Recent streamflow trends

Water Freedom System

Survive Global Water Shortages

Get Instant Access

Rivers integrate the hydroclimatic variables within the watershed they drain. Streamflow results from the interaction of the hydroclimatic variables in both time and space, and watershed physiography exerts temporal influences on transformation of the residual precipitation into streamflow in a specific drainage basin. Precipitation and temperature are major climatic factors determining runoff, and both are influenced by ocean-atmosphere processes to produce streamflow variability on interannual and decadal time scales (Robertson and Mechoso, 1998). However, the water balance (Equation 2.13) indicates that annual precipitation variability drives annual streamflow variability and changes in precipitation are amplified in runoff changes. For this reason it is surprising that the observed increasing frequency of extreme high precipitation is not being detected in annual peak flow data and the duration of low-flow events (Katz et al., 2002). In North America the evidence of increase in extreme precipitation events may be masked in streamflow records by concurrent advances in the timing of peak spring-season flows by as much as one to four weeks due to the earlier onset of spring snowmelt (Regonda et al., 2005; Stewart et al., 2005; Hodgkins and Dudley, 2006). Low-flow indicators in the United Kingdom show little evidence of sustained change since the 1960s and are characterized by relative stability (Hannaford and Marsh, 2006).

Annual mean streamflow is the mean flow for a given year, and observed annual streamflow serves as a pertinent indicator of interannual variability (Anctil and Coulibaly, 2004). A time-series of annual mean streamflow displays the annual streamflow variation related to hydroclimatic variability resulting from both natural and human influences within the watershed. The extent of human impact on the land and vegetation in many large watersheds has had important hydrologic implications that mask natural processes. Therefore, identifying natural hydroclimatic variability in streamflow records is most successful for watersheds where human impacts are minimized. Changing

1900 1920 1940 1960 1980 2000

Year

Fig. 8.16. Annual mean streamflow for the Susquehanna River at Wilkes-Barre, Pennsylvania, for 1899-2005. (Data courtesy of the U.S. Geological Survey from their website at http://waterdata.usgs.gov/nwis/.)

conditions within the watershed during the period of streamflow observations is especially likely in regions where dam construction and water diversions for irrigation are common. Pre-instrument data periods are assessed using reconstructed streamflow discussed in Section 8.3.

8.12.1 The Susquehanna River

The Susquehanna River (Fig. 8.16) rises as the outlet of Otsego Lake in central New York at an elevation of 360 m. It flows southeast across Pennsylvania and through Maryland to enter Chesapeake Bay. It is the longest river entirely within the United States that drains into the Atlantic Ocean. The watershed at Wilkes-Barre, Pennsylvania, (41° N) covers 25 900 km2 of the Allegheny Plateau, and the elevation of the gauge at Wilkes-Barre is 155 m. The Susquehanna River is unregulated above Wilkes-Barre as it flows through dairy farm land and a former anthracite coal industrial area in the ridges of northeastern Pennsylvania. The mean annual streamflow at Wilkes-Barre is 385 m3 s~\ the highest annual mean streamflow is 622 m3 s_1 in 1978, and the lowest is 175 m^1 in 1965.

The Susquehanna River annual mean streamflow displays a generally similar pattern of wet and dry years as the precipitation record for Williamsport, Pennsylvania (Fig. 8.17) 90 km west of Wilkes-Barre. However, the details of the two time-series are different in that the years of maximum and minimum precipitation at Williamsport are not coincident with the maximum and minimum Susquehanna River streamflow at Wilkes-Barre. Both high and low flows lag precipitation maxima and minima by a year or more, and the nature of the streamflow response is related to conditions during intervening years.

Year

Fig. 8.17. Annual precipitation for New Meadows Ranger Station, Idaho, and Williamsport, Pennsylvania, for 1896-2002. (Data courtesy of NOAA's National Climate Data Center and the Oak Ridge National Laboratory, Carbon Dioxide Information Analysis Center from their website at http://cdiac.ornl.gov/epubs/ndp/ ushcn/usa_monthly.html.)

Year

Fig. 8.17. Annual precipitation for New Meadows Ranger Station, Idaho, and Williamsport, Pennsylvania, for 1896-2002. (Data courtesy of NOAA's National Climate Data Center and the Oak Ridge National Laboratory, Carbon Dioxide Information Analysis Center from their website at http://cdiac.ornl.gov/epubs/ndp/ ushcn/usa_monthly.html.)

Annual mean streamflow for the Susquehanna River displays drier conditions for the first half of the record compared to the second half. Thirty-five years before 1953 have annual mean streamflow less than the period of record average. The second half of the record contains both the highest and lowest annual mean streamflow, there is greater variability among annual values, and the consecutive years of above and below average streamflow are longer in the more recent record. Land-use changes involving vegetation removal may account for some of the increased variability, but the Williamsport precipitation record displays a generally similar pattern of wet and dry years and increasing precipitation variability in the recent record.

8.12.2 The Salmon River

The Salmon River, in the Pacific Northwest of the United States, drains a region of high mountains and deep canyons in central Idaho. The watershed area at White Bird, Idaho (46° N), is 35 230 km2 and elevation differences of over 2120 m occur between the river's headwaters in the Sawtooth Mountains and its confluence with the Snake River a few kilometers below White Bird. Snowmelt from the high mountains in the headwaters area provides a significant portion of the Salmon River streamflow, but winter precipitation at lower elevations contributes to streamflow as well. The Salmon River is the longest free-flowing river in the contiguous United States, and considerable effort has been devoted to preserving the natural state of the Salmon River watershed. The Middle Fork of the Salmon River is one of the premier recreational rafting and kayaking rivers in the world.

8.12 Recent streamflow trends 299

500-

450-

150 1920

500-

450-

1940

1960 Year

1980

2000

Fig. 8.18. Annual mean streamflow for the Salmon River at White Bird, Idaho, for 1920-2005. (Data courtesy of the U.S. Geological Survey from their website at http:// waterdata.usgs.gov/nwis/.)

1940

1960 Year

1980

2000

Fig. 8.18. Annual mean streamflow for the Salmon River at White Bird, Idaho, for 1920-2005. (Data courtesy of the U.S. Geological Survey from their website at http:// waterdata.usgs.gov/nwis/.)

Cattle, forest products, gold mining, tourism, and hay are the primary economic activities in the Salmon River watershed. Diversions to irrigate 66 000 ha occur upstream from White Bird. The Salmon River annual mean streamflow at White Bird (Fig. 8.18) is 313m3s_1, the highest annual mean streamflow is 506 m^1 in 1997, and the lowest is 165 m^1 in 1931 and 1977. The drier setting of the Salmon River watershed compared to the Susquehanna River is evident in that the mean annual streamflow for the Salmon River is less than that for the Susquehanna River even though the Salmon River watershed area is one-third larger.

The first 47 years of the Salmon River streamflow record are drier than the more recent years largely due to the dominance of below average years from 1923 to 1946. Only six years during the 24 years have annual mean streamflow greater than the period of record average. Since 1958, year-to-year annual mean streamflow displays greater variability with abrupt changes from high values to low values. However, the 24 years from 1963 to 1986 are dominated by annual mean streamflow greater than the average, and this period includes the second to fifth highest values and one of the lowest values in the record. The longest consecutive departure from the average is the eight dry years from 1987 to 1994. Another six consecutive years of below average flow begins in 2000. The relative dryness in the first half of the record and the absence of multiyear droughts from 1950 to 1987 are characteristics shared with the entire Columbia River Basin (Gedalof et al., 2004).

The New Meadows Ranger Station (NMRS), Idaho (45° N), is 100 km south of White Bird and just outside the western boundary of the Salmon River watershed. However, the record for NMRS (see Fig. 8.17) provides the longest precipitation time-series available for a station in or near the Salmon River basin. The predominantly dry years in the 1920s and 1930s in the Salmon River streamflow are evident in the NMRS precipitation data, but the low flow in 1931 is not present in the precipitation data, which have low values in 1924 and 1925. The precipitation spikes in the more recent record for NMRS do not occur in the same years as the high streamflow values for the Salmon River, but the NMRS precipitation data have a sequence of dry years from 1985 to 1994 that coincides with a similar period of below average flows for the Salmon River.

Comparing streamflow time-series for the Susquehanna River (see Fig. 8.16) and the Salmon River (see Fig. 8.18) reveals almost no agreement in the years of highest and lowest flows. The single occurrence of similar conditions is that both watersheds have notable low flows in 1931. The dominant period of below average flows in the first half of the century begins 18 years earlier for the Susquehanna River and lasts 2 years longer for the Salmon River. More recently, the Susquehanna River has below average flows in the 1960s while below average flows in the Salmon River are more prevalent in the late 1980s and early 1990s and the early 2000s. Flows are consistently above average in the Susquehanna River during the 1970s, but the longest run of above average flows for the Salmon River occurs in the 1990s.

8.13 Recent lake level trends

Lake levels provide a hydroclimate indicator that is an integrating factor of precipitation and streamflow anomalies for the land area contributing runoff into the lake. Changes in inflow and evaporation from the lake will affect lake volume and water level in ways related to lake geomorphology.

The lake response to changes in atmospheric, land surface, and subsurface water fluxes is expressed by Equation 6.14. A lake's role as a natural hydrocli-matic indicator on time scales of months to years is most valuable when the lake and watershed are relatively undisturbed (Street-Perrott, 1995), but many lakes and/or their watersheds have experienced significant human influences. Since few continuous instrumental records of lake levels are available before 1875, selecting examples where disturbances are minimized is important so that natural responses are not masked.

8.13.1 Lake Champlain, Vermont/New York

Lake Champlain (42° N) is an exorheic lake in the Champlain Valley between the Green Mountains of Vermont and the Adirondack Mountains of New York in the northeastern United States. The lake area is 1130 km2, and it extends across the United States-Canada border into Quebec. Lake Champlain is drained northward by the Richelieu River which joins the St. Lawrence River

1908 1928 1948 1968 1988 Year

Fig. 8.19. Annual precipitation for Burlington, Vermont, for 1908-2002. (Data courtesy of NOAA's National Climate Data Center and the Oak Ridge National Laboratory, Carbon Dioxide Information Analysis Center from their website at http:// cdiac.ornl.gov/epubs/ndp/ushcn/usa_monthly.html.)

1908 1928 1948 1968 1988 Year

Fig. 8.19. Annual precipitation for Burlington, Vermont, for 1908-2002. (Data courtesy of NOAA's National Climate Data Center and the Oak Ridge National Laboratory, Carbon Dioxide Information Analysis Center from their website at http:// cdiac.ornl.gov/epubs/ndp/ushcn/usa_monthly.html.)

east of Montreal. Several cities are located around the lake, but small craft, ferries, and lake-cruise ships constitute the contemporary lake traffic.

The Lake Champlain Basin covers 21300 km2, and the largest portion of the watershed is in Vermont. The watershed's hydroclimatology is influenced by the lake's location on the eastern edge of the continent, by the frequency of Arctic, Pacific, and tropical air passing over the region, and by the presence of mountains on both the east and west sides of the watershed. Precipitation amounts vary from 76 cm near the lake to over 127 cm in the surrounding mountains, which rise to elevations exceeding 1300 m. Annual precipitation on the lake's east shore at Burlington, Vermont (44° N), averages 85 cm (Fig. 8.19). The annual values range from 128 cm in 1998 to 58 cm in 1914.

The average depth of Lake Champlain is 19.5 m, but the greatest depth is 122 m. The mean annual lake level is 29.1 m. The highest annual mean lake level is 29.6 m in 1976, and the lowest is 28.6 m in 1915 and 1941 (Fig. 8.20). The distinctive feature of the 97-year time-series is the low levels from 1930 to 1968. Only 6 of these 39 years have an annual mean lake level higher than the 97-year mean annual value. The mean annual lake level for 1930 to 1968 is 28.9 m compared to the mean of 29.2 m for the 36 years from 1969 to 2005.

The lake level data indicate hydroclimatic fluctuations in the Lake Champlain Basin have been modest for the past century. The change in the average lake level from the dry years of 1930 to 1968 to the wet years of 1969 to 2005 is 0.3 m. Furthermore, the variability of the annual mean level indicated by the standard deviation is the same for the two periods (0.17). These characteristics portray a dampened response compared to the variation evident in the Burlington precipitation data.

Year

Fig. 8.20. Annual mean lake level above sea level for Lake Champlain at Burlington, Vermont, for 1908-2005. (Data courtesy of the U.S. Geological Survey from their website at http://waterdata.usgs.gov/nwis/.)

Year

Fig. 8.20. Annual mean lake level above sea level for Lake Champlain at Burlington, Vermont, for 1908-2005. (Data courtesy of the U.S. Geological Survey from their website at http://waterdata.usgs.gov/nwis/.)

8.13.2 Great Salt Lake, Utah

Great Salt Lake (41° N) in northern Utah is the fourth largest endorheic lake in the world and the largest of this type in the United States. It is a small remnant of prehistoric Lake Bonneville that covered much of present-day western Utah and parts of neighboring states. The arid region west and southwest of Great Salt Lake is part of the former Lake Bonneville lake bed, and the low, flat landscape is known as the Great Salt Lake Desert. The Bonneville Salt Flats occupy 180 km2 of extremely level salt beds on the western edge of the Great Salt Lake Desert.

The Great Salt Lake historic surface area is 4400 km2, but the lake's area fluctuates substantially due to its shallowness and changes in its water balance. The watershed area is 34 600 km2, and three major rivers discharge into the lake. All three rivers receive runoff directly or indirectly from the Uinta and Wasatch mountains that form the eastern watershed area, where peaks rise to elevations exceeding 3600 m. The three rivers provide the majority of the lake's water supply, and they provide water for some irrigation and urban use. The majority of the lake's watershed is in an arid region of small mountain ranges and broad basins that produce little runoff. Watershed precipitation varies from less than 30 cm annually near the lake to 140 cm at the highest elevations in the eastern mountains. Winter moisture accompanies cyclonic systems formed over the Gulf of Alaska, summer thunderstorms result from moisture inflows from the Gulf of California, and spring and fall moisture is carried over the region by westerly winds flowing from the Pacific Ocean.

The historical average elevation of the Great Salt Lake is 1280 m. At this elevation, the lake's average depth is 4 m and the maximum depth is 10.7 m.

1847 1867 1887 1907 1927 1947 1967 1987 Year

Fig. 8.21. Annual mean lake level above sea level for the Great Salt Lake at Saltair Boat Harbor, Utah, for 1847-2005. (Data courtesy of the U.S. Geological Survey from their website at http://waterdata.usgs.gov/nwis/.)

1270

1847 1867 1887 1907 1927 1947 1967 1987 Year

Fig. 8.21. Annual mean lake level above sea level for the Great Salt Lake at Saltair Boat Harbor, Utah, for 1847-2005. (Data courtesy of the U.S. Geological Survey from their website at http://waterdata.usgs.gov/nwis/.)

The lake level fluctuates seasonally about 0.5 m from the highest level in May to July to a low level in October through November. Even modest changes in the lake's level cause relatively large changes in its area and volume due to its shallowness. The salinity of Great Salt Lake varies between 5% and 27% as the lake level changes. The lake's chemical composition is similar to sea water at a salinity of 5%.

The annual mean lake level elevations for historic Great Salt Lake (Fig. 8.21) show that the levels range from highs of nearly 1276 m in 1872 and 1873 and again in 1986 and 1987 to a low of 1270.2 m in 1963. Declining levels predominate from 1873 to 1963 except for the period 1909-29 when lake levels are above the average. The abrupt lake level increase from 1963 to 1986 implies a relatively wet period for these 23 years.

Ogden Pioneer Power House (Ogden PPH) (41° N) at 1326 m in the Wasatch Mountains east of Great Salt Lake is a representative site for precipitation in the runoff-producing area of the Great Salt Lake watershed (Fig. 8.22). Mean annual precipitation of Ogden PPH is 53 cm and ranges from a low of 29 cm in 1928 to a high of 109 cm in 1983. There is a small increasing trend in annual precipitation over the 128 years, and the trend contrasts markedly with the pattern of the Great Salt Lake levels, which show a declining trend to 1963 and recovery after 1963. Another evident contrast is that the Great Salt Lake levels do not show a response to individual years of above or below average precipitation at Ogden PPH. For example, the large precipitation amount received at Ogden PPH in 1983 does not produce an identifiable single-year increase in the Great Salt Lake level in 1983 nor does the meager precipitation at Ogden PPH in 1928 produce a single-year lake level decline in 1928. However, groups of wet years at Ogden

1875 1895 1915 1935 1955 1975 1995 Year

Fig. 8.22. Annual precipitation for Ogden Pioneer Power House, Utah, for 1875-2002. (Data courtesy of NOAA's National Climate Data Center and the Oak Ridge National Laboratory, Carbon Dioxide Information Analysis Center from their website at http:// cdiac.ornl.gov/epubs/ndp/ushcn/usa_monthly.html.)

1875 1895 1915 1935 1955 1975 1995 Year

Fig. 8.22. Annual precipitation for Ogden Pioneer Power House, Utah, for 1875-2002. (Data courtesy of NOAA's National Climate Data Center and the Oak Ridge National Laboratory, Carbon Dioxide Information Analysis Center from their website at http:// cdiac.ornl.gov/epubs/ndp/ushcn/usa_monthly.html.)

PPH are associated with increasing lake levels and groups of dry years result in declining lake levels. In both cases, a lag of about three years is evident in the lapse between the cumulative precipitation increase or decrease at Ogden PPH and the rising or declining Great Salt Lake level.

Review questions

8.1 What are the hydroclimatic ramifications of short records, a changing climate, and the risk of extreme events?

8.2 What are the relative strengths and weaknesses of proxy data as hydroclimatic indicators?

8.3 How are documentary data and tree rings used to improve knowledge of the hyrdroclimate of a place or region?

8.4 How are teleconnections a valid aid for understanding hydroclimatic variability?

8.5 What is the relationship between El Nino and the Southern Oscillation?

8.6 How is El Nino warming of the equatorial Pacific Ocean thought to project hydroclimatic anomalies to distant regions of the globe?

8.7 What are the similarities and differences in the atmospheric patterns identified as the NAO and the PNA?

8.8 What is the hydroclimatic relevance of temperature and precipitation changes over time with respect to streamflow?

8.9 How are lakes useful for studying long-term hydroclimatic relationships?

Was this article helpful?

0 0
Waste Management And Control

Waste Management And Control

Get All The Support And Guidance You Need To Be A Success At Understanding Waste Management. This Book Is One Of The Most Valuable Resources In The World When It Comes To The Truth about Environment, Waste and Landfills.

Get My Free Ebook


Post a comment