HYDROLOGY & PRECIPITATIONFrank S. Kwapnioski P. E., Ed J. Dekleva, Jr., Nebraska Public Power District, P.O. Box 310, North Platte, Nebraska 69103, (308) 532-9200.
For almost 20 years, there has been a myth circulating concerning the extent to which Platte River flows have been depleted by man's development efforts. Many people have believed this myth and entities ranging from the United States Fish and Wildlife Service to National Geographic magazine have stated as fact that man has depleted the flow of the Platte River by 70% and more throughout the period that changes to the river have occurred. An often quoted source of this myth was a paper written by Garnett P. Williams in 1978 entitled The Case of the Shrinking Channels - the North Platte and Platte Rivers in Nebraska. Since its publication, this document has been used by many to support the belief that the Platte River in general has dramatically changed as a result of man's use of its water.
Unfortunately, this study is still a work in progress regarding the specific myth of Platte River flow depletion, its origins and its propagation; but some very definite conclusions regarding Platte River flow change are already apparent. Funk and Wagnalls defines a myth as a "false opinion, belief or idea." Webster defines myth as an "imaginary thing". Both of these definitions are appropriate when describing prevailing views on flow changes in the Platte River and their causes.
This study is neither intended to confront the existence of, or identify the amount of man- made depletion to Platte River flows, but to simply challenge the perception that Platte River flow has been significantly depleted because of man's development of the basin. This study examines several of G. P. Williams' conclusions and the methodology he utilized in producing his report relative to more complete and up-to-date data.
In his report, Williams states: "It is impossible to isolate and identify the influence of any one factor (storage reservoir, irrigation diversion, and so forth), because (1) several such factors have operated in varying degrees simultaneously, and (2) measurements of water discharge are scarce for most gaging stations prior to the 1930's. Nevertheless, it might be possible to determine roughly the effects of certain major influences, such as big dams." In this statement, Williams is vague at best, yet this presumption of the impacts of man-made Platte River development has been utilized by many entities to misrepresent the Williams report as forming a direct correlation between said changes and the potential "major influences such as big dams." It must be duly noted that Williams made it clear that exact causes cannot be determined, but it "might" be possible to determine "roughly" the effects of certain influences.
One of his most compelling representations in support of the myth that man has significantly impacted river flows are the graphs on page 32, shown as Figure 1 in this report. Williams illustrated both annual peak flow and mean annual flow graphically to indicate the effects of development on these flows. In approximately 1940, the flows at North Platte begin to decline significantly and apparently remain low for the remainder of the study period. This location however is inappropriate to use to represent the general effects of development on the basin because these values do not include the North Platte Hydro returns. At face value, they apparently prove substantial flow depletion. Simply updating his Overton data while duplicating his methodology suggests other serious problems with his conclusions. A graph, showing the updated information, is presented in Figure 2. This graph is for demonstration purposes only, and does not suggest any concurrence that the methodology is appropriate for its purpose. With this finding, a more rigorous analysis of his methodology along with a challenge of his assumptions appeared to be warranted. It was this expanded scope that converted this study from a finished product to a "work in progress".
HYDROLOGY & PRECIPITATIONWilliams used the following statement to dismiss climate as a possible contributor to the observed flow decline in the Platte River:
"Analysis of U.S. Weather Bureau precipitative data for North Platte, Kearney, and Grand Island reveals that there have been no significant long-term trends or changes in total annual precipitation for these cities during the period 1900-74. Therefore, any observed long-term changes in streamflow probably are not attributable to changes in climate." (Williams, 1978)
This observation is localized and generally irrelevant. The precipitation records utilized in the Williams study were examined for only three of the lower basin precipitation stations. The Platte River basin covers approximately 90,000 square miles and encompasses three states, with a total length of approximately 680 miles. (Letter from the Secretary of War, 73d Congress, 2d Session, Jan. 3, 1934) His analysis was inadequate for several reasons including; choosing three localized precipitation gages within the study area to make an assessment of climatological stability for the entire basin is inappropriate because this approach neglects such important factors as changes in upper basin precipitation, the runoff associated with overland flow and groundwater flow contributions because of climate changes in other areas of the watershed excluded from the Williams' study area. In a 1905 Department of Agriculture Study titled Investigation of Water Rights of Interstate Streams: The Platte River and Tributaries, the author, R. P. Teele, (Teele, 1905) observes that "The precipitation is high in the extreme altitudes, drops down near the base of the mountains, then gradually increases to the east." This statement helps to further explain that precipitation characteristics in the Platte River basin are not homogeneous, and emphasizes the inappropriateness of the localized site selection.
Further analysis of the Platte River basin precipitation data reveals that Williams' contention that there were no significant long-term precipitation trends was false. A brief examination of the annual precipitation graphs at the sites he chose (Figures 3, 4, 5 and 6) does not reveal any long term trends, but this research has found that obvious and evident trends in the data set do exist. Because of the significant cumulative effects of long term droughts and wet periods, it is inadequate to analyze the effects of trends in terms of annual variations. To properly analyze the effects of climatic conditions such as precipitation, it is important to consider cumulative impacts of the annual variations. When one considers than, on average, 50 to 60 million acre feet of precipitation falls on the basin annually and only about 5 million acre feet has ever accumulated as runoff at Overton, it becomes apparent how large the natural background for water consumption really is. This point also emphasizes why the cumulative departure from average analysis is necessary to define the component of precipitation actively affecting stream flows. It has been further identified that the effects of the identified precipitation trends are significant in terms of observed flow changes in the river. (Figures 7, 8, 9 and 10)
In extending this analysis, 11 precipitation gauges throughout the basin were chosen to generally represent the watershed rainfall. For each station, the mean annual precipitation was calculated from monthly values. The mean value for each station was subtracted from the individual annual total precipitation values to determine a "deviation from mean" for each station for each year. A running total of the deviations was tabulated, thus yielding the "cumulative mean deviation".
The cumulative mean deviation for each station was developed for use in this analysis, and the basin-wide mean cumulative departure was also developed. Although this method helps to identify precipitation trends and their potential affect on the basin, the nature of this method does not produce a quantifiable value with the graphs. Most correctly, the trends in the graphs are valuable in identifying wet periods and dry periods in the data set.
From this analysis, three very interesting and very important observations relative to Williams' work were discovered:
Both wet and dry departures from normal tend to accumulate over time and produce a net cumulative effect on basin flows. This effect becomes quite apparent when comparing the difference between the first two time periods shown above. The net difference is 2.3 inches. Although this does not seem to be a significant difference, one must consider that this average difference will impact the entire basin. The entire Platte River basin covers approximately 90,000 square miles. For this discussion and analysis, it is estimated that approximately 61,000 square miles of the basin contributes to the flows at Grand Island. Using this area value, a change of 2.3 inches will produce a volume of about 7.5 million acre-feet per year. This implies that during the 1900-1929 time period, approximately 7.5 million acre-feet of additional water was available as precipitation each year when compared to the 1930-1940 time period. It is this huge difference in water supply between these periods that is responsible for the majority of stream flow impacts observed in the record.
When the cumulative deviation graphs are compared with streamflow graphs for comparable time periods, another important relationship is apparent. Basin streamflow seems to be closely correlated with precipitation over time.
A premise of the Williams report is that precipitation has remained somewhat constant, but flow at Overton, Nebraska has been steadily declining since the late 1920s or 1930s. However, when basin precipitation cumulative departure is compared with flow at Overton, as one would intuitively expect a definite correlation between precipitation and flow can be made; and the effects of man's development are relegated to the background in respect to items of significance.
HISTORIC FLOWSA general perception has been held that flows are generally lower since man's development of the watershed area than prior to such development. The argument made is generally weak because of another myth pertaining to the time period during which development occurred. The Williams report identifies that the flows in the Platte River had been declining since the early 1930s, allowing the misperception that a majority of development impacts had taken place during that time. Information is available that substantial development had occurred since the mid to late 1800s. Although the information from the very early time periods is missing measured flow data, a good deal of precipitation data and recorded testimony as to the characteristics of the Platte River basin is available. This information is contrary to the theory that the early Williams flow data might somehow represent some sort of predevelopment baseline.
Teele, 1905, provides a great deal of testimony of low flow times during which various reaches of the river went completely dry – well before any development had occurred. An intent of Teele, in part, was to determine the effects of surface irrigation development on riparian water right holders downstream.
"To determine this, in the absence of the records of the flow of the stream in the early years, it has been necessary to collect the testimony of settlers and travelers as to early conditions." (Teele, 1905)
"There are yet living in the basin of the Platte a great many people who went to that section before irrigation began. Statements as to the condition of the river have been secured from a number of those early settlers." (Teele, 1905)
"Bruce Johnson, of Greeley, Colo., came out in 1859. He went up and down the South Platte two different years during the sixties, but did not remember which years. Both of these years the river was dry from Julesburg to the junction, and in one of these years - he thought it was in 1864 - the river was dry from the junction to Fort Kearney, and farther down for all he knew. The river in the years mentioned was completely dry." (Teele, 1905)
"Hon. Eugene F. Ware, formerly Commissioner of Pensions, Topeka, Kans., who was with the United States troops along the Platte River in the sixties, stated that in 1865 he saw the buffalo pawing in the bed of the Platte to obtain water." (Teele, 1905)
"W. A. Paxton, of Omaha, Nebr., who has had cattle ranches on the Platte since the early sixties, states that the river went dry in 1863, 1866, and 1871. . ." (Teele, 1905)
"The testimony of the parties living in the Platte Valley show that the river was dry in places, and for short periods at least, during the years 1863, 1864, 1865, 1866, 1871, 1873, 1874, and 1875. For these years we have definite statements. For the years 1863 and 1864 we have a large number of statements that the river was dry. In addition to these definite statements as to particular years, we have the statement of J. J. Armstrong that it usually went practically dry. . . Mr. John Bratt, of North Platte, states that is was his opinion that it went dry in the sixties as now. It seems, from these statements, that there can be no question that from the time of the earliest travelers the river has gone dry during the summer, in some years at least, and that it has always been very low in the late summer, and that in the summers of 1862 to 1866, before there were enough diversions in Colorado to have any effect upon the flow of the stream, the river went dry." (Teele, 1905)
This information directly challenges the misperceptions or misrepresentations that the early period in Williams' data (i.e. 1900-1930) somehow represents a predevelopment baseline. Indeed this is not the case because basin development was an on-going gradual process beginning even earlier than Williams' study period and continuing to the present.
PEAK DISCHARGESThis section of the Williams report discusses several aspects of the changes in annual peak discharges in the Platte River. Specifically, the average peak discharges are said to have decreased with the closure of each major dam. "Discharges at some or all of the stations downstream from Lewellen have been influenced since the mid-1930's or shortly thereafter by Lake McConaughy, the Sutherland Canal, the Tri-County Canal, Johnson Power Plant releases…, and other features." (Williams, 1978) This statement is inconsistent with Williams' contention that it "might" be possible to determine "roughly" the effects of certain major developments, such as big dams. Additionally, Lake McConaughy was not completed and could not have produced significant impact before 1941, well after the majority of depletion suggested. The other mentioned reservoirs, Sutherland and Johnson, cannot be considered as capable of "major influences". Despite these inconsistencies, many have taken this position as fact, thereby propagating the myth.
Also, Williams inappropriately omitted data between 1971 and 1974 from his analysis because "in 1971, 1973, and 1974, the various reservoirs had to bypass abnormally large inflows from spring snowmelt, thus producing unusually high peaks for stations in the study reach (Bentall, 1975, p. 7-9)". (Williams, 1978) Williams did not consider in his report the effects of snowmelt and precipitation in the upper basin for the period of the study, except where such influences produced both increased peak and annual flows. In fact, upper basin runoff significantly contributes to downstream flows as described in Simons and Associates 1990 District Joint Deficiency Response, Appendix IV - Hydrology (Simons, 1990). "Water production is thus about 10 times larger per unit area than that in the plains, indicating that a very substantial percentage (approximately 90%) of the water that is produced above the confluence of the North Platte and South Platte Rivers is derived from mountain snowpack." (Simons, 1990)
A review of more current data indicates that no lasting downward trends in peak discharges have been realized. Figure 15 depicts peak discharges at Overton.
MEAN ANNUAL DISCHARGESIn regard to perceived changes in mean annual discharge, Williams speculates that; "By withholding or diverting water which otherwise would flow downstream, dams also lower the mean annual discharges at downstream stations, unless other factors influence the flow." (Williams, 1978) It must be understood that withholding water in a reservoir after its initial fill does not alone significantly influence mean annual discharges. To do so would imply that the reservoir is of near infinite capacity and perpetually able to retime flows. This, of course, is not the case. Once a reservoir fills, its capacity to further capture flows is limited to the amount it is drawn down the previous year. Although greater losses can be associated with reservoir evaporation and seepage, the effects of these losses are slight in relation to annual flow volumes. In fact, most of the seepage loss is generally returned to the river through direct groundwater return flows.
A portion of any water that is diverted for irrigation or storage helps to enhance the local ground water table. The effect of building this groundwater mound is a greater groundwater return to the river in development areas. Although the majority of gains closely follow cumulative precipitation patterns, these gains have gradually and steadily increased in the study area examined by Williams (and throughout the developed basin) throughout the life of the projects as shown in Figure 11. It is conceivable that flow declined during the early operation of each project, because some of the diverted and stored water was being directly contributed to the water table, with little immediate return to the river. However, the most current data indicates that contributions from local gains to the river have significantly increased throughout the basin since then.
As mentioned, Williams omitted the data during the 1971-1974 time frame because of unusually high flows. One must wonder whether it was appropriate to omit the data. We can begin to answer this question by updating the information presented in Williams by including the latest available data. A graphical representation of the data is shown in Figure 2. Although we neither endorse nor recommend Williams' methods as appropriate to identify depletion, the same methods were used for the purpose of consistency. Using the same logic and curve-fitting methodology as Williams, one can see that the conclusions reached by Williams may have been premature. The extended long term trend indicates that the average discharges have continued an upward trend as was suggested by the data omitted in Williams, 1978.
VEGETATIVE ENCROACHMENT / CHANNEL NARROWINGOften, low flows have been associated with vegetative encroachment and subsequent channel narrowing. One of the arguments presented to support this position is that the lack of peak discharges that could periodically remove the young seedlings from the banks of the river channels are reduced. Additionally, bed aggradation and degradation caused by flow reductions could contribute to channel changes, but this effect is only relevant in a channel that is not in equilibrium.
In a 1996 report by W. Carter Johnson entitled Channel Equilibrium in the Platte River, 1986-1995, he concludes "Recent measurements from aerial photographs strongly support my earlier research findings that the Platte River is in dynamic equilibrium, with the exception of a short reach of the lower Platte River near Grand Island. This reach was immediately downstream of an intensively cleared and disked section between Alda and Grand Island. Channel area in the 7 sampled Platte River reaches upstream of Grand Island either did not change or increased between 1986 and 1995." (Johnson, 1996) (See Figure 14).
As has been shown, the mean annual flow in the Central Platte River has been generally increasing since the completion of the Williams report. With these flow increases in mind, another long held myth comes into question, the concern that the habitat of the Platte River continues to degrade. Unless Johnson's contention, that the channel widths have been stable or increasing for many years, is correct, the contention that channel vegetative encroachment is somehow related to man created flow reductions would be illogical.
Figure 14 represents W. Carter Johnson's more recent finds related to channel encroachment. As would be expected if a relationship between flow and channel encroachment exists, the increased flows appear to be decreasing channel encroachment. However, if the recent Trust and Audubon finds are correct, this would suggest that channel encroachment is not a function of flow reduction. Either way, it would appear that the alarm generated by the myth of 70% man made flow depletion in the Platte is unsubstantiated.
PRELIMINARY CONCLUSIONSAlthough this study is a work in progress, enough research has been completed to effectively present several preliminary conclusions:
To quantify flow depletion, one must effectively choose a baseline which
somehow identifies predevelopment conditions. Williams implies that his 1900- 1930 time period could have been representative of predevelopment flow, but an extensive review of available data clearly shows that development had been occurring as early as the 1860s. Also, current data shows increasing flow which would not be possible if the reduction was development induced.
The initial time period, 1900-1930, was very wet compared to prior or later time
periods as used in his analysis in terms of precipitation. The significance of these precipitation differences indeed had a substantial effect on Platte River flows.
This research is ongoing, and the initial results are considered preliminary. There is no mistake, however, that there are significant problems associated with the misuse and misinterpretation of the Williams report. This situation, combined with flaws in Williams' methodology, logic, and conclusions, has led to the propagation of the myth of quantified Platte River flow depletion. Flow depletion in the Platte River during certain time periods cannot be denied, however the effects of climate can be closely associated with much of this flow reduction. It would be premature for us to attempt to quantify a percentage of flow depletion because of man- made influences in this river basin, and no attempt to do so should be implied from this ongoing study.
REFERENCESWilliams, Garnett P., 1978, The Case of the Shrinking Channels – the North Platte and Platte Rivers in Nebraska, Geological Survey Circular 781, 45 p.
Secretary of War, 1934, Platte River, Colo., Wyo., and Nebr. – Letter from the Secretary of War; 73d Congress, 2d Session, House Document No. 197, 533 p.
Teele, R. P., 1905, Water Rights on Interstate Streams: The Platte River and Tributaries – Results of Investigation; U. S. Department of Agriculture, 95 p.
Johnson, W. Carter, 1996, Channel Equilibrium in the Platte River, 1986-1995; Department of Horticulture, Forestry, Landscape, and Parks, South Dakota State University, 24 p.
Simons, R., 1990, District Joint Deficiency Response Appendix IV (Hydrology)
Return to 1997 Platte River Basin Ecosystem Symposium
Last updated by Darren A. Jack on 4/28/97