71-2142 PAWLING, John W . , 1929MORPHOMETRIC ANALYSIS OF THE SOUTHERN PENINSULA OF MICHIGAN. Michigan State University, Ph.D., 1970 Geography U n iv e rs ity M icro film s, A XEROX Com pany , A n n A rb o r, M ich ig an COPYRIGHT 1969 By John W. Pawling MORPHGMBTRIC ANALYSIS OF THE SOUTHERN PENINSULA OF MICHIGAN ! i l V' ji By John W • Pawling A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Geography 1969 ABSTRACT MORPHOMBTRIC ANALYSIS OF THE SOUTHERN PENINSULA OF MICHIGAN By John W* Pawling The results of this study incorporate a morphometric description of the major topographic compartments of the southern peninsula of Michigan and the eleven glacial landforms which char­ acterize its terrain* The primary configuration of these landforms occurs within a gross delineation of two dissimilar highlands sur­ rounded by a discontinuous peripheral lowland. Of the highlands, the larger northern highland has greater relief, slope, and eleva­ tion in comparison to the lower and less rugged southern one* Each highland and the encompassing lowland is defined on the basis of continuous hypsometric characteristics above or below the 750 footisohypse* Because of the absence of previous morphometric research in the study area, the use of traditional terrain dimensions such as local relief, average slope, and average elevation is adopted in support of the geometrical characterizations of morphometric regions and glacial landform types. 2 1 /2 A grid network of 6798 -minute unit areas, conterminal with the peninsular shores, serves as a matrix of data locations for the computer manipulation of the Blevation-Relief Ratio, and a new terrain parameter, the Comparative Relief Index* T h e latter index, devised for the p u r ­ pose of delineating homogeneous relief regions, failed to achieve significant results because of the lack of abrupt regional changes in local relief* Bach of the morphometric highlands is divided into four levels, called morphometric regions, with continuous altitudinal values* Bach level is further compartmentalized into morphometric subregions based on concentrations of stated classes of relief within core locations* Transitional belts of relief units not com­ mon to either of two proximate relief cores are shared, equally and arbitrarily, with each core area; the combination of relief cores and their anomalous units provides an index of relief heterogeneity and is advocated as a differential index in formalizing morpho­ metric subregions* Thirty-four isolated and distinctive relief cores, in con­ junction with their aggregates of unlike relief values, are identi­ fied in an effort to differentiate the changing topographic surface of the peninsula* Because of the good correlation between increas­ ing slope and relief, and the frequent appearance of certain landform types with given relief slope criteria, both slope and p r e ­ vailing landform characteristics are stated as additional descrip­ tions of the morphometric subregions* It is not possible to describe the peninsular lowland as a flat lacustrine plain, in its entirety, nor is it always possible to e s tab­ lish correlations between increased elevation and more rugged relief oj steeper slope* Relatively flat till and outwash plains character­ ize portions of the northern and southern highlands and some of the most rugged topography of the study area is found at lowland elevations along the northwestern littoral* The principal physio­ graphic lineaments of the study area include the amphitheater-like rim of morainic heights which outline northern highland and the northeast-southwest trending moraines of the eastern and western faces of the southern highland* The two highlands are set apart by the trough of the Grand-Maple rivers which rival the incisements of the Manistee, Muskegon, and A u Sable rivers in the north­ ern highland* In contrast, the clay bottomlands of the eastern side of the peninsula commonly have a relief of less than 25 feet and an average slope of less than 1/2° per six-square-mile unit area* For sake of comparison, the morainic heights of the northern highland and the drumlin fields of the Traverse City lowland frequently have slopes in excess of 4° and relief in excess of 350 feet* In terms of the landform type of the largest areal extent, and within the constraints of generally subdued topographic surface, recession­ al moraines are the principal relief-makers of the peninsula* The future of regional morphometric research, implicit in meeting the exigencies of this work, rests in the innovation of flat— land parameters, automated procedures of data acquisition from topographic maps, and increased accessibility to computers* The principal deficencies in applying morphometric theories are concerned with formalizing the ideal unit area concept, delimiting morphometric subregions on the basis of integrated or weighted terrain datay and devising a rigorous method for determining the class intervals of the data. ACKNOWLEDGEMENTS The author wishes to acknowledge the applied virtues of integrity, perseverance, and dedication which this research repre­ sents. To his major professor, Dieter H. Brunnschweiler, the author expresses his appreciation for the rigorous standards of scholarly writing and cartographic execution which Professor Brunnschweiler imposed at each stage in the development of this thesis topic. The perseverance to complete this research arose out of seven years* encouragement and continuing financial assistance from the au t h o r 1s parents, Mr. & Mrs. John A. Pawling. To his father, the writer is grateful for the admonishments and persistent criti­ cisms of the unfinished task to his mother, for the quiet and certain knowledge that the task would one day be complete. Dr. H e n r y H . Michael, To the author is grateful tor his critiques of the intermediate drafts of this work. Finally, the author is in­ debted to Betty Famiglietti for typing the final manuscript and its careful proof-reading. In dedication, the author offers the merits of this work to his son, John Scott Pawling, with the hope that he will discover in his time the personal satisfactions of scholastic endeavor* Any errors contained in this work are the responsibility of author alone and should not be attributed to any other person or persons. TABLE OF CONTENTS Page ACKNOWLEDGEMENTS ii LIST OF TABLES vi LIST OF ILLUSTRATIONS viii Chapter I. II. MORPHOMETRY AND ITS APPLICATION TO MICHIGAN *S SOUTHERN PENINSULA Purpose and Methodology The Field of Geomorphometry Description of the Study Area The Unit Area Concept in the Study of Genetic Landform Types and Morphometric Regions Status of Topographic Coverage CLASSIFICATION OF TERRAIN PARAMETERS Rationale for the Selection of Parameters Average Elevation Maximum and Minimum Elevation Local Relief Average Slope 'I'U i, U C f? 1 ,-.»»•>! 4-^ .— > » » ■ " * T >« V «A 1 ^^ D 1 22 ^« iV M v a , v Comparative Relief Review of Diagnostic Indices III. MORPHOMETRIC ANALYSIS OF LANDFORM TYPES Introduction Primary Genetic Types Ground Moraine (Till Plains) Recessional Moraines Outwash Format ions Complex Terrain (Unclassified Formations) Lacustrine Plains Lake Bed Formations Minor Lacustrine Features Linear Landform Types Drumlins Sand Dunes Minor Linear Features iv 59 Chapter General Comments on the Histograms of Landform Types IV. V. TERRAIN REGIONALIZATION FOR SOUTHERN MICHIGAN Introduction The Problem of Terrain Regionalization The Hypsometric Factor The Relief-Slope Factor T he Genetic Factor Analysis of Morphometric Provinces The Lowland Province The Southern Highland Province The Foreland Region of the Southern Highland The Intermediate Uplands of the Southern Highland The Uplands of the Southern Highland The Hillsdale Highland of the Southern Highland The Northern Highland The Foreland Region of the Northern Highland The Intermediate Upland Region of the Northern Highland The Upland Region of the Northern Highland Assessment of Regionalization and Comparison With Other Works 89 EVALUATIONS AND PROSPECT OF TERRAIN REGIONALIZATIONS BY MORPHOMETRIC METHODS 125 Rurifipsti ons for Futurp Rpsparrh Th e Need for an Automated Morphometry APPENDIX 133 BIBLIOGRAPHY 140 v LIST OF TABLES Page Table 1. 2 . 3. 4. 5. 6 Area of Surface Formations of Southern Michigan: Summary of Genetic Types after Helen M. Martin (op. cit.) 6 Status of Topographic Coverage for the Southern Peninsula of Michigan 19 Comparisons of Area and Traverse Lengths at Varying M a p Scales 42 Derivation of the 4:1 Conversion of 50-foot Contour Counts 44 Frequency of Landform Types in Unit Areas 60 . Characteristics of Drumlin Fields 80 Characteristics of the Sand Dune Littorals 81 Key to the Identification and Nomenclature of Terrain Subregions 95 Table of Hypsraetric Levels Selected for the Differentiation of Terrain Regions 98 10. Classes of Local Relief and Average Slope 99 11. Local Relief Classes as a Percentage of Average Slope Classes 7. 8 . 9. 11a. Percentages of Unit Areas in Specified Classes of Slope and Relief in the Northern and Southern Highlands 12. 13. 101 114 Number of Unit Areas in the Various Regions and Subregions and Percentages of Units Occurring in Specified Classes of Slope 134 Number of Unit Areas in the Various Regions and Subregions and Percentages of Units Occurring in Specified Classes of Local Relief 135 vi Table 14. 15* 16. 17. Page Number of Unit Areas in the Various Regions and Subregions and Percentage of Units per Landform T y p e in Given Regions and Subregions 136 Percentages of Specified Classes of Average Slope Pound in the Various Regions and Subregions 137 Percentages of Specified Classes of Local Relief Found in the Various Regions and Subregions 138 Percentages of Specified Genetic Landforms Pound in the Various Regions and Subregions 139 vii LIST OF ILLUSTRATIONS County Reference Map . ......... . . . . . . . . . 9 Recessional Moraines lO Principal Drainage Systems Outwash . . . . .. 7 . . . . Ground Moraine: Till Plains .. . . . . . . . . . 11 12 • Lacustrine Formations 13 Unit Ar e a Reference Ma p 16 Grid Network of Unit Areas, Traverses, and Spot Elevations . . . . . . . . . . . . . . . . . 18 Topographic Map Coverage 20 Average Elevation . . . . . . . . ................ 28 Maxinun Elevation 30 Minimum Elevation 31 Local Relief 35 Average Slope 41 Elevation-Relief Ratio .............. • ........... 48 Selected Blevation-Relief Ratios 49 Inequalities of the Blevation-Relief Ratio . • • • 53 Comparative Relief and Regional Development 53 Comparative Relief Ground Moraine: Till Plains viii • • • Figure Page 21* Recessional Moraine 22« Outwash . 66 68 23. Clay Lake Bed . . • . . » » • • • . • • • • • • « 73 24. Sandy Lake Bed 74 • • • • . • • • • • • • • • • • • 25* Selected Formation Types 76 26. Complex Terrain (Unclassified Unit Areas) 27. Average Elevation: Survey of Unit Areas • * • • 73 • • • • . 86 28. Local Relief: Survey of Unit Areas . • • • • • • 87 29. Average Slope: Survey of Unit Areas . . . . . . . 88 30. Terrain Regions 92 31. Terrain Subregions • • • . • • • • • • • « . • • 94 32. Lowland Subregions • • • • • • • . • • • • • • . 105 33. Foreland Subregions 108 34. Intermediate Upland Subregions 112 35. Upland Subregions 119 36. Highland Subregions • • • • • » » • • • • • • • » 120 CHAPTER I MORPHOMETRY AND ITS APPLICATION TO MICHIGAN fS SOUTHERN PENINSULA Purpose and Methodology Geographers are committed to the description of landscape in time from both a physical and a cultural point of view* The physical landscape consists in part of topographic surfaces which represent a variety of individual landforms; because of the re­ lationship with other physical or cultural phenomena, these surfaces must be the subject of objective description in any geographic regionalization: "Descriptive landform analysis, as an objective statement of facts, should be the first step in any type of land­ form study * * * once the descriptive analysis is completed, we can utilize the information for either genetic or functional studies." Zakrzewska (1967: 131-132) The purpose of this dissertation is to provide a morpho­ metric analysis of the landform types and the terrain regions of the southern peninsula of Michigan. The dimensional reality of the prevailing glacial landform types can be defined according to measurements of slope, relief, and elevation. The study area ex­ cludes the northern peninsula of Michigan because this region is a physically separate appendage of the state of Michigan; furthermore, the geographical outline of the southern peninsula exhibits a true peninsular shape and comprises the largest peninsula of the con­ terminous United S t a t e s * T h e primary task of this study is the selection of proper terrain parameters needed to produce an objec­ tive and systematic analysis of southern Michigan landform types and their regional characteristics* The literature on quantitative techniques used in landform analyses is extensive* Neuenschwander (1944) offers a bibliography of 640 items , for the period before 1944, and Carr and Van Lopik (1962) provide a review of 326 studies appearing in the interim* A n overwhelming number of these references are concerned with parameters which reveal minor details and individual features of a topographic surface, and a review of the more specialized quan­ titative measures would serve no purpose here* Several studies, concerned with the more limited litera­ ture of morphometric regionalization, apply a single terrain index to relatively large-scale study areas such as a single drainage basin, a mountain massif, or an isolated valley* This study em­ ploys a multiple-factor approach: six terrain parameters were found to be satisfactory in identifying the principal morphometric c om­ partments of the largest peninsula of the conterminous United States* These parameters are concerned with the classical mor p h o ­ metric indices of maximum and minimum elevation, average elevation, local relief, and average slope* The morphometric analysis of glacial landforms, first mapped by Leverett and Taylor (1924) and ^ ^ F l o r i d a is larger (55,000 sq* mi*) than the southern peninsula of Michigan (40,000 sq* mi*); however, t h e present study area is slightly larger than peninsular Florida alone (excluding the panhandle area)* revised by Martin (1955), forms an important link between a strict­ ly quantitative and a genetic evaluation of the study area* Although Hoy and Taylor (1963) suggest an even more com­ prehensive approach to include change in slope direction, other researchers commonly employ only a single factor in developing topographic regionalizations* Examples of the latter include the use of relief in Illinois and Ohio by Calef (1953) and Smith (1935), respectively, and the analysis of slope in New England and Illinois by Raisz & Henry (1937) and Thoman (1955), respectively* Hammond (1955) and Pike (1963), alone, apply integrated terrain factors for Missouri and southern New England in order to develop geometrical concepts of regional terrain surfaces* The Field of Geomorphometry Geomorphometry may be defined as the subfield of geomor­ phology which employs dimensional values in describing landforms* Statistical characterizations of terrain formations, leading to a higher order of regional description, is a major objective of no r - (2A phometric research*' ' Discrete terrain characteristics have to be isolated, evaluated, and translated into numerical statements in order to express the surface geometry* Wood and Snell (1960:1) state this objective as follows: "Because numbers cam be manipulated and carry a preciseness of definition that qualitative terminology lacks, quantitative analysis has increased the value of terrain studies for both theoretical and practical purposes •" (21 ' 'The shorter term "morphometric" will be used in place of geomorphometric in the remainder of this work* 4 Goldberg (1962:537) comments further that: “The problem, therefore, is to d i s ­ cover a method of terrain analysis which will provide both a realistic and a quantitative description of world landforms, the one serving as a key to the o t h e r «w Although improved quantitative expressions for topographic models are clearly within the realm of advanced theoretical mathe­ matics (topology), the task of characterizing landforms and landform regions can be accomplished by conventional geometrical relation­ ships extracted from topographic maps. It is probable that these improved parameters will utilize basic terrain dimensions, such as slope, elevation, and relief, in either new relationships with each other or as primary data in newly devised mathematical concepts. Description of the Study Area The topography of the southern peninsula consists of a peripheral lowland of varying width surrounding two highlands, a northern and a southern, which are separated by the Grand River corridor. The structure, materials, and orientation of landforms occupying these surfaces are mainly the result of widespread depo­ sition during the later substages of the Wisconsin stage of glaci­ ation. Glacial drift is unevenly distributed over the peninsula varying in depth from 100-300 feet in the southern highland to a maximum of 1000 feet in t he northern upland. The total area of these drift materials amounts to approximately 40,000 square miles and incorporates a maximum regional relief of surface of Lake Brie 1200 feet from the (565 feet) to the morainic heights (1706 feet) near Mesick, Michigan. The underlying bedrock surface of a structural basin has had little effect in controlling the surface configuration of landforms although buried cuestas may be related to morainic heights along the outer perimeter in the northwestern sector of the peninsula. Drift is deepest in the interlobate moraines of Otsego County and the West Branch Moraine of Ogemaw and Clare counties. No similar correlation can be made between the thick­ ness of the drift mantle and the primary topographic heights of Hillsdale County; however, secondary heights in the Kalamazoo Moraine and the interlobate moraines of Oakland County have some­ what thicker drift accumulations. In every case, the 100-300 feet of glacial overburden of the southern highland points to a bedrock surface which is uniformly closer to the present topographic sur­ face than is the case in the northern highland. The "Map of the Surface Formations of the Southern Penin­ sula of Michigan,** Martin ( 1 9 5 5 ) , summarizes more than fifty years of field work in identifying and locating the various glacial formations of the study area. Table 1 lists the glacial landforms, differentiated in this work, and indicates the fragmentation and areas of the various surface formations. The formation parcels, in turn, determine the prevailing landform type in the identifi­ cation of unit areas in the morphometric analysis of glacial landform types (Chapter 3). {31 ' ‘'Hereafter referred to as M.S.F. Map 6 TABLE 1 Area of Surface Formations of Southern Michigan: * Summary of Genetic Types after Helen M. Martin (op. cit.) Regional Type Recessional Moraines Number of Parcels** Area in Sq. Miles Average Parcel** Size (Sq. Miles) Percent of Total Area 681 11,860 17.5 30 Outwash and Glacial Channels 70 9,130 130.4 24 Sand Lake Beds and Spillways 38 6,595 173.6 18 270 5,270 19.5 14 Clay Lake Beds 49 4,062 82.9 11 Waterlaid Moraine 24 550 19.5 1 Ponded Water or Interior Lake Formations io 441 44.1 1 9 275 30.5 1 20 119 5.4 --- 2 45 22.7 ——— Q /1 '7 55*0 Ground Moraines (Till Plains) Drumlins Sand Dunes Boulder Belts A U T -*■»— ' •3 0 lOO Determined with use of Bruning Areagraph Set No. 4850, degree of precision stated to be at 97% of area surveyed. ** Parcels are defined as irregular-shaped land units of homogeneous genetic landforms. The surface formation map clearly depicts the pattern of recessional and interlobate roaorines as arcuate residuals of the ice lobes of Lakes Michigan, Huron, Saginaw, and Erie. The best linear expressions of these moraines occur along the outer fringes of the northern and southern highlands; hundreds of small, isolated 7 COUNTY REFERENCE MAP EMMET f RESQUC I SLE ANTRIM OTSEGO OSCODA ■ENZIE ALCONA IO SC O MASON LAKE OSCEOLA CLARE HURON OCEANA MIDLAND , A V MECOSTA TUSCOLA I KENT 41 OTTAWA IO N IA EATON ALLEGAN INOHAM AVNE CALHOUN C AS S JACKSON BRANCH LENAWEE •I M Pig. 1 •I morainic parcels dominate the interlobate areas of the southern highland and dozens of irregular tracts occur within the rim mo­ raines of the northern highland (Fig* 2)* cipal drainage systems of moraines* it If the map of the prin­ (Fig* 3) is superimposed on the pattern becomes apparent that moraines control the locally dominant heights and are* in fact, adjusted to their locations in alternating outwash belts* Till plains (Fig* 5) are typically associated with mo­ raines and occupy major tracts in Gratiot* Montcalm, Clinton, Shiawasee* and Genesee counties* Their segmented linear pattern is not surprising here* but* where tills are found in association with outwash and lacustrine formations, their pattern and outline are highly irregular* Later* the morphometry of till surfaces will be shown to vary from nearly slope-less surfaces to hills of rugged relief; however, the generalization that till formations occur mainly at intermediate elevations in moderate slopes and relief will also be demonstrated* Lacustrine surfaces (Fig, 6 ) occupy nearly one-third of the peninsular study area (Table 1), With the exception of inte­ rior proglacial lake beds and certain sandy river bottoms of the southern highland* fully 98% of the lacustrine surfaces occur below the 750 foot-contour on the peninsular lowlands* Not shown in Figure 6 , owing to the small scale of the map, is a discontinuous belt of lake-border sand dunes extending from the Leelanau Peninsula to Benton Harbor trated in Fig, 6 (Fig* 25)* Clearly illus­ , however* is the glacial Grand River spillway now occupied by the Grand River and its tributary from the Saginaw REC ES S I O N A L M O R A I N E S 10 Fig. 3 11 O U T W A S H Fig. 4 12 G R O U N D Fig. 5 13 LACUSTRINE FORMATIONS Leg end | | (USED f l l f I m MTEILilB ISM IIE H IIIIIE I m CLIT Lite IEDI Ull 1(11 ill 1 ^ EUE IIICII1 C IIIIH I 'l u l l ll (Irl NnlliP Fig. 6 L 14 Basin, the M aple River; this narrow corridor provided a westward outlet for a succession of impounded glacial lakes (Saginaw, Maumee, and Whittlesey) to the east* Outwash plains (Pig* 4) extend over approximately onefourth of the study area (Table 1). They are particularly evident at higher elevations in the northern highland where they spill over the Lake Border Moraine and occur as major landforms, at lower elevations, in Newago, Lake, Manistee, and Benzie counties* The outwash trains of the southern portion of the peninsula are found in conjunction with both moraines and tills, however, and their areal pattern is more tenuous than the more extensive, blocky tracts in the northern highland. The topography of the western half of Charlevoix and Antrim counties, including the Old Mission and Leelanau peninsu­ las, is dominated by drumlins which are among the locally most prominent landforms of the peninsula* Bskers, locally conspicuous microfeatures of the study area, are found almost exclusively to the south of the Grand-44aple line* The Unit Area Concept In the Study of Genetic Landform Types and Morphometric Regions A grid network of 6832 unit areas was superimposed on the National Topographic Maps of the study area in order to obtain terrain dimensions of the landform types included in each unit* The decision to use the 2 1/2-minute grid was made after the 5minute and 7 1/2-minute grids, used in previous studies, failed to provide a fine-mesh reticulation suited to the restricted areal 15 extent of most individual landforms of the study area* A standard** size unit area was chosen in place of the variable-size landform parcels of the surface formation map (Table 1 and Figs* 2-6) because of the bias toward increased ranges of terrain data taken from the larger parcels in each landform category* In addition, a carto­ graphic problem arose in transferring the spectrum of parcel sizes and shapes between the larger-scale topographic and the smallerscale surface formation maps* All morphometric data are obtained from a standard 2 1/2minute sample area* These data, in turn, are used as the basis of terrain descriptions given for the various landform types* As mentioned earlier, the reticulation of this uniform grid is compat­ ible with the grain or texture of topography in the study area* However, the accuracy of terrain description for unit areas is based upon the accuracy and completeness by which they are rendered on the surface formation and topographic base maps* It is obvious that such insufficiencies, should they exist, cannot be controlled by the author• Bach unit area of the uniform grid comprises a 2 1/2-minute rectangle aggregating 6*12 square miles between 42° and 43° North Latitude, 6*02 square miles between 43° and 44° North Latitude, and 5*93 square miles between 44° and 45° North Latitude* This slight reduction in area is a result of the polar convergence of meridians with increasing latitude* A 14*4 mile system of linear transects, in the form of a diamond outline superimposed on the unit1s diag­ onals (Fig* 8; Panel 2), is part of the physical design of each unit area* Spaced as northeast-southwest and northwest-southeast lines *88 miles apart, these transects facilitate the acquisition 16 UNIT AREA REFERENCE MAP i t f a i l i a u i i i i i i i i i i i i g a i i i i i i i i i i i i « i § « M * i i i i bbibbb a a a a a a i ! ! ! ! ! ! ! ! ! ! ! ! • i a k i « a a « a l i a i j a a E i a a i l i B 4 Ba ■ i * a i « a i M a « i i t a a i i a la i a n t J i > t t a a > j a « M ! * i t * B i i n ••••••■ •■ •■ ■ ■ •■ ••■ ••■ ■ a B iii* * ia iiia n a iia B ia a ta * a a ia iB a ia a a a iia a « ia a iiiB B il» a B i« a r ■ ■ • ■ • • ■ ■ ■ ■ ■ ■ a iiiiiiiH B B a ii* * a a a n a a a ia ja a « * a * iiB iii« a a ia a t a n aaaaaaaa • a a a i a a i a i a a a ■ ■ ■ • « ■ » * • ■ • • ■ « » « • i B a i t a i a a u a a a a a i a a a a i f l a a i i a B i a t i a a a i a a a a a a B M M a a 11 ••a « S a » » ii» riia a i5 i« a iia ii a a a a a lia a la a i ta iR v a a a a iiB ^ a B B iB iia a a r iiB B a a ia a a a a B t aaaaaaaaaaaaai •^■■aaaaaaaai a a a a a a a a a a a a a ia a a a a a a a a a a a a •B aB aaaaaB afea^ aaiB B B aiaB rB aaaaaiaaaaaaiaB B B B iaaa aaaai r a a a n i i i a n a a a i a a a i a i ; * " ■■■•*£ a a a a ia a iiB iiB a a a a ia a iB lia M M i*B B iia a ta ftia a a a a iB a a B B ia raaaaaB B aM B aB aaa aaaaaarf .■SiiS i B B B B i a a B 11 b i b b b b b b b b b r b * 5 5 5 5 1 1 ! ■ 5 5 M , M ', , , i l a a a i B B t a a a a a a a a B a a a a i i b b b b b b b b b b b b i a a B a a B a a a ■■ a a a a a a i B f l M . ! ! ! * S H t l B B t i i » » « iB ia a a *iB B iia a a a a a a a ta a a » ia a » a a a a ta B a ia ia a B a a a B a » B a a B r« a a a a B a a « a a a ia a a a i a a a B ia a a aaaaaa iB B a a * B a B B a a * * * * * f * * * f i- * * a * * *!!!!!!!1 558185 • ^ • ■ b l « i i 4 « * » U a t » i t * t « « ahiiaa«ii . A ■ i ^ ■ a * # » * : * & ' A ' * ? f 1 7 a * t*t or 4§J #! |«CM i f iC * ««I ««■ **» * • I V«r*,* K laa ff t t s. #*r* « t * , Tig. 15 E S S E N T IA L L Y AVERAGE ! l ir I m IS - 34 S3 — 44 ■ <1 - 34 ■ $3 - 14 ■ IS - 74 ■ IS ir a i n □ II 1471 LEVEL ELEVATION 49 SELECTED E L E V A T I O N - R E LI EF RATIOS PERCENT o f LAN D • • ESSENTIALLY ABOVE A V E R A 6E [~W1 II ir lilt m R R ir n n n rw n ' i s '>t • •• ■ '> ; ■ ft • • ■ ■ " •■ ar' • . v v r . . • . • % ' . •• ,•••*••••„ . • T • • • I M . # • • .**s\ r: •:' ‘— • • • • • . M- • ••••• ••*• ■ 'J • • • • • • *_ « r ♦ •# •*+ •ft• • • H+ • R■ »|I •** « ♦ •IMt ♦i ••• ••R* •• •• • R «• ♦ ♦ •: • • • • H ••••— .—• ^ — •• • • fl ••. •• • •• H * L -. *>i• . — _ ■ > J _■ _■ ■ Ji m • *“ . P r • ■ ■ •■ • . v.-- * . # “ •m a ii n n * V • . Fig. 16 .i ■ I* . f* ■> mm .5 / * • h j LEVEL ELE VA TIO N 50 bias the B-R index in a unit area with a single peak and produce an abnormally low E-R ratio* However, this negative aspect of the r a ­ tio did not come into play because of the absence of isolated peaks significantly higher than other nearby summits. The infrequent appearance of such discrepancies is explained by the fact that 82% of the unit areas have a local relief of less than 200 feet. The B-R ratio is particularly sensitive to the difference between average and minimum elevation values in the absence of marked local relief. The utilization of minimum elevation, than maximum elevation, rather takes into consideration the probable e x i s ­ tence of a local base level within the unit and the likelihood of essentially level surface at lower elevations. The range of B-R ratios, represented in Fig. 16, is broken down into three categories: (1) 0%-34%, with lowland surfaces b e ­ low the average elevation prevailing, (2) 35%—64%, with intermedi­ ate surfaces predominating at or about the average elevation, and (3) 65%-100%, with upland surfaces above the average elevation predominating.^4 ) Approximately 26 % of the unit areas occur in the first category. This fact indicates that one-fourth of the peninsular sample units contains some topographic heights which are isolated features of the terrain surface. Unit areas of the first category appear to be widely scattered throughout the study area, although minor concentrations occur in the dune belts of the western littoral and on the highland plains near Houghton Lake. (1 4 ) v 'The sections labeled "No D a t a ” on Fig. 16 occur in areas for which large scale topographic coverage was unavailable. It was not possible, at map scales smaller than 1:63,360, to obtain the four spot elevations needed to establish a valid average el e ­ vation statement. 51 Only 4% of the study units occur in the third category (65%-100%), and they feature a preponderance of highland surfaces, above the average elevation, in comparison of their respective cells. It is not surprising that the v-shaped profiles suggested by this category are concentrated along the inner portion of the Detroit-Huron lowlands where several rivers are entrenched on the southeastern slopes of the southern highland (see Fig. 16). The unusually high proportion of unit areas in the inter­ mediate category (left blank on Fig. 16), amounting to 70% of penin­ sular area, leads to the conclusion that upland and lowland distri­ butions within unit areas tends to concentrate about the 50% modal value. This is a clear indication that anomalous terrain features, i.e., single or isolated summits jutting up from a lowland surface, or, conversely, marked depressions on highland surfaces, are rela­ tively insignificant. The lack of evidence for such terrain anomalies may be re­ lated to the preponderance of unit areas with subdued relief. Ad­ ditionally, even slight errors in estimating spot elevations may lead to changes of 15% to 25% on the E-R ratios when relief values are less than 25 feet. local relief values. Only 165 unit areas, however, exhibit such Because of the concentration of E —R values in the middle register, this parameter supports the conclusion that marked relief contrasts of surface formations are the exception rather than the rule in depicting the peninsular landscape. Comparative Relief The Comparative Relief index, hereafter called the C-R index, was developed in an attempt to test a method for identifying 52 contiguous regions of contrasting relief. The index expresses the degree of relative change by comparing the local relief of a central unit area with the combined differences of local relief values in the eight unit areas surrounding it (Fig. 18, Panel A). This re­ lationship may be expressed as follows: CR LR where (CR) ^(LR) (JW) is the Comparative Relief index, the local relief of the central unit, and the sum of the positive and negative change of local relief in eight surrounding units. The C-R index should be able to delineate regions of little internal change of relief as well as lines or 2 ones of higher index values within the reticulated grid network of unit areas. Such zones (Fig. 18, Panel D) would distinguish adjacent regions with dissimilar but internally consistent relief values. The two regions of dissimilar internal relief, one with consistently reduced and the other with consistently higher relief values, will produce higher C-R values along their borders than within either of the two regions. The C-R index is a sensitive parameter in detecting subtle changes in local relief such as occur on the subdued terrain of lacustrine formations. tivity. Panel A of Fig. 18 demonstrates this sensi­ The central unit (dashed lines) has a local relief of 10 feet and is surrounded by eight unit areas ^ach of which have a local relief of 20 feet. The sum of the difference in relief b e ­ tween the central unit and the surrounding ui-.Vfcs amounts to 80 feet which, when divided by the local relief of the.- central unit, p r o ­ duces a relatively high index value of 8 .0 . A secondary use of this index is found in its ability to 53 t he of Inequalities E le v a tio n - R elief ES SEN TI AL LY LEVEL TERRAIN SURFACES, Ratio . Elevation "V E* - A°°oV°° * LACK < TERRA IN 900- 800- LEVEL SURFACES Averajje Elevation 700 _ 800 - 700 _ = . 50 200 Fig. Comparat i ve Relief 17 and (A) Regional Devel opment (C) (B) 110 110 110 1100 1100 j n o o 110 100 110 1100 tooo j n o o 110 110 i 110 1 1100 1 i i o o [ n o n .... S g X = 8 0 ft. I 1 1 C R = 8,0 CR s 8.0 CR = 8.0 S b X = 8 0 0 ft. S 8 X = 8 0 ft. CD) 10 10 10 100 100 100 1.0 1.0 27 27 1.0 1.0 10 10 10 100 100 100 1.0 1.0 27 27 1.0 1.0 10 10 10 100 100 100 1.0 1.0 27 27 1.0 1.0 10 10 10 100 too 100 1.0 1.0 27 27 1.0 1.0 Zone of Local T r a n s i t i on Zone Rel i ef CR Fig. 18 of Tr ansi t i on Index 54 RELIEF COMPARATIVE Ind ex e l R e la tiv e Change: lo c a l R e lie f □ M- M ■ I I - «J 1U and Abava * if if if Fig, 19 if 55 delineate rugged terrain of homogeneous internal relief. The higher A expected relief summations (^LR) are offset b y the larger absolute relief values common to the units of such regions* For example, Panel B of Fig* 18 demonstrates that regions of high, but similar internal relief (lOO feet) produce a lower C-R value (0.8) and co n ­ firms the coherent relief character of the region. Other relief summations, on the same order of magnitude (Panel A), will result in substantially higher index values in regions of subdued relief* The fact that the same C-R value may characterize two d i f ­ ferent local relief summations constitutes an inherent weakness of this parameter: tral unit with the C-R value of 8*0 could represent either a cen­ 10 feet of relief surrounded by units with 20 feet relief (Panel A) or a central unit with relief of 1000 feet sur­ rounded by units with relief of llOO feet (Panel C). Obtaining a relative percentage of change in local relief obscures the magni­ tude of absolute change between the central unit and the eight units surrounding it. The index is not suited to the delineation of relief boundaries where (1 ) the horizontal distance required for a change of relief is great, and (2 ) where there is a lack of h o ­ mogeneous internal relief within an area. The results of applying the C-R index to the study area are shown in Fig. 19. Because 84% of all unit areas have an index value of less than 5, the C-R parameter failed to produce an indi­ cation of a transition zone (see Panel D, Fig. 18) between two h o ­ mogeneous regions of unlike relief. The high frequency of low values (61% of all units have a value of less than 3) is due to the A overriding character of the reduced composite change (?LR) for O 56 terrain of generally subdued relief* Littoral unit areas, as cen­ tral units, do not exhibit high C-R values because the sunnation of relief change does not include surrounding units if they had more than 50% water surface* In summary, it is not unexpected that the map of Compara­ tive Relief characterizes the peninsular study area as a geometri­ cal surface with few regions of sharply contrasting relief; however, it is possible that transitions between unlike homogeneous relief regions are so gradual that the Comparative Relief parameter is unable to localize these changes* Rev i e w of Diagnostic Indices Five of the seven parameters discussed in this chapter (excluding the E-R ratio and the C-R index) represent the extraction of specific geometrical identities from the topographic surface of the study area* Bach of these parameters succeeds in isolating a discrete terrain dimension but fails, individually, to portray the topographic diversity of the peninsular landscape* T he usefulness of any one of the diagnostic maps is based on its association or correlation with one or more of the remaining parameters and, to­ gether, the five diagnostic indices verify the existence of certain topographic features which are not evident on a topographic map of a comparable scale* Certain physiographic features such as land corridors, topographic fronts, highland cores, etc., identified as a raapcontinuum on one or more of the morphometric maps, are more clearly represented by an areal-symbol in the grey-tone mosaic than by a line-symbol at the scale of the included maps (1 :2 ,0 0 0 ,0 0 0 , approx­ 57 imately). The contour lines of a topographic map of similar scale, with a contour interval of lOO feet, would serve only to designate the major hypsometric levels of the peninsula: most of the p romi­ nent topographic features identified in the preceding chapter are not readily identified on such a map. The map of average elevation is essentially a simplifica­ tion of the small-scale topographic map, but a judicious selection of class intervals of the data has produced a representation of the amphitheater-like highlands, the secondary relief cores, and topo­ graphic fronts of the two major highlands. As a diagnostic index of drainageway locations, the minimum elevation map provides an ex­ cellent basis for comparing the longitudinal profiles of the major rivers of the study area; also, the map of maximum elevation effec­ tively isolates the peninsular summit areas which are even less o b ­ vious on the topographic map described above. A prior knowledge of maximum and minimum elevations, in unit areas, is essential to an interpretation of regional differ­ ences in local relief. Increased relief values may not always be due to higher maximum elevations (which normally increase toward the interior), but rather to the influence of minimum elevations which fail to increase toward the interior. Similar values of m a x ­ imum elevation in both a peripheral and an interior position could produce disparate relief increments based on the change, or lack of change, of minimum elevation. Therefore, all three indices must be integrated in order to assess the significance of any one of the parameters. Local relief and average slope, highly correlated but ne v ­ ertheless ordered abstractions of topographic roughness, reflect 58 the attribute of contour—density in topographic source maps* They fail, however, to capture the hypsometric configuration, topographic heights, and major drainageways of the peninsula* of the terrain subregions — The continuity based on topographic texture — is best served by the combined application of these two parameters, and evidence is presented in Chapter 4 in support of these being the primary diagnostic criteria in the geomorphic regionalization* CHAPTER III MORPHOMBTRIC ANALYSIS OF LANDFORM TYPES Introduction The morphometric analysis of the major glacial landform types constitutes one of the two major goals of this study* The classification of these landforms was first given and their distri­ bution mapped by Leverett and Taylor (1915) and revised, on the basis of additional field work and air photo analyses, by Martin (1955). In the present study, each landform type is evaluated on the basis of data derived from average elevation, average slope, and local relief (Figs* 27-29)* The range and prevalence of terrain data is discussed under separate sub-headings for each genetic landform type* Regional anomalies and the relationship of glacial processes to specific landform types are sought from an anlaysis of the appropriate mor­ phometric parameters* The histograms of elevation, slope and relief (Fig* 27-29) summarize a primary, lacustrine, an d linear classifi­ cation of the existing landforms to facilitate this analysis* The description of glacial landforms is based on a survey of only those units in which more than 90% of the area is repre­ sented by a single landform t y p e * ^ ^ The locations of these units ^ ^ T h i s requirement limits the analysis to 37*4% of the 59 TABLE 5 Frequency of Landform Types in Unit Areas Regional Type Unit Areas: 90% or more Coverage Percent of All Unit Areas Unit Areas: 50% - 90% Coverage Percent of All Unit Areas Total Number Unit Areas Percent of All Unit Areas Recessional Moraines 507 7.45 1084 15.83 1591 23.40 Outwash and Glacial Channels 539 7.93 798 11.74 1337 19.67 Sandy Lake Beds and Spillways 486 7.15 486 7.15 972 14.30 Ground Moraines 300 4.41 623 9.16 923 13.58 Clay Lake Beds 346 5.01 272 4.00 618 9.09 Waterlaid Moraine 20 •29 55 .81 75 1.10 Ponded Water 38 .56 79 1.16 117 1.72 127 1.87 127 1,87 •Eskers 98 1.44 98 1.44 ♦Sand Dunes 67 .99 67 .99 ♦Boulder Belts 21 .31 21 .31 852 24.60 852 12.53 ♦Drumlins No Prevailing Landform Type TOTALS 3401 6798** 3397 (1 ) 1 0 0 .0 0 % v 'Based on "Map of Surface Formations of the Southern Peninsula of Michigan,” Martin (1955)* (2 ) . . ' 'Lacustrine formations of ponded lakes in the interior of the study area* * Included on the basis of a trace appearance to a 100% coverage in unit areas* ** Does not include 34 unit areas with 50% or more of water surface* 61 are presented in Pigs* 20-26; these maps should be compared with maps illustrating the areal extent of the same landforms shown in Figs* 2-6. Counts of unit areas with dominant glacial landforms are summarized in Table 5* Groups of unit areas with noted percentages (50%-90%, 90% or more) of a given formation are expressed as a pe r ­ centage of all units of the study area* Primary Genetic Types The eleven glacial landform types, (2 ) derived entirely from the M*S*F* map, have been grouped into a three-fold classification of primary, lacustrine, and linear surface formations* Glacial formations occupying units to the extent of 90% or more account for 2549 of the 6798 units of the study area* They are analyzed, ac­ cording to type, in Table 5 and Figs* 27-29* Bach genetic classi­ fication contains four specific formations, including those "un­ classified" under the primary category* The primary classification includes four of the areally most extensive landforms of the study area; 63% of the sample units are assignable to the primary group which includes ground moraines, recessional moraines, outwash, and complex or mixed topographic forms* The latter category represents composites of at least three landforms so distributed that none constitutes as much as 50% of a 6798 units of the study area* Drumlins, sand dunes, and eskers have no coverage requirement for the purpose of this survey* (2 ) ' 'A twelfth or "unclassified" category of mixed landform types is discussed on p* 6 8 . 62 unit area* Because unclassified forms occur in nearly one-fourth of the sample u n i t s , they are included in the primary group* GROUND MORAINE (Till Plains) A sample of 300 units containing till plains gives morphometric validity to this second most common landform of the pen­ insular study area (see Table 1)* Fig. 27 indicates a 73*6% con­ centration of the sample at elevations of 700-950 feet* Nearly three—fourths of the till plains have a local relief range of 25-125 feet and fully 92% of the units have a relief of less than 200 feet* Ninety percent have average slopes of less than 3° with an approximately equal distribution in Classes 2-5 (Fig* 29)* Approximately 70% of the units representing ground moraine are located south of the Saginaw—Grand River demarcation, at eleva­ tions of less than 950 feet; the second maximum (1100-1250 feet) of the bimodal curve (Fig* 27) represents morainic drift fields of the northern highland* They account for 10% of the sample and are found principally in Osceola and Missaukee counties* (3) Another 10% of the sample occurs at 800-950 feet level along the southern fringe of the northern highland (Montcalm and Ionia counties)* An isolated area of till plains, at 750-850 feet, lies to the west of Alpena and comprises approximately 8% of the sample* With the exception of low relief and gentle slope on the till flats to the northwest of Lake Houghton, slopes in excess of ( 3) ' 'Statements pertaining to average elevations of a regxon or county are based on work maps (not included) with a fifty-foot interval; the 100 foot-interval of Fig. 10 utilizes six grey-tones, the maximum number capable of reproduction in this work, to repre­ sent the included range of elevation* 63 GROUND M O R A I N E : TI LL PLAINS Area Coverage 90% or more H 50% to 90% 1X1 Not Included W a te r 1 : ! I 13 Fig. 20 Area 64 2° and relief of 100-300 feet express the surface characteristic of till on the high plains of the northern highland* In contrast, more than half of the ground moraine west of Alpena have a relief of less than 100 feet and average slope of less than 2 °. An east-west trending belt of till plains, at general ele­ vations of 800-900 feet, extends from the southern fringe of the northern highland (Ionia County) across the valley of the Grand River and eastward to Lapeer County* It accounts for another 20% of the sample and contains 70% of the till areas with a relief of 50-100 feet* Yet another 20% of the 300 unit sample is located north of this belt, bordering the Saginaw lowland at general ele­ vations of 700-800 feet* The remaining 30% of the sample is found along the flanks of the Thumb Upland (Hillsdale to Oakland counties) at elevations of 850-950 feet; the majority of these units have re­ duced slopes of l°-2°* More than 90% of this group has intermediate relief of 75—125 feet* RECESSIONAL MORAINE Recessional moraines encompass the largest area (507 units) of steep slope and high local relief (see Tables 1 and 5)* imately 20% Approx­ of the moraines sampled have average slopes in excess of 5°; in contrast only 3% have slope of less than 1°* Relief of 100-250 feet characterizes 60% of these hilly formations* Only drumlins have as much as 5% of their units in the upper third of the relief register (400 feet or more) while moraines are ranked second with 4*2%* However, recessional moraines account for nearly one-half (43%) of all units in this register, whereas the smaller drumlin sample makes up only 14%* 65 Although moraine formations occur at all elevations above 600 feet* more than one-half (58%) are in the range of 750-1050 feet. All moraine units above these elevations* comprising 35% of the sample* are concentrated in Osceola and Otsego counties; these moraines are associated with the highest elevations of the entire peninsula* T he belted morainic hills of the southern highland lie at much lower elevations (Figs. 2 and 10). The Kalamazoo Moraine* for example* has average elevations of only 650-750 feet; the lower portions of this morainic system are thus positioned on t he Muskegon lowland* The interlobate moraines of Oakland County have the great­ est local (250—300 feet) south of the Saginaw-Grand River line* occurs at 900-1100 feet* compared with lesser relief of the morainic Irish Hills It (100-200 feet) (Hillsdale County) at elevations of 1050-1200 feet. While 2°-4° slopes characterize the moraines on the Thumb Upland* the interlobate moraines of Barry and Kent counties exhibit less relief (100-150 feet) at lower general elevations (700-950 A feet) and surprisingly steep slopes in excess of 4” . This regional characteristic is due to the finer terrain texture of the morainic belts in Barry and Kent counties; the higher average elevation is thus the result of larger areas in slope* per unit* rather than significantly steeper slopes for individual landforms* Although the Thumb Upland receives 2” -4” greater average annual rainfall* the interlobate complex of Barry and Kent counties have a more complete development of higher-order tributaries to the Grand River and produce greater annual runoff totals (Ash, 1954). The shorter rivers and smaller drainage basins of the Thumb Upland 66 MORAINE RECESSIONAL aaaaaaa n ia a a a i • ■■ i ait'*a i isaaaaiB bi BBB BBB aba aaa a a a i a a a a i t l l l i i l i a i aaaa aiaaiataiaatanaaaa B BBBB a aa Area aa a i a a a a a a a a a a a i i i l i a n . a i a iiiaiiaaiiiaaaaiaaBBiiaaai a b b a ii a a a i i i a i i a i i i i i ••■■■■ C ove r a g e 111BI 11 B11B BIB11 • 11 ■ •■ ■ « • B la aaaii a a i i a i i i a i i i f M iaaai [" •I aaa m a a *a b «i a ibibii b b b i b b b a a a «■ a i B IB B 8 B B B B B B B B 1 B B B B BBIIIIIB* B B B IB IB B B B B I B IB B I B B aiaaaaB aB B iaiB B B ii laat aaB a *a « aa a aB iibbbbbibbiii ib b b b a iiiaaiaaaaaiaiBBta iiaiiia aar b a a aaaaaiai a a a iB B a a a aa a a a iibbi aaaaa a ibibbbbibbbbbi B a a B B ia a a a iB B atia aaiaa m bibbbb B B aaaiB aaatl aaaaa a a aaB aflaaaeaiai 90% or mod 50% (o 90% ■aaaaaaaa •a Not \Z\ a ibbbi a a a B a a a B i a i B a i aaiBiaaaaaaiaia taaa a a a i B B B B bbbbb a ffaaaaaaataiBi aaa Watar a a a a a bbbbibb a a B B B B a a a i a i. . . . a t B B a a a a a bbb « <■■■■■■ • aaBaaBBBBBBB aaa B BBBi BBBBi a aaataaaaaia a laaB aiiiaaiiiiaiaiB taaaaaaa It* ■■ ■i* II IB l« I B I 1 B ibbb bibb B a B B a B B B B B B v a a B a a B B a a B a a *■ ■ ■ ■ ■ ■ iii i a i aa ii a t aiai m i a i i B a B B i i n i i B a a i i i i i B a i i i l B i a i a *bb ■■ •• ■■■! • a aaaabbbbbb b • ■ ■ ■ ■ i i m i b b aaBBaiBiiaaaiaiBiiaBaBBBBBaiaaaiBB bbbibb bbbbb «■ ■iiiaiianaaiaf a a iiiiiib n a a i BiBaiBaaaiBiBaaaaiaaiaaBBaiaaa ataa « a » b b a a a a a a a a a a a bb l a a a a t B B t a a a a a a u n i a aBBiaaiiiiaBiaaaaiaiaatiiBBaai aaa i i a i B i a M i B B ■■ a i B B a a i i i a a a a i a i a a a B a a a i i i i i a a a a aaa b i aiaiiaaiiiiaBaiiaBiiBiiitii ■ aa •llaaaaaaiiai bi ib tana biBbbibii a a a laaaaaiB aaaB iaaaataB aai a* b ib b * ... ■t a i n t ■aaaaai i a * in a aa a a a a laaaaaB B aaaB aiB ataaaiaaitiaaaa | ■ ■ ■ i,!.!!;, .............■ »** ■ ■■■■■■;: ■aaaa a a a a a inaai aai aaataaaai a « a a aaaaa an aaaaaaaaaaaaa iiiaiaiittifiBB ■ ■ , iifiiftiiiii aiB aaaaa a a ai ibbbi ................. laaaaiaaaiaaaaiaiaaiM ■ aaaaaaaaa a lataai .................. , ,,l, , , !, S!!!.!!!i.!iiiiir taaaaaa aaaaa bib a a a iiaaaB aB taB iaaB aiiB B ai l l B aB aaaiiaaiB B ii a bibbbbiibbbbbbiib -- . . . . . . . . . . . . . . . ■< ■ ■ ■ ■ ■ ■ a ■ a aaa a aaaiI ab b aaaaaa H* aaa aaaaaaan aaaaaa aaaaaaaaa aaa aa a a a a a a a a a rtf? ■l i i a a a i n n a i i f i iia a a a ia iia a ia ia aaa n a a n a a a ia 1 1 a a i a a i aa laaaaa a a ia ia a a a a ia a aaa aaaa aaa m ir m a aa ibbb ■n tBaBaaaaBiiaiaaaai b b a a a a a a ................... ... aB IIB B llllM IB B B I M i ic if Fici. 2 1 r If Indudod Ai m 67 are in marked contrast to the intricate channel network of the Grand River system. Morainic areas of less than 700 feet elevation, constitu­ ting 2% of the same, are restricted to the western lowlands of the peninsula — - half of them occur in the Outer Kalamazoo Moraines and the other half in the lowlands east of Ludington and Manistee* The footings of all these recessional moraines were well within the reach of the fluctuating levels of glacial lakes in the Lake M ichi­ gan basin. OUTWASH FORMATIONS Outwash plains and recessional moraines constitute the pre­ vailing landforms of the northern highland and its associated low­ lands; 79% of the outwash and 69% of the moraines, from a composite sample of 1046 units, occur north of the Maple-Grand River line. Outwash trains parallel the major northeast-southwest belts of hilly moraines over most of the highland. Parallel outwash belts bound the Roscommon Moraine, at lower elevations, along its inner or northwest facing marnin. In contrast, the topographic heights of the northwestern rim are characterized by outwash formations, rather than morainic hills, with 82% of the units bearing a relief of 100-200 feet. However, the northwestern rim is flanked by outer and, of course, lower morainic belts of even greater relief. The long western highland rim, capped with morainic hills, is without a similar inner belt of outwash trains. The largest continuous area of these fluvioglacial materials, 30% of the total sample, lies to the west of the highland rim. In turn, only 30% 68 OU TWAS H 4 90^ 5i Not 01 In c lu d e d W a te r bb h■ ) Fig. 22 i: Bi m o te Area 69 of this regional aggregate has relief in excess of 150 feet and only 13% less than 75 feet* By contrast, nearly half (47%) of the ou t ­ wash plains of the southern highland and more than half (52%) of the Roscommon outwash trains have a relief of less than 75 feet. The comparative analysis of outwash and till plains p ro­ duces an unexpected result: outwash, i.e., a fluvioglacial landform, manifests steeper slopes and greater relief than till plains (a constructional landform resulting from the uneven deposition of glacial debris by an ablating ice mass)* More than one-third (36%) of the outwash terrain has a relief of 125-250 feet in comparison to 22% for the till fields. Slightly more than half (57%) of the outwash areas, but fully three-fourths of the till plains have a local relief of less than 125 feet. Similarly, one fourth of the fluvioglacial units have an average slope in excess of 3° while only 11% of the till do. Also, the former occur over a greater range of hypsometric levels than till plains; they appear nearly twice as frequently as till (34% to 20% respectively) at elevations above 1000 feet. COMPLEX TERRAIN (Unclassified Formations) Unit areas with a mixture of genet landform types consti­ tute a major topographic category in this study. Such formations are characterized by at least three different landform types so that no one formation constitutes as much as 50% of the unit's area. The term 'complex terrain' is used for the lack of a better term. Approximately one-fourth (852 units) of the total sample (3401) consists of mixed topographic forms, a distinctive, though 70 arbitrary, landform "type” of the peninsular study area. Approximately 80?6 of the unclassified unit areas have an average elevation of 650-1050 feet and local relief of 50-250 feet and result in negatively skewed distributions in their respective histograms (Figs. 27 and 28)• A more balanced distribution occurs in the range of average slope classes (Fig. 29). It is not pos- wible to account for these variations on the basis of morphoroetric differences because of the diversity of landforms present. Fig. 26 indicates no significant areal concentration of these mixed f o r ­ mations although unclassified units are notably absent in areas of extensive lacustrine deposition. Lacustrine Plains The lacustrine category includes sandy lake beds, clay lake beds, ponded water formations, and waterlaid moraines. Ponded water formations, unlike the other three types, occur as inland, proglacial lake deposits; they are the residual products of fluvially eroded lake deposits. Unlike sand and clay lake bed deposi­ tion, these impounded finger lakes produced sedimentation basins down-slope from and perpendicular to the ice front. However, simi­ lar processes of quiet-water deposition occurred in both instances; yet, the mantling effects of sedimentation in the finger lakes is incomplete in comparison to flatter lake bed topography at lower elevations. Water-worked moraines, non-lacustrine in origin, are included in this grouping because of their genetic relationship to surrounding lacustrine areas and because of their similar morphometric character (Fig. 6 ), There are 894 lacustrine units representing all four terrain 71 types; this amounts to 14.3% of the sample. Morphometric character­ istics of low elevation, the lack of apparent slope, and reduced re­ lief are typical of each lacustrine type described below. LAKE BED FORMATIONS Extensive regions of lake bed topography prevail in the eastern and western lowlands of the peninsula. lake bed formations of the sample occur in Because 50% of all the category of '*90% or more coverage,*' the concentration of units in those lowlands produces the largest continuous "type-terrain** surface of the study area. No other genetic formation appears in a higher percentage or has a greater number of units in the "90% or more" category, nor is any other formation type more concentrated in a restricted sector of the study area. Because of these two factors, lake bed topog­ raphy is the most extensive and most continuous terrain surface of the peninsula (Figs. 23 and 24). ------ --------- Nearly twice as many of the clay bottomlands have a local relief of less than 50 feet as compared to the sandy beds. Part of this difference is due to the inclusion of glacial spillways in the sandy lake bed category; these spillways, often occupied by underfit streams, occur at higher elevations in the southern high­ land (Figs. 6 and 24). slopes exceeding 2 °, The fact that 16% of the sandy beds have three times the percentage of clay beds, is partly due to the increased erosional capacity of streams with steeper gradients at elevations above 750 feet. occur at elevations of less than 750 feet; and 95% of the clay bottoms do so. Lake, beds usually 85% of the sandy beaches 72 Because three-fourths of the clay floors and 56,5% of the sandy bottoms have average slopes of less than 1 °, lake bed t o pog­ raphy is essentially a geometrical plain except w here stream banks occur* This combination of a genetic landform type and a d i s t i n c ­ tive terrain surface will be used for the delineation of specific lowland subregions in Chapter 4* Figure 23 illustrates the absence of clay lake bed t o p o g ­ raphy on the western lowland while Figure 24 confirms the presence of sandy lake bed surfaces on both sides of the study area* This east—west contrast of lowland surfaces is probably due to the v i r ­ tual absence of limestone and shale in the source regions of the Lake Michigan lobe whereas the S aginaw and bake Brie lobes o rigi­ nated in regions rich in argillaceous shale and calcerous materials. MINOR LACUSTRINE FEATURES W a t e r l a i d moraines and interior proglacial or poided water formations comprise less than 1% of all unit areas and stand in marked morphometric contrast to one another* This contrast is implied b y the very different processes from which they originate* Waterlaid moraines are landforms produced at the margin of a rapidly melting ice lobe* Un l i k e the sub-aerial recessional m o r a i n e , the waterlaid moraine (subaquatic in origin) has been largely destroyed by the sorting action of currents and waves in shallow water at the margin of the ablating ice lobe* Ponded water f o r m a t i o n s , in c o n ­ trast, represent brief lacustrine phases of ice- or moraine-impounded spillways* As previously noted, these spillways have greater r e ­ lief and slope characteristics than any of the other three lacustrine 73 Mot Fig. 23 In c l u d e d 74 SANDY LAKE BED Not Included W a te r Fig. 24 Area 75 formations. The proglacial mantle of lacustrine sediments, confined to the spillway courses of former glacial streams, have been affected by greater rates of stream erosion* Four-fifths of the ponded water sample are located above 750 feet where the effectiveness of stream erosion is enhanced b y greater stream gradients. Secondary relief and slope characteristics become evident when the interior lake formations are compared with waterlaid moraines. the former have average slope of do. 0 Less than 30% of °-l°; m o r e than 80% of the latter Similarly, more than half of the ponded water formations d i s ­ play a local relief in excess of 100 feet, whereas only 10% of the water-worked moraines m a y be consigned to this register. W a t e r l a i d moraines and lake beds share similar morphometric characteristics: subdued relief at lower general elevations and a marked lack of steep slope and slope direction change (Figs. 27-29). These similarities, in addition to the enclave position of such moraines on the lake bed plains of the Detroit lowland (Fig. 6 ), account for their inclusion as a minor lacustrine feature. The distinction of well-stratified and relatively poorly sorted drift deposits, in lake beds and waterlaid moraines respec­ tively, provides a poor basis for differentiating them into two separate types since their morphometric dimensions provide little indication of their topographic differences. However, the identi­ fication of different depositional materials is significant with respect to the processes of differential erosion which have, to a degree, and will continue to produce contrasting terrain surfaces. 76 SELECTE FOR M ATION TYPES lim r S irfic i FsraM isit: I t Ir a i R ig iir ia iil (Tries Ii 100% C m r i|S ) Boulder | S I O ltir Sand S irfsti C itings Belts Dunes f ir e it im : > 50 % fisgiiriB iit psr lliil me 13 Fig. 25 Area ir is 77 Li n e ar L a n d f o r m Types T h e linear c a t e g o r y of M i c h i g a n l a n d f o r m s i n c l u d e s d r u n l i n s , eskers, s a n d du n e s j a n d b o ulder t rains* B a c h of these landforms, w i t h the e x c e p t i o n of b o u l d e r b e l t s * is i d e n t i f i e d in u n i t a reas on the b a s i s o f t h e **trace to 100 % coverage requirement; cre i t e r i o n is i n t r o d u c e d h e r e for the first time* m e t r i c a l l y u n d e f i n e d s y m b o l is u s e d on th e M a r t i n this Since a plani(op* cit.) sou r c e map, it is n o t p o s s i b l e to d e t e r m i n e w i t h e x a c t i t u d e t h e a r e a l e x ­ tent o f d r u m l i n s * sand dunes* or bo u l d e r b e l t s in a s p e c i f i e d unit* The p r e s e n c e of a s i n g l e symbol met the s t a n d a r d of t h e " t r a c e " appearance* In n o case w e r e s u f f i c i e n t s y m b o l s p r e s e n t to i n d i c a t e a 90% or m o r e c o v e r a g e in an y of t h e s i x - s q u a r e - n i l e u n i t areas. Li n e a r l a n dforms* w i t h th e e x c e p t i o n of t h e bo u l d e r trains of M o n r o e County* are conspicuous topographic features* The "linear" d e s i g n a t i o n p r o p e r l y i d e n t i f i e s the t r e n d — line c h a r a c t e r i s t i c of the four types; t h e t e r m " sand dune** m a y i m p l y a c o n i c a l form* but* at the m a p scales u s e d in this study, th e c o a l e s c i n g d u n e s of the we s t e r n l itteral c o n s t i t u t e a d e f i n i t e ridge-line^ T h e p e r t i n e n t g r o u p of units compr i s e s o n l y 4 * 6 % of the study area* B e c a u s e of the " t r a c e to 100%** a s s i g n a t i o n * drunlins* eskers* a n d s a n d d u n e s m a y b e e a s i l y r e c o g n i z e d on t o p o g r a p h i c maps as s i n g l e g r a i n s or i n d i v i d u a l ter r a i n c o m p a r t m e n t s . A morpho­ metr i c a n a l y s i s m a y be m a d e of i n d i v i d u a l l a n d f o r m r e p r e s e n t a t i v e s and t h e s e r a n d o m l y s e l e c t e d samples m a y b e u s e d to g i v e sta t i s t i c a l v a l i d i t y to the entire c o l l e c t i o n of a given l a n d f o r m type Bury, 1960); h o wever* genetic landf o r m s (Salis- the m o r p h o m e t r y of a c o l l e c t i o n of given (as i n d i v i d u a l grains) m a y p r o d u c e the same I 78 COMPLEX (UNCLASSIFIED W TERRAIN UNIT AREAS) l Willi ■■•■■■in* aaaaaaaaaa aaiin liit ■ ■ ■ ■ liaaiaiB aaaaaaaaaaaaaB B atB iaa a a •aaatB iaaaaaaaa aaaaaaaa aaiaiiafaaitllMit aa.................... a a aaaaa..............a a a a In n firltii •■■aaaa a a a a ataaaaB «iaB B ai«aaaaaaia«iM aaB aB aaB aaiaaaaaaaa aaiaiiaaa aaaaaaaaaaaa aaaaaaaa a aa a aaa a a a aaaaaia aaaaaaaa a aa a a a a a a a ■ aaaaa a aaaaaataaaa t ■a a a taaaaaai B l B t l l B i B B B B a i B f a a a a a a aai aaaaaa a a a a a aiataaaaiaaaiaaa aaaaaaaaaaaaiaa aaaaaaaa ■ ■■ ■ i ti a a iaa iiiaatiaiiiiiii laaiaaai naai a aa a aa laiaaaaaaiiiaBaaalaaiiaaiB aaaaB B aaaaaaaa a a a a a aaaaiaaiaaa a a a aaaaaaa a aai ■ •»!“ •aaaaaiaatifli iiatiaiiaaaiaaii ntaiiaiiiatiiMaaaataaaaaaaafi ■ •ill aaaaaaaaaaaaaaa a aa a a .............. iaa «• aaaaaaaa aiiaiiaaiaaaa aiB aaiaaaaaaiB B ia aaaiaiaiiii iitaa a a a a •■aaaaaaaaa aaaaaaiaaaaaa aaaaaaaa aaaaaaaaaa aainaa ..................... tBaaaaiiaataaiaaaiiia aaiaaaiaiiaiaaBBiii ■■■•B aaaaaaaaaaaaaaaaaaai aaaaaaaaaaiB a a aaaaaiaiaaaaaaa a a a» a ■llaaaiaaaa aaaa lllli ifili _ _ _ _ _ _ _ _ aaaiaiiai li il all* •■•••***l*ll*llll,ll'B J{J2!!!JS!!a! >laaaBaaaaiaaa>* i !iiaaiaaa » ir a itaaaaaioaaaaaaaaaaaaa a a a a a a aaaa aaaaaaaaaaa a aaaaaaaaan •aaiiaaiaaaaata a a aaaaaaa...................aaaa a a a ......................aaaaa lailllllaaiiaataaaa* aaaaaa ataai a IIW IIW II i i w n m n I i i w u m i n 11w m i ii i m u h aa^a-aa* aiaaaaaaaaB aaaaaaaaaaaaaaaa aiaaii a a aa a ...'-SSSSSiiiiiaaa ■ aaaaaaa laiiriaaaaai aaaaaaaa aaaaaa a aaaaaaaaa....................................................................am aaaa aaaaaaaaaa a a aaaaaa a a a aaaaaiaaiaaaaaaBiM aaaaaaaaaaaa a am i a aaaa a aaaaaaa a — ■ ■ aa ai aaaa aaaimaa a a a •« a a a aaaaai a a a a aaaaa aaaaaaiiaaaa****** a itna • a aaaamma at aa.................................aa aaaaai •■a arfaaiaaiiaaaai*a**a** *** __ - i l B # aaB l a a a i a i a a a a aaa tr riffl QQSI II* II- IS’ Fig. 26 ir Vitfe I f P rU iB im l F i r a i t i l i I |M : * 5 0 1 1 m l C m r i |i 79 result as the m o rphometry o f unit areas which contain the same landform. DRUMLINS D r u m l i n fields occur in t h r e e widely separated sectors of the study area: (1) Leelanau, Antrim, and C h a rlevoix counties, (2) Presque Isle County, a n d (3) Branch County, Individual d r u n ­ lins are r e d u c e d in size, area, and topographic expression in the order enumerated; Figs. 1 and 25. more than half these regional concentrations are represented in The d r u m l i n fields of the largest area constitute (55%) of the sample and are m o r p h o m etrica lly d i s ­ tinct from separate formations in P resque Isle and Branch counties* The graphic analyses of average elevation, local relief, and aver­ age slope (Figs. 27-29) provide some measure of the overall m o r p h o ­ metric character of all drumlin groups, but the data of T a b l e 6 underline their marked regional differences. F i g u r e s 11 and 12 support the statement that max i m u m e l e ­ vations of 9 0 0 — 1000 feet are characteristic of all three drumlin groups; d e c r easing slope an d relief factors for tii« groups in Presque Isle and Br a n c h counties are related to higher minimum e l e ­ vations — 700-800 for the former and 900-1000 feet for the latter* Sixty-two percent of the drunlins of the Leelanau-Antrim-Charlev oix complex have minimum elevations of less than 6 00 feet and all of the units in this region exhibit local relief of more than 250 feet* In srun, the interpretation of the data indicates increased relief and slope values at lower elevations for the L e e l a n a u — TABLE 6 Characteristics of Drumlin Fields AVERAGE ELEVATION LOCAL RELIEF Leela­ nau Presque Isle Branch Presque Isle Branch 25 ft. — M M 50 — — 75 ■■ tm mm . . 600 ft . Leela­ nau Presque Isle Branch Under Under Under 100 Leela­ nau AVERAGE SLOPE 2 1 M M 1/2° 1° 1 650 15 1 — 4 8 700 22 6 — 9 8 750 19 18 2 800 9 3 mm 850 1 10 1 1 1/2° 2 M M — 2 2 1 2 2 « M — 2° 1 3 12 M M 3° 1 12 5 1 4° 15 1 mm 10 5° 13 — — 150 1 9 200 7 10 250 14 6 — 900 300 14 1 mm 950 — m m 5 10° 36 Mm M M 400 26 — 1000 — mm 3 69 39 19 500 7 — Unit Area Count Unit Area Count Unit Area Count 69 — 39 19 69 39 19 — 81 An t r i m —C h a rlevoix ssimple; on the other hand, the Presque Isle and Branch units are at higher elevations and manifest less relief and slope than the north-western drumlin fields* SAND DUNES S a n d d u n e formations occur in more than half the littoral units of the Lake Michigan shore to Berrien County). A s previously noted (51*6%) of (Leelanau Peninsula (p* 76), these coalescing dunes constitute a ridge-line of higher elevation, slope, and relief than dunes extending inland to the base of the northern and southern highland* T ABLE 7 Characteristics of the S a n d D u n e Littorals L O C A L RELIEF Mi c h i g a n Littoral AV E R A G E S L O P E Huron Littoral Michigan Littoral Under Under 25 ft. 50 Huron Littoral — 2 1/2 4 .o A ° 3 — 4 4 6 ° 6 1 3 3° 11 1 19 1 4° 8 2 250 io 1 5° 5 300 2 io ° 13 1 400 4 500 2 48 18 75 1 5 lOO 1 -- 150 9 200 Unit Area Count 48 1 1/2° 2 w 2 18 82 The morphometric dimensions of the windworked dunes along the Lake Michigan littoral contrast sharply, in slope and relief, with substantially lower values of the dunes along the Lake Huron littoral (Table 7). Bach of the sand dune units has the adjacent lake level as its minimum elevation; the higher average elevation (Fig* 10) and the greater relief and slope values of the Lake M i c h ­ igan group are the result of higher maximum elevations (Fig* 11)* More than one-third (37*5%) of the Lake Michigan sample has relief in excess of 200 feet while this is the case for only one-sixth (16*6%) of the Lake Huron aggregate* In addition, a similar ratio prevails, 37*6% to 16*6% respectively, when slopes in excess of 4° are considered* MINOR LINEAR FEATURES Eskers and boulder belts account for only 119 unit areas or less than 1% of the total sample* Boulder belts are difficult to recognize at even the largest available map scales; they are essentially microfeatures without topographic expression, and their surface expression is probably not related to the genesis of their deposition* Although eskers are probably the best defined landforms of the study area, on a local basis, they are the least susceptible to quantification at the scale of the unit area grid used in this study* In no case did eskers constitute as much as 50% of a unit and, due to their restricted areal extent, it is clear that values of slope, elevation, and relief assigned to eskers are mitigated by the presence of other landform types* 63 G e n e r a l Comments on the Histograms of L a n d f o r m Types Figures 27, 28, and 29 constitute a graphical r e p r e s e n t a ­ tion, b y classes of slope, relief, and elevation, of the three genetic classifications of glacial landform types occurring in the sutdy area* W i t h the exception of sand dunes, eskers, drumlins, and recessional moraine, each of the landform types prevails in the form of extensive geometrical surfaces w i t h little eviden ce of t o p o ­ graphic grain in unit areas; the morphometric values obtained for the remaining landform types, w h i c h have essentially homogeneous surfaces, provide v a lid summations of the range and concentration of terrain data for such "extensive1* landform types* "Intensive** landform types, dunes, and recessional moraines, i.e., eskers, drumlins, sand are locally dominant formations which determine m o r p h o m e t r i c summations of slope and relief in (4) unit areas* A limited check of m o rphometric values assigned to these unit areas p r o v i d e d little contrast to values obtained by evaluating slope and relief characteristics of sharply-defined map representations of the same formations* square-mile 2 1/2 Th e choice of the six- -minute unit area size is crucial to this finding; experimentation w i t h the 24 square—m i l e 5-minute unit area and the 36 square-mile 7 1/2—mi n u te unit area, at an early stage in the re­ search, failed to achieve good agreement with the morphometric character of individual landforms, in s i t u , and the aggregates of landforms in unit areas* R e c e s s i o n a l m o raines and outwash, with reference to the ^ ^ N o t e statement of exception regarding eskers (p. 81)* 84 histograms, constitute the most common landforms above lOOO feet of average elevation and the presence of till plains on the northern highland accounts for v i r tually all of the remaining upper-level surfaces* G r o u n d moraines, recessional moraines, and outwash flats also account for mo r e th a n 9096 of all intermediate surfaces 1000 (800- feet) although certain minor topographic features, e*g*, eskers, proglacial lacustrine formations, at this level* an d drumlins are also concentrated Clay and sandy lake beds, major components of the inventory sample, and minor areal groups of sand dunes, waterlaid moraines, and boulder belts occur in 95% of all cases at elevations of less than 800 feet* C l a y lake b e d formations constitute the largest areal cate­ gory of landforms without visible slope or relief (Figs* 28 and 29). Recessional moraines contain the largest area of steep slope (3° or more) a n d strong relief (more than 200 feet) although the smaller drumlin sample exhibits a greater percentage in the foregoing relief category* Paradoxically, a greater percentage of the recessional moraine sample has steeper slopes t h a n the drunlin sample* This is no doubt related to the gentle back-slope characteristic or drum- lins in the absence of a counterpart slope in the typical morainic formation* O u t w a s h formations not only occur over a greater range of average elevations than ground moraines (till), but also have more pronounced relief and slope characteristics* This attribute may be due to the role of differential erosion in sculpting the fluvioglacial materials into landforms of greater topographic expression as well as to the fact that the outwash summit elevations of the 85 northwestern rim of the northern highland have experienced pro­ nounced dissection. Sand dunes produce strongly-marked local land­ forms, but they have less relief and slope than drumlins -- due possibly to the transformation of a nucleated series of coastal dunes into an arete-like ridge of coalescing dunes with a net loss of steep slope. Ponded water formations (partially filled glacial channels) share similar morphometric values be thought of as form-reversals troughs. with eskers which, (on a smaller scale) in turn, may of proglacial Sandy lake-bottom topography approaches the morphometry of either proglacial lake forms or esker-marked topography; as a result, sandy lake bed terrain exhibits significantly greater re­ lief and steeper slope than clay lake bed topography. The raicrorelief produced by elevated beach ridges and wavecut cliffs along the inner or western reaches of the Saginaw Lowland is not susceptible to morphometric analysis from a study of even the largest scale topographic base maps (1:24,000). Although the M.S.F* map contains symbolic representations of these locally conspicuous features, their planimetric outlines are indeterminant to the point of precluding the acquisition of un­ ambiguous values of slope and local relief. 86 AVEBAGE >90% ELEVATION: COVERAGE, Per Unit Area, of P R I M A R Y GR OUND M O R AINE of survey INDICATED G ENETIC G E N E T I C M O R A IN E unit TYPES areas (E X C E P T UNEAR) T Y P E S OUTW ASH U N C L A S S IF IE D P tftC C N TO F U N IT A M A t L A C U S T R I N E CLAY LA KE I S » SEO SAND II I C L A S S E S G E N E T I C LAKE • It IE D I It I t l WATER i AVERAGE © P LINEAR • AND n m JIS m m 570-- SO* *00-- *4* m ra 650-- •** 700-- 74* 750-- 7** H E 09 m m ra •00 -- S4* ■50 -- •** 900 -- *4* *50 -- *** 1000 --104* t M ORAINE It II T Y P E S BOU LDER BELT 1 i I KLKVATION A V I ft A O K of I OUNI EL C L A S S P S WA T E R L A I O E L E V ATION G E N E T I C D R U M L IN i T Y P E 8 P ONOCD mi DE 09 1050--10** 1100--114* 1150--11** 1200--124* 1250--12** 09 mi mi in 1200--114* 1150--11** 1100 --144* Jw<*> Fig. 27 87 LOCAL >90% BEL I EF : COVERAG E, Por U n it P R I MA R Y GR OUND unit GENETIC areas TYPES (EXCEPT LINEAR! G E N E T I C U N C L A S S IF IE D OU T W A S H MORAINE M O R AI NE q< survey Aroa, of INDICATED P E R C E N TO P U N IT A fttA S I S L O C A L L A C U S T R I N E CLAV LAKE BED SAND G E N E T I C LA K E BED I* IS T Y P E S PONDED WATER WATER LAID MORAINE L I N E A R O R UM L I N C L A m m s s tu m 0 ----25----SO ----75----100----125----- • s 24 4* 74 OS 124 140 of m m m fiol un DU L O C 150 ----- 174 175 —— If f 200----- 224 225 ----- 240 250----- 274 275 ----- 200 R IL IIF A L Q5] D3 Hftl fiel Dll [Te] 300 ----325 ----350 ----375 ----400 ----425 --- Fig. 28 so R E L I E F 324 340 374 300 424 440 in PIET Hal 450• [201 475 rail 500 • 474 4** 524 88 A V E R A G E SLOPE s u r v e y of unit a r e a s • > 90% COVERAGE, Per Unit Area, of IN D IC A TE D TYPES ( EXCEPT LINEAR) PRIMARY GROUND OUT W A S H M O R AINE P C R C IN TO f u ftir a r e a s GENETIC LACUSTRINE aC L A Y SANO l a k e T YP PONDED LAKE E S WATERL A I D WATER MORAI NE i i LINEAR AVE RAGE SLOPE GENETIC TYPES SAND DRUMLI N I A L I 1 ■ . CLASSES of I . « AVERAGE [ T ] o ° - V [ T ] 1° - 1 1 2° □ □ G H lS °-2 ° V - r Fig. 29 J, 1 BO U L D E R DUNE l 5 SLOPE [ T ] 2 ° - 3 0 [ 6 l l 3 0- 4 0 II . in m PT1 I ' . ' DEGREES 4 ° - 5° 5 U- 1 0 U BELT CHAPTER IV TERRAIN REGIONALIZATION FOR SOUTHERN MICHIGAN Introduction The following discussion is concerned with the delineation of morphometric regions and subregions for the southern peninsula of Michigan, The identification of essentially homogeneous terrain surfaces must be organized in a manner which affords recognition of the three primary topographic levels of the study area: (1 ) a d i s ­ continuous peripheral lowland with subsections of unlike relief and different glacial landforms, (2 ) a northern highland characterized by greater extremes of topographic roughness than (3) a southern highland of relatively lower average elevation. Consequently, the recognition of morphometric compartments -- called regions and subregions — is based on the a priori recognition of these primary topographic provinces* The identification of morphometric compartments is based on relatively small changes in elevation and relief because only restricted fluctuations of the morphometric data characterize the subdued topography of the peninsula. The limited range of the data is not unexpected in view of the youthfulness of the post-glacial erosion surface, the topographic simplicity of the pre-glacial surface, and the virtual absence of differential erosion in the 89 90 unconsolidated drift materials. The Problem of Terrain Regionalization T h e initial problem in the identification of terrain regions is concerned with the selection of appropriate class intervals for the purpose of grouping elevation, relief, and slope data into frac­ tional subgroups* The range and frequency of all terrain data must be accommodated in a sequence of value—classes so that collections of unit areas of one class may be compared with collections of unit areas of different class* This intent is explicit in Hammond's (1954:36) statement: "Their purpose [class or group boundaries] is to promote objectivity in classifying a piece of terrain, to make possible the com­ parison of different regions, and to insure the comparability of results by different people* N o attempt has been made, as yet, to employ statistical analysis as an aid in choosing the most valuable boundary values* This could possibly vary depending on the regions and the purpose for which the clas­ sification is made*1* Consequently, the determination of specific class limits may be arbitrary, according to the purpose of the regionalization, but this procedure need not be a subjective one* Success in the selection of appropriate class intervals of the terrain data can only be measured by the validity of the regions or subregions which they represent; the primary goal in devising class limits for the data used in this study was to obtain results which would delineate previously undefined regional terrain types* The three main topographic levels of the study area, as 91 described earlier, consist of two highlands of differing dimensions and a discontinuous, peripheral lowland* The vertical component of this peninsular landscape is taken as the primary agent for regional differentiations unit areas with a stated average eleva­ tion range are sought to establish the altitudinal planes of each highland* W i t h only one or two exceptions, the selected class in­ tervals produce continuous planes of a single, rigorously defined altitudinal level* Pour altitudinal levels are defined for each highland* They are based on unequal elevation intervals because of the di f ­ ferences in general elevation of the two* Each level is called a "terrain region” for descriptive purposes, and a nomenclature su g ­ gests the increasing elevation of each level: the three-dimensional character of each highland and the aforementioned increase in ele­ vation is implied in the terms (1) Foreland, (2) Intermediate U p ­ land, (3) Upland, and (4) Highland* The determination of elevation values for the four hypso­ metric levels of each highland is both arbitrary and systematic in nature; it is precisely stated in observable values of elevation, and justified on the basis of the identification of such unlike regional topographies as the Cadillac Highland and the Lansing Foreland* Because the altitudinal designations serve to compart­ mentalize the cross sectional levels of the two highlands, they are called ‘•morphometric” regions in the sense that they delineate sub­ sections of a landform province in terms of discrete and objective terrain measurements* Bach of the thirty-four morphometric s u b regions is located entirely within one of the five regional levels (including lowlands)* 92 TERRAIN REGIONS L • g •w d □ L o w la n d F o r a la n d In ta rm a d la ta U p la n d Upland ar 14 ir Fig* 30 ■3" 93 The identification of a subregion is based upon the recognition of a group of unit areas with a given class of local relief in a limited area of a given altitudinal level* The grouping of such unit areas is not based on the existence of a continuous or homogeneous relief pattern; Figs* 32-26 verify core areas of nearly homogeneous relief surrounded by zones of transition consisting of unit areas with anomalous relief values* The problem of devising subregional boundaries is resolved by laying out boundaries which bisect the transition zone, approximately, of anomalous relief values between core areas of unlike but nearly homogeneous relief* Th e resultant subregional boundaries have, as a minimum, at least one-third of all the included unit areas in the same re­ lief class and at least one subregion has three-fourths of its unit areas in a single relief c a t e g o r y * T h e identification of a p r e ­ dominant relief class, in subregions, serves to adequately differ­ entiate sectors of the various hypsometric levels and leads to such valid subregional topographies which, for instance, distinguishes the Lansing Foreland from the Boyne City Foreland (Map 23, 6D and 2B)« Tiit-* diiuwaluus unii aieas of ail subregions are depicted in Figs* 32 — 36 according to relief value and grid location within the subregion; this presentation is more objective than an alternate method which "absorbs** all unit areas of a different class of relief* A method for summarizing the total relief differential of all such anomalous ^ ^ I t should be noted that Hammond (19S7) meets this problem by use of the arbitrary but objective '*four-contiguous-unit-areas" rule: anomalous unit areas, up to a maximum of four contiguous units, may be absorbed in the delineation of a homogeneous relief subregion* 94 TERRAIN SUBREGIONS Fig. 31 TABLE 8 Key to the Identification and Nomenclature of Terrain Subregions Regions Code Lowland: 1A IB 1C ID IE IF 1G Subregions Detroit-Huron Saginaw Alpena Cheboygan Traverse City Ludington Muskegon Prevailing Relief Class: Percent of Unit Areas_______ 1462 % 1* 48% 2:41% 4:33% 4:48% 3:41% 3:33% Southern Highlands Adrian Salem Sandusky Lapeer Lansing Schoolcraft Wayland Niles 3:57% 2:72% 2:62% 3:60% 2:74% 2:55% 4:43% 4:36% 3B 3C Ann Arbor Jackson Kalamazoo 4:48% 2: 70% 3:39% Upland: 4A 4B Pontiac Addison 4:73% 2:44% Hiahlandi *»A H i 1 1 c rfa l e Foreland: 2A 2B 2C 2D 2E 2F 2G 2H Intermediate Upland: 3A Northern Highlands Foreland: 6A 6B 6D 6B Mount Pleasant Onaway Posen Boyne City Sparta 2:49% 3:61% 2:71% 5:44% 4:48% 7A 7B 7C Muskegon Manistee Au Sable 4:48% 4:52% 4:51% 8A 8B Grayling Houghton Lake 4:58% 2:45% 9A 9B 9C Gaylord Cadillac Ogemaw 4:66% 4:57% 4:56% 6C Intermediate Upland: Upland: Highland: 96 unit areas, not studied here, should be developed for the purpose of measuring the relief heterogeneity of a given subregion; this index could serve just as well as relief homogeneity in the identi­ fication of morphometric subregions* Average slope classes and prevailing surface formations are included in the description of morphometric subregions although they are not determining factors in delineating these compartments* Be­ cause slope classes have a good correlation with relief categories (Table 11) and because specific genetic landforms are known to occur within selected classes of both slope and relief, these areal char­ acteristics are included in a fractional key of terrain factors supporting the subregions set forth on Figs* 32-36. The Hypsometric Factor The change of altitude in various parts of a land mass constitutes a primary basis for differentiating areal segments of that land mass* Because the average elevation parameter sets forth the principal or first-order topographic formations of any study the S a p O jl t e i i alii iey i u i i b (F 3C) uelituiis the e x L eii t o jl the two major morphometric features of the peninsula: a northern and a southern highland. Veatch (1953) utilizes the 750-foot isohypse to distinguish between the highland and lowland areas of the peninsula* This iso­ hypse is also used in this study (Table 9) because rapid increases in slope and relief, to the interior, occur along this line of d e ­ marcation -- a correspondence which is best illustrated (Figs* 13 and 14) along the southeastern margins of the northern and southern 97 highlands* The delineations of the two highlands are compact in form with few outliers separated from their nuclear cores (Fig* 30)* In contrast to the continuum of the foreland regions, the highland and upland regions occur as non-continuous morphometric levels of the northern and southern highlands; these darkened areas of Fig* 30 stand out as peninsular heights (or islands) and a com­ parison of the northern and southern highland confirms the greater areal extent of higher elevations in the northern highland. T h e cross sectional profiles of the northern and southern highlands reveal important differences in their general or pre­ vailing elevations more than 75% of the northern highland has average elevations in excess of 1000 feet whereas this is true for less than 25% of the southern highland* For the purpose of creating meaningful internal differentiations, the two highlands are divided into foreland, intermediate upland, upland, and highland levels; because of the profile differences of the two highland provinces-i different isohypses are used in each highland to identify these internal compartments* The choice of these critical isohypses is entirely subjective, but seems justified in view of their delinea­ tion of such regional topographies as the primary and secondary cores of the southern highland, the continuous nature of the SaginawMaple trough, the riverine incursions of the northern highland, and the morainic rimland of both highlands* 98 TABLE 9 Table of Hypsometric Levels Selected for the Differentiation of Terrain Regions Northern Highland Regions UAC* Southerrv Highland UAC* Foreland 750- 899 ft. 1268 750- 849 ft* Intermediate Upland 900-1099 ft* 490 850- 949 ft. 1080 ■ 193 Upland 1100-1249 ft. 502 950-1099 ft. 696 Highland 1250-1468 ft* 261 1100-1206 ft* 44 Totalx *UACx 2521 2013 Totalx Unit Area Count The regional terminology of Table 9 is descriptive of a series altitudinal surfaces of increasing elevation; however, the terms are also objective statements of absolute elevations and are used in this dual context in the remainder of this chapter* In addition, they relate to the physiography of the peninsula insofar as they represent logical divisions of the two highland provinces* The principal hypsometric belts of the peninsular study area are iilusf2 ) trated on Pig* 31 and listed on Table 13 by individual subregions*' ' The Relief--Slope Factor A six-fold classification of local relief intervals was (2 ) The arabic numeral indicates the gross morphometric type, the figure " 1 ” specifies lowlands, and the accompanying capi­ tal letter identifies subregions based on a concentration of a given class of local relief* Similarly, forelands are indicated by the figure n2,n intermediate uplands by ”3**, and so on* A list of urban place names and river basins is included as a guide to the geographic location of the subregion within the study area* used for the purpose of identifying terrain subregions (Table 10 ). Because of the good correlation (see below) between local relief and average slope, local relief classes alone were used in the de­ lineation of these subregions* The consolidation of relief units into relief subregions is based on the grouping of proximate unit areas of the same relief class so that the group of like relief units represented at least one-third of all the units included in the subregion* The success of this approach is based upon the recognition of significant breaks in the regional patterns of maximum-minimum elevation change* The compartmentalization of the prevailing subdued relief of the study area, and the included anomalous unit areas, is shown in Figures 32-36. The first three increments of the relief classification have intervals of 50 feet each while the latter three are disparate and reflect, in ascending order, the decreasing number of unit areas involved (see 'Total,** Table 13)* Class equal concentrations of unit areas — respectively — 2 , 3, and 4 have almost 29*5%, 27*7%, and 26*4% and Class 1 (14*4%) occurs in a 1:2 ratio to these categories but twice as often (2 :1 ) as the combined upper categories in classes 5 and 6 * TABLE lO Classes of Local Relief and Average Slope Classes: 1 Local Relief (feet) 0-49 Average Slope (degrees) 0-.99 2 50-99 1-1.99 3 4 5 6 100-149 150-299 300-499 500-622 2-2*99 3-3.99 4-4.99 5-10 100 Only 7*3% of the units have a relief value in excess of 300 feet and the use of Class 6 relief (in excess of 500 feet), constituting less than 1% of all units, succeeds in producing the areas of maximum relief within the study area* Because only 14*4% of the units have a relief of less than 50 feet and because the lowland province occupies 37% of the peninsula, it was anticipated that lowland compartments would be recognized on the basis of either Class 1 or Class 2 relief and no other* Subsequent investigation of the regional patterns of relief in the lowland regions proved that more than one-third of all the included unit areas occurred in the relief categories of Classes 3-5* study area, only four subregions With respect to the entire are marked by assemblages of less than 40% of the included unit areas in a single prevailing relief class — in no case does the predominant relief class constitute less than one-third of all such units* More importantly, 20 of the 34 subregions are characterized by the same relief class over 48% or more of their areas (Table 13)* Although slope characteristics have not been used to dif■- f p r p- n t * i A tp th p f f l o r n h n m— p- t— r— i r* - ■- — -------— ----------f — between the increasing orders of r>r> «* . z* /ia a H ww* - i n n ■» « •+W «W local relief, defined above, and average slope* The high correlation between slope and relief increments is obvious from Table 11* Thus, 77*8% of all unit areas with a relief of less than 50 feet also have an average slope of less than 1/2°. The fact that 1 *1 % of this relief class also has an average slope of 2° - 3° is an indication that the Wentworth method produces a "roughness index** rather than discrete measure- 101 TABLE 11 Local R elief Classes as a Percentage of Average Slope Classes . Local Relief Classes in Feet Average Slope__________ 0—49_____ 50-99 100-149 150-399 300-499 ° 77.8 20.6 1 /2 ° - 1 ° 39.4 49.7 8.1 2.7 9.0 58.9 23.5 2.3 41.5 1.1 0 °—1 /2 1/2 ° 1°-1 1 / 2 °- 2 1 ° 2°-3° — mm ms 18.5 •3 — 37.3 18.5 •3 mmmm 15.9 36.3 44.8 1.9 sees 4.4 21.7 65.6 7.7 •1 — .5 8.1 68.2 23.2 — -- .4 1.3 31.5 60.7 6.0 0 0 H 1 m 0 aoem 0 >n i 0 •2 .5 3°-4° ments of real slopes. •9 .5 500-623 It follows , then that low relief values do not ordinarily produce topographic knobbiness. The slope factor is not expressed in descriptive terms such as ••steep,” "gentle," or “undulating” because (1 ) it was necessary to use topographic maps with different scales and contour intervals; (2 ) of the contrasting detail of contour— line expression on maps compiled over a period of sixty years, and (3) of the difficulty of obtaining repeatable contour-counts in areas of topographic roughness* T he Genetic Factor Glacial landform types are included in the description of discrete subregions because they convey an impression of the physi— 102 cal appearance of the landscape* However, genetic landforms are not used as criteria to delimit terrain regions* Their often diverse morphometric character yields too wide a range of local relief values, as is the case in a consideration of drumlin fields which vary from 50—450 feet in local relief and provide little evidence of cluster­ ing in a given relief class (Fig* 13)* Nevertheless, genetic land­ forms frequently produce slope and relief values which can serve as the basis for connoting the physiographic aspect of delineated mor­ phometric subregions* Analysis of Morphometric Provinces The identification of morphometric provinces is based on the utilization of the 750 foot-isohypse to delineate three firstorder provinces of approximately equal areas (Fig* 30): (1) Lowland: 32*1% of the study area, (2) Northern Highland: 30*98 of the study area, (3) Southern Highland: 37*08 of the study area. Bach of the highland provinces consists of four morpho— metric regions, i*e*, a foreland, an intermediate upland, an upland, and a highland in a sequential order of increasingly higher eleva­ tion surfaces (Table 9)* Bach of these second-order regions may include from one to eight subregions based on coherent groupings of units with similar slope and relief factors* Thirty-four third- order subregions, or compartments, have been identified and desig­ nated after names of important settlements or, in sparsely settled areas, by major hydrographic features* A fractional key, described below, is used in conjunction with Figs* 32-36 to quantify the surface geometry of each of the 34 subregions* 103 The key also indicates the number of unit areas in each subregion. The placement of each terrain factor is illus­ trated in the following examples 2B 36 The individual digits stand for the following characteristics: (2 ) (B) (M) (.41) (/2 ) (.72) (3) (.47) (36) hypsometric factor, subregion code, prevailing surface formation, percentage of unit areas with prevailing surface formation, prevailing relief class, percentage of unit areas with prevailing relief class, prevailing class of average slope percentage of unit areas with prevailing slope class, number of unit areas in the coded subregion. The prevailing landform type for each subregion is specified according to the following codes M T = a* Moraine Till O L = Outwash a Lacustrine D U = = Drumlin Unclassified Bskers and sand dunes have been omitted because neither occurred as a dominant genetic type in any s u b r e g i o n . T h e lacustrine formations, for the purpose of this category of code, consists p r i ­ marily of lake plains of either clay or sandy materials (87% of all lacustrine units of the study area) but also includes smaller areas of boulder belts, upland or interior lacustrine formations, and waterlaid moraines. (3 ) ' 1It must be pointed out that the identification of pre­ vailing landform types is based on counts of unit areas with 50% or more coverage of the indicated landform. Unit areas with no single formation amounting to 50% are entered in the unclassified category. 104 The Lowland Province The Lowland Province of the southern peninsula of Michigan, in spite of its generally low absolute elevation, is diverse in both terrain morphometry and topographic configuration. Seven low- land compartments may be identified at elevations of less than 750 feet on the basis of contrasting patterns of regional relief and slope* Marking the periphery of the study area, these lowlands en­ circle the peninsula with the exception of the Emmet County littoral in the extreme north, which is without a discernible lowland at the scale of the study map (Pig* 32)* The lowlands of the Lake Michigan borderland contrast markedly with the eastern flatlands of the Lake Huron-Erie coastt in no sense of the word may the former be termed "flatlands* and it is equally misleading to infer generally lower elevations for the latter* The Saginaw Lowland (IB in Pig* 31) is the largest of the lowlands and is related genetically to the third largest lowland, the Detroit-Huron (1A); both lowlands consist of extensive lacus­ trine plarns formed *>y the «ulaiing ice masses of the Huron, tirie, and Saginaw lobes* It is not surprising, therefore, that both lowlands display relief of less than 100 feet over more than 85% of their respective areas* Furthermore, slopes of less than 2 ° char­ acterize more than 95% of each lowland* In contrast, the Traverse City Lowland (IE) is marked by exceptionally strong relief and is one of only two subregions which have as much as 42% of their respective areas in the steepest slope category (5°— 10°)» No other subregion has the abundance of 105 LOW LAND SUBREGIONS 1 D ^ M a i i 2 3 LOCAL RELIEF n Feet) 1E D^Lii130 1 0 ^ ^ 7 4 [ | 1 0 0 - 1 4 9 3 0 0 - 4 9 9 | | 5 0 0 - 6 2 2 ,F M ^ l11,J7TmI 1 B L ' V » 1 831 -f" C 3 -43" lG k^ll 496 1A l ^ i l 4Ia 42* Fractional Specified • 6" •3' •S’ Fig. 32 Cada ia Taat 106 relief energy of the Traverse City Lowland: 87% of its unit areas have local changes in elevation in excess of 150 feet. Unlike the other six lowlands, most of the Traverse City Lowland is non-lacustrine in origin and exhibits a distinct linear grain due to the presence of large complexes of drumlin hills and numerous inter* vening water-filled depressions* T he Muskegon (1G) and Ludington Lowlands (IP} have p r e ­ vailing slope and relief intermediate in value between the flatness of the Saginaw Lowland and the roughness of the Traverse City Lo w ­ land* The lake—border portions of these lowlands are marked by north— south belts of interlocking sand dunes which produce most of the slope and relief noted above* Proglacial sandy lake bottoms characterize much of the Muskegon Lowland, whereas outwash plains are the most common landforms of the Ludington Lowland; post­ glacial erosion in both instances has produced nearly identical slope characteristics (Table 15, IF, 1G ) • Th e Lake Border reces­ sional of southwestern Michigan, extending from New Buffalo to Grand Rapids, is a provincial anomaly; with the exception of the Caro Moraine of Tuscola County, no other lowland contains a sizeable recessional moraine* The Cheboygan and Alpena Lowlands (ID and 1C), the two smallest lowlands, are characterized by rolling lake plain topog­ raphy with approximately half of each lowland having relief of 100—300 feet and slopes of less than 2 °. These values are somewhat higher than those for similar topographic forms in the south (1A, IB), but significantly lower than values obtained from lacustrine formations of the Muskegon and Ludington Lowlands* Six major lakes (Grand, Long, Douglas, Black, Burt, and Mullet), complemented by 107 six large lakes of the Traverse City Lowland (Charlevoix, Torch, Elk, Leelanau, Crystal, and Glen), form a crescentic array of waterfilled troughs at lowland elevations along the northern coast of the peninsula* The Southern Highland Province Compared with the Northern Highland, the Southern Highland is characterized by lower average elevations and reduced slope and relief values* It contains slightly less than one-half (48%) of the total peninsular relief of 50-99 feet; however, less than 3% of this available relief is in excess of 300 feet (Table 16). Al­ though the combined area of all the lowland provinces accounts for nearly two-thirds of the available flatlands (less than peninsula, forty-two percent of the next slope class 1 °) of the (l°-2 °) occurs in the Southern Highland and serves to distinguish it from the Northern Highland (Table 15)• The Foreland Region of the Southern Highland Forelands comprise 56% of the Southern Highland Province (c f • Table 12, Number of Unit Areas) and the changing patterns of slope and relief, over such a large area, necessitated eight thirdorder terrain compartments to adequately portray these differences* Certain non-contiguous subregions, on the other hand, are remarkably similar in the case of slope and relief; for instance, the Sandusky (2C) and Schoolcraft (2F) subregions have a local relief of less than lOO feet over more than 80% of their respective areas* The 108 FORELAND SUBREGIONS LOCAL 11n I | RELIEF FeptI 1 0 0 - 1 4 9 011/144 3 0 0 - 4 9 9 44 6E 6 Aiy^l369 7 s> 2C X I 2£ 1^174 2 GWfg" 2BM11ZU1,6 NOTE f r *r ** ft nal Specif ied 2F2 lLLil I0S 2A xmii6. 2Ha M n m «« 14 as Fig. 33 13 f nrie in is t ext 109 fact that they also have more than 90% of their surfaces in average slopes of less than 2 ° further emphasizes the similarity of these two compartments* Stratified lacustrine deposits and the tenuous recessional moraines of the Port Huron system characterize more than half of the Sandusky subregion, whereas two-thirds of the Schooleraft com­ partment is made up of outwash. plains Except for the Houghton outwash (3B), the Schoolcraft outwash plains are morphometrically the most subdued examples of this landform type for the peninsula as a whole* Adjacent outwash trains at higher elevations in the Jackson subregion (3B) , including the M arshall—Climax—Coldwater triangle, occur consistently with 150-199 feet local relief* The Lansing Foreland (2B), second largest subregion of the 14 compartments of the Southern Highland, consists of the east-west trending moraines of the Saginaw system and the most ex­ tensive till plains of the study area* Because less than 1% of these formations have relief in excess of 150 feet and because a relief component of 50-99 feet prevails over 75% of the area, the Lansing Foreland is distinguished by abundant but only moderate changes in local elevations* (2 G ) , and Niles In contrast, the Lapeer (2D), Wayland (2H) subregions exhibit relief in excess of 150 feet in more than one-third of their unit areas* The topography of these compartments provides a sharp contrast to the more moder­ ate relief of the conterminous Sandusky, Lansing, Schoolcraft, and Adrian (2A) subregions (Fig* 33)* The Adrian (2 A) and Northville (2B) compartments represent a relatively narrow zone of altitudinal and relief transition be­ tween the nearly flat Detroit-Huron Lowland and the dissected south­ eastern rim of the Ann Arbor Intermediate Upland (3A)* This transi­ tional belt is an area of rapidly rising values of elevation, re­ lief, and slope and is marked by the topographic dominance of the Defiance Moraine* As mentioned earlier,this morainic system rises abruptly out of the lacustrine plains and produces one of the most distinct physiographic boundaries of the southern peninsula* The Intermediate Uplands of the Southern Highland Three intermediate uplands (Fig* 34) make up one-third of the Southern i Highland area and occur at elevations of 850-949 feet* The Jackson compartment (3B) is the largest of these subregions and is characterized by a surface geometry similar to that of the Lansing Foreland, at lower elevations, to the north* Because three- fourths of the Jackson subregion has slopes of less than 2° and a relief of less than 100 feet, it contrasts sharply with the adjacent Ann Arbor (3A) and Kalamazoo (3C) compartments* Both of the latter have more than 7096 of their units in slopes of 2° or greater and more than 75% of their units have relief in excess of 100 feet* Although morainic hills are the dominant formations in both the Kalamazoo and Ann Arbor subregions, they occur at the lowest elevations (except for the Lake Border and Caro Moraines) and are among the most subdued examples of their genetic type* Significantly greater slope and relief values are obtained for similar proportions of moraine and outwash types in the Boyne City Foreland (6 D) and the Cadillac Highland (9B) of the Northern Highland (Tables 12 and Ill T he Uplands of the Southern Highland The Pontiac (4A) and Addison (4B) uplands (Fig* 35) are characterized by intermediate slope and relief categories. The Pontiac subregion has 73% of its area in relief of 150-299 feet and the Addison tract has 76% of its area in relief of 50-149 feet* These two compartments* at elevations of 950-1100 feet* comprise only 9% of the area of the Southern Highland* (This is only half of the corresponding hypsometric level* 900-1100 feet* the area of the Northern Highland*) The larger complex of interlobate moraines in the Pontiac subregion accounts for the fact that slopes in excess of 3° occur at a ratio of 2s 1 compared to the Addison compartment* the Kalamazoo moraines Although (3C) are topographically rougher* with 21% of all slopes in excess of 4°* the morainic hills of the Pontiac compartment have the greatest share of pronounced relief (80% of all units in excess of 150 feet) of any subregion of the Southern Highland* The differentiation of the eastern and western sectors of the Addison compartment (Fig* 35) is apparent in the greater relief and more extensive moraines of the eastern half as compared to the reduced relief (less than 150 feet ) and substantially larger of outwash and till of the western half* areas The Hillsdale Highland of the Southern Highland The Hillsdale Highland (5A, Fig. 36), delimited by a basal elevation of 1100 feet and a summit elevation of 1284 feet* 112 INTERMEDIATE UPLAND SUBREGIONS LOCAL RELiEF 11n F e e t ) 7C “ V iV ' 137 I I 100-149 150-299 300-499 500-622 7A M -ffr1 1,9 3BaVRlZt»*' i N0TF frictional code is specified in l e s t II aa H Fig. 34 ■3 I 113 is less rugged than the interlobate Pontiac subregion* The pres— ence of moderate relief (84% of all units in the range of 100-299 feet) and moderate slope (93% in the l°-3° range) makes it compar­ able to the Lapeer subregion (2D) in the foreland region (Tables 15 and 16)* The small number of units (44) of the Hillsdale Highland attests to lower absolute elevation of the Southern Highland when compared to the 751 units of the Northern Highland above the llOO feet isohypse* The utility of adopting a higher basal elevation (1250 feet) to classify the highlands of the Northern Highland is supported by the face that highlands of both the Northern and the Southern Highland are then characterized by morainic formations over 55-59% of the resultant areas* Such agreement would not be possible if identical hypsometric categories were applied to both of the first-order provinces. The Northern Highland The Northern Highland has both a larger area and a greater amount of upland/highland surfaces than the Southern Highland. Table 11a summarizes the greater concentration of higher slope and relief values (Classes 5 and 6 ) in the Northern Highland. 114 TABLE 11a Percentages of Unit Areas in Specified Classes of Slope and Relief in _____ The Northern and Southern Highlands Slopes Class 1 0 -1° Class 2 l°- 2 ° Class 3 2 -3° Class 4 3 —4 Class 5 4 —5° Class 6 5°-10° Northern 13.1 27.1 43*5 55.3 72.1 81.1 Southern 21*4 42.7 39.2 28.2 14.6 3.2 UAC* 1609 2101 1488 767 402 465 Relief (ft*) 0-49 50-99 100-149 150-299 300-499 500-622 Northern 11.1 19.3 38.1 56.1 83.5 Southern 14.4 47.7 36.7 24.3 2.3 UAC* 961 2016 1552 1801 473 96.6 ww 29 *UAC = Unit Area Count The Northern Highland contains 72.1% of all Class 5 and 81% of all Class 6 slope and relief units of the study area; the Southern High­ land does not exceed 15% of either class* Conversely* the Southern Highland includes between 62% and 64% of all units in Class 1 and Class 2 of local relief and average slope* The fact that neither highland province includes more than 50% of the lower three classes of slope and relief is an indication of the greater frequency of these three categories in the peninsular lowlands* The Foreland Region of the Northern Highland The foreland region of the Northern Highland* consisting of five separate subregions* comprises the largest second-order region of the study area and accounts for peninsula* 20% of the area of the The size and diversity of these subregions is considerable* 115 an indication of the heterogeneous and transitional character of the land surrounding the highland core of the peninsula. The western foreland is divided into two compartments, Sparta (6 E) and Boyne City (6 D), on the basis of contrasting slope and relief factors. Of the two, the Sparta subregion contains con­ siderably less relief and less topographic roughness! 51% of its units have; a relief of less than 150 feet and only 25% have slopes of more than 3°. The Boyne City area shows 83% and 75%, respective ly, in the same registers of relief and slope. In spite of the quite similar mix of landform types in both subregions, the morphoroetric character of the two subregions is clearly different. The Boyne City compartment (6 D) consists of six separate parcels or segments, the greatest fragmentation of any subregion, extending over 175 miles from Levering in the north to Shelby in the south. Although this subregion is bound together by certain hypsometric characteristics, a northern component (from Petoskey to Rapid City) of extreme ruggedness must be distinguished from a southern component (Rapid City to Shelby) of slightly less rugged slope and relief conditions (Fig. 33). Sixth largest of the 34 subregions of the study area, the Boyne City subregion has greater surface roughness than any other compartment, followed, in order, by the Traverse City (IB) and the Ogemaw (9C) subsections. This finding is supported by the fact that no other subregion exceeds the proportion of total area in rugged slope and relief which char­ acterizes the Boyne City compartments 38.5% and 54.6% of its units occur in Classes 5 and 6 (combined) of relief and slope, respec­ tively (Tables 12 and 13).^ (4'The ) ruggedness of terrain in the northern component is reflected in the construction of ski runs in the uBoyne Mountains•'* It is interesting that these and other winter sport areas are not 116 The northeastern foreland is divided into two contrasting terrain surfaces: the poorly-drained till plains of Posen (6 C), with relief of less than lOO feet over 82% of its area, and the lake and swamp-studded till and outwash plains of Onaway (6 B) with a relief in excess of lOO feet over 95% of its area* The Posen compartment contains fewer inclined surfaces and less relief than the nearby A l p e n a Lowland (1C) while the Onaway compartment is differentiated from the adjacent Mt, Pleasant Foreland (6 A) on the basis of greater slope and relief (Fig, 33), Isolating the Onaway subsection serves to emphasize its prevailing outwash formations which do not occur in the adjoining subregions (6 C, 1C, ID), The southeastern foreland of the Northern Highland is made up entirely of the Mt, Pleasant subregion (6 A)« This compartment is the largest morphometric unit of either highland province, com­ prising 2,450 square miles, approximately, and ranks as the third largest subregion of the study area. The subdued or undulating topography of this compartment is indicated by the fact that (1 ) no unit area has a relief in excess of 300 feet or a slope of more o than 5 , and (2) approximately two-thirds of all units have a relief of 50— 149 feet and slopes of l°-3°. A preponderance of the Mt, Pleasant subregion consists of undulating till and outwash plains as well as extensive lacustrine bottomlands with a relief of 50-149 feet. Two intracompartmental anomalies are apparent from an inspection of Figs, 13 and 14: all units with a slope of less than 1° (1) and a relief of less than 50 feet are associated with the nearly flat sandy lake plains near located on the topographic heights of the Northern Highland, but marginal to it. 117 Alger, and (2) all units with a relief in excess of 150 feet, 11.1% of the compartment area (Table 13), contain prevailing sur­ face expressions of the Gladwin, Owosso, or Port Huron moraines* These findings illustrate the control of various terrain dimensions by the presence or absence of specific landform types; however, these results apply to the Mt* Pleasant subregion alone and, because of the range of parametric values for any given land­ form type, one may not assume the same topographic expression of slope and relief for moraines, till and outwash, and sandy lake plains in other compartments of the study area* The Intermediate Upland Region of the Northern Highland The intermediate upland region is made up of three subregions which bear the names of three major river systems: the Muskegon (7A), the Manistee (7B), and the Au Sable (7C)* Bach of these rivers, because of headward erosion, is cutting back into surfaces of higher elevation and collectively they mark the greatest impact of fluvial erosion in the study area* These riverine incur­ sions manifest themselves as ringer— like indentions of the Northern Highland in the cases of the Manistee and Au Sable rivers (Fig* 34)* The Au Sable forms the longest continuous penetration while the Manistee occupies the deepest valley of the peninsula (Fig. 10)* The Manistee and Au Sable compartments, in that order, feature drainage surfaces of increasing slope and relief when com­ pared with the Muskegon subregion* The Muskegon segment contains an even distribution, approximately, of relief values on either side of the 150 foot-interval; in this respect, it is comparable to the A n n Arbor 118 (3A) subregion* This morphometric parallel may indicate a similar degree of dissection of the recessional moraines common to the two intermediate upland surfaces* Although all three subregions of this province have about the same relative area of moraines (37%-39%), the Au Sable area has more than twice as much outwash (47%-*20%) as that of the Manistee* The latter, however, has 20% of its area in sandy lake plains formed during the high-water stages of glacial Lake Michigan* The morpho- metric contrast between the sandy lake plains and the A u Sable outwash plains is discernible in Pig* 14 which shows a relief of less than lOO feet for the former, but of 100-300 feet for the latter* Although isolated segments of the Lake Border and Port Huron moraines (Lake Michigan Lobe) exceed 500 feet in local relief in the Manistee subregion, the West Branch (or Roscommon) Moraine produces a relief of 150—500 feet in 93% of the Au Sable unit areas* The corresponding frequency for the Manistee district is 73%, high in itself, and reinforces the morphometric uniformity of the Muske­ gon (7A) — Manistee (7B) — Au Sable (7C) relief slope ranking mentioned abov« (p* 116), The Upland Region of the Northern Highland Accounting for only 10% of the Northern Highland area, this second-order highland includes the Gaylord (9A), Cadillac (9B), and Ogemaw (9C) subregions at average elevations 50-250 feet above the highest average elevation of the Southern Highland* Inter- lobate moraines occupy 55%-59% of each subregion and, together with the Traverse City and Boyne City compartments, produce the most 119 UPLAND SUBREGIONS LOCAL RELIEF I I n Peel I n 0 - 4 9 □ 5 0 - 9 9 n 100 - i n 1 5 0 -2 9 9 ■ n OH/i 6 ■5 14 Fig. 35 •3 in is tent 120 HIGHLAND SUBREGIONS LOCAL RELIEF I I n Feet ) 9A ,n 1 | | ] | 0 - 4 9 5 0 - 9 9 | 1 0 0 - 1 4 9 1 5 0 - 2 9 9 MH/UI 3 0 0 - 4 9 9 9C * L m » 39 | ] 5 0 0 - 6 2 2 NOTE _- at M iyi.i.T;rt» 3.il 44 as I specified in 84 F i g .36 83 text 121 rugged topography of the study area* Although the Cadillac Highland has 3096 of its units in slope of more than 5° and includes the highest elevation of the peninsula (1706 feet), the Ogemaw sector has slightly more area in relief of 300 feet or more and slope in excess of 4°* The Gaylord Highland, with less broken topography, has considerably less area in 4° slope, or 1596-19%, and relief of more than 300 feet, or 18%—25%, than either the Cadillac or Ogemaw Highlands* Assessment of Regionalization and Comparison with Other Works A major objective of this study concerns the differenti­ ation of the topographic surface of the southern peninsula of Michigan into morphometric provinces, regions, and subregions* The morphometric provinces a discontinuous, peripheral lowland surrounding a northern and a southern highland — are established by use of the 750 feet-isohypse to designate an upper and a lower register of changing average elevation values over the entire peninsula* T U a auc a ai. O k -w*. V. -J M t, 1 M — --------: ----- j^/A» U V U i C V b ----- .1 . ..i v - -1 UtfbCl A U W U *- * - of four dissimilar altitudinal levels, called secondary morpho­ metric regions, and labelled with an identifying set of descriptive terms* Homogeneous altitudinal levels, based on average elevation in a mosaic of unit areas, facilitate the study of the gross fea­ tures of the peninsula considered as a physiographic entity* Morphometric subregions, a third-order of terrain general­ ization, are based on the changing relief and slope characteristics within the peripheral lowland and at each of the altitudinal levels 122 of the two first-order highlands* Because of the limited ranges of relief and slope data within the generally subdued terrain of the peninsula (and the resultant narrowly-defined class intervals of the data), the subregions display only diminished concentrations of the same class of relief and slope increments in their respective area-mosaics* However, the identification of relief-slope nodes may be verified from an inspection of the regional maps; the same maps point up the difficulty in constructing boundaries be­ tween such nodal areas when the change in relief and slope char­ acter is only gradual across significant horizontal distances* In such cases, subregional boundaries are drawn in a manner that equally apportions the anomalous unit areas to either of the morphometric nodes• In concept, the present work is closer to Hammond's (1957) approach than to any other in the literature of regional morpho­ metry* Hammond seeks the delineation of landform regions through the integration of slope, local relief, and profile characteristics* Slope description is based on the differentiation of steep and gen­ tle slopes; profile traits are determined by the appearance of a minimum of two-thirds of all gentle slopes in either the upper or lower register of a unit's local relief* A third profile category allows an approximately equal distribution of gentle slope in both registers of relief* Hammond overcomes the problem of creating terrain regions with homogeneous phenomena by absorbing all anomalous units, up to a maximum of four contiguous units, within any one of the 62 possible regional combinations of three terrain parameters (slope, relief, and profile). 123 The approach used in the present work seems to be more objective in stating the percentage of prevailing classes of relief and slope and the mapping of the locations of all anomalous units. Although Hammond includes the percentage of stoniness of Missouri soils as a morphological indicator, this parameter is not used in delineating landform regions. In contrast, the eleven gla­ cial landform types of the present study greatly facilitate the physiographic description of morphometric compartments. Veatch's (1957) brief outline of the physical regions of the southern peninsula, without the benefit of morphometric param­ eters, is remarkably consistent with the results of this study. This is undoubtedly the result of his use of Leverett's earlier work in differentiating glacial formations of the study area, his careful analysis of terrain types, and his rich field experience and knowl­ edge of the soilt and vegetation patterns of the peninsula. Veatch's list of 21 subregions, compared with the 34 subregions of the pres­ ent study, is notable for its lack of compartmentalization within the province of the Northern Highland. As an integrative morphometry, the present work may be com­ pared with Pike's (1963) thesis which incorporates eight weighted morphometric parameters in delineating 10 generic landform regions for New England. However, Pike's work is based on isoline con­ structions within a discontinuous network of 142 sample points representing regions of presumably homogeneous terrain character. These representative samples serve to quantify 100 square mile (average) terrain surfaces whereas the standard six square-mile unit area of this study provides a total inventory of the peninsular landscape. Calef's and Newcomb's (1953) slope map of Illinois, a single-factor morphometric study, applies the Wentworth formula to irregularly shaped tracts of unlike slope characteristics* These -tracts were isolated on topographic maps in a purely arbitrary m a n ­ ner as areas of either greater or lesser contour densities which contrast with other proximate slope tracts* Because of the diffi­ culty in assigning transitional slope zones to either the higher or lower adjacent slope tract, the authors conclude that the arbitrary or "irregular1' unit area method is virtually useless in areas of minor slope change (which characterizes much of the present study area). It has been shown, in Chapter 3, that recessional moraines are consistent relief-makers in the array of glacial landform types common to the morphometric subregions of the southern peninsula; conversely, lacustrine landforms are often without visible slope or relief* Till and outwash formations in the study area display a variety of morphometric values but tend to cluster below the mean of such ranges of the data* It was also demonstrated that moraine-and- outwash combinations of both highland provinces have steeper slopes and greater relief than the moraine-and-till associations of other locations of either highland* Finally, the coincidence of reduced values of slope and relief with the lower elevation of lacustrine features contrasts sharply with the normally greater values ob­ tained for the non-lacustrine regions of the peninsula* CHAPTER V EVALUATIONS AND PROSPECT OF TERRAIN REGIONALIZATIONS BY MORPHOMETRIC METHODS Geomorphic processes of mainly a glacial character have sculpted the peninsular landscape and a variety of constructional landforms and terrain regions remain in place as topographic relics of a past climate* The two-fold objective of the present study seeks a morphometric assessment of the genetic landform types and the terrain regionalization of the southern peninsula of Michigan* Parametric characterizations (slope, relief, and elevation) of eleven glacial landform types relate to the morphometry of ter­ rain surfaces rather than to the morphometry of individual landform types* This is accomplished by reference to the smallest standard graticule, forming a network of more than 6,800 unit areas* Because of the relatively narrow spectrum of landform dimensions found in the morphometric data, it is extremely important that grid cells be small enough to effectively isolate the limited areas of slope and relief anomalies which give geometric contrast to the abundance of subdued topographic surfaces: many maps of the study area, at simi­ lar scales, depict an undifferentiated southern peninsula as simply a 'plains' area. 125 126 The relatively small 2 1/2-minute graticule also minimizes the displacement of regional boundaries caused by placing such boundaries along unit perimeters rather than as.isolines plotted The resultant mosaic patchwork of quadrangular between data points* units produces a continuity of the morphometric data which compares favorably with that on contour or hypsometric maps of a similar scale* Small scale terrain subregions, differentiating the penin­ sular landscape according to precise combinations of landforms with known dimensions, are produced from large scale topographic maps of a 5:1 ratio to the included morphometric maps (1:2,000,000 scale research maps to 1:62,500 scale topographic maps)* Consequently, it was necessary to use area symbols in a grid network at the m i ­ croscale because point or line symbolizations, in the fine-mesh data network of unit areas, become area symbols at the scale of the completed maps* It is true that many of the subregional differentiations are based on patently minor fluctuations of the slope and relief ■4 ^ u u ^ • w«* 9 W ♦ a w 4 ^ ^ 4 1 ^ ** 1 ** 0 asm a a a j . x / m A *■ A AA ♦ w + 1a w tia w + I* A w *4 *5 *■ A i c s t* 1 + u A «A + i i w a : : %» n K < * —a n 4 a a u u a . c ^ n • X M iiO a u u not represent totally homogeneous terrain regions within the nar­ rowly defined class intervals of the diagnostic indices* However, this is the product of a subdued topographic surface without marked ranges in the morphometric data; it is also the concentration of slope and relief characteristics in area (rather than the exclusion of anomalous terrain examples) which provides the best approach to the delineation of significantly different terrain compartments within the constraints of only limited fluctuations of the data. 127 The adoption of the unit area as a reference frame is a conventional approach in the study of regional morphometry; however, the rectangular format is deficient in dealing with the varying sizes and shapes of non-geometrical parcels of homogeneous landform types, slope, or relief* When rectangular grid units are used, however, the topographic grain of the textured landform types is the determining factor in the selection of a proper unit-size; the determination of unit-size seems more critical when dealing with extensive surface formations of restricted changes of the included terrain dimensions* Although Michigan landform types with a definite shape or grain, viz*, eskers, drumlins, sand dunes, and kames, are commonly smaller than any other which has been applied to an extensive geo­ graphical area in the literature of regional morphometry* Despite the fact that a limited check of unit areas with textured landforms failed to produce significant parametric differences from the aver­ age of all the grains in a given unit area, additional research is required to establish the relationship between grain-size and the size of unit areas* The determination of a proper unit-size, in application to a diverse topographic surface, remains as a vexing problem in the literature of systematic morphometry (Thompson, 1959). The results of this study are founded on the recency, accuracy, and scale of the appropriate series of topographic maps and the M.S.F* map (Martin, 1955) of the surface formations of the southern peninsula* Thrower and Cooke (1968) have discovered subdued drumlin a n d river terrace formations on updated (1959) 128 versions of older (1902) topographic maps drawn with f ield-sketching techniques* They note that **a great deal more is said about tech­ nique and form than the quality of the contouring itself (in slope studies) Despite the improved expression of contours on topographic maps produced by photogrammetric methods, the requirement of contour accuracy remains far below the capabilities of modern photogrammetric plotters* Unchanged in more than half a century, the United States Geological Survey requirement for the placement of contours states that as many as 90% of the data points along a contour line may contain an error of as much as one-half the value of the contour interval; the remaining 10% of data points, by implication, may sustain errors of even greater magnitude* The M.S.F* map (Martin, 1955) was drawn from topographic and county-road work maps, utilizing a scale of 1:63,360; and then reduced to a scale of 1:500,000* The map identities for the various glacial landform types was first accomplished by Leverett (1915) from one-inch to the mile reconnaisance maps; his published map, x •*50,000, was revtsed and updated on the baste cf **detaalcd field work and the use of air photos•” Area, profile, and genetic characteristics of certain gla­ cial landforms are crucial to the findings of this study* For instance^ the areal extent of microfeatures, such as eskers, is so limited as to warrant study only as discrete, individual landforms* The difficulty in differentiating recessional and ground moraines ^ ^ P e r s o n a l correspondence: Helen M* Martin, August, 1963* 129 is yet another problem and the determination of such boundaries introduces a subjective element in the genetic classification of area landforms. Suggestions for Future Research There is a need to discover the morphometric character of specific formations on a chorographic basis* This approach might be extended to include, for instance, a detailed morphometry of the moraines of a given substage of the Wisconsin glaciation and their changing dimensions at different locations of the peninsula* The derived terrain data of these or other formation types could lead to certain conclusions regarding the rates of glacial ablation in various locations, and, possibly, the influence of a previous gla­ cial or non-glacial topographic surface on a subsequent, superposed surface* Because most of the parameters utilized in the present study were developed for the purpose of quantifying "alpine" ter­ rain surfaces, their application to the lacustrine flats of the eastern coastal plain has produced relatively narrow rAnnp^ of the morphometric data* With the possible exception of parameters deal­ ing with the order and texture of drainage (Strahler, 1957), there is a notable lack of terrain indicators which could serve to detect the subtle topographic differences which occur in different sections of these flat bottomlands* Of the parameters used in this study, the index of average slope displayed the greatest sensitivity to these intraregional differences* A perusal of the appropriate topographic maps supports the 130 contention that (1) the bank slopes of individual stream courses, and (2) the abrupt, but minor, changes in elevation along the face of the various beach benches make up the principal diagnostic fea­ tures of extensive portions of the lacustrine plains* Because of the lack of specific techniques designed to quantify these kinds of subtle and extremely localized changes in slope and relief, the need for the innovation of flat-land parameters is crucial to future research* It is conceivable that such indices might express these "concentrated" relief and slope changes in a ratio to the total increment of slope and relief for the entire unit area* A study of the peninsula's major drainageways is needed to complete the analysis of its gross landforms* Such a study should include a longitudinal survey of trunk valleys using the mean valley depth parameter (Pike, 1961), This indicator, actually an adaptation of Wentworth's formula, provides an index value (in feet) of the average depth of valley surfaces below the crest elevations of bounding ridge lines* In practice, 10-mile traverses (perpendicular to the channel) are analyzed for counts of slope direction changes and contour crossings« A comparison of s” cb index values, vaken at predetermined intervals along the longitudinal stream profile, should provide a formulatory basis for characterizing the basin character of each major stream or river; comparisons with the p r o ­ file characteristics of other discharge channels should produce meaningful regional differences* The Need for an Automated Morphometry The tedious and time-consuming nature of collecting 131 morphometric data is either directly stated or implied (on the basis of parameters used) in this and each of the regional studies mentioned in the text. It is unfortunate that a direct correlation exists between the size of the study area and the complexity of the parameters used; studies involving larger areas were based either on sampling techniques or on relatively simple terrain parameters* Re­ search should be initiated to develop new parameters which can make use of unsophisticated terrain data to describe more abstract char­ acteristics of the topographic surface; these new methods might provide such indices as the total area of a deformed surface com­ pared to the area of the geometric plane surface of the unit itself* In the absence of new concepts, ordinary data can be manip­ ulated quickly and accurately with the use of automated equipment* The quaint ity and precision of computer-produced maps provide a powerful tool for the analysis of data and serve as a logical a d ­ junct to the present calculations of local relief, average slope, the Index of Comparative Relief, and the Elevation-Relief Ratio* Future research will come to rely on the speed and load-bearing ^ 4 ^ 1 kC ** kXA US Vk A m -i T-m 1 « k .£ * /M V n •»4* M UW U ^n fi♦ r* kkW M « kW k4 M U S U W k * TMtn K k i M « W U W % * cards or magnetic tapes* It is now feasible for computers to produce a multitude of work maps on ordinary line printers utilizing a variety of class intervals, representative symbols, and numerical transformations of the data* The perception of the various data distributions in a sequence of computer-produced maps will allow the researcher a choice of options not available previously because of the time/ cost limitations in producing the maps or calculations by hand* 132 The SYMAP computer program, composed at the Laboratory for Computer Graphics (1968), can produce choropleth and isopleth maps on ordinary line printers and allows the user much flexibility in imput and output specifications* The recent introduction of the calculating computer adds a new dimension to computer-produced graphics: the coded data of a matrix imput is utilized in a threedimensional representation given either as a histogram format or a topographic surface* These programs have meaningful cartographic applications in the presentation of morphometric data, but the actual production of research maps is limited at the present time because of the lack of access to computers with very large memory cores, expensive coordinate plotters, and a technical staff needed to operate and maintain the diversity of program variations* However, t h e feasi­ bility and increased tempo of morphometric research is virtually assured as cost-reductions occur and as the availability of trained programmers increases* The application of primary scanners to conduct contour counts and of densitometers for assessing contour densities, however, might be even more fundamental to morphometric research than com­ puter analysis of data* The time-consuming nature of data-collection remains as the largest single impediment to increased morphometric research* 133 APPENDIX Morphometric Data for Regions, Subregions, and Landform Types -In Unit Areas Containing 50% or more of the Indicated Genetic Landform Type 134 TABLE 12 : HUMBER OF UNIT AREAS IN THE VARIOUS REGIONS AND PERCENTAGES OP UNITS OCCURRING IN SPECIFIED CLASSES Subraaion: Number of Unit Areas: 0-0.99 AND SUBREGIONS OF SLOPE Classes of Averaae Slope: (In Degrees) 4-4.99 2-2.99 3-3.99 1-1.99 5-10 TOTAL ---1.6 41.5 3.1 2.6 100. -- >• -- " --2.0 <• " 1A IB 1C ID IE IF 1G Subtotal (412) (631) ( 74) (123) (130) (127) (496) 2193:32.1* 82.0% 69.0 23.0 30.9 11.5 14.2 15.7 15.1% 27.0 52.7 30.6 4.6 41.0 40.4 2.2% 5.1 18.9 20.3 6.2 25.2 25.2 0.7% 1.9 4.1 4.9 23.1 10.2 11.1 ---1.3 1.6 13.1 6.3 5.0 2A 2B 2C 2D 2E 2F 2G 2K subtotal ( S9> ( 36) (258) ( 72) (327) (105) (203) (115) 1185:17.3% 10.1 11.1 58.5 -28.2 33.3 4.6 12.2 60.9 27.8 40.3 41.7 53.6 58.0 20.6 23.7 29.0 47.2 1.2 55.5 18. 1 6.7 35.6 40.9 ___ 11.1 -2.8 0.1 1.9 23.7 16.5 --2.8 -------13.3 1.7 3A 3B 3C Subtotal (241) (351) (104) 696S10.2X *.1 6.9 1.0 68.6 16.4 45.6 22.2 25.0 21.2 2.3 36.5 2.5 -14.3 0.8 -6.7 U 4B Subtotal ( 67) (126) 193: 2.8% 2.4 9.0 46.0 47.7 34.9 32.8 14.3 9.0 1.6 1.5 .8 5A Subtotal ( 44) 44: -- 25.0 68.2 6.8 -- -- 6A SB SC 6D SB Subtotal (369) (279) ( 38) (357) (225) 1268:18.6% 24.7 2.5 13.1 3.7 6.7 42.3 23.7 60.6 7.0 24.9 26.0 36.6 23.7 15.1 40.9 6.2 22.6 16.5 19.1 0.8 10.0 2.6 15.7 6.6 42.0 1.8 7A 4.2 2.B 7C Subtotal (189) i1 /ei (137) 502: 7.4% ... 19.6 11.4 6.6 34. a 29.0 11.7 24.3 17.6 27.0 13.8 17.6 23.3 3.7 21.6 31.4 8A SB Subtotal (423) ( 67) 490: 7.2% 6.4 55.2 25.8 31.8 23.6 6.0 16.8 11.3 16.1 9A 9B 9C Subtotal (111) (111) ( 39) 261: 3.8% -... 2.7 20.7 28.9 18.9 15.4 1B.0 16.2 16.2 16.2 29.n 28.2 33.3 15.4 TOTAL 6832:100% 1488 30.8% 767 11.8% 1o __ 27.B .6% 1609 23.5% 16.2 7.7 2101 30.8% Expressed in the subtotals. -- 402 5.9% __ 4.6 -- «—— 13.5 465 6.8% •' •• 135 HUMBER OF UNIT AREAS IN THE VARIOUS REGIONS