Q _ T'— — 7 # , 7 W, —— — — __ —7. , , 4 §; i 7 , _ — .. AN ANALYSIS OF HYDROLOGIC SIMULATION AND GROUNDWATER WITHDRAWAL PATTERNS IN EASTERN LANSING, MICHIGAN A Research Paper for the Degree of M. A. MICHIGAN STATE UNIVERSITY RONALD L. VAN TIL 1977 I II II III IIIIIIIIIIZIIIIIIIIIIIII 293 02 III 3 ABSTRACT AN ANALYSIS OF HYDROLOGIC SIMULATION AND GROUNDWATER WITHDRAWAL PATTERNS IN EASTERN LANSING, MICHIGAN BY Ronald L. Van Til Piezometric surface values for 42 Saginaw Forma- tion wells in the eastern Lansing area were used to analyze piezometric simulations projected by the Wheeler- Vanlier electric analog model. Following hydrologic analysis, yearly and seasonal groundwater withdrawal pat- terns for 1975-1976 were mapped and correlated with the 1976 piezometric surface. Significant differences between actual and analog-projected piezometric conditions from 1964—1976 remained unexplained within the existing model design. Consequently, the model was considered unreliable for future piezometric simulation and further investigation was recommended. Actual groundwater withdrawal patterns for 1975- 1976 indicated proportionally larger and more evenly distributed withdrawals throughout Michigan State Univer- sity than in East Lansing and Meridian Township, with Ronald L. Van Til localized piezometric declines in areas of concentrated pumpage in central East Lansing and north central Meri- dian Township and more uniform piezometric declines throughout Michigan State University. Actual withdrawal procedures for 1975-1976 were shown to reflect both hydrologic and jurisdictional circumstances throughout the eastern Lansing area. AN ANALYSIS OF HYDROLOGIC SIMULATION AND GROUNDWATER WITHDRAWAL PATTERNS IN EASTERN LANSING, MICHIGAN BY Ronald L. Van Til A RESEARCH PAPER Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Geography 1977 TO Richard A. Van Til and Lois A. Carter ii ACKNOWLEDGMENTS I would like to eXpress my appreciation to Dr. Jay Harman and Dr. Gary Manson of the Department of Geography, Michigan State University, and Dr. Grahame Larson and Dr. Harold Stonehouse of the Department of Geology, Michigan State University, for their advice and guidance in the preparation of this manuscript. Additional research assistance was provided by Robert Nestle of the Michigan State University Department of Engineering, Floyd Twenter of the United States Geo- logical Survey, and Duane Warren of the City of East Lansing. Special appreciation is extended to Ralph DeVitto, whose careful judgment and attention to carto— graphic detail simplified my task considerably. iii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . LIST OF FIGURES O O O O I O O O C O O 0 INTRODUCTION . . . . . . . . . . . . . REVIEW OF LITERATURE . . . . . . . . . . METHODS O O O O O O O O O O O O O O Groundwater Hydrologic Principles . . . Analog Simulation of Groundwater Hydrology . Data Collection and Analysis . . . . . PRESENTATION OF FINDINGS . . . . . . . . . Eastern Lansing Piezometric Conditions . . 1975 Analog-Projected Piezometric Surface . . . . . . . . . . 1976 Actual Piezometric Surface . . . Eastern Lansing Groundwater Withdrawal Patterns . . . . . . . . . Total Groundwater Withdrawals: 1975 and 1976 O O O O O O O O O O O 0 Seasonal Groundwater Withdrawals: 1975 and 1976 . . . . . . . . . . DISCUSSION OF FINDINGS . . . . . . . . . Analog Model Simulation Accuracy -. . . Analog- Projected and Actual Withdrawal Rates . . . . . . . . . . Precipitation Recharge . . . . . . Hydrologic Variables and Model Design . Summary . . . . . . . . . . Groundwater Withdrawal Patterns . . . . East Lansing-Meridian Township . . . . Michigan State University . . . . . Summary . . . . . . . . . . . iv Page vi vii 11 ll 13 16 22 22 27 3O 33 36 41 47 47 48 52 55 57 58 59 61 63 Page CONCLUSIONS . . . . . . . . . . . . . 65 LIST OF REFERENCES . . . . . . . . . . . 67 Table 1. LIST OF TABLES 1975 analog-projected and 1976 actual piezometric values for the eastern Lansing area . . . . . . . . . 1964 and 1976 actual piezometric values for the eastern Lansing area . . . . . Production well total groundwater with- drawals for 1975 and 1976 for East Lansing, Meridian Township, and Michigan State University . . . . . Water system total groundwater withdrawals for 1975 and 1976 for East Lansing- Meridian Township and Michigan State University . . . -. . . . . Production well seasonal groundwater with- drawals for 1975 and 1976 for East Lansing, Meridian Township, and Michigan State University . . . . . Water system seasonal groundwater with- drawals for 1975 and 1976 for East Lansing-Meridian Township and Michigan State University . . . . . Projected and actual average daily with- drawal rates from 1965-1975 for East Lansing, Meridian Township, East Lansing-Meridian Water and Sewer Authority, and Michigan State University . . . . . . . . . . Annual precipitation, annual departure of precipitation, and cumulative departure of precipitation for 1965-1975 for Michigan State University, East Lansing, Michigan . . . . . . . . . . vi Page 23 25 34 35 37 38 50 53 LIST OF FIGURES Figure Page 1. Eastern Lansing area production wells of East Lansing, Meridian Township, and Michigan State University . . . . . . 17 2. Eastern Lansing area analog-projected 1975 piezometric surface . . . . . . . . 28 3. Eastern Lansing area actual 1976 piezometric surface . . . . . . . . . . . . 31 4. Eastern Lansing area total groundwater withdrawals: 1975 and 1976 . . . . . 39 5. Eastern Lansing area winter groundwater withdrawals: 1975 and 1976 . . . . . 42 6. Eastern Lansing area spring groundwater withdrawals: 1975 and 1976 . . . . . 43 7. Eastern Lansing area summer groundwater withdrawals: 1975 and 1976 . . . . . 44 8. Eastern Lansing area fall groundwater withdrawals: 1975 and 1976 . . . . . 45 vii INTRODUCTION The upper subsurface bedrock of Michigan's cen- tral lower Peninsula is characterized by a complex association of lithologic units of Pennsylvanian age known collectively as the Saginaw Formation. Composed primarily of alternating beds of differentially permeable sandstone and shale with thin interbedded strata of lime- stone and coal, the Saginaw Formation is the primary hydrologic system underlying the lower central Michigan region. In the Lansing Metropolitan area, the Saginaw Formation serves as the principal source of municipal and institutional water supply. Although lithologic characteristics have been determined for the Saginaw Formation, initial hydrologic investigations have provided only a generalized view of the Saginaw Formation as an artesian system. An analy- tical approach was therefore utilized in 1967 in an attempt to determine specific hydrologic characteristics and more exact aquifer response to variance in patterns of municipal and institutional groundwater withdrawal. Using existing hydrologic data, an electric ana- log model was constructed by Wheeler (1967) and further refined by Vanlier and Wheeler (1968) to simulate known 1 past aquifer piezometric conditions in the greater Lansing area. Following adjustment and subsequent agreement bet- ween model projections and documented past hydrologic conditions, model simulation accuracy was assumed. Pro- jections of future aquifer piezometric response to anti- cipated groundwater withdrawals were then made in an attempt to forecast future hydrologic conditions through- out the area. Completion of the electric analog model provided a more-detailed understanding of the specific hydrologic characteristics of the Lansing aquifer system. Model projections of aquifer response to anticipated future aquifer withdrawals further suggested considerable vari- ance in the distribution and configuration of future groundwater reserves. Significant among the 1967 model- analysis projections was the delineation of the eastern portion of the Lansing MetrOpolitan area as an area of potential maximum groundwater decline. Rapid expansion of the aquifer's cone of depression was hypothesized throughout the area in response to anticipated increased groundwater withdrawals by the City of East Lansing, Meridian Township, and Michigan State University. The need for a more precise understanding of pro- jected aquifer response in the eastern Lansing Metropoli- tan area has been demonstrated by the continued decrease since 1967 in available groundwater supplies and substantial projected increases in future groundwater withdrawals. Recorded 1964—1976 piezometric surface values further indicate less extreme piezometric declines than projected by the Wheeler-Vanlier analog model. A research project was therefore designed to analyze the predictive accuracy of analog model simulation and the patterns of municipal and institutional groundwater withdrawal for East Lansing, Meridian Township, and Michigan State University. The primary investigative hypothesis formulated for the study was that differences between actual and model-projected piezometric declines from 1964-1976 would exceed the model—stated predictive accuracy range of i 10 feet. To test the hypothesis, an initial com- parison was made between analog-projected and actual piezometric pressure surfaces for the years 1975 and 1976. Following analysis of analog piezometric simula- tion, yearly and seasonal groundwater withdrawal patterns were determined for East Lansing, Meridian Township, and Michigan State University. Finally, an assessment was made of the combined effect of groundwater withdrawals on area hydrology. REVIEW OF LITERATURE Prior hydrologic investigation of the Saginaw Formation in the Lansing Metropolitan area has emphasized the importance of a regional artesian aquifer system as a primary source of domestic and industrial water supply. Hydrologic investigations have been undertaken by Stuart (1945), Mencenberg (1963), Firouzian (1963), Carroll (1963), Wheeler (1967), Vanlier and Wheeler (1968), and Wood (1969). The first comprehensive investigation of the groundwater hydrology of the Saginaw Formation in the Lansing area was undertaken in 1945 by W. T. Stuart. Saginaw Formation sandstones were described as predomin- ately well rounded and well sorted quartz grains lightly cemented together to form interstices equivalent or larger than the grain sizes themselves. The connecting passages between grains were identified as generally parallel to the bedding planes, with considerable vertical connection within beds and restricted connection between beds. The sandstones of the formation were included within the low to medium permeability range, with higher overall transmissibility values as a result of substan- tial sandstone thickness. Divergent transmissibility and 4 storage values were attributed to variations in sandstone type and manner of sedimentation, with average transmis- sibility and storage values calculated as 23,000 gpd per foot and .000382, respectively. The overall groundwater flow into the area of diversion was estimated as between five to nine million gallons per day, with maximum flow from the south and minimum flows from the east and north. Although direct and solution channel recharge were specified, downward and lateral percolation in areas of contact between formation sandstones and overlying glacial deposits was hypothesized as the greatest probable source of aquifer recharge. Stuart concluded that conditions of groundwater equilibrium in the Lansing area prior to 1935 had been followed by a decline from 1935-1945 in area groundwater levels as a direct consequence of increased municipal and industrial pumpage. The total area influenced by ground- water withdrawals in 1945 was estimated as forty-eight square miles. Stratigraphic characteristics of the Saginaw Formation in the Lansing area were correlated in 1963 by F. Mencenberg. The Saginaw Formation was broadly des- cribed as consisting principally of alternating sandstones and shales deposited in terrestrial and marine environ— ments during the Pennsylvanian Period, with extensive and generally continuous basal sandstones and incontinuous cyclical strata formations. The Saginaw group was defined hydrologically as an artesian system, with occasional elevated or perched water tables associated with discontinuous and unrelated clay lenses in the overlying glacial drift. The genera- tion of artesian pressures within the aquifer was attribu- ted to the presence of confining shaley aquicludes, basin- ward formation dip, and general surface topography. An electrical resistivity study of the Saginaw Formation in the Lansing area was conducted in 1963 by J. Carroll. Lithologic profiles were constructed along a southeast to northwest traverse in the Lansing-East Lanshx; area and an east-west traverse in southern Lansing in an attempt to discern the lithologic characteristics of the Saginaw Formation and the composite thickness of the over- lying glacial deposits. Resistivity measurements were found to approximate actual well profile data for glacial deposits, although inaccuracies were encountered in the analysis of deeper bedrock formations. Hydrologic characteristics of the Saginaw Forma— tion in the Lansing area were investigated in 1963 by A. Firouzian. Average transmissibility for the Saginaw Formation using flow-net analysis was estimated as 23,628 gpd per foot, while direct and indirect recharge were estimated as three million gallons a day and twenty-eight million gallons a day, respectively. Wide areal variations in transmissibility and recharge values were noted. A decline in the piezometric surface of as much as ninety feet in the Lansing area was related by Firouzian to increased aquifer pumpage from 1945-1963, with the deepest cones of depression over areas of maximum pumpage. The area of effective recharge was extended to one hundred and twenty square miles, or 2.5 times the original recharge area estimated by Stuart in 1945. A comparison of actual pumpage and estimated aquifer recharge values led Firouzian to conclude that the Lansing artesian aquifer was in a state of near-equilibrium in 1963. An analytical approach to the investigation of the hydrology of the Saginaw Formation was developed in 1967 by M. Wheeler. Idealized assumptions inherent in earlier theoretical studies were minimized through the utilization of an electric analog model designed to simu- late actual aquifer conditions. Leakage through the upper confining beds of the aquifer was identified as the primary source of aquifer recharge in the Lansing area. Leakage was described as nearly uniform and responsible for a reduction in hori— zontal groundwater movement through the aquifer, thereby lessening the overall importance of transmissibility. The artesian storage coefficient was found to have no effect on long-term drawdowns and the water-table storage coefficient of a small dewatered portion of the aquifer only a small effect. Increased importance was given by Wheeler to the storage coefficient of overlying glacial drift, as addi- tional groundwater withdrawals were found to have incmmsed leakage from overlying glacial deposits to the aquifer, thereby significantly lowering water levels in the drift. Area river courses were further identified as important but generally finite sources of aquifer recharge. Local- ized high recharge areas hydrologically associated with the Mason Esker were identified in northern and southern Lansing. Additional analyses of Saginaw Formation analog simulation in the Lansing Metropolitan area were under- taken in 1968 by K. E. Vanlier and M. Wheeler. A series of future analog piezometric projections were developed to illustrate the importance of analog model application in area groundwater planning and development. Model pro- jections indicated a continuation of the decline in the overall cone of depression in the Lansing Metropolitan area through 1985. In addition, groundwater declines were hypothesized to increase most rapidly in depth and extent in eastern Lansing from 1968—1975 in response to substantial expected increases in withdrawals from East Lansing, Meridian Township, and Michigan State University. Similar though less severe groundwater declines were projected for the southern and western Lansing area. Although groundwater supplies were termed adequate until 1985 by Vanlier and Wheeler, the need for water management and development programs was suggested for post-1985 area water demands. Additional alternatives for future groundwater development were outlined, includ- ing more extensive use of glacial aquifers, artificial aquifer recharge, diversion of high streamflow, develop- ment of more-distant well fields, and importation of water from the Great Lakes. The groundwater geochemistry of the Saginaw Forma- tion in the Upper Grand River Basin was investigated in 1969 by W. Wood. Glacial drift and soils were found to contribute substantially more dissolved solids to Saginaw Formation groundwater than atmospheric precipitation, water migration from adjacent formations, or mineral solution within the formation. Lower dissolved solids concentrations in the Saginaw Formation were attributed to processes of osmosis and ultrafiltration of ground- water by shale members of the formation. Although a state of dynamic ion filtration equilibrium was suggested in hydrologically undeveloped areas of the Upper Grand River Basin, with the continuous removal of dissolved solids concentrations by local cell flow and lateral groundwater movement, the likelihood of 10 increased dissolved solids concentrations was hypothesized in areas of high aquifer pumpage. The probability of increased dissolved solids concentrations was recognized in the Lansing Metropolitan area, with possible consequent decreased groundwater recharge to the Saginaw Formation from overlying glacial deposits. METHODS Groundwater Hydrologic Principles Fundamental principles of groundwater hydrology for the Saginaw Formation are based on concepts of poro- sity, permeability, transmissibility, and storage as they occur in an artesian aquifer system. An aquifer can be defined as a hydrologic unit capable of storing and transmitting water in appreciable quantities. In the presence of underlying and overlying impermeable or semi-permeable strata, an aquifer is con- fined and made artesian. Subsurface aquifer water move— ment is a direct function of the number, size, shape, arrangement, and interconnection of rock pore spaces, or interstices, and the overall hydraulic gradient. The extent and nature of subsurface water move- ment within an artesian aquifer is determined primarily by overall values of porosity and permeability. Porosity can be described as the comparison of interstitial volume to overall rock volume, typically expressed as a percent- age. Permeability is described as an expression of water movement or transmission through an aquifer, as determined by overall interstitial arrangement and interconnection and the occurrence of structural faults, cracks, or joints. ll 12 The coefficient of permeability is expressed quantita- tively as the rate of flow in gallons per day through a cross—sectional area of the aquifer one square foot under a hydraulic gradient of one foot per foot. Aquifer transmissibility represents an overall expression of the permeability of an aquifer. Transmis- sibility is defined quantitatively as the flow of water in gallons through a cross-sectional area one foot wide and extending the entire saturated thickness of the aquifer under a hydraulic gradient of one hundred percent. The coefficient of storage represents the value of water released from or taken into storage per unit aquifer area per unit head, or pressure, change. Under natural conditions, water levels within an artesian aquifer fluctuate within an equilibrium range in response to natural processes of recharge and discharge. A piezometric, or imaginary pressure surface is formed at some level above the confined artesian system. This sur- face is defined as the level to which confined artesian groundwater would rise in a well tapping the aquifer formation. The piezometric surface serves as an indication of both the general configuration and extent of ground-. water supplies below. When artificial groundwater with- drawals are initiated, the piezometric surface is lowered in the vicinity of the wells and a cone of depression is 13 formed. If continued artifical discharge exceeds natural processes of recharge, both the groundwater levels and the piezometric surface in the affected area are lowered and widened. At any given time, therefore, the elevation and configuration of the piezometric surface serves as an indication of overall aquifer conditions. The amount of groundwater which can be withdrawn from an artesian aquifer is dependent on the overall capacity of the aquifer formation and the nature and extent of artificial discharge. Groundwater pumpage in excess of total natural recharge may substantially lower and even- tually deplete groundwater reserves by creating a continual state of groundwater disequilibrium. Analog Simulation of Groundwater Hydrology The hydrologic functions of the Saginaw Formation artesian aquifer were simulated through the use of an electric analog model develOped by M. Wheeler in 1967 and further refined by Vanlier and Wheeler in 1968. An electric analog model simulates hydrologic func- tions and piezometric response by directing an electrical current through a network of capacitors and resistors arranged in a representative fashion. In model design, resistors are used to simulate aquifer transmissibility or permeability, while capacitors are used to simulate aquifer storage capacity. Once adjusted, the movement and storage of electrical current in the model network is considered l4 analogous to hydrologic changes in the aquifer system. Both hydrologic characteristics and piezometric response to various past or future pumpage or discharge schemes can then be determined in greatly reduced time. Design of the analog simulation model for the Saginaw Formation in the Lansing Metropolitan area was accomplished through successive analyses. An initial con- ceptual model of the aquifer was developed using trans- missibility data collected in 1945 by W. T. Stuart. Where unavailable, values of aquifer transmissibility were esti- mated on the basis of sandstone thickness and average permeability. Using these data, low transmissibility zones surrounding the Lansing Metropolitan area were identified and used to delineate natural aquifer boundaries. Following definition of a discrete aquifer, an analysis of the model in steady—state was undertaken to determine aquifer hydrologic conditions prior to develOp- ment. Hydrologic data for 1900 and initial model analyses were used to determine preliminary estimates of natural recharge and discharge values. Although initially over- estimated, general recharge or leakage rates were subse- quently determined for area river segments and eleven regions within the Lansing Metropolitan area. Completion of steady-state analysis thereby provided a more specific understanding of aquifer recharge-discharge functions and potential leakage from overlying glacial deposits. 15 Following steady-state analysis, time-dependent variables were introduced for transient model analysis. A condition of aquifer equilibrium was assumed for 1935 in response to the reported stabilization of the artesian piezometric surface to 1935 groundwater withdrawals. As a result, 1935 conditions were used as a reference base for additional transient analyses. Groundwater pumpage values were determined for all production wells in the Lansing MetrOpolitan area and grouped into a series of pumpage centers to facilitate analysis. Division of groundwater withdrawal rates into step-functions further simplified analysis with arbitrarily identified time periods between 1935-1964. Following the completion of actual 1945 and 1964 piezometric surface maps from existing hydrologic data, groundwater withdrawals for 1935-1945 were imposed on the analog simulation model. Analyses indicated a signifi- cant difference between model-projected and actual aquifer piezometric drawdowns. Significant movement of groundwater from the overlying glacial material into the aquifer system was subsequently identified, suggesting the increased importance of the groundwater storage capacity of area glacial deposits. The identification of areas of high glacial-material recharge associated with the Mason Esker in northeastern Lansing led to further modification of model recharge values. Following 16 additional minor revisions, the analog simulation model was considered representative of the Saginaw Formation aquifer system in 1964. The completed analog model was considered a significant predictive tool in forecasting future piezo- metric surface response to groundwater withdrawals in the Lansing Metropolitan area. Caution was emphasized, how— ever, in the extension of model analyses based on the developed condition of the aquifer in 1964. Additional errors identified as inherent in the analog model included the possible inaccuracy of electrical components, basic hydrologic assumptions of aquifer equilibrium and storage, and previously recorded hydrologic data. Within these limitations, projections of future aquifer piezometric response to anticipated area groundwater withdrawals were expected to be accurate within : 10 feet. Data Collection and Analysis Hydrologic data for the period 1975-1976 were collected for the analysis of analog piezometric simula- tion and patterns of municipal groundwater withdrawal in the eastern Lansing study area. Forty-four municipal and institutional well sites for East Lansing, Meridian Township, and Michigan State University were used for hydrologic analysis (Figure l). Twenty-five municipal production wells are Operated by l7 .huflmum>flcb mumum cmoanoflz can .mwnmc3oa cmflpwnwz .mcflmcmn ummm mo mHHOS cofluosponm mmum mcflmcmn cumummmnl.H musmwm O O: 0...........mom: _ o. ........... 3m: 4.. ..... 952.6% 2:33: I ..... . . 0523.. tow mutm . 44w? .\ .CI\ I 3v... -'~. «I on; «a u e I. . 3m: oz< $55230... 259mm: 625244 hmdw no m44w3 zo_ho:oomm o_ no. 52: a. 6:30 .32 :2 22.3 33.2.; 52:00 E uo