Bell & Howell Information and Learning
300 North Zeeb Road, Ann Arbor, MI 48106-1346 USA
800-521-0600
UMI The Potential of Soil Survey Data
in a Quantitative Evaluation of Surficial Geology
Mapping in Northern Maine

by

Rosalia Evans
Thesis submitted to the
Eberly College of Arts and Sciences
at West Virginia University
in partial fulfillment of the requirements
for the degree of
Acknowledgements iv
Table of contents v
List of figures vi
List of tables vii
Introduction 1
Purpose 5
Research objectives 7
Study areas 8
Previous research 11
Research materials and methods 14
Soil survey data 14
Converting soil mapping units to geologic units 22
Soil-surficial geology framework development 25
Prescott’s map data 27
Analysis of Prescott’s map data 28
Genes’s map data 31
Analysis of Genes’s map data 33
Kite’s point observations 35
Analysis of Kite’s point observations 37
GIS overlay results 42
Kite’s point data compared to the derivative map’s classifications 42
Kite’s point data compared to Prescott’s surficial geology map 46
Kite’s point data compared to Genes’s surficial geology maps 48
Prescott’s map units compared to the derivative map’s classifications 50
Genes’s map units compared to the derivative map’s classifications 51
Analysis of Kite’s point data compared to Prescott, Genes, and the derivative map 52
Analysis of the derivative map compared to Kite, Genes, and Prescott 55
Conclusion 73
Bibliography 80
APPENDIX 1 Soil series descriptions 83

5 Area and percentage of area of surficial geology map units on the derivative surficial
geology map derived from soil surveys 30
6 Area and percentage of area of surficial geology map units on portions of Prescott’s
ground-water favorability map 30
7 Area and percentage of area of surficial map units on portions of Genes’s Fort Kent
Stockholm, Van Buren, and Eagle Lake quadrangles 34
8 Descriptions of the landforms upon or in which Kite’s point observations are located 39
9 The origins of Kite’s data at the 248 point stations 39
10 Sub-categories of Kite’s, Prescott’s, Genes’s, and the derived map’s surficial geology
combined for comparison purposes 41
11 Kite’s point data compared to the derivative map’s surficial geology classifications 43
12 Agreement between Kite’s point data and the derivative map’s classifications 43
13 Kite’s point data compared to Prescott’s surficial geology 47
14 Agreement between Kite’s point data and Prescott’s mapped units 47
15 Kite’s point data compared to Genes’s surficial geology 49
16 Agreement between Kite’s point data and Genes’s mapped units 49
17 Prescott’s map compared to the derivative map at Kite’s point locations 51
18 Agreement between the derivative map and Prescott’s map 50
19 Genes’s map compared to the derivative map at Kite’s point locations 51
20 Agreement between the derivative map and Genes’s map 52
21 The probability till mapped by Genes and Prescott, and classified on the derivative
map is interpreted by Kite as the same 54
22 The probability alluvium mapped by Genes and Prescott, and classified on the
derivative map is interpreted by Kite as the same 54
23 The probability ice-contact stratified drift mapped by Genes and Prescott,
and classified on the derivative map is interpreted by Kite as the same 54
24 The probability till, ice-contact stratified drift, and alluvium mapped by Genes and
Prescott, and classified on the derivative map is interpreted by Kite as the same 55
25 The probability till mapped by the geologists is classified on the derivative map
as the same 56

38 Gene’s till interpretations arranged according to percent of slope compared to the
derivative map’s till classifications 66
39 Prescott’s till interpretations arranged according to percent of slope compared
to the derivative map’s till classifications 67
40 Kite’s alluvium interpretations arranged according to percent of slope compared to
the derivative map’s alluvium classifications 68
41 Gene’s alluvium interpretations arranged according to percent of slope compared
to the derivative map’s alluvium classifications 69
42 Prescott’s alluvium interpretations arranged according to percent of slope compared
to the derivative map’s alluvium classifications 70
43 Kite’s ice-contact stratified drift interpretations arranged according to percent of
slope compared to the derivative map’s ice-contact stratified drift classifications 71
44 Gene’s ice-contact stratified drift interpretations arranged according to percent of
slope compared to the derivative map’s ice-contact stratified drift classifications 71
45 Prescott’s ice-contact stratified drift interpretations arranged according to percent
of slope compared to the derivative map’s ice-contact stratified drift classifications 72
INTRODUCTION

The late Wisconsin Laurentide Ice Sheet’s final retreat from the St. John River
Valley between approximately 12,000 and 11,000 YBP (Hughes et al., 1985; Thompson
and Borns, 1985; Lowell and Kite, 1986b) left a landscape rich in glacial deposits.
Commonly stratified, unconsolidated deposits in the St. John chronicle the Quaternary
geologic history of the area. This transported surficial material is dominated by till,
outwash, ice-contact stratified drift deposits, and alluvium. The unconsolidated glacial
deposits were derived locally and from the Canadian Shield (Genes et al, 1981).
Subsequent surficial processes and soil genesis during post-glacial time resulted in the
formation of 73 different soil map units in the two study areas along the St. John River in
northern Maine.
Surficial geology maps are becoming more important as environmental issues
related to human population growth increase. The earth's resources, formed by or

areas has a controlling effect on the soil properties that lead to the identification of a
particular soil. The parent material associated with particular soils of the St. John River
Valley (table 1) was obtained from Arno (1964).
Roy Charles Lindholm (1993) determined that compiling a derivative geology
map based solely on soil distribution portrayed on soil survey maps may be a valuable
aid to geologists or others in the absence of surveyed geologic maps. The northern
Maine study areas are covered by soil survey maps (Arno, 1964) as well as surficial
geology maps (Prescott, 1973) (Genes, undated; 1978A; 1978B; 1981). Unpublished
field-observed point data were also available for the areas (Kite, 1996). These rich
sources of soil and surficial geology data afford a unique opportunity to investigate the
soil-surficial geology relationships in northern Maine. 3

Table 1
PARENT MATERIAL SOIL SERIES DRAINAGE
THICK TILL gravelly, stony Plaisted
Howland
well
moderately well
gravelly Caribou well
fine, stony Mondara
Burnham
Easton
Washburn
poor
very poor
poor
very poor

moderately well
poor
very poor
ALLUVIUM
Mixed Alluvial Land
Hadley
Winooski
poor/very poor
well
moderately well
MADE LAND
Made land
SWAMP
Peat & Muck under water
OUTCROP
Steep Rockland &
Thorndike
well

Table 1. Parent materials associated with particular soils in the St. John River
Valley, northern Maine, study areas. (Arno, 1964) 4

Hydrologic Atlas (Prescott, 1973) and Maine Geological Survey Maps of the Surficial
Geology of Fort Kent, Stockholm, Van Buren, and Eagle Lake quadrangles (Genes,
undated; 1978A; 1978B; 1981). Unpublished records of surficial deposits obtained
through extensive field work conducted by J. Steven Kite from 1977 to 1989 in
Northeastern Aroostook County, Maine, collected for genetic-stratigraphic interpretation
provide a third source of surficial geology data for this project.
A map portraying the geographic distribution of the surficial geology of the areas
based solely on information derived from soil survey maps was compiled. Agreement
among the derivative map, Prescott's map, Genes’s maps, and Kite’s point data was
assessed. Kite's point records of surficial geology deposits were used to determine the
position for all comparisons.
67

RESEARCH OBJECTIVES
In order to fulfill the purpose set forth in this study, the following objectives have
been identified:
Primary objective
1. To employ the multi-layer analytical capabilities of a geographic information
system (GIS) to examine soil-surficial geology relationships in Northern Maine.

Secondary objectives
1. To compile a derivative surficial geology map of selected areas in the St. John
River Valley in Northeast Aroostook County, Maine, derived solely from soil survey data
as represented in the Soil Survey of that area (Arno, 1964).
2. To compare the derivative surficial geology map, Prescott's (1973) surficial
geology map, Genes's (undated; 1978A; 1978B; 1981) surficial geology maps, and the
geographic distribution of surficial geology deposits obtained through field observations

(Arno, 1964).
Soil formation is also a reflection of the length of time the soil has been
developing (Arno, 1964). Soil age is indicated by horizon development. The oldest soils
in the area have developed in till or outwash. These soils have developed since the Late
Wisconsin glaciation, and have strong A, B and C horizons. Soils on stream terraces
have weakly developed horizons. Soils in recently deposited alluvium have very little or
no horizon development.
The topography along the St. John River consists of narrow floodplains (Arno,
1964). Few of these floodplains extend beyond one half mile in width. The moderately 9
sloping ridges in the area are 800 to 1400 feet above sea level with the steepest slopes
extending in a northerly direction toward the St. John River.
The area has a cool, humid continental climate (Arno, 1964). The ground is
frozen between November and April denoting a long cold winter. Precipitation averages
34 inches per year, with much of it falling as snow.
1011
PREVIOUS RESEARCH

Part of the literature relevant to this study concerns the areas’ geographic
location in relation to late Wisconsin history. This area has received much attention
concerning the deglaciation pattern of the Laurentide Ice Sheet (Lowell, et al., 1990).
Theories exist regarding the expected distribution of glaciofluvial deposits that may be
confirmed or refuted, in part, by the additional evidence obtained by this research as to
where these deposits are in relation to a postulated ice cap divide extending over the
area (Fastook, et al., 1980; Hughes, et al., 1985). Support for this glacial history has

1994).
Less time is involved in the production of a geologic map compiled from soil
survey sheets versus one compiled employing conventional mapping techniques.
Conventional mapping techniques involve extensive research and observation in the
field. Conventional soil mapping data for soil surveys include: steepness, length and
shape of slope, stream size and speed, kinds of vegetation, types of rock, bore hole
data, comparisons to local and regional soil profiles, and comparisons to aerial
photographs (Arno, 1964). Data generated for surficial mapping include: location type,
coordinate system, landform, elevation, photos, sediment composition of geologic units,
age, depositional environment of the unit, sample analysis including thickness, texture,
color, HCl reaction, unit sorting, bed sorting, strength, roundness and sphericity,
sedimentary structures, secondary feature development subsequent to original
deposition, underlying contact units, and directional data (Kite, et al., 1986; Kite, 1994).
Since soil surveys contain descriptions of each soil type and the parent material in which
it formed, using these surveys as a source for geologic information involves merely
identifying the soil type formed in a particular parent material. 13
The area surveyed in relationship to the amount of time involved in conventional
mapping indicates a degree of intensity (Kite, 1982). Prescott (1973) in compiling his
Groundwater Favorability and Surficial Geology of the St. John River Valley, Maine
surveyed at a pace of approximately 16 km
2
to 24 km
2
per day (assuming a five-day
work-week) with perhaps a reduction in this intensity of 8 km
2
to 12 km
14
RESEARCH METHOD
Soil Survey Data
The study areas contain smoothly sloping soils in till derived from shale and
limestone, smoothly sloping soils in till derived from acid rock, irregularly sloping shallow
to moderately deep soils in till derived from calcareous rocks, irregularly sloping soils in
till derived from acid rock, nearly level to sloping soils of the flood plains and terraces,
and nearly level to gently sloping poorly drained and very poorly drained soils in firm till.
The landscape is comprised of geomorphic features that include floodplains, bedrock
outcrop, terraces, eskers, deltas, kames, kame terraces, and hummocky, blanket and
veneer surfaces. A total of 73 different mapping units were identified by the Soil
Conservation Service (Arno, 1964) (table 2) in the two study areas. These units
represent 28 soil series (table 3). Appendix 1 describes the soil series.
The 73 mapping units are located on portions of 17 Soil Survey of Aroostook
County, Maine, Northeastern Part (Arno, 1964) map sheets at a scale of 1:20,000. Aerial
photographs comprise the base maps for the soil data. Each paper map sheet covers
approximately 38.7 km
2
. The two study areas lie within 47

°20’00’’ and 47

°00’00’’ N
latitude, and 68

°40’00’’ and 67

°45’00’’ W longitude. The soil data were collected on the

CoA Conant silt loam 0 to 2
CoB Conant silt loam 2 to 8
CoC Conant silt loam 8 to 15
DaB Daigle silt loam 2 to 8
EaA Easton and Washburn silt loams 0 to 8
EaB Easton and Washburn slit loams 2 to 8
EsB Easton and Washburn stony silt loams 0 to 8
FhA Fredon and Halsey silt loams 0 to 2
FhB Fredon and Halsey silt loams 2 to 8
HaA Hadley silt loam level
HaB Hadley silt loam undulating
HoA Howland gravelly loam 0 to 2
HoB Howland gravelly loam 2 to 8
HoC Howland gravelly loam 8 to 15
HvB Howland very stony loam 0 to 8
HvC Howland very stony loam 8 to 15
MaA Machias gravelly loam 0 to 2
MaB Machias gravelly loam 2 to 8
MaC Machias gravelly loam 8 to 15
MbA Madawaska fine sandy loam 0 to 2
MbB Madawaska fine sandy loam 2 to 8
MbC Madawaska fine sandy loam 8 to 15
Md Made land
MhB Mapleton shaly silt loam 0 to 8
MhC Mapleton shaly silt loam 8 to 15
MhD Mapleton shaly silt loam 15 to 35
Mn Mixed alluvial land
MoA Monarda and Burnham silt loams 0 to 2
MoB Monarda and Burnham silt loams 2 to 8
MrB Monarda and Burnham very stony silt loams 0 to 8

Sb Steep rockland and Thorndike materials
SgA Stetson gravelly loam 0 to 2
SgB Stetson gravelly loam 2 to 8
SgC Stetson gravelly loam 8 to 15
SgD Stetson gravelly loam 15 to 25
SgE Stetson gravelly loam 25 to 45
ThB Thorndike shaly silt loam 0 to 8
ThC Thorndike shaly silt loam 8 to 15
ThD Thorndike shaly silt loam 15 to 25
ThE Thorndike shaly silt loam 25 to 45
TkB Thorndike very rocky silt loam 0 to 8
TkC Thorndike very rocky silt loam 8 to 15
TkD Thorndike very rocky silt loam 15 to 25
TkE Thorndike very rocky silt loam 25 to 45
Wn Winooski silt loam

Table 2. The 73 soil mapping units in the northern Maine study areas and associated
slope (Arno, 1964).
17
Table 3


Washburn
firm till
Glacial till derived from
weakly calcareous or acid
shale or slate

Daigle
firm till

Glacial till derived from
acid or weakly calcareous
shale
Thorndike &
Steep
Rockland
thin till

Glacial till derived from
calcareous rock
Mapleton
thin till

Glacial outwash and
gravelly material on
terraces
Stetson
gravelly

Machias
gravelly

18
cases stream forks, and locating these on the soils base of the aerial photographs to use
as tic points as the soils map sheets contained no spatial referencing.
Once digitized, the covers were appended into the final coverage representing
the study areas (figure 3). This vector format was edited interactively to add polygon
labels that identify the soil units, to repair open and closed polygons, and to connect like
soil polygons to adjacent units. The purpose of this outline map of the digitized soil
sheets is not for analysis or intrepretation, but merely as a depiction. Any map of the
soils data at the scale this format allows would be impossible to represent in a
meaningful way. A legend of the 73 soils on the depiction would be too extensive to
include.
The described conversion process introduces various sources of positional error
from the original maps or from digitization or from registration that may impact the overall
accuracy of the database and the end product. The source soil documents used in the
conversion may be distorted resulting in map boundary adjustments. This distortion may
be due to paper shrinkage or expansion. Because the original soils maps are no longer
available, distortion may also result from photocopying these maps for digitizing. The
original drafting may also be flawed (Maling, 1989). Aerial photographs taken in 1942,
1947, and 1960 and used in the construction of the soil base maps in 1962 may no
longer be accurate. The soil boundaries were drawn by the Soil Conservation Service on
these photographs because they show woodlands, buildings, field borders, trees, roads,