significantly different values for about 7 percent of the state. The same methods produced 50-year and 100-year flood flows that were significantly different for about 24 percent of the state. Flood-flow statistics computed using Log-Pearson distribution and PSU-IV were not significantly different in any regions of the state. These findings are based on a statistical comparison using the t-test on signed ranks and graphical methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.