Advancing Water Resources Research and Management

 HYDROLOGIC IMPACT OF CLIMATIC CHANGE
IN THE COLORADO FRONT RANGE

William C. Rense

Professor of Geography
Geography Earth-Science Department
Shippensburg University of Pennsylvania
1871 Old Main Drive, Shippensburg, PA  17257-2299
Phone: (717)477-1662, Fax: (717)477-4029
E-Mail: wcrens@wharf.ship.edu

ABSTRACT: This paper analyzes possible changes in stream flow in the Front Range, Colorado, by applying the Thornthwaite Water Budget technique to the drainage basin of Boulder Creek. The Front Range is a mountain region of extreme environment ranging from steppe grassland to Alpine tundra. Rapid population growth has put increasing pressure on the environment and hydrology of the region. Using climate data gathered by the Institute of Arctic and Alpine research for the years 1952 through 1969, water budgets were calculated for the Boulder Creek drainage on a sequential “real time” basis rather than from a generalized “mean.” The mean annual flow as calculated using the water budget predicted the discharge of Boulder Creek to within 1% of the gauged flow for 1952 through 1969. Stream flow was also calculated for Boulder Creek under several different climatic scenarios of “global warming” and the “Altithermal”, the Little Ice Age, and full Glacial conditions. For the global warming scenarios, temperatures were increased by 2° C and precipitation was allowed to vary from the same as the modern climate to 10% more and 10% less. Stream discharge in Boulder Creek decreased by 11% with no change in precipitation, decreased by 30% with a 10% decrease in precipitation, but actually increased by almost 8% with a 10% increase in precipitation. Assuming modern precipitation values and a 1°C decrease in temperature during the Little Ice Age, stream flow increased by almost 5%. Under full glacial conditions, discharge from the lower 60% of the basin alone would vary from 150%+ to less than 50% of the modern discharge depending upon which precipitation scenario is used.

KEY TERMS: water budget; climatic change; stream discharge

INTRODUCTION

     Climatic change and its effects on both human and natural environments has long been of interest to scientists and planners. Today, with concern about global warming being brought about by the anthropomorphic contribution of so-called “greenhouse gases” into the atmosphere, issues involving the effects of climatic change have taken on a new urgency. Furthermore, in order to better understand the future impact of climate change as well the past development of the planet, there is considerable scientific interest in past changes, especially those related to the Glacial periods of the Pleistocene, the  Altithermal period of warm climates about 6,000 years B.P. (Benedict and Olson, 1978), and the recent cold episode known as the Little Ice Age.

     The Front Range of Colorado has seen enormous population growth in recent decades. The so-called “Front Range Corridor, running for more than 100 kilometers north and south of the city of Denver, is one of the more rapidly growing regions of the United States and now contains almost three million people. The principal source of water for this burgeoning population is the Rocky Mountains lying immediately to the west. The Front Range not only serves as a major source of this water, but it must also absorb the environmental impacts of this urban corridor in terms of recreation and general development.

     This paper uses the Thornthwaite Water Budget methodology to calculate surpluses and thus estimate stream flow under various conditions of climatic change within the Front Range. The focus is on the relatively well-instrumented drainage of Boulder Creek, west of the city of Boulder.    Scenarios involving “global warming” (CO₂), the Little Ice Age and full late Pleistocene Glacial conditions are considered.

METHODOLOGY

     John Marr, in his seminal work on the ecology of the Colorado Front Range, founded the Institute of Arctic and Alpine Research at the University of Colorado with the goal of pursuing ecological research in mountain and high latitude environments. Marr (1961) recognized four ecosystems on the East Slope of the Front Range and placed climatic instrumentation in each of these with each site chosen to be representative of that ecosystem. These four ecosystems were, with increasing elevation, the Lower Montane Forest, the Upper Montane Forest, the Subalpine Forest and the Alpine zone. The Bioclimates of the Front Range have been discussed more recently by Greenland, et al (1985). The climatic stations first established by Marr in the early 1950’s were designated as A-1 for the Lower Montane, B-1 for the Upper Montane, C-1 for the Subalpine, and D-1 for the Alpine location. These stations all lie within the drainage of Boulder Creek. These original designations will be used in this paper. The purpose was to develop the data from Marr’s four stations in order to calculate a climatic water budget for the Boulder Creek drainage and then model runoff from this basin under different climatic scenarios

Data Selection

     All four of Marr’s stations were operated from the early 1950’s until about 1970 by the Institute of Arctic and Alpine Research. From that date onward, data were collected consistently only from the two higher locations, C-1 and D-1, and those two stations remain in operation to this date. The lower two stations, A-1 and B-1, were not maintained continuously after 1970 and, in most years, only precipitation and no temperature data were collected. However, a complete data set of mean monthly temperature and total monthly precipitation exists from 1952 through 1969 for all four ecological zones. Therefore, this paper uses a 17-year period of complete instrumentation encompassing Water Years 1953 through 1969 for analysis.

     The data used in this paper were collected some years ago by this writer by using original field and in-house data sheets. Published data were not used.

Thornthwaite Water Budget

     The Thornthwaite water budget technique has been demonstrated to effectively model stream flow and runoff (Mather, 1978; Mather, 1981; Carter, 1958). This writer (Rense, 1999) has shown that the Thornthwaite method is more accurate if it is calculated in “real time”, that is, on a continuous month to month basis, rather than if it is calculated from long term “means.” The key hydrologic factor in the water budget is “Surplus.” Surplus theoretically represents the water remaining in a landscape after soil moisture and evapotranspiration demands have been satisfied. Surplus, therefore, represents runoff. Other terms within the water budget have ecological application and significance (Greenland, et al, 1984). In order to be conservative, the important “heat index value”, which determines the rate of evapotranspiration at any given temperature, was recalculated for each climatic scenario used in this paper.

     This paper utilized the climatic data from the original Institute of Arctic and Alpine Research data set (1953 – 1969) to calculate water budgets under various climatic scenarios for the Boulder Creek Drainage. Each station was assumed to be representative of its entire ecosystem. The Boulder Creek drainage was digitized and the area of each ecosystem was determined. Then, using the calculated surplus and area, the total discharge in acre feet could be estimated for each ecosystem and, therefore, for the entire basin.  Under the modern climate (WY 1953 – 1969), the Thornthwaite method calculated the average annual discharge of Boulder Creek to within 1% of the gauged flow (Boulder Creek at Orodel).

     Calculations in this paper were done on the “real-time” year to year basis rather than from the long-term mean. As described previously, this “real time” methodology is preferable to the use of “mean” values as it provides for both better accuracy from the water budget calculations (Rense, 1999) and is also a true reflection of actual weather conditions  month by month within the basin. Thus, the 1953 – 1969 data set can be presented in a realistic, time sequential manner under all climatic scenarios merely by changing the temperature and precipitation values.

Possible Sources of Error

     There are several possible sources of error in this study. First, the City of Boulder withdraws water from Boulder Creek as its headwaters are the watershed for Boulder’s supply. This means that the “real” flow of Boulder Creek is larger than the gauged flow and that this study is probably underestimating flow by some amount. Second, despite the good instrumentation available, there are still only four climatic stations in a rugged mountain basin of over 250 square kilometers (100 square miles). Third, there are significant problems with precipitation collection at the Alpine station (D-1), the most important when it comes to stream flow. This is a zone of high wind motion in winter and much drifting. Under these conditions, it is difficult to accurately collect precipitation and it must be assumed that error exists in the data set from this zone. Other sources of error include the effect on annual stream flow from the presence of small glaciers within the basin and a storage reservoir. The water impounded in this reservoir was used for hydroelectric generation within the basin, but there are evaporative losses from the reservoir surface and the effect on timing of runoff as water is stored and released. The exact impact of these possible errors is not fully known.

CLIMATIC SCENARIOS

     Three major climatic scenarios were investigated. First was a condition of “global warming” as might exist under conditions of increased carbon dioxide in the atmosphere, or as might have existed some 6,000 years ago during the Altithermal. Second was a scenario for the recent Little Ice Age. The third scenario was for the full glacial climate of the late Pleistocene.

Global Warming Scenarios

     For a global warming scenario, as might exist under carbon dioxide forced warming or as might have existed during the Altithermal, a conservative increase in temperature of 2°C was selected (Houghton, 1997; p. 95). Then, three scenarios of precipitation were considered – warming with no change in precipitation, warming with a 10% increase in precipitation, and warming with a 10% decrease in precipitation. Water budgets were then calculated for the water years 1953 through 1969 for each of these three scenarios and runoff estimated from the resulting surpluses. Of course, there is uncertainty about the effects on global warming on both seasonal temperature change and on precipitation. Temperature may increase an average of 2°C, or more or less. But, the increase might be more in winter and less in summer, not equal in all months as was assumed for this paper. Under global warming, precipitation may stay the same or increase or decrease. For this paper, it was assumed that changes in precipitation would occur equally in all months of the year. In reality, some season may become drier and other seasons wetter with global warming. However, to try to model all the possible permutations was beyond the scope of this paper. It was felt that a 10% plus or minus range in precipitation would conservatively cover changes that could occur with warming.

Little Ice Age Scenario

     The Little Ice Age is one of the more interesting climatic periods since the end of the Pleistocene. The Little Ice Age was documented, at least indirectly, in many parts of the world through historical records, art, records of glacial advances near villages or passes, and so forth. In North America, though, there were few such records. Therefore, following Lauritzen and Lundberg (1999), who identified a 1°C decrease in temperature during the Little Ice Age in Norway, a similar change in temperature was assumed for the Colorado Front Range. Precipitation was assumed to have remained the same as the modern climate. Water budgets were calculated for the Boulder Creek drainage under these cooler conditions.

Full Glacial Scenarios

     In a recent paper, Leonard (1989) identified the equilibrium lines of the Front Range glaciers during the late Pleistocene. From these equilibrium lines, he developed scenarios of climate that would have explained the intensity of glaciation. He speculated that the late Pleistocene glaciers could have been supported either with a cooling of 8.5°C and precipitation the same as today, or a 10.4°C cooling and precipitation half of today’s levels. Under these temperature conditions, a water budget would be superfluous at the subalpine and alpine zones since subfreezing conditions would exist virtually year-round. Runoff from these elevations would be dependent on the relative accumulation/ablation balances of the glaciers and snowfields. Therefore, water budgets were computed only for the lower ecosystems of Boulder Creek, the Upper and Lower Montane zones, using Leonard’s two scenarios of temperature and precipitation. No attempt was made in this paper to estimate the significant contribution to runoff from glacial ablation in the Subalpine and Alpine zones.

DISCUSSION

Modern Discharge, Calculated vs. Gauged

     The USGS gauge at Orodel indicated an annual mean discharge for Boulder Creek during the 17-year period of WY 1953- WY 1969 of 59,824 acre feet. The water budget calculation for the same period estimated a discharge of 60,310 acre feet, less than 1% above the gauged flow. (See discussion of possible sources of error above.) However, on a year to year basis, there was some variation between the gauged and calculated flows. For example, the water budget estimated a discharge of only 21,931 acre feet in WY 1954 when the gauged flow was 34,589. In 1965, the water budget estimated a flow of 97,320 acre feet when the gauged flow was only 85,193. These variations are probably due to such factors as groundwater storage, melt rates of snowfields in the alpine zone, storage or release schedules from the reservoir on Boulder Creek and errors in precipitation collection.

     The Water Budget technique also allows an estimate to made concerning the contribution to flow in Boulder Creek from each ecosystem. Although common knowledge is that “all flow” comes from snowmelt in the Alpine and Subalpine zones, it appears that the lower two Montane zones are more important than may be thought. An average of 19,519 acre feet were contributed annually over the period of this study, almost 1/3 of the calculated discharge of Boulder Creek. The relative importance of the Montane zones is greater during wet years and reduced during dry years.

Global Warming  Scenarios.

     Figure 1 shows the annual calculated discharge of Boulder Creek under the global warming scenarios as compared to the modern calculated discharge for each water year 1953 through 1969. The major drought years of 1954 – 1956 and the unusually wet years of 1957 and 1965 are very obvious.  It can also be seen that stream flow will decrease under all scenarios of global warming except one, that of increased precipitation.

(click on image for full page view of graph)

     With a 2°C increase in monthly temperature and no change in precipitation from modern values, annual discharge of Boulder Creek is estimated to decrease by 6,474 acre feet to a flow of 53,845 acre feet, 89.5% of the modern value. This is a significant decline. The Lower Montane zone shows no surplus under this scenario in two years, 1954 and 1963. However, if precipitation is assumed to decrease by 10% with global warming, the impact on runoff is dramatic. Under this scenario annual average discharge declines by 17,794 acre feet to only 42,525 acre feet, 70.5% of the modern value. Such a decline would have major implications to water supplies in the adjacent urban corridor of Colorado. In the complete gauged record of Boulder Creek from 1907 through 1969, a runoff value as low as this has occurred only five times, in 1910, 1954, 1955, 1964 and 1966.

     Should precipitation in a globally warmed environment increase by 10%, then the losses resulting from higher temperatures and evapotranspiration would be offset and the calculated discharge of Boulder Creek would actually increase. A wetter, warmer world would result in a calculated discharge of 64,981 acre feet, an increase 4,662 acre feet and 108% of the modern calculated annual discharge.  Total calculated discharge of Boulder Creek under all climatic scenarios except full glacial is given in Table 1. Table 2 shows the calculated runoff as a depth of precipitation (surplus) by ecosystem.  

Table 1. Total Calculated Average Annual Discharge of Boulder Creek Under Various Climatic Scenarios
(Acre Feet and Percent of Modern Value)

Table 2. Runoff Depth (Surplus) in Inches by Ecosystem and Climate Scenario

Little Ice Age

     The small amount of cooling postulated for the Little Ice Age in the Front Range, 1°C, would result in a slight reduction of annual evapotranspiration and an increase in surplus and runoff. As can be seen from Table 1, calculated discharge from Boulder Creek increased by almost 5% during this period of colder climate. This amounts to an additional 2,877 acre feet per year. Some 62% (1,790 acre feet) of this additional flow is contributed by the Lower and Upper Montane zones, where the cooler climate has a bigger impact on evapotranspiration and surplus than it does in the colder Subalpine and Alpine zones.

Full Glacial Scenarios

     A calculation of the total Full Glacial discharge of Boulder Creek lies beyond the scope of this paper due to the complexity of accumulation and ablation in the large areas of glacial accumulation above elevations of about 2800 meters. However, some estimates are possible of runoff from the modern Lower and Upper Montane Zones. Earlier in this paper reference was made to the work of Leonard (1989) who estimated climatic factors in the Front Range under full glacial conditions in the late Pleistocene.   Leonard used the equilibrium lines of the Front Range Glaciers to model the possible combinations of temperature and precipitation necessary to support these glaciers. Leonard developed two possible scenarios. The first assumed that precipitation was at modern values and that ablation season temperatures were 8.5°C. colder than today. The second scenario was one of assumed dry glacial climate in which precipitation was only half the modern value and temperatures were 10.4°C. colder than today. Leonard felt that either of these two scenarios could support the late Pleistocene glaciers of the Front Range.

     Under either of these temperature regimes, a water budget analysis would be inapplicable to the Subalpine and Alpine zones. Temperatures here would be at or below freezing most of the year, so the concept of “evapotranspiration” becomes meaningless. However, calculations can be done for the modern Lower and Upper Montane. Under Leonard’s first scenario of modern precipitation and a temperature decrease of 8.5°C, runoff from the Montane zones together increases by 155% above the modern value, a discharge of over 30,000 acre feet per year. The modern combined discharge is about 19,519 acre feet and the geomorphic implications of the glacial runoff is obvious. However, under Leonard’s “cold and dry” scenario, runoff would have decreased dramatically to 39% of the modern discharge, only 7,646 acre feet per year.

SUMMARY

     Calculations using the Thornthwaite water budget indicate that climatic change has a significant impact on the hydrology of the Colorado Front Range. Global warming scenarios with only a modest 2°C increase in temperature show that runoff from the Boulder Creek Basin would decrease by more than 10% with precipitation  unchanged from the modern value, and would decrease by almost 30% with a 10% decrease in precipitation. However, a 10% increase in precipitation would result in a 7%  increase in runoff.  During the Little Ice Age, runoff in the Front Range was probably greater than today, an increase of almost 5% in the Boulder Creek Basin. Full Glacial scenarios are more difficult to calculate, but runoff from the Montane zones would be dramatically greater or less than today depending upon the temperature/precipitation values that are used.

REFERENCES

Benedict, J. B. and B. L. Olson, 1978. The Mount Albion Complex: A Study of Prehistoric Man and the Althithermal.
Research Report No. 1, Center for Mountain Archeology, Ward, Colorado.

Carter, D. B., 1958. The Average Water Balance of the Delaware Basin. Drexel Institute of Technology, Laboratory of
Climatology, Publications in Climatology XI (3): 249 – 269.

Greenland, D;  J. Burbank; J. Key; L. Klinger; J. Moorhouse; S. Oaks; and D. Shankman, 1985. The Bioclimates of the
Colorado Front Range. Mountain Research and Development 5(3): 251 – 262.

Greenland, D.; N. Caine and O. Pollak, 1984. The Summer Water Budget and its Importance in the Alpine Tundra of
Colorado. Physical Geography 5(3): 221 – 239.

Houghton, J., 1997. Global Warming: The Complete Briefing, 2^nd Ed.  Cambridge University Press, Cambridge, U.K.

Lauritzen, S. and J. Lundberg, 1999. Calibration of the Speleothem Delta Function on Absolute Temperature Record
For the Holocene in Northern Norway. The Holocene 9(6): 659 – 669.

Leonard, E. M., 1989. Climatic Change in the Colorado Rocky Mountains: Estimates Based on Modern Climate at
Late Pleistocene Equilibrium Lines. Arctic and Alpine Research 21(3): 245 – 255.

Marr, J. W., 1961. Ecosystems of the East Slope of the Front Range in Colorado. University of Colorado Studies,
Series in Biology, No. 8, University of Colorado Press, Boulder, Colorado.

Mather, J. R., 1981. Using Computed Stream Flow in Watershed Analysis. Water Resources Bulletin 17(3):
474 – 482.

Mather, J. R., 1978. The Climatic Water Budget in Environmental Analysis. D. H. Heath Co., Lexington Books,
Lexington, MA: 39 – 65.

Rense, W., 1999. Improving the Accuracy of the Thornthwaite Water Budget Stream Discharge Estimations.
Paper presented at the annual national meeting, Association of American Geographers, Session 3320,
March 24, 1999, Honolulu, Hawaii.

home page
Maintainer: AWRA Webserver Team
Copyright © 2000 American Water Resources Association