GEOLOGICAL SURVEY CIRCULAR 268

PRELIMINARY REPORT ON THE GEOLOGY AND GROUND-WATER RESOURCES OF THE MATANUSKA VALLEY AGRICULTURAL AREA ALASKA

UNITED STATES DEPARTMENT OF THE INTERIOR Douglas McKay, Secretary

GEOLOGICAL SURVEY W. E. Wrather, Director

GEOLOGICAL SURVEY CIRCULAR 268

PRELIMINARY REPORT ON THE GEOLOGY AND GROUND-WATER RESOURCES OF THE MATANUSKA VALLEY AGRICULTURAL AREA, ALASKA

By Prank W. Trainer

Washington, D. O, 1963 Free on application to the Geological Survey, Washington 25, D. C.

CONTENTS

Abstract...... 1 Ground water> Continued 'Introduction...... 1 Water-bearing materials...... 18 Location and extent of area...... 1 Bedrock...... 18 Purpose and scope of investigation. 1 Till f'hardpan")...... , 18 Previous investigations...... 3 Outwash sand and gravel...... 19 Method of well numbering...... 4 Windblown sand and silt...... 20 Acknowledgments...... 4 Hydrology of physiographic units., 20 Geography...... 4 Lazy Mountain...... 20 Climate...... 4 Bodenburg terrace area...... 21 Topography and drainage...... 6 Matanuska-Knik flats and Vegetation...... 7 alluvial plains...... 23 Culture...... 7 Palmer terrace...... 23 Geology...... 7 Eska Creek-Matanuska Agri­ Mesozoic and Tertiary rocks...... 7 cultural Experiment Station Quaternary deposits...... 8 area...... < 23 Glacial deposits...... 8 Little Susitna-Goose Bay Till ("hardpan")...... 8 morainic area...... 25 Outwash sand and gravel.. 12 Quality of water...... < 26 Fluvloestuarine de­ Water-level fluctuations...... 26 posits...... 14 Construction of wells...... 29 Nonglacial deposits...... 14 Dug wells...... 29 Windblown deposits...... 14 Driven wells...... 29 Other deposits...... 15 Drilled wells...... c.. 29 Perennially frozen ground Utilization of ground water...... 31 (permafrost)...... 16 Public supply...... 31 Postglacial weathering and Domestic and farm supplies.., 31 erosion...... 16 Construction materials...... 31 Quaternary history...... 17 Sand and gravel...... 31 Ground water...... 18 References cited...... 31 Occurrence...... 18 Unpublished reports...... 32

ILLUSTRATIONS

Plate 1. Map showing surficial geology of the Matanuska Valley agricultural area and location of wells...... 'In pocket Figure 1. Index map showing the location of the Matanuska Valley agricultural area...... < 2 2. Physiographic units in the Matanuska Valley agricultural area...... 3 3. Generalized section of the Matanuska Valley agricultural area showing stratigraphic units...... 4 4. Generalized section showing perched ground water in windblown sand and silt...... < 22 5. Map and section showing distribution of older glacial deposits known from well logs...... < 22 6. Graphs showing water-level fluctuations in wells...... < 30

TABLES

Table 1. Climatological data...... 5 2. Logs of wells showing older glacial deposits, 11 3. Analyses of ground water...... 27 4. Data from mechanical analysis of sediments.., 28 5. Records of wells...... 34

PRELIMINARY REPORT ON THE GEOLOGY AND GROUND-WATER RESOURCES OF THE MATANUSKA VALLEY AGRICULTURAL AREA, ALASKA

By Prank W. Trainer

ABSTRACT This report describes the geology and A table giving records of 333 wells is ground-water resources of an area of present included in this report. The locations of and potential agricultural development iii the wells are shown on the geologic map. south-central Alaska. The agricultural area lies on a wide valley floor, most of which is formed by glacial deposits. Nonglacial un- INTRODUCTION consolidated deposits include windblown material distributed generally over the agri­ Location and Extent of Area cultural area and slope deposits along the valley walls. Small bodies of perennially The Matanuska Valley is a papf of the frozen ground (permafrost) are present in lowland lying north of the Cteaat Range in some bogs. south-central Alaska. The valley of the Matanuska River and the lowland extending Till ("hardpan"), possibly of late westward from it to the Susitna River are in "Wisconsin (Mankato) age, occurs at the" sur­ the Matanuska and Tfasilla districts as de­ face or beneath surficial outwash gravel de­ fined by P. S. Smith (1939, pi. 3).I/ The a- posits in a large part of the valley floor. rea described by the present report, here­ In several widely separated localities th.e after termed the Matanuska Valley agricultur­ till is known to be underlain by older gla­ al area, is best known as including the site ^ cial gravel, and the presence of an older of agricultural colonization undertaken by /V^O till beneath this gravel is suspected. the Federal Government in 19*9-. It lies be- ' \ ^ Several types of outwash deposits, most of tween the Talkeetna Mountains on the north them formed during glacial recession in this and the Chugach Range on the south (fig. 2). area, are differentiated on the geologic map It is bounded on the north by the Talkeetna which accompanies the report. Associated Mountains and the Little Susitna River, and with glaclofluvial deposits of existing on the south by the Knik River and Knik Arm. streams are estuarine deposits of glacial It lies between Eska Creek on the northeast silt. The topography developed on the uncon- and Goose Bay on the southwest. As tnus de­ solidated deposits is due chiefly to glacial fined, the area lies approximately between deposition, large-scale stagnation of ice, 148°55 f and 149°50' west longitude and between and trenching of glacial deposits by outwash 61°25 f and 61°45' north latitude; it covers streams. about 350 square miles. (See fig. 1.) Most wells in the agricultural area obtain water from gravel. Supplies sufficient Purpose and Scope of Investigation for domestic and farm use are generally avail­ able wherever the gravel is saturated. Only Studies in the Matanuska Valley agri­ a little is known of the quantities of water cultural area were nade by the writer during available. Till in this area is relatively the 1949 and 1950 field seasons and part of imp^rHie_abl_ej_mosjL> wells in till obtain water the 1951 season, as a part of the investi­ from included thirtor gravel layers. Bedrock gation of ground-water resources of Alaska here appears to be^a relatively poor water-­ begun by the U. S. Geological Survey in 1947. bearing material. The purpose of the studies in the area was to map the water-bearing materials and to The area is divided into six physiograph­ determine the occurrence, availability, and ic units to facilitate description of the quality of ground water in the area. The occurrence of ground water. need for the compilation and interpretation of geologic and hydrologic data became in­ Recharge of ground water is chiefly creasingly important after colonization in from precipitation, but parts of the area 1935, and this need has increased during the receive drainage from adjacent mountain slopes. postwar period of continuing settlement. Fluctuations of the water table as much as Many homes and farms are completely dependent several feet were observed during the period upon wells for their water supply, and home­ 1949-51. steaders taking up new land in undeveloped areas have lacked data on the availability of Chemical analyses show that the ground ground water. More extensive utilization of water ranges from moderately hard to very ground water, possibly including irrigation, hard but is suitable for general domestic undoubtedly will come in the future. and farm uses. _!/ See page 31 for list of references, During the 1949 field season a well in­ Knik quadrangles of the Corps of Engineers ventory and a water-level observation program of the U. S. Army. were begun. Locations and descriptions of existing wells were recorded. The altitude Three test wells were drilled by the of the land surface at each well was deter­ Geological Survey, using the jet-percussion mined with an altimeter from benchmark con­ method. trol. Depths to water levels in wells were measured wherever possible with a steel tape, Samples of unconsolidated materials ex­ and periodic measurement of water levels in posed in the area were collected for labora­ selected wells wes begun. This program is tory study. Mechanical analyses were made being continued as of the date of writing of by the writer, using sieves for the coarser this report (spring 1952). In addition to fractions < and the hydrometer or the pipette the well inventory, preliminary geologic method for the finer fractions. (For a de­ field work was done in 1949; the geology was scription of methods in general use for mapped on areal photographs during 1950 and mechanical analysis of granular materials, 1951. Data were transferred from the photo­ see Am. Soc. Testing Materials, 1950.) The graphs to a base map with a vertical sketch- penneability of small undisturbed samples master. The base used for the surficial- was determined in the field with a variable- geology map (pi. 1} was taken from parts of head permeameter (Wenzel, 1942, p. 64). the Button, Matanuska, Eklutna, Houston, and

54

I40W 132"

Figure 1. Index map showing the location of the Matanuska Valley agricultural area 2 Data representing 333 wells are tabula­ stone made a number of water-level measure­ ted in the well records which are part of ments. this report. Included are the locations and a brief description of the topographic situa­ tion of each well, and such information as Previous Investigations is available on the depth of the well, the water level, the yield of the well, and the type of water-bearing material. The loca­ No published reports describe in detail tions of wells are shown on plate 1. the geology of the Matanuska Valley agricul­ tural area, although parts of the area are discussed in several publications. Martin The ground-water investigation was made and Katz (1912) describe that part of the under the general direction of A. N. Sayre, area in the vicinity of Moose and Bska Creeks, chief of the Ground Water Branch of the Water and Landes (1927) describes the district be­ "Resources Division of the Geological Survey. tween the Knik and Matanuska Rivers, includ­ The field work was supervised by D. J. Ceder- ing part of the Chugach Range. The geology strom, district geologist of the Ground Water of the general region is discussed briefly by Branch. M. J. Slaughter, G. W. Whetstone, Capps (1940) and the physiography by Martin and Mrs. Arline Day, of the "Water Resources (1942). Bockie (1946) gives the most comr Division at Palmer did much to facilitate the plete description of the physical geography field work. E. C. Casey, D. C. Phillips, of the agricultural area. Karlstrom (1950) Clifford Shaw, Mr. Slaughter, and Mr. Whet­ includes the area discussed in the present

150° 00' 149°40' 149° 00'

EXPLANATION Lozy Mountain 5. Eska Creek-Experiment Station area rt^\«s ^* Eska Creed Bojdenburg terrace area o. Coacpicuouo ridges and-hills underlain Matanuska-Knik flats and ' hy. grrmoln- ^ alluviol t«w>«rce &LM/tl' f b TiUoovei'td uj Iliiiicgrovel X 4. Palmer terrace a., XiH oBtfei-ed-by-tTfirrgTavel i. 5a &_Little Susttmr- Goose Bay morainic area 61° 401

61° 20' 150*00' 149° 40' 149° 20' 149° 00'

5 10 Miles I ! I

Figure 2. Physiographic units in the Matanuska Valley agricultural area.

3 report in a map of the larger area bordering 'Without exception, residents of the area Cook Inlet. Other papers, including those by willingly permitted access to wells on their Black (1951), ttockie (1942), and Tuck (1938), property or provided information regarding treat of special problems of the geology of them. Henry LaRose, A. R. and Thomas Moffitt, the area. and James and Albert Frey, drillers, describ­ ed their experience in the Matanuska Valley and gave the writer, much valuable informa­ Method of Well Numbering tion, T. B. Bourne and Associates, Inc., consulting engineers, provided data obtained Wells described in this report are de­ during construction of a test well for the signated by simple consecutive numbers; blank city of Palmer. The owners of the observa­ numbers have been left to accomodate wells tion wells listed in the well records permit­ that may be constructed in the future. In ted use of their wells for this purpose, and sequence the well numbers follow approximately J. C. Baldwin, Henry LaRose, F. B. Linn, the physiographic units or subareas describ­ Loren McKechnie, G. E. Murphy, Oscar Tryck, ed in this report. (See p. 20.) The first and Noel Woods made periodic water-level meas­ wells listed are on the lower slope of Lazy urement . Mountain (pi. 1). Wells east of the Matanus- ka River and north of the Knik River are list­ ed next. Then follow, in order, wells on GEOGRAPHY the terrace at and south of Palmer, wells be­ tween Eska Creek, Four Corners, and the Climate vicinity of the Matanuska Agricultural Ex­ periment Station, and wells located in the The climate of the eastern part of the remainder of the agricultural area to the Cook Inlet lowland, which includes the Mata­ west and southwest. nuska Valley agricultural area, is the result of a combination of marine and continental influences. The lowland lacks both the high Acknowledgments rainfall of coastal areas and the temperature extremes of the interior of Alaska, For many Yourtesies the writer is indebt­ ed to W. A. Jtyckie, Soil Conservation Ser­ Although climatic data have been collect­ vice, Portland, Oreg., to C. W. Wilson and ed at several localities in the agricultural T. H. Days Soil Conservation Service, Palmar, area in recent years, the only extended rec­ and to D. L. Irwin and &. H. Mick, Alaska ord is that for the Alaska Agricultural Ex­ Agricultural Experiment Station, Palmer. periment Station near Matanuska. Selected James Hurley made available copies of well data for this locality are presented in table logs from the files of the,Alaska Rural Re­ 1. habilitation Corp. The Matanuska Valley Fair Association permitted the use of storage The departure from the mean annual precip­ space. itation and the seasonal distribution of precipitation are significant. In a given Special thanks are due the late Kirk year the total precipitation may be about one- Bryan, and M. P. Billings, K. F. Mather, third greater or less than the mean. The H. C. Stetson, and C. E. Stearns, of Harvard mean total annual snowfall is about 3-^ foet, University, for their discussion of and many but the annual departure from the mean may suggestions regarding the writer's work. be as great as half this amount. In most

Oe(dune)

Nonglaciol deposits I Oe Windblown sand and silt Ogp Pitted outwosh deposits I Qt Talus deposits Om Ground moraine (including inferred older till) [-QUATERNARY Oof Alluvial-fan deposits Oml Lateral moraine J Of Frost-disturbed deposits MT Bedrock }MESOZOIC AND TERTIARY

Glacial deposits Qfe Fluvioestuarine deposits Ogr Modern alluvial-plain deposits Qga Older alluvial-plain deposits

Figure 3. Generalized section of the Matanuska Valley agricultural area showing stratigraphic units. Table l.--Climatological data* for Alaska Agricultural Experiment Station, near Matanuska, Alaska

Jan. Feb. Mar. Apr May June July Aug. Sept. Oct. Nov. Dec. Annual Seasonal 2/

Precipitation in inches: Mean, 1920-48 O Q"7 O T*, 0.56 O A.O O CQ 1 1 1. 1 QK 2 QC 2 CC I ric 0 07 1 c. ^4.

*7^ »»ri -i -i Maximum, 1939-48 1.04 QQ 1 71 . xw 3 6 4.81 3 AQ 2 1 74. Ol 1 ^

Minimum, 1939-48 OK O7 .14 f)O 1 7 .16 AK .51 .39 .10 .05 11.07 Snowfall (unmelted snow}, in inches: Mean, 1936-46 rj c 5 -z 6 ' Q 2 Q 0 0 0 0 5 n 8 "t 8 r\ 44.0 4.^ ^ -1 Maximum, 1936-46 ?1 4 in t. 1 t. O 17 ^ 5 3/Tr~/ *-x. i Q r\ IS 7 O7 c. on t. 77.4

ZC A Minimum, 1936-46 0 0 0 0 0 0 1.5 OQ c. Temperature, in degrees Fahrenheit: Mean, 1939-48 10.6 Of, "*. ?4. ft 17 o 4.7 n C.K f. c.7 O 54.6 frc c TO Q 1C C Maximum, 1939-48 PS 7 in -> "*."*. t> 44.4 t^l 4. Kn c. CQ A t>o n *\o n t.Q R o a ngf ff f.Q n ^ Minimum, 1939-48 -4.0 1 f. O 4.^ O CO C CC f~\ CO C 4.4 ft 1 ^ Q -1 O f.O Q 15.8 33.0 M»

I/ Data from U. S. Department of Agriculture (1941) and U. S. Weather Bureau (1936-48). Data for 1949-51 are incomplete, 2/ Seasonal snowfall is that recorded between July 1 of one year and June 30 of the following year. 3/ Tr., less than 0.1 inch. years the winter and spring are relatively (pi. 1), the local relief J,s generally not dry. On the average about two-thirds of the more than 100 to 200 feet. Features provid­ annual precipitation occurs during the 5- ing greater local relief include Bodenburg month period June-October. Butfre, which is almost 800 feet higher than the surrounding lowland, and several similar There is also a wide range in departure hills of rock. The bluffs along the Mata­ from mean temperature. This is best illus­ nuska River north of Palmer rise 200 to 300 trated by the length of the growing season. feet above the river channel. The last spring frost commonly occurs in late May. the earliest autumn frost in late August Most of the valley floor, extending or September. During the 10-year period westward from the Matanuska Mver north of 1939-48, however, the length of the growing Palmer, is a gently rolling surface. In season ranged from 67 to 151 days. much of it, the hills and valleys have a southwest trend; this orientation is shown Midsummer temperatures in the agricul­ mo.st conspicuously by the series of lakes tural area range from 45 to 70 F; tempera­ whose axis passes south of Wasilla and less tures as high as 80° P are unusual. The strikingly by a second series of lakes to winters are moderately cold; periods during the northeast. In the northwestern part of which the temperature reaches -20 to -30 P the agricultural area, west and northwest of are usually short. The freezeup in autumn Pittman, the hills and valleys trend south- comes in October or November. Seasonal frost southwest. Two tracts, one between Eska and commonly reaches depths of 6 feet or more. Moose Creeks and extending 2 to 3 miles west The ground begins to thaw in April or May, of Moose Creek, and the other between Palmer but seasonal frost may persist beneath the and the Agricultural Experiment Station, are surface in protected spots as late as July. characterized by irregular hills and swales and conspicuous ridges. The local relief The distribution of rainfall and the in these tracts is as much as 150 feet. The danger of late spring and early autumn frost ridges are most conspicuous near the experi­ are responsible for a measure of uncertainty ment station, where they are continuous and of crop yields in the Matanuska Valley agri­ parallel. cultural area. A conspicuous belt of hills which rise The dominant wind of the agricultural 50 to 150 feet above the surrounding country area, known locally as the "Matanuska wind, 11 extends southwestward past Pittman. A chain is from the northeast. It is an autumn and of similar hills borders Big Lake on the winter wind. During storms it may blow more south and extends southwestward out of the or less continuously for periods of several area described in this report, beyond which days; Weather Bureau records indicate that it curves to the south and then to the south­ gusts reaching velocities of 50 miles per east, ending at Goose Bay. hour or more occur during the more severe storms. The "Knik wind," oceanic air from Palmer is on a wide, flat-topped bench. the south moving down the Knik Valley, is Similar benches lie east of the Matanuska relatively warm. During late winter and River south of Wolverine Creek, between the spring it brings mild weather and, together Knik and Matanuska Rivers, and along part of with rain, may remove much of the snow cover the top of the bluff overlooking Knik Arm. from the agricultural area before the ground Smaller benches north of Palmer and through­ begins to thaw. out the rolling country to the west are less conspicuous. The rolling country and benches north of Knik Arm are separated from it, and Topography and Drainage from the low-lying flat ground near it, by a conspicuous bluff 50 feet or more high which The Matanuska Valley agricultural area extends from Goose Bay eastward and grades lies in a wide, flat-floored valley formed by into the bluff of the Matanuska River near the merging of the Matanuska and Knik Valleys Matanuska. at the eastern end of Knik Arm. Figure 2 shows the principal physiographic units. The Most of the area drains into the Mata­ valley is bounded by rugged mountains which nuska and Knik Rivers, but several small rise abruptly above its floor. In the Chu- streams flow directly into Knik Arm. The gach Range, at the southern edge of the val­ Little Susitna River drains part of the ley, Pioneer Peak rises to an altitude northern section of the area. The drainage greater than 6,300 feet; several other peaks in many interstream tracts is poor because surpass 4,000 fset, and altitudes of 3,000 of the irregular topography and the vegeta­ feet are common. Along the northern edge of tive cover. There are large areas of swampy the valley, peaks in the Talkeetna Mountains ground, and shallow lakes occupy many of reach altitudes of 3,000 to 5,000 feet. the hollows. The oriented lakes west of Pitt­ man and the two southwestward-trending series Although the altitude of the valley of lakes near Wasilla are among the prominent floor ranges from tide level on Knik Arm to features of the valley floor. 1,000 feet at the base of "Wishbone Hill The Knik River floods annually in July The population of Palmer is estimated at or August when Lake George, impounded by about 800. Wasilla is much smaller, and only Knik Glacier, is drained as a result of its a few families remain in Matanuska and Knik. overflow and the resulting erosion of the The farm population of the area is 2,000 to ice along one edge of the glacier. 3,000; it is distributed chiefly around Pal­ mer and, to a lessar extent, around Wasilla. Vegetation The agricultural area is traversed by, the main line of the Alaska Railroad, which passes through Matanuska and Wasilla north­ In its natural state most of the area ward to Fairbanks. A branch of the railroad discussed in this report was forested. White extends from Matanuska through Palmer to spruce, aspen, cottonwood, and birch are Jonesville, on Eska Creek. The Glenn Highway characteristic of the better drained soils. begins at Anchorage, 48 miles southwest Of Willow is found on all types of deposits. Palmer, and extends through Palmer and into Black spruce is common only in bogs. Alder the interior of Alaska. Daily bus service is common both -in moist spots on the lowland is maintained between Wasilla, Palmer, and and with willows on the mountain slopes bor­ Anchorage. All the settled sections of the dering the valley. The altitude of tree line valley lie on a road net maintained by the depends upon exposure; locally it is above Alaska Road Commission. Air travel has long 2,000 feet. Trees in this area are shallow been popular in this area, as elsewhere in rooted and are easily blown down. Windfalls Alaska, Several small local fields have been are common in forests composed of older used, and a new airport was completed at trees. Palmer in 1950.

Fire, probably in part natural but large­ Development of agriculture in the area ly accompanying settlement and railroad con­ has continued since establishment of the struction, has burned over many parts of the agricultural colony; dairying and vegetable valley floor. Extensive burned areas have growing are the most important types of farm­ become reforested. ing. The history of the agricultural colony is the subject of a recent study by Stone The ground cover in the forest consists (1950). of various shrubs, herbs, grasses, and other small plants. Mosses and grasses are charac­ teristic of poorly drained areas. Pireweed is the commonest plant on newly burned land. GEOLOGY Pireweed and grass may persist for long peri­ ods on burned land near the tree line. Mesozoic and Tertiary Rocks

The flats along Knik Arm are, or recent­ The character of the rocks underlying ly have been, subject to tidal flooding; the greater part of the Matanuska Valley over most of their area they bear only small agricultural area is unknown. The writer salt-tolerant plants. The wide alluvial estimates that bedrock is exposed at the flood plains of the Matanuska and Knik Rivers surface in less than 1 percent of the area; are practically bare of vegetation because at elsewhere the bedrock is covered by unconsol- some time during every season or two the idated deposits whose thickness is known at gravel bars either are submerged or are re­ relatively few places. Exposures of bedrock moved and rebuilt during the channel shifting in the area are indicated on plate 1. that accompanies flooding. The middle slopes of the mountains flanking the valley bear a The bedrocks exposed in and adjacent to cover of moss and low or prostrate shrubs; the agricultural area have been described by near the summits there is no vegetative Martin and Katz (1912), Capps (1940), and cpver. Landes (1927). The Talkeetna Mountains, to the northAcomposed mostly of igneous rocks, predominantly granitic intrusives (Mesozoic?) Culture and to a lesser extent lava and tuff. A belt of Cretaceous and Tertiary sedimentary rocks As a result of the finding of gold in forms the south flank of the mountains. the Talkeetna Mountains, the settlement at Mesozoic rocks in the Chugach Range, to the Knik was established in 1898 on the site of south, include granitic intrusives, meta­ an Indian Village and Russian mission. In morphosed sedimentary rocks fchiefly slate 1916 the Alaska Railroad was extended through and argillite), and greenstone. the Matanuska Valley, and where it crossed the trail between Knik and the Talkeetna Cretaceous sedimentary rocks extend down Mountains the community of Wasilla was estab­ the Matanuska Valley to Moose Creekj they are lished. Matanuska grew at the junction of sandstone (including graywacke) and shale. the main line and the spur line leading to Conglomerate and sandstone (graywacke) ex­ the Matanuska Valley coalfields. After the posed in small hills south of Palmer may be establishment of the agricultural colony in 1935 the center of population of the valley shifted toward the community of Palmer. the southwestward extension of these rocks. obtained by the mechanical analysis of sam­ Conglomerate, sandstone, shale, and coal of ples of the unconsolidated sediments are Tertiary age are exposed in the Eska Creek- given in table 4. "Wishbone Hill-Moose Creek area. Tertiary coal-bearing rocks also occur at Houston, just beyond the northwestern corner of the Glacial Deposits area described in this report. Till ("hardpan") Wishbone Hill, at the northeastern cor- , ner of the agricultural area, is'a Synclinal ^ Till is a clastic unconsolidated rock hill held up by the Tertiary Eska conglomer­ deposited directly by or from glacial ice ate. Martin and Katz (1912, p. 72-75, with little or no modification by running pis. 15, 16) describe the straight front of water. It is dominantly unsorted, consisting the Talkeetna Mountains as representing a of rock fragments ranging from clay to large zone of faulting; they believe that the course boulders. The distinction between a till of the Little Susitna River is approximately slightly sorted during deposition and a poor­ along the fault downstream from the point ly sorted gravel is in many instances arbi­ where the stream emerges from the mountains. trary. Till is one end member of a continuous Recent work by F. F. Barnes, of the U. S. series of materials; the other end member is Geological Survey, shows the presence of coal- well-sorted outwash sand and gravel (Flint, bearing Tertiary rocks north of the Little 1947, p. 103). Susitna River; these Tertiary rocks, with other evidence, suggest that the mountain Till at the surface in the agricultural front rather than the stream coarse marks area is in the form of ground moraine, a the western extension of the fault (Barnes, glacial deposit which usually occurs as a ''~4 F. JC.. personal communication, 1952). Martin relatively thin mantle on the underlying and Katz (1912, p. 74) suggest also that the material. Ground moraine may consist of relatively straight front of the Chugach material deposited beneath glacial ice, or of p^i Range, to the south, ^30 may be due to fault­ debris derived, upon melting, from the gla­ ing, but they do not find enough evidence to cial load within or upon the ice. Though the form a definite conclusion. Exposures along ground moraine is predominantly till, layers the Matanuska River and Moose and "Wolverine of washed material are present in it, and Creeks show that the folded sedimentary rocks thin sand and gravel deposits locally mantle strike northeastward and are faulted. The its surface. Over much of the area mapped available data are insufficient to justify by the writer as ground moraine (pi. 1) the conclusions regarding the structure of the till presents the original surface of glacial sedimentary rocks underlying most of the deposition, modified little or not at all by valley floor to the west or their depth of erosion. burial beneath the overlying unconsolidated deposits. Till in this area is commonly gray or blue gray. It is composed mainly of subangu- lar to rounded stones in a matrix of mixed Quaternary Deposits sand and silt. Table 4 shows the grain-size distribution of the fragments smaller than 2 Unconsolidated deposits of both glacial millimeters in diameter, in two samples of and nonglacial origin cover the bedrock of till. Very little clay is present in these the valley floor in most of the area describ­ samples. Poor sorting is shown by the dis­ ed in this report. The glacial deposits con- tribution of material over a wide range of 0 /*-»( sist of tillj «¥ outwash sand and gravel, grain sizes (as compared, for example, with and fluvioestuarine deposits in and along the lesser range of sizes in windblown sand). Knik Arm. The existing Matanuska and Knik Stones in the till range from granules to Glaciers lie beyond the limits of this area; boulders; they consist of the greenstone, but, because the Matanuska and Knik Rivers de­ slate, schist, and felsic intrusive rocks rive much of their water and sediment from characteristic of the adjacent mountains and these glaciers, the modern deposits of these of the sedimentary rocks exposed in the Mata­ streams are considered glaciofluvial. The nuska Valley. nonglacial deposits include windblown mate­ rial, which mantles most of the agricultural Silt-rich till is compact and tough. It area, and alluvial fans, talus, and frost- is difficult to excavate and is known local­ disturbed deposits chiefly along the walls ly as "hardpan." of the valley. (See fig. 5.) Irregular roughly horizontal streaks and It has not been shown whether the exist­ layers containing more stones or more silt ing glaciers are remnants of the more exten­ than the underlying and overlying till can be sive Pleistocene Matanuska and Knik glaciers, observed in some exposures. They are common­ from which most of the glacial deposits in ly a few feet thick and may extend laterally this area originated. Separation of the a hundred feet or more in well-exposed sec­ Pleistocene and Recent in epochs in this tions. Fractures which may be faults cut the area is therefore difficult. In this report till in some exposures. In the bluff above all unconsolidated materials overlying Ter­ the river 1 mile north of Palmer the walls of tiary or older rocks are designated simply several such fractures are separated by a few Quaternary deposits. inches of silt and sand laminated parallel to the walls. These fractures are most reason­ The geologic map (pi. 1) shows the dis­ ably explained as tension fractures filled tribution of bedrock exposures and unconsoli­ with sorted sediment by water. The fractures dated deposits exclusive of swamp deposits and the irregular layers consisting of sandy and the mantle of windblown material. Data till may be recognized in dry weather because they remain more damp than adjacent silty 40 feet above the level of Tflasilla Lake. The till. Because of the limited exposures of ground moraine north of Lake Lucile and north­ till, it is not possible to draw conclusions west of Wasilla is marked by entrenched regarding the occurrence of fractures and stream-cut channels which stand above the irregular sandy layers in the agricultural level of the lake. The high-level bedded area as a whole. gravel and stream channels are moat reason­ ably explained by the assumption that the In some exposures layers of sand or basins of Lake Lucile and lasilla Lake were gravel occur within massive till. The layers occupied by blocks of stagnant ice. Streams are commonly a few inches to a few feet thick. flowing along the boundary between the ice Most of those the writer has seen are composed and the adjacent ground moraine deposited of medium to coarse sand or sandy pebble gravel; where they locally flowed across the gravel; they appear to form relatively narrow moraine, they cut channels. Melting of the stringers enclosed in compact till. Some of last ice left the lake basins in much their these show sharp changes in grain size and present form, with bedded gravel and channels thickness within short distances. perched upon the surrounding hillsides. The pattern of gravel islands and spits in the Similar thin sandy and gravelly streaks lakes east and northeast of'"Was ilia is best in till have also been found in many wells ir explained by this ice-block hypothesis; an the area. These layers seem to be similar to alternative explanation of the lake basins lenses and stringers of imperfectly sorted and accompanying features as having been material found in till in the United States formed by erosion alone is unsatisfactory. (Meinzer, 1923, p. 285). The writer believes All the other basins in the two southwest- that these deposits were laid down by small trending series of lakes near TSasilla (see subglacial streams which flowed temporarily pi. l) are probably ice-block holes. upon till beneath the ice before being cover­ ed by additional till from the overlying ice. The ground moraine in the tract north The sorting of sand in these layers, which is and west of Pittman is characterized by north- better than that in most other sediments northeastward-trending parallel, elongate analyzed, is shown by the analyses of two hills and valleys; most of the valleys are samples in table 4. The better sorting may now occupied by lakes or marshes (pi. 1). be due to deposition from a confined stream Horizontally bedded gravel deposits mantle the flowing under hydrostatic pressure. Evidence hillsides above some of the lakes. Discon­ available from outcrops and from wells tinuous eskerlike ridges of till, as much as suggests that these layers are of limited and 20 feet high and several hundred feet long, irregular areal extent. li'e upon the ground moraine in this tract. Many of these ridges are covered by several In a few exposures slightly sorted feet of gravel; locally, as at the gravel pit material which resembles till may be observed 1^ miles west of Pittman, gravel has been resting upon massive till. The till-like deposited between the till ridges and has material shows neither the bedding nor the partly or completely covered them. Till sorting of outwash gravel. The best exposure ridges at the lake north of the railroad and seen by the writer is in a gravel pit on the 3 miles southeast of Houston extend into the lower slope of Lazy Mountain, about half a lake. The ridges are interpreted as fillings mile northeast of the Matanuska River bridge. formed in short crevasses in stagnant ice. There the slightly sorted material rests upon Their presence over most of the tract under the underlying massive till along an irregu­ discussion shows that the form of the morain- lar but distinct surface. This slightly al surface is due to glacial deposition and sorted material is best explained as super- that the surface has been modified only glacial till: it is probably composed of slightly by erosion. The extension of ridges, debris which lay upon the surface of the ice into one lake and the presence of high-level and became slightly sorted before being let bedded gravel beside others are taken as down upon the massive till beneath as the ice evidence that the elongate valleys are ice- melted. Exposures of till in the agricultur­ block holes. The orientation of the elongate al area are not sufficiently numerous or ex­ parallel hills of ground moraine is attributed tensive to show the areal importance of super- to the presence in the ice of elongate alter­ glacial till. nating zones of debris-laden and relatively clean ice. Debris from dirty ice formed hills; The till of the Matanuska Valley agricul­ the last remaining clear ice between bands of tural area is relatively impermeable. Only dirty ice separated into blocks around which the layers of sorted material yield water gravel deposits were laid down. Final melt­ freely, and these in small quantities. Poor ing of the last blocks left the existing lake surface drainage is characteristic of tracts basins. underlain by tillj marshes are common, even on high ground. Some lakes, including Wasilla Over the remainder of the ground moraine Lake, appear to be perched on till. Contact in the western part of the agricultural area springs are present along hillsides in local­ the orientation of hills and valleys is much ities where saturated gravel lies on till. lessv regular and conspicuous than that near Pittman. The valleys, most of which are Many of the depressions on the ground Interpreted as ice-block holes, show a south- moraine in the western part of the agricultur­ westward trend like that of the large glacial al area, or on ground moraine covered by thin drainage channels west of Wasilla. The ice- gravel, are bordered by high-level deposits block holes have the form of dimples in the of horizontally bedded gravel. Such beds, ground moraine; they are most reasonably ex­ three-eighths qf a mile west of "ffasilla, are plained as cavities formed by the melting of relatively clean ice, whereas the adjacent The till deposits described in preceding higher moraine Is the result of deposition paragraphs are considered to form a single from ice having a heavier load of debris. sedimentary unit deposited during one glacia- tion. Several natural exposures and about 35 The writer concludes that the form of wells show older glacial deposits lying be­ the ground moraine in the western part of the neath this till. The older deposits are main­ agricultural area Is due to widespread stag­ ly glaclofluvlal but in 1 well such deposits nation of Ice, and that the distribution of rest upon what may be still older till. debris in the Ice may have controlled' the po­ sition and form of hills and valleys on the In the bluff along Knik Arm east of moraine. The elongate, parallel zones of Goose Bay the surface till rests upon gravel. dirty Ice inferred to have formed the paral­ (See section 1.) In several exposures along lel hills northwest of Pittman may have been the Matanuska River north of Palmer the till folded medial moraines such as may be seen lies beneath surficial gravel but rests upon on the Malaspina (Washburn, 1935) and Bering older gravel. (See section 2.) Most of the Glaciers along the Gulf of Alaska. If similar wells that pass through the near-surface till zones of debris were present In Ice over the into older deposits are in an area lying remainder of the western part of the agricul­ within a few miles west, northwest, or north tural area, they left no conspicuous topograph­ of Palmer, but others are near the Agricul­ ic expression; any that were present may tural Experiment Station and about 3 miles have been broken by renewed movement of part east of Wasilla. Table 2 presents logs of of the terminal zone of the nearly stagnant three such wells. The first of these Is glacier. about three-eighths of a mile west of the Matanuska bluff and 2-J- miles north of Palmer; Till Is present also in a hilly belt the "blue mud and gravel" 63 to 101 feet be­ which extends southwestward through Pittman neath the surface (table 2) is correlated toward Big Lake and In hills south of Big with the buried till In the Matanuska bluff Lake. Near Pittman the till Is overlain by (section 2). The older gravel shown by the pitted gravel deposits. This hilly belt may drillers' log.may represent the advance out- represent a medial moraine in the Ice, or, if wash of the gTacier which deposited the over­ part of the terminal ice moved past stagnant lying till, the outwash associated with the ice lying northwest of Pittman, the hills may retreat of an earlier glacier, or both. mark the juncture of the two Ice masses. Several of the logs, including the second in Hills of till south of Big Lake are continu­ table 2, record red or brown gravel in which ous to the west with the band of hills men­ the color may represent the effects of weath­ tioned on page 6, which is arcuate toward ering; in addition, the gravel locally is the west and which extends from the vicinity slightly consolidated. The gravel may there­ of Willow and Nancy, northwest of Houston, to fore have been exposed for some time after Knik Arm near Goose Bay. On the basis of its deposition before it was covered by the study of aerial photographs, T. N. V. Karl- overlying till. Thus it is reasonably con­ strom (personal communication, 1949), of the sidered to represent an earlier glaciation. Geological Survey, Interpreted this arcuate The blue mud from 198 feet to bedrock at belt of hills as the end moraine of the last about 226 feet in the first log in table 2 glacier which lay over the agricultural area. may be till deposited during this"earlier The writer believes this interpretation to be glaciation. The gravel at Goose Bay, as correct. shown in the section, may represent the re­ cessional outwash of the earlier glaciation Over most of the agricultural area east and the advance outwash of the glacier which of Wasilla and west and north of the Matanus­ deposited the overlying till; if this inter­ ka River the till is covered by gravel. The pretation is correct, the vegetation that thickness of the gravel ranges -from a feather- formed the peat grew during the period between edge to about 100 feet; in general the depos­ the retreat of one glacier and the advance of its are thicker toward the east. the next.

1. Section in bluff north of Knik Arm about one-quarter mile east of Goose Bay Feet (estimated) Windblown sand - 3 Till, gray-brown, sllty; includes a few layers of poorly sorted gravel- 15 Pebble and cobble gravel, sandy; interbedded horizontal sand lenses as much as 1 foot thick and 40 feet long show crossbeddlng 20 Peat, brown-black, slabby, containing compressed twigs and stems of wood- 2* Silty, gray-- >- - Sand, silty, somewhat Iron-stained. « - 3 Pebble and cobble gravel, sandy, conspicuously iron-stained; locally consolidated Covered - -Beach deposits- Total-

10 2. Section in bluff west of Matanuska River about 3 miles northwest of Matanuska River Bridge Feet (estimated) Windblown sand - 20 Pebble and cobble gravel, sandy; horizontally bedded 40 Till, gray, silty 40 Pebble and cobble gravel, sandy and silty;, slightly consolidated; locally deformed beneath overlying till - 60 Covered - 70 River ____ Total 230

Table 2. Selected drillers' logs showing the presence of older glacial deposits in the Matanuska Valley agricultural area [Logs through courtesy of Alaska Rural Rehabilitation Corp.]

Feet Interpretation AHRC tract 152, well 9 (abandoned; near USGS 505) 0 - 15 15 - 65 63 - 101 /?«z i ni 101 - 116 116 - 155 Slightly consolidated frT»Q TTO T /»oTnon^ orl _-___-.. _. ______155 - 178 1 RR T7fi 178 - 181 181 - 187 181 - 187 187 - 198 198 - 215 1 QQ OOC<9 Blue mud and shale rock, pipe stopped at 226 feet on 2-foot ledge of hard 215 - 226 226 - 510 Well abandoned - dry hole.

ARRC tract 14 (USGS 578) Tops oil . 0 - 5 Sand and gravel 5-75 Sand, coarse 75 - 90 Gravel, mud, and sand; water blue and mucky; will bail out 90 - 110 Till- 90 - 174 Glacier mud and gravel, blue and thick; water seeping 110 - 128 Glacier mud and sand- 128 - 159 Blue mud' ______... . 159 - 174 "Water, gravel; water stands to 75 feet; red gravel 174 - 177 Oxidized gravel- 174 - 177 Water, gravel -. 177 - 180

ARRC tract 95 fUSGS 564) 0 - 5 5-28 Till 28 - 86 28-40 CxT*fl n 1 t" o wfl ciVi______- ______40-45 45 - 55 55 - 59 59 - 64 64 - 65 65 - 73 73 - 77 ______(jo_ _ ..______73 _ 77 77 - 86 Water gravel; 60 feet of water in pipe 86 - 88 Older gravel 86 - 88

11 The thickness of the near-surface till, associated with channel-and-fill structure. in wells that pass through it, commonly ranges Faults are present, in some exposures, partic­ from 10 to about 60 feet. The thickest sec­ ularly in thin beds of sand, and are attrib­ tion of till known to the writer is in well uted to slumping of the deposits. 258; till extends from 10 feet beneath the surface to the bottom of the well, which was Thin layers of silt are included in the at a depth of about 150 feet in November 1951. sand_and_gravel. Other deposits, such as '1Till7 peat7wfine-Sraine(i sediments deposited A bench on the slope of Lazy Mountain is in ponds, are present locally. interpreted as a lateral moraine deposited along the contact between glacier and valley The outwash materials in the agricultural walls. Gullies cutting the bench expose* till area are relatively permeable. Where the resting upon bedrock. This bench lies between land surface is underlain by sand and gravel, 1,700 and 2,100 feet above sea level. Two it is generally well drained except where miles north of Palmer the upper surface of till lies beneath the sand and gravel at shal- the till deposited beneath the same glacier _tin and Katz (1912, p. 70-71) is at an altitude, of about 300 to 400 feet. Ascribe gravel-covered benches to an alti­ The thickness of the ice over this part of tude of about 2,200 feet, where the valley of the valley may therefore have been about 1,600 Moose Creek emerges from the high mountains. feet. A discontinuous bench which appears They suggest that these deposits were formed essentially similar to that on Lazy Mountain in ponded water within the tributary valley , extends along the slope of the Talkeetna at a time when the main Matanuska Glacier / Mountains westward from Moose Creek and be­ stood across the mouth of the tributary _,--'' yond the Little Susitna River. Over a dis­ valley. tance of 6 miles this bench slopes westward from about 2,500 feet above sea level to The geologic map (pi. 1) shows that the about 2,000 feet, or about 80 feet per mile. ground moraine west and northwest of TWasilla is crossed by many gravel-floored valleys. The age of till deposits in the Matanuska These are drainage channels followed by gla­ Valley agricultural area can be discussed cial meltwater. North of Knik Arm, and near most conveniently after the other glacial the experiment station and near Palmer, far­ deposits have been described. ther easty are similar deposits beneath ter­ races. Throughout the agricultural area are other smaller tracts underlain by gravel de­ Outwash Sand and Gravel posits of meltwater streams. All these de­ posits are mapped as alluvial-plain deposits. Outwash deposits, formed by streams that Many of them are considered to have been derive part or all of their water and sedi­ formed in close association with glacial ice ment from glacial ice, cover a large part of and were somewhat modified by its melting; the Matanuska Valley agricultural area. They they are ice-contact deposits. Others, like include not only deposits formed in associa­ the flood-plain deposits of the modern Mata­ tion with the ice that once lay over this area nuska and Knik Rivers and deposits in terraces but also deposits such as those of the exist­ near these streams, are similar although ing Matanuska and Knik Rivers, formed at some formed farther from the ice or not modified distance from the ice. by its melting; these are proglacial deposits. The deposits of the modern flood plain of the The mechanical composition of samples of Little Susitna River are considered progla­ outwash sand and silt and the finer fraction cial deposits although the stream is only of gravel are shown by the data given in table partly of glacial origin. Little information 4. The absence of large proportions of silt is available regarding the thickness and com­ and clay is well shown by these data. These position of these deposits in the western samples are probably representative of most part of the agricultural area, but the form of the outwash deposits of the agricultural of the topography, the presence of small area. Some stream-laid deposits are so poor­ hills of till which protrude through the ly sorted, however, that they resemble till. gravel, and a few exposures suggest that in The other extreme of sorting is represented general the gravel is relatively thin (per­ by openwork gravel composed of pebbles or haps 20 to 30 feet thick or less). West of cobbles of approximately equal size and with­ Wasilla and the experiment station most of out interstitial finer materials. Such a the alluvial-plain deposits are in valleys gravel consisting entirely of pebbles is known cut in the ground moraine, and till probably locally as "pea gravel." lies beneath all these gravel deposits. The stones in the gravel represent all Southwest of the experiment station the rock types found in the Matanuska Valley alluvial-plain deposits are about 30 to 70 and in the surrounding mountains, although re­ feet thick; they are terraced. At Palmer and latively soft and weak sedimentary and meta- beneath the lower terraces north of the city morphic rocks naturally are less common than (pi. l) these deposits are commonly 30 feet massive igneous rocks. The sand consists thick or less and rest on till or bedrock. predominantly of grains of quartz and dark Meager evidence suggests that the deposits minerals and fragments of schist and green­ of the terraces on the lower slope of Lazy stone* Mountain, near Wolverine Creek, are also re­ latively thin. Deposits beneath the exten­ Bedding is well or moderately well de­ sive terrace south of Palmer are thicker; veloped in many exposures of sand and gravel; many wells 75 to 100 feet or more deep do not but, where exposures are extensive, the beds reach till or bedrock, and 2 wells (table 5, are generally seen to pinch out laterally. wells 160 and 161) 175 and 200 feet deep may Crossbedding is commonly observed, in places be in gravel throughout their entire depth. 12 Nothing is known of the maximum thickness of 'tween Wasilla and Kings Lake, may be super- gravel in the area between the Matanuska and glacial sorted debris let down upon ground Knik Rivers; several wells about 50 feet deep moraine by the melting of the last ice. Out- and 2 about 110 feet deep, all near Bodenburg wash deposits in the tract west and north of Butte, do not pass out of the gravel. The Pittman probably are chiefly alluvial-plain thin gravel deposits, such as those of the deposits and fillings around the edges of ice- terraces immediately north of Palmer, seem block holes. Reconnaissance field work was to represent deposition during terrace cutting, done in that part of the agricultural area, not extensive deposition after such cutting. however, and because of lack of exposures and poor topographic expression no attempt to The' maximum thickness of alluvial de­ differentiate the deposits is made at present. posits beneath the existing Matanuska and Knik River flood plains is not known. Stream The tract characterized by crevasse measurements by the Water Resources Division fillings, extending northeastward from the ex­ of the Geological Survey show that the depth periment station, grades northward and south­ of channel scour at the Matanuska River bridge ward, into areas of irregular ridges and de­ during high water is at least 23 feet. Three pressions. Cuts along the Palmer-Four Corners telephone poles were driven 19 to 21 feet in­ road expose horizontally bed,ded gravel on to flood-plain gravel beside the approach to hilltops between depressions. The depressions the Knik River bridge without encountering are interpreted as ice-block holes. These bedrock. -*wsts are prominent in gravel deposits north and west of Palmer and extending south and The alluvial-plain deposits are composed west as far as Matanuska and Wasilla. of silt, sand, and pebble and cobble gravel. Where it is exposed, the material is hori­ Pitted topography is conspicuous also zontally bedded and fairly well sorted. Iso­ between Eska and Moose Creeks; there relief lated boulders are present in many locali­ as much as 150 feet is common. The large de­ ties, however, particularly in reworked de­ pression that forms the lake basins south of posits or in those laid down near the ice. Wishbone Hill consists of ice-block holes. Gravel in a tract north and west of the The ice-block origin of the lake basins Matanuska River and extending southwestward near Wasilla and northwest of Pittman has from eskerine deposits west of Moose Creek already been discussed (p. 9-10). was deposited by aggrading streams from ice which stood near Moose Creek. The extension Because of the number, size, and distri­ of this deposit southwestward was partly laid bution of ice-block holes over most of the down by the same streams, but there the form Matanuska Valley agricultural area west of of the deposit was modified by the melting of the Matanuska River, th& writer concludes buried blocks of ice. The whole deposit may that stagnation of ice and the resultant have been a valley train. pitting of the ground moraine and outwash de­ posits were the dominant factors in the forma­ An elongate southwest-trending tract tion of the topography. between Palmer, Four Corners, and the experi­ ment station is characterized by sinuous Throughout much of the agricultural area ridges. Near the experiment station the glacial deposits have been modified by stream ridges have a relief as much as 150 feet and erosion. Small terraces are present in many are conspicuously parallel; several of them, localities; in some there are several terraces approximately level with each other, grade standing one above another. Terraces are most into a flat surface at the experiment station. conspicuously developed along the Matanuska Clean sand and clean cobble gravel are exposed River, between the Matanuska and Knik Rivers, in a gravel pit in one of the ridges. Rookie and above the bluff overlooking Knik Arm. (1942, p. 365) has interpreted these ridges I Several of these terraces north of Knik Arm as crevasse fillings, deposits formed in cre­ Ol and also thAse along the Matanuska River near vasses between narrow ridges of ice which ' and north or Palmer are pitted; the terraces later melted, leaving the interridge valleys. were formed while blocks of stagnant ice The writer believes this interpretation to still lay buried in the gravel. Pitting of be correct. terraces is most conspicuous in the narrow tract west and north of Palmer (pi. l); these Sinuous and irregular ridges that cover terraces give way to pits toward the southwest, much of the tract immediately east and west All the higher terraces along the Matanuska of Moose Creek and extending about 3 miles River near Palmer, down to and including that west of that stream are probably eskers. on which Palmer is situated, are attributed Similar ridges near Pittman, where they are to the action of meltwater from ice which lay associated with hills of till and with gravel up the Matanuska Valley. Some of them may be deposits modified by the melting of buried equivalent to terraces near Eska Creek, and ice, also are considered eskers. Elsewhere possibly to those east of Kings River or in the agricultural area, and particularly about 19 miles upstream from Palmer. Terraces about ij- miles northwest of Tfasilla, there north of Knik Arm may be remnants of alluvial are isolated sinuous ridges which contain plains of both the Matanuska and Knik Rivers. gravel and which also probably are eskers. Several terraces lower than the one on Some deposits of outwash sand and gravel which Palmer is situated are preserved on were not differentiated during mapping. Most both sides of the Matanuska River near the of these deposits undoubtedly were stream highway bridge and between the bridge and the laid. Some of them cap small isolated hills Knik River. Some of these were formed by the of older deposits. Some, as the deposits be- Knik River. They are not pitted. 13 At many places along the Matanuska River Gary (?) or Tazewell(?) substage, older than low tree-covered terraces stand a few feet 15,000 years (Flint and Dewey, 1951, p. 286). above the highest deposits of the present alluvial plain. Alluvial fans of many streams The writer concludes tentatively that flowing into the Matanuska River, such as the glacial episode during which the surfi­ those of Moose and Eska Creeks, have been cial and near-surface till in the Matanuska eroded by the river and trenched by their Valley agricultural area was deposited corre­ own streams. These low terraces and fans, sponds approximately in time to the Mankato, which appear to be at least as old as the cot- although it may have begun sooner and lasted tonwoods growing upon them (about 100 years?), longer than the Mankato in the Midwest; and suggest that the Matanuska Mver has recently that the peat at Goose Bay and the Eagle been eroding Its alluvial plain. Mver dates from the preceding warmer inter­ val. The postulated older till cannot be Most of the preceding discussion has dated on the basis of available evidence, but been concerned with deposits of sand and grav­ it was probably deposited during the glacial el at or just beneath the land surface, and episode preceding the last. the general thickness of these deposits has been described. In the discussion of the till of the Matanuska Valley agricultural Fluvioestuarine Deposits area older gravel deposits which underlie the younger till were mentioned. The older gravel Along and in Knik Arm, glacial silt seems to differ from the younger gravel chief­ brought into brackish water oy the Knik and ly in having slightly cemented layers or Matanuska Wvers is being deposited in the streaks. In exposures in the Matanuska bluff form of beaches and bars. North of Knik Arm (section 2) the older gravel seems to be and west of Matanuska is a flat surface which slightly consolidated because of binding by stands 20 to 30 feet above mean sea level. silt; deposits of silty sand and gravel at This flat is underlain by tough, relatively the base of the bluff, formed by rainwash, impermeable gray silt. Streams crossing it are as firmly consolidated as gravel in the are tidal; to the west they are nearly bank- bluff. Slmil'ar slight consolidation was ob­ full at the average high tide (the tide range served in gravel in the bluff at Goose Bay. at Anchorage, farther west on Knik Arm, is- In addition, however, gravel at Goose Bay about 30 feet), and probably the flat is part­ locally has been cemented by iron oxide (sec­ ly covered at very high tides. Farther east, tion 1); some of the gravel is sufficiently near Reedy Lake, there are fresh-water bogs cemented that masses of it which have fallen underlain by gray silt which resembles the from the bluff remain as boulders on the beach estuarine silt seen farther west. Stream- below. Slightly consolidated gravel, in lay­ laid sand forms the surface of the flat at ers that alternate with loose gravel, is also Matanuska and near the point at which Wasilia suggested by some well logs which record old­ Creek enters the flat. The boundary between er glacial deposits (table 2). The thickness stream and fluvioestuarine deposits is thus of the older gravel is known at only one lo­ established only within broad limits (pi. 1). cality, ARRC tract 132 ('table 2); there the gravel is 97 feet thick and rests on what may On a similar flat at the Eklutna CAA be older till. station, south of Knik Arm, the ailt ranges from 4 to more than 11 feet thick; a well Peat in gravel beneath till at Goose about 20 feet deep passes through the silt Bay has been described briefly (section 1). and obtains water from underlying gravel. A sample of wood from this peat, together The deposits beneath the flat north of Knik with samples from other peat deposits in Arm may be of similar thickness, but no infor­ Alaska, was analyzed for the Geological Survey mation is available. It is possible that the by the Lamont Geological Observatory of center of the estuary formerly was much deep­ Columbia University. The age of the wood, as er than at present, and that thick deposits determined by radio-carbon dating, is 19,100 were laid down there during the postglacial +_ 900 years (Kulp and others, 1952, p. 412- rise of sea level. 413). A peat sample collected by Ernest Dobrovolny, of the Geological Survey, beneath Bars in the modern estuary reach a level till on the south bank of the Eagle River, somewhat lower than high tide. It seems across Knik. Arm from Goose Bay, was dated at likely that this relation existed during for­ 14,300 +. 600 years (Kulp, 1952, p. 263). mation of the flat north of Knik Arm, and that Correlation of surficial till deposits at the flat was formed during a stand of the sea Goose Bay and the Eagle River is justified several feet higher, relative to the land by their proximity In the same valley. If surface, than that of the present. Later dep­ the average values for the radio-carbon dates osition during flooding by very high tides are accepted, therefore, It is necessary to probably has built up the flat somewhat and postulate that the ice-free period preceding smoothed irregularities in its surface. the advance of the ice that laid down the till overlying the datod peat deposits must have lasted at least 4,800 years in this part of Wonglacial Deposits the region. Windblown Deposits Prom previously determined radio-carbon dates Flint, and Deevey (1951, p. 263) con­ Windblown deposits mantle the surface of clude that the time maximum of the Mankato the Matanuska Valley agricultural area, with ice advance in the State of Wisconsin was the exception of alluvial plains along exist­ about 11,000 years ago. Samples of wood from ing streams, some recent terraces, flats sub­ tills of tbo Cary(?) substage are older than ject to tidal flooding, and a few steep slopes 17,000 years, and samples from tills of the in bedrock. Over most of the agricultural 14 area this mantle consists of silt or sandy uous deposits farther west are probably of silt (loess). Sand is present in several about the same age. All these deposits are local areas, generally as dunes. These wind­ postglacial. The windblown material was prob­ blown deposits have been discussed by Tuck ably derived chiefly from bare alluvial (1938), Rockie (1946), and Black (1951). plains of the Matanuska and Rnik Rivers. Near the Matanuska River modern deposition of Samples of loess and dune sand (table 4) windblown dust continues at a measurable rate show the excellent sorting characteristic of (Tuck, 1938, p. 649). It is unlikely, however, windblown deposits. that deposition is now significant in the western part of the agricultural area. Cobbles and pebbles present in the lowest foot or so of the silt in many exposures of Windblown sand and silt in the agricul­ loess are attributed to overturn of near- tural area are relatively permeable; except surface gravel, when the loess was thinner where they are underlain by impermeable mate­ than at present, as a result of the fall of rial, they are well drained. trees and the tearing up of the roots (Lutz and Griswold, 1939). Other Deposits Samples of sand and silt examined under the microscope consist chiefly of quartz The rounded upper slopes of Lazy Moun­ grains which, except for the larger grains, tain are covered by a mantle composed of a are fresh and angular. Chips of dark rock are heterogeneous mixture of angular rock frag­ included among the sand grains. ments and fine-grained material. In appear­ ance it differs from sandy till in the weather­ Interbedded with windblown sand and silt ed character of its fragments and in its lack are layers of light-colored volcanic ash. of compaction. The deposit is attributed to TYhere the deposits are thick, near the Mata- breaking and transportation of rock material nuska River, five £- to|-inch layers are by frost action. Similar deposits formed from present; two are together near the base of underlying unconsolidated materials have been* the section, two are higher, and one is by recognized at a few localities on the valley itself in the upper part of the section. In floor. The coarse frost-disturbed material a thinner section of loess west of Palmer it is relatively permeable, the silty material is not possible to distinguish all five lay­ less so. ers or any closely spaced pairs. West of Wasilla, ash was recognized only in thick At an. altitude of about 2,800 feet on silt deposits which evidently had been re­ the southwest slope of Lazy Mountain the frost- worked; elsewhere the ash, if present, is disturbed mantle contains rounded granitic masked by weathering. boulders unlike the bedrock exposed on Lazy Mountain. These boulders may represent an Many sections of windblown material also older glaciation during which ice covered contain roots, fragments of wood and bark, Lazy Mountain. and %- to 1-^-inch layers of woody debris. In some exposures the loess contains small Talus deposits are present beneath rock shells, and most exposures are characterized cliffs along the sides of the valley; they by darker and lighter brown-color bands due are best developed beneath Pioneer Peak. to weathering. They are composed of angular rock fragments of a wide range of sizes. The deposits have The dunes are composed of inclined beds. been formed by rockfall, rockslide, and ava­ Some of the sand and all the loess deposits lanche (snowslide). They are relatively per­ are bedded parallel to the land surface. meable . Old dunes covered by weathered sand and Deposits of silty poorly sorted and poor­ silt are present southwest of Pish Creek and ly permeable material transported by slump­ above the west bluff of Moose Creek* Modern ing, rainwash, or mudflow are present local­ dunes extend from the Knik Uiver near Jim ly along bluffs cut in unconsolidated mater­ Creek about a mile northwestward toward Boden- ial. Most of these deposits consist of silt burg Butte. A ridge of horizontally bedded encrusting the faces of bluffs and of silty windblown sand capped by cliff-head dunes gravel in layered or massive accumulations be­ that probably are still being formed extends low bluffs. Some of them closely resemble northward about 2 miles along the west Mata- till. nuska bluff, beginning about 1-g- miles north­ west of the highway bridge near Palmer. Several small alluvial fans are present west and south of Lazy Mountain. They are The horizontally bedded sand along the composed of poorly sorted sand and gravel in Matanuska bluff is locally at least 40 feet which many of the stones are angular or sub- thick. It grades westward into loess. Near angular. Irregular bedding and channel cut- Palmer the loess is 3 to 5 feet thick. and-fill structure were observed at one lo­ Thick deposits of loess are present also near cality on the slope of Lazy Mountain. These Bodenburg Butte. At Wasilla the loess is deposits are relatively permeable. commonly 18 inches thick or less. Qver the part of the agricultural area west of Wasilla Lake ramparts ridges of sand and gravel it is generally less than 12 inches thick. built along lakeshores by ice push are pres­ ent at many of the lakes in the agricultural Near Palmer the loess and sand may be area. Commonly they are a few feet high. At correlated by means of the layers of volcanic some lakes several ramparts are present. ash they contain; the topographically contin­ 15 Beaver dams may be seen along many of the time since its formation; if it had been streams, or at the downstream ends of ponds, been thawed for long intervals, weathering of in the agricultural area. Most of the dams the loess and its organic contents should the writer has seen are abandoned, breached, have become well advanced. The evidence ob­ and tree covered. tained does not permit more detailed dating of the formation of the frozen ground. Deposits of reworked gravel, sand, and silt occur along the channels of existing The writer saw no evidence suggesting nonglacial streams. Many of the lakes of that postglacial perennially frozen ground this area are being filled by the deposition has been widespread in the agricultural area, of peat in the water near shore. Peat is al­ or that in bogs such as those cited it has so being deposited in poorly drained tracts extended very far beyond the present borders throughout the area. The thickest peat known of che bogs. to the writer is a section 2.7 meters thick measured by "W. S. Benninghoff, of the Geologi­ cal Survey; this section lies to one side of Postglacial Weathering and Erosion the axis of the bog, so that a greater thick­ ness is probably present. The bog is about Postglacial weathering and erosion in 2% miles southeast of "Wasilla. According to the agricultural area began in each part of D. L. Irwin (personal communication, 1949), the area as soon as it became exposed by the deposits of calcareous marl, formed by the melting of the glacial ice. plant Chara. are present in many lakes. Scattered clam shells may be seen on the Greenstone and closely jointed shale, bottoms of some of the lakes, but the writer where they are exposed, have generally under­ has not seen accumulations of these shells. gone deeper weathering than massive graywacke and clean sandstone. The only glacial polish the writer has seen preserved on bedrock is Perennially Frozen Ground (Peimafrost) on Bodenburg Butte, in spots where the polish has been protected until recently by the loess Perennially frozen ground (permafrost) cover. Disintegration of bedrock' is chiefly was found in three bogs. On plate 1 these by frost wedging, but weakening and expansion bogs are located as follows: 2-g- miles south­ due to chemical changes probably are important east of Tfl/asilla, 2-3/4 miles east-southeast of contributory factors. Erosion of bedrock is "Wasilla, and 2-£ miles west of the experiment unimportant within the agricultural area. station. A fourth locality, in a bog three- Cutting of rock walls by the Matanuska fttver quarters of a mile south of Palmer, is de­ upstream from Palmer, and by Moose and Wolver­ scribed by Dachnowski-Stokes (1941). No doubt ine Creeks, may have occurred during degla- there are many additional localities in the ciation; possibly the gorges are partly in­ agricultural area in which small poorly herited from a time preceding the last gla- drained areas are underlain by thin bodies of ciation. Modern erosion of bedrock is most perennially frozen ground. evident along the walls of the valley, where talus is still being formed by rockfall, A pit dug in the bog about 2^- miles rockslide, and snowslide. southeast of Wasilla exposed frozen peat underlain by frozen wind-deposited silt (loess) Over much of the agricultural area till 21 inches thick, in turn underlain by non- and gravel are protected to some extent from frozen saturated gravel. The frozen loess weathering by the loess mantle. Postglacial contains crystals, veinlets, and small erosion of till and gravel has not been im­ irregular masses of ice, together with many portant except along the Matanuska and Knik well-preserved twigs and other bits of wood. Rivers and along Knik Arm. The conspicuous The same loess, where it overlies the gravel terraces described in earlier pages of this on the hilltop adjacent to the bog on the report were formed chiefly during deglacia- south, is 18 to 20 inches thick; .in some sec­ tion, although cutting of the lower terraces tions on the hilltop the loess contains poor­ appears to have continued to the present. ly preserved woody material. The writer be­ The most conspicuous modern erosion is by lieves it likely that the presence of many undercutting and slumping of the Matanuska wood fragments was characteristic of the loess bluffs south and northeast of Palmer. There in general during its deposition, and that is much erosion of channel deposits by the they are not peculiar to the loess now found Knik Mver during its annual flood. beneath the frozen peat. It seems likely that wood in the loess beneath the bog was No evidence of important erosion was preserved by being frozen at some time sub­ found in most loess sections, but a greater sequent to deposition, and that wood in un­ thickness of deposits in many valleys and a frozen loess nearby was largely destroyed by lesser thickness on adjacent hills suggest weathering. that erosion of the loess has occurred. The topography does not seem sufficient to cause The perennially frozen ground in this differential original deposition by the wind. bog rests on glacial deposits and involves The writer believes that throughout the period loess; it is therefore postglacial. It prob­ of deposition newly fallen dust was washed or ably was not formed until the loess mantle blown from hillsides into adjacent valleys. had reached approximately its total thick­ Folding of older layers of volcanic ash and ness in this part of the agricultural area; of weathering zones may be seen in many ex­ if it had been present during deposition of a posures of loess. This deformation, which large fraction of the loess, silty peat prob­ occurs on flat tracts as well as on slopes, ably would have been deposited in the bog. is attributed to frost action during the It has remained frozen for all or almost all earlier part of the period of loess deposi- 16 tion. Chemical weathering of loess is shown near-surface deposits of the Matanuska Valley by alternating bands of lighter and darker agricultural area. A brief summary of the brown silt in many exposures. These bands development of the topography of the valley are commonly parallel to the present land sur- floor, as already interpreted in this report, facej they probably were formed during periods follows. of temporarily slower dust deposition before being buried by further deposition. Over The last ice tongue to lie over the most of the agricultural area west of Wasilla agricultural areas, a large glacier formed by . the loess is so iron stained that'ash layers the merging of the^Uatanuska and Knik Glaciers J>\ and color bands, if they were present, are of that time (and possibly including ice from completely masked except in thicker deposits farther up the Matanuska Valley), extended a in valley bottoms. few miles west of what is now Big Lake, where its end moraine is preserved as an arcuate The soils of the area have been describ­ band of hills. Deglaciation over most of the ed by Rookie (1946) and by Kellogg and valley floor was by stagnation. The behavior Nygard (1951). In the eastern part of the of the Knik Glacier at that time is not known. area, particularly near Palmer, soil is de­ It has been suggested that during melting of veloping on windblown sand and silt which is the ice the topography formed depended large­ relatively thick; soil formation at no time ly on the distribution of rock debris in the has progressed far because of continual addi­ ice. Water from the melting ice cut shallow tion of fresh parent material to the land valleys across the ground moraine and deposi­ surface. In the western part of the area, ted gravel in them. The last fragments of where the loess is commonly less than a foot ice that remained, blocks lying in depressions thick, soil-forming processes have affected on the moraine, were partly surrounded by not only the loess itself but, in many sec­ gravel deposits before they melted entirely. tions, the uppermost part of the underlying Farther east, masses of stagnant ice were glacial material. Kellogg and Nygard (1951, surrounded or buried by thicker gravel depos­ p. 72) believe that podzolization (formation its. Melting of the buried ice blocks and of podzols or forest soils) is the dominant collapse of the gravel covering them led to soil-forming process but that podzols have formation of the irregular topography charac­ been formed only where deposition of windblown teristic of this part of the valley floor. material is slow enough that podzolization can keep pace with addition of new material. The glaciation upon the valley floor was accompanied by glaciation in the surround­ Water erosion of the loess mantle is un­ ing mountains. The larger mountain valleys important. The high permeability of the held tributary glaciers which joined the main material and the presence of the vegetative ice stream. In many of the smaller mountain cover (and perhaps the low rainfall intensity) valleys were glaciers which did not join make surface runoff negligible* Wind erosion, larger glaciers; their moraines may be seen except on bare alluvial flats, was insigni­ in valleys above Eklutna Lake, up the Little ficant prior to the introduction of agricul­ Susitna Canyon, and up the valleys of Moose ture in this area. At present wind erosion and Wolverine Creeks. Many valleys and moun­ is a serious problem in some cleared agricul­ tain slopes which were not glaciated were tural land, particularly in the path of win­ affected by frost action due to the same cold ter storms moving down the Matanuska Valley. climate that produced the glaciers, and these were covered by a mantle of frost-dis­ turbed debris. Rock glaciers now preserved Quaternary History in some of the mountain valleys may have been formed at this time. There is little evidence of the form of the preglacial topography in this region, Meltwater from ice farther up in the but the mountains must have presented a dif­ Matanuska Valley flowed southward upon leav­ ferent appearance before the glacial over- ing the mountain valley northeast of the site steepening of slopes. Oversteepening is most of Palmer. This change of course may have pronounced along the front of the Chugach been because of ice which lay in the way to Range, where several prominent spurs were the west. The meltwater streams repeatedly truncated. Bodenburg Butte and the other bed­ made channels and trenched them, producing rock hills between the Matanuska and Knik the conspicuous terraces which border the Rivers probably are remnants of the preglacial present Matanuska River. divide separating those streams. During or after deglaciation streams The writer has found evidence for two, flowing down the mountainsides bordering the at the most, glacial episodes in this area. valley began to build alluvial fans. Beneath The older, less well established, of these steep slopes talus began to accumulate. episodes, which the writer believes resulted These deposits are still being formed today. in deposition of the now-buried older till Probably during deglaciation, and certainly and gravel, has no surface expression in the afterward, wind carried dust from the bare topography of the valley floor. alluvial plains and deposited it over the sur­ face of the valley floor. Deposition of wind­ The deposits of the younger ice form the blown dust has continued to the present; the surface over most of the agricultural area. silt forms the parent material of most of the Nonglacial processes, during and after melting cultivated soil in the agricultural area. of the last ice that 1 ay over the valley During the early part of the period of dust floor, nave modified the deposits left by the deposition there was at least one period ice and formed nonglacial deposits. characterized by frost disturbance of the loess. Local bodies of perennially frozen The geologic map (pi. 1) and the section ground were formed in bogs; they have not given in figure 2 summarize the surficial and been dated but are probably later than this 17 frost action. Water-Bearing Materials The gradients of the terraces that stand Bedrock at about the level of the present Matanuska River and Knik Arm suggest that the water Although several wells have penetrated body into which the meltwater streams flowed bedrock in the Matanuska Valley agricultural stood considerably below present sea level. area particularly in the vicinity of Palmer, During deglaciation sea level rosej salt water only a few have obtained water from it. Hell partly filled the lower reaches of the Mata- / 70 (well records, table 5), on the farm of nuska-Knik Valley and an estuary,,Knik Arm, Victor^ Palk, Jr., near Bodenburg Butte, is was formed. Glacial silt carried' into Knik in rock from 36 feet below the surface to its Arm was deposited as bars. Some of these total depth of 110 feet. The rock is probab­ bars, which now stand above the level of ly greenstone. The driller's log reports average high tide, suggest that at least water at several places, chiefly below 65 once during postglacial time sea level was feet; it is probably derived from fractures somewhat higher, relative to the land surface, in the rock. than it is now. Activity of the streams con­ tinues, and low terraces along the Matanuska The Lester (123), Gregerson (120), River suggest that the river has recently Knapp (136), Mehan (376), and Cope (142) wells, been eroding its alluvial plain. in or near Palmer, penetrate bedrock, although the Knapp well, at least, may derive its water from overlying gravel. The old slaugh­ GROUND WATER terhouse well (144), in rock, and the old hospital well (135), which may have penetrat­ Occurrence ed rock, obtained salt water (se,e section on "Quality of water"). Two other wells, proba­ Below a certain level in the near-surface bly also in Palmer, obtained salt water which part of the earth, the pores or interstices may have come from bedrock. between fragments of unconsolidated sediments and the fissures and other openings in bedrock Bedrock is at best a poor water-bearing are saturated with water. The upper surface material in the Matanuska Valley agricultural of this zone of saturation is known as the area. In most drilled wells which penetrate water table. Water occurring below the water bedrock, the bedrock part of the hole serves table is ground water. Between the water merely to collect and hold water derived table and the land surface there is commonly from the overlying unconsolidated material. a zone of aeration in which the pore spaces Wells such as those on the Bugge and Thuma and other openings are not saturated. farms (128 and 145), which are seated on bed­ rock below the water table, have proved satis­ Ground water is derived from rain, from factory. If bedrock is found near the sur­ the melting of snow, and from surface bodies face during well construction, it will proba­ of water. The water moves downward through bly be wiser to move to one side, space permit­ the zone of aeration until it reaches the ting, and begin a new well rather than to water tablej it then migrates from higher to undertake blasting in a dug well or more lower parts of the saturated zone until it difficult drilling in a drilled well. is discharged naturally through seeps and springs into streams or lakes, or by evapora­ tion, and by transpiration of plants. Arti­ Till ("hardpan") ficial discharge may take place from wells and improved springs, which are excavations About 25 wells have obtained water from extending below the water table. till in the Matanuska Valley agricultural area. Many other wells have passed into or Locally, fine-grained or dense material through till and derive their water from which does not permit easy passage of water gravel lying over or beneath it. (that is, which is relatively impermeable) may be surrounded by relatively permeable The till is relatively impermeable. material. If the upper surface of such an Field tests made with a variable-head permea- impermeable mass is below the land surface meter suggest that its permeability is of the and is sufficiently extensive, it may hold in order of 1/10,000 that of outwash sand in this the permeable material above it a body of area. The permeability of thin sand layers perched ground water that stands higher than in the till, on the other hand, is comparable the general water table in the surrounding with that of surficial outwash sand. "Where area. Streams, marshes, and lakes are present till lies near the land surface, bodies of at places where the land surface intersects surface water or bodies of ground water in the water table; these features are therefore gravel may be perched above it; where till commonly,considered to mark the local posi­ lies at and immediately below the water table, tion of the water table. Bodies of surface water in reasonable quantity cannot be obtain­ water may be perched upon impermeable materi­ ed from the upper part of the saturated zone. al, however, and caution must be used in in­ terpreting a lake or stream as an indicator In almost all wells that obtain water of the local water table. from till, the water occurs in sand or gravel

18 layers within the till. These permeable lay­ Outwash Sand and Gravel ers are commonly 1 foot or less in thickness, as in the Nash (6), Venne (22), Moore (230), In the Matanuska Valley agricultural and Bailey (343), wells, but there are some area sandy gravel and subordinate sand and thicker water-bearing zones. The Withey well clean gravel are of such permeability as to (474) obtains water from a 2-|--foot zone of be good water-bearing material wherever they sandy material. A Geological Survey test occur below the water table. Most wells ob­ hole (4) in till penetrated 1 foot of coarse tain their water from these materials. sand, 2 feet of gravelly material, and 1 foot of fine and medium sand before passing into Over a large part of the agricultural till again at a depth of 26 feet. "Water area ground water is present under water- seeps into well 2, at the Lazy Mountain table conditions. Perched water bodies cause Childrens Home, at several levels; the water­ apparent local irregularities in the level of bearing material may be superglacial till the water table. On ground moraine in the rather than layered sand or gravel, however. western part of the area lakes, marshes, and bodies of ground water in gravel are common­ Water obtained from near-surface till, ly perched on till. In some horizontally as at the Lazy Mountain Childrens Home, is bedded gravel deposits ground-water bodies probably derived from precipitation falling are perched on silty layers. The Brown (42), upon the land surface. The writer believes Rippy (43), and Bastian (44) wells, at the that water obtained from included sand or east end of Bodenburg Butte, obtain water at gravel layers is derived from the till itself a depth of 28 to 35 feet. The Gallagher by downward percolation, and that the quanti­ well (47), a few hundred feet west of the ty of water obtainable depends not only upon Bastian well, passed through a thin body of the permeability of the till and the size of water at about 34 feet. The static (non- the well which collects the water but also pumping) water level in the Gallagher well upon the roof area of the sand or gravel lay­ was 53 feet below the land surface in Septem­ er. ber 1951. It is likely that all these wells reached or passed through the same body of The importance of fractures cutting till, water; the impervious layer beneath it may be and sandy or stony layers in it, in the move­ buried channel floored with silt. Perched ment of ground water through the till cannot water, in bodies as much as a few feet thick, be estimated because of the inadequacy of was found at several other levels in the exposures. Gallagher well (Prey, J. D., personal communi­ cation, 1951). Springs issuing from till are uniirpor- tant. Seepage from thin sand or gravel lay­ Apparent irregularities in the water ers in till may be observed in some exposures, table may also be due to the presence of rela­ as in the east bluff of the Matanuska River tively impermeable material at and below the about half a mile north of the highway bridge. level of the water table. A till layer in The flow of water from seeps the writer has gravel beneath part of Vasilla lies at the observed is not sufficient for more than a water table over an area of at least 1 to 2 very small supply, but the water may present acres. It was found in the P. Swanson (513), a drainage problem if the till is to be exca­ Wasilla Hotel (519), G. Swanson (520), and vated. Teeland Store (521) wells (Fabian, Jack, personal communication, 1949). In several The yields of wells in till are small of these wells water is reported to have at best, but a well penetrating one or more risen as much as several feet when the base water-bearing layers may provide a modest of the till was penetrated. Till in the Berg- supply for a household or for a limited num­ man well (531) is beneath the water table. ber of livestock. The Cook well (338) yielded Till was not found in wells surrounding those about 50 gallons of water per day in 1949 but listed above. was dry during the dry summer of 195t>. The Nash well (6) yields about 100 gpd regularly, Many of the small streams flowing across and the Kibbe well (586), 150 gallons. Geo­ gravel deposits appear to be perched. Water logical Survey test well 3 (well 4 in table in the Carson well (576) stands 33 feet below of well records) was pumped steadily at the the surface; a creek a few hundred feet away rate of 30 gallons per hour over a 3-hour is only about 6 feet lower than the ground period, with a drawdown of 19 feet. The re­ surface at the well, or 27 feet above the covery of water levels in wells in till may water table at the well. Well 412, on the be alow; an extreme example is the Bradley Kirchner and Menk property, is 20 feet from well (607), a large-diameter dug well, in Wasilla Creek and 6 feet above it; the water which the water level required 7 days to re­ level in the well is about 7 feet lower thftn cover after 250 gallons had been pumped in the bed of the stream, and according to re* 45 minutes. ports, movement of ground water at the time the well was dug was in the direction oppo­ The development of ground-water supplies site that of stream flow. The water level in from till, even in limited amounts, must be TJSGS test well 2 (well 57) is about 30 feet considered because of the current need for below the bed of Bodenburg Creek, 50 feet water supplies in areas of till that are al­ away. It seems likely that the beds of these ready settled and because of possible future streams have been rendered relatively imper­ need in large unsettled areas of potential meable by a "seal" of silt in the gravel over agricultural land that are underlain by till. which they flow, although they probably feed the ground-water body to a slight degree.

19 In many wells that pass through a buried they are permeable and permit rapid infiltra­ stratum of till and into gravel beneath, tion of water that reaches the land surface. water rises into the well to a level higher Information obtained during irrigation experi­ than the base of the till. In several such ments (Wilson, C. W., personal communication, wells (for example, the Bryant (287), Gaylord 1951) shows that on 2 farms, 1 at Palmer and (294), Palmer (363), Benson (364), and Hemmer the other near Matanuska, infiltration of (398) wells, west and northwest of Palmer) water into agricultural soil may take place the static water level is higher than the at the rate of at least one-third inch per upper surface of the till. In the Palmer hour for several hours, without perceptible test well (363), on a hill which rises 30 to runoff. 40 feet above the surrounding surface, the water rose within 20 feet of the surface. Small perched bodies of water occur in Figure 5 summarizes the available information eolian sand and silt on the Holtet property, on the depth and thickness of the till stra­ along the Matanuska bluff about 4 miles north tum and the water level in wells along a sec­ of Palmer. The water is present only beneath tion west and northwest of Palmer. It is evi­ depressions, which are interpreted as pits dent that imperfect artesian conditions are formed by the melting of buried blocks of developed locally where the buried till stra­ glacial ice. The impermeable layer which re­ tum is present. tains the water is probably the residue of unsorbed glacial debris which was left upon With the exception of seeps from sand melting of the ice or the deposit formed in layers in till, all the springs seen by the a pond which may have occupied the pit during^" i writer derive their water from gravel. The melting of the ice. By prd^fecting with a A^l springs occur in three general situations: pesthole auger Mr. Holtet has found water in ' (1) in saturated gravel below the water table, many of the pits on his farm; no test holes exposed by recent stream erosion, saa/it at &JU on hilltops or ridges between pits struck springs along the Matanuska bluff south of water, however. The sand and silt range from Palmer and near Matanuska; (2) in topographic 9 to 14 feet in thickness; the perched water depressions where the wate.r table intersects bodies are from 3 to 7 feet thick. Most indi­ the land surface, as is probably the case at vidual pits cover a few acres or less. Figure Brazil Springs northwest of Palmerj and (3) 4 is a section through a typical pit. It is at the contact of saturated gravel and under­ evident that the water is derived only from lying till, as on the hillside above (east of) precipitation received within the pit itself, the mouth of Pish Creek. and that the quantity of water present in a single pit is small. Perched water in eolian Only one well in the Matanuska Valley deposits has not been reported elsewhere in agricultural area has been pumped at a rate the agricultural area; it is probably present greater than 100 gpm. This is well 363, only in restricted areas near the Matanuska drilled as a test well for the city of Palmer. River where the eolian material is thick and It yielded 118 gpm with 35 feet of drawdown, is underlain locally by impermeable material. after 16 hours of pumping (Bourne, 1952). Therefore its specific capacity was about 3.4 gpm per ft of drawdown. This well was fin­ Hydrology of Physiographic Units ished with a screen. It obtains water from sand and sandy gravel beneath a buried stra­ As an aid in the discussion that follows, tum of till. USGS test well 1 (well 37) the writer has divided the Matanuska Valley yielded 44 gpm with a drawdown of 5.6 feet agricultural area into physiographic units. after 4^ hours of pumping. It is a 3-inch The differences in the land surface in these open-end well. In view of its small size, units are a reflection of major differences lack of a screen, and higher specific capaci­ in their geology'. These differences cause ty, it must penetrate material more productive significant changes in hydro logic conditions than did well 363. These are the only wells from one unit to another. The location and in the agricultural area that have been pump­ boundarils of the physiographic units are ed at a rate greater than about 20 gpm. Farm indicated in figure 2. Geographic names are wells are usually pumped at rates of a few applied to the units only for convenience of gpm for onlv short periods. According to reference in this report. Blvers (1&50, p. IT), the Hemmer well C398) has yielded 5.65 gpm with a drawdown of 7 feet, in a 24-hour test. Lazy Mountain (Unit 1) Most wells in the agricultural area, and The -western slope of Lazy Mountain and all those producing more than about 200 gpd, the tract that extends westward from it to obtain their water from outwash sand and the Matanuska River (hereafter termed the gravel. There is no reason to doubt that lower slope) constitute physiographic unit ! yields of wells in sandy gravel in the agri­ The higher slopes were not glaciated and are cultural area can be substantially increased Qomposed of rock covered by slope deposits. by use of well screens and proper development An intermediate section of the slope consists practices. of lateral-moraine and alluvial-fan deposits. The lower slope, which is more than half of the unit, is composed of ground moraine or Windblown Sand and Silt gravel-covered terraces cut into it. The windblown sand and silt which mantle The talus and frost-disturbed deposits glacial deposits in the Matanuska Valley agri­ on the upper slopes of Lazy Mountain are re­ cultural area lie above the water table and latively permeable. Streams' which derive generally are not saturated. They are impor­ their water from these materials, and possibly tant in the hydro logic cycle, however, because in part from fractured bedrock beneath them, 20 continue flowing down the west slope of the All the ground water present in deposit? mountain during prolonged dry weather. The on this slope is derived from precipitation alluvial fans at the foot of the steep slope upon it. are composed of relatively peimeable sand and gravel. The lateral moraine and the lower slope are relatively impermeable till. Seeps Bodenburg Terrace Area (Unit 2) occur along the lower edges of the alluvial fans where the underlying till reappears at The land surface in most of physiographic the surface. unit 2, which is the tract between the Mata­ nuska and Knik Rivers and the mountains to The lower slope is poorly drained and, the east, consists of stream-laid sand and in places, marshy. Wells in the till have sandy gravel. Bodenburg Butte and smaller ranged from unsuccessful to moderately suc­ hills of bedrock, and a few hills of till, cessful. USGS test well 3 (well 4) and the protrude through the gravel deposits. Beechlk (5) and Nash (6) wells penetrated thin water-bearing layers in the till. A The part of this area mapped for this re­ well at the Lazy Mountain Childrens Home (2) port is well drained; to the southeast the obtains water from slightly permeable near- surface is lower and marshy. Bodenburg Creek, surface till. According to reports a well east of the Anchor age-Palmer highway near the begun on the Copenburg property, about a Knik River, appears to be a "perched" stream quarter of a mile north of well 4, is dry at (p, 19), During the annual late-summer a depth of about 50 feet. The success of a flood of the Knik River the water table near well in the till is dependent upon its pene­ the river fluctuates in response to changes trating a sorted layer that is water-bearing. in river level. The location, thickness, and extent of these layers cannot be predicted on the basis of Whatever ground water is in the rock available evidence. The probability of find­ comprising Bodenburg Butte and similar hills ing water-bearing material in till near the occurs in fractures in the rock. Only one surface seems to be highest if the well is well (70) obtains water from the rock, and located in one of the shallow valleys that there is no information regarding the amount trend down the lower Lazy Mountain slope. It of water available in general from rock in is unlikely that the average well in the till this area. The lake on the east end of will yield more than a modest household or Bodenburg Butte is spring-fed; the water may farm supply. be derived partly from fractures in the rock, but it probably comes also from the glacial Small seeps occur along the bluff about deposits mantling the rock. half a mile north of the Matanuska Bridge. Similar seeps from till along the hillside Till is present in two hills left by overlooking the lower ground south of the erosion in the northern part of this area lower slope of Lazy Mountain yield enough and under the higher sloping ground below water to hinder excavation or to make drain­ (south of) the lower slope of Lazy Mountain. age of completed excavations difficult. In these areas the till will make acquisition of ground-water supplies relatively difficult* In the triangular tract between the Mata­ nuska River and Wolverine Creek several The lower slope of the mountain wall east terraces indent the original till surface. of this area is mantled by relatively permea­ Each of these terraces is covered by gravel. ble alluvial-fan and talus deposits. The Stewarj^ well (1), on the uppermost ter­ race, passes into till 10 feet beneath the The remainder of the Bodenburg terrace surface. No other wells have been constructed area is underlain by permeable gravel. Small on these terraces, and the thickness of the perched bodies of ground water are present gravel deposits is not known. By analogy locally (p, 19) southeast of Bodenburg Butte with terraces on the other side of the Mata­ near its east end. Ground water probably nuska, however, it seems likely that till is moves southward and southwestward through near the surface on the higher terraces here. most of this area. North of the bedrock hills Exposures along the river and along Tfolverine near Bodenburg Butte, and also on the terrace Creek suggest that bedrock is near the sur­ southwest of Bodenburg Butte, the water table face on the lower terraces. Nonetheless, stands close enough to the land surface to this terraced tract is the most promising area be accessible to suction pumps. Geological for development of ground-water supplies on Survey test well 1 (37) yielded 44 gallons of the slope of Lazy Mountain. Where a choice water per minute from an open-end 3-inch cas­ of location is possible, the most favorable ing having no screen. Similar quantities of is toward the center of a terrace, away from water probably can be obtained from wells In its inner and outer edges. gravel elsewhere in this area, and larger quantities might be expected if screens were Small gravel-covered terrace remnants used and the wells developed. are present on the slope above fnorth of) the Matanuska River bridge. A spring on the Ground water in this area is derived Clark property discharges 3 to 5 gpm and is from precipitation upon the land surface and reported (Clark, Prank, personal communica­ from runoff from the mountains to the east. tion, 1949) to have a fairly regular flow. For quantity of water available and the The writer believes it unlikely that a cost of its acquisition, this physiographic deeper gravel stratum is present beneath unit is probably the most favorable part of till on the slope of Lazy Mountain. the Matanuska Valley agricultural area for the development of ground-water supplies. 21 Dry hote Dry hole Impermeable layer Perched /water c5-\ / |7 ->-" O O ° o ° °';^r-~"^:~^^rr'°??;='Fqc^^f2f:2r^:^^ .% °"dsiit S0n6. o On O O « o o o o o o ° ° ° 0 o / ° o ° / ° o - 0 o o ° o o O 0 o 0 ° o o o o

0 ° o O O Unsaturated gravel ° ° + ° o _ 0 ° d o o ° O o o o

O o " O o o o o oo o ^° 0 0 o°(? (>0 o 0

Figure 4. Generalized section showing perched ground water in windblown sand and silt.

61° 149° 25' 40'

EXPLANATION o Well that passes through till stratum

Well that does not reach till stratum

492°

«40 437 435° ° °442

61°,_____ B Matanusko 32f'149°25'

I*

EXPLANATION Wotir level

Fiu.-are 5. Map and section showing distribution of older glacial deposits known from well logs, Matanuska-Knik Flats and .Alluvial gravel, till(?), or bedrock, were "dry holes." Plains (Unit 3) A sixth, which reached bedrock at 72 feet, is reported (ARRC log) to have obtained "sulphur The flats bordering Knlk Arm, together water" at 121 feet. Two colony wells (135 with existing alluvial plains of the Matanus- and 144) at Palmer obtained salt water; two ka and Knik Rivers, constitute an area in other colony wells, probably in Palmer, also which the water level stands no more than a obtained salt water. In three of these wells few feet beneath the land surface. "Water in salt water was reached at 140 to 160 feet gravel beneath estuarine silt at the Eklutna above sea level. In the old slaughterhouse CAA station becomes brackish during very high well (144) salt wate'r was reached at a depth tide (Tffesternbarger, Kenneth, personal communi­ of 569 feet, or about 340 feet below sea cation, 1951). Hater beneath the similar flat level; this well is reported to have pene­ north of Knik Arm is certainly brackish or trated shale and limestone from 18 to 590 feet salty at least part of the time. Large quan­ below the surface. The difference in chemi­ tities of water are undoubtedly available from cal composition of water from two of the the alluvial deposits along the Matanuska and wells, 134 and 135( see section on "Quality Knik Rivers, but recurrent flooding renders of water"), and the absence of salt water in settlement of such low areas unwise. The other wells in the vicinity of Palmer suggest dommunity of Matanuska is said to have been that the highly mineralized water is of local abandoned because of flooding by the Matanuska occurrence. It probably has been trapped in River. Flooding and channel shifting along the rock since a time when this area was these braided streams would prevent installa­ covered by marine or estuarine water; it is tion and use of infiltration galleries in the not modern salt water from Knik Arm. alluvial deposits. The Bugge (128) and Thuma (145) wells are seated on bedrock. In the Thuma well, Palmer Terrace (Unit 4) which is 24 feet deep, 2 feet of saturated gravel overlies the bedrock. In the Cope The Palmer terrace is the single exten­ well (142), about half a mile east and at the sive terrace upon which Palmer is situated. same altitude, the water stands about 75 feet Permeable gravel underlies the surface of most below the surface. In the Moffitt well (229), of the terrace, and except for a few depres­ about half a mile south and 10 feet lower, sions containing ponds the surface is well the static level is about 85 feet below the drained. surface. Water in the vicinity of the Thuma well thus appears to be perched on bedrock. In general, farm wells in the area south In other wells in or near Palmer the water of Palmer have been successful. Over most of stands about 30 to 40 feet beneath the land the terrace the water stands below the limit surface. In wells within a radius of a mile of suction lift. Locally it is necessary to to the southeast and south (Green (138), drill somewhat below the level of the water Cope (142), Stock (152), Moffitt (229), and table before water is brought into the well; others) the static level is 75 feet or more this seems to be due to the presence of local beneath the surface. The writer believes that layers of relatively impermeable silty materi­ water-bearing gravel in the immediate vicini­ al. The water table slopes south and south­ ty of Palmer is not only very local and thin east toward the Matanuska River, and seeps but rests on bedrock or till so that the occur along the base of the bluff. Seasonal water is perched above the level of the water fluctuations in the elevation of the water table in the thicker gravel to the- south. table are less than in most of the rest of the New household wells in Palmer may or may not agricultural area, and the maximum expectable be successful, depending upon chance location, fluctuation is probably not more than a few but it is unlikely that large quantities of feet. It is possible that in a few places water can be obtained there. rock or deep-lying till will be penetrated before the water table is reached, but this is Ground water in the Palmer terrace is not so likely as it is at or north of Palmer. derived not only from precipitation upon its There are no data from which to estimate the surface but from water moving beneath.the sur­ possible yields of wells. face from the higher ground to the north and northwest. Subsurface conditions are more irregular near and north of Palmsr. Till and bedrock are at or near the surface in many places. Eska Creek-Matanuska Agricultural Surface exposures and well logs suggest that a Experiment Station Area (Unit 5) partly buried bedrock ridge or series of knolls extends westward from the Matanuska River Unit 5 may be considered to comprise bridge. Other bedrock hills, some of them three smaller tracts (fig. 2), in each of buried in gravel, are in Palmer or just south which gravel deposits as much as 50 to 100 of the city. The Bugge (128) and Felton (130) feet thick rest on older glacial deposits 01? wells have been in use for many years (the on bedrock. This unit is conspicuously dif­ Bugge well since 1914) and have yielded ferent from those already described, in which supplies sufficient for many families; they the surface is largely of ground moraine dr penetrate water-bearing gravel. Three other of extensive terraces, and from the physio*- wells (now abandoned and filled) dug by indi­ graphic unit lying to the west, in which the viduals in Palmer are reported to have been surface consists of ground moraine or thin successful, and during establishment of the gravel deposits. agricultural colony four successful colony wells, 37 to 46 feet deep, were constructed in Conspicuous ridges and hills in the alrea gravel. Five other colony wells in Palmer, in between Eska and Moose Creeks and extending 23 about 3 miles west of Moose Creek (unit 5a) that the till is discontinuous. The writer are underlain by gravel. The surface is well believes that the ground water in unit 5b is drained. Bedrock is exposed along the Matanus- most reasonably considered as a single body. ka River and Moose Creek and in Wishbone Hill Most wells that pass through the till stra­ just north of this area. Till has been found tum obtain water under artesian pressure; in some wells (such as, 253, 256, 257, and 258) many wells do not reach till, however, and and probably lies near the surface over much these encounter water-table conditions. It of the tract. Little is known of the hydro- seems likely that the till is absent locally, logic conditions in unit 5a. Pew wells have and that at such places the body of unconfined been begun and most of them have not been water is connected with the confined water completed. The water stands 75 feet below beneath the till stratum nearby. the land surface in the Boulter well (253); the water is derived from what may be either The local occurrence of water in wind­ .a sorted layer in till or gravel beneath the blown sand near the Matanuska bluff has al­ till. The water level may be nearer the sur- ready been described (p. 20). face on the lower terraces along the Matanus- ka River, or in depressions farther from the Brazil Springs, at the bend of the Pal- river. Recharge of ground water in this area mer-Fishook road about 3 miles northwest of is from local precipitation and from runoff Palmer., provided the Palmer water supply for and underground flow down the slopes to the many years. The water issues from gravel at north, except in an area east of Moose Creek the base >of a small hill. It probably reaches which is separated from Wishbone Hill by a the springs by movement down (southwest) the deep valley and which probably receives water small valley that crosses the Henry Laltose only from local precipitation. property (near wells 314 and 315) and in which there are two small lakes (pi. l). On Between the area just described and the the basis of the local topography and of rec­ Pa liner-Was ilia Road the topography is irregu­ ords of wells nearby (302, 315, 317, and 318) lar but more gently rolling than near Moose the writer believes that the small valley may 'Creek. This part of the physiographic unit be floored by till at shallow depth but that (unit 5b) is bounded on the west by an ar­ the hills are composed of gravel. If this bitrarily-chosen line which lies to the east interpretation is correct, the springs proba­ of Finger Lake and the lakes north of it. bly/nark the intersection of the water table East of this line the till is generally cover­ by a slight topographic depression. Spring ed by thick gravel; to the west the till is flow is probably about 150 to 200 gpm when exposed or is covered by thin gravel. The the water table is at its average position; surface is commonly well drained, although during the dry season of 1950 and in 1951 the some of the deeper depressions contain lakes. spring flow declined so markedly that the Many of the wells in this part of the unit existing pipeline was extended north to Car­ pass through surficial gravel into a buried negie Creek to obtain surface water. stratum of till ffig. 5 and table 2). The surficial gravel is about 20 to 100 feet The part of this physiographic unit des­ thick. The underlying till is 15 to 90 feet cribed in the preceding paragraphs grades thick; it is probably absent locally. Most southward and southwestward into a more roll­ wells that pass through the till obtain water ing surface having more conspicuous ridges from gravel beneath it; in some of these wells and hills. This more rugged tract (unit 5c) the water is under artesian pressure (fig. 5J. extends from an arbitrarily chosen line be­ Several wells on the Owen Moffitt farm (ARRC tween Palmer and Four Corners to the bluff tract 132) were "dry holes" even though they overlooking the flats west of Matanuska. The penetrated the lower gravel to greater depths surface is well drained. The only stream is than successful wells nearby. This may be Wasilla Creek. Lakes occupy several depres­ due to local relative impermeability of the sions near Matanuska. Springs along the base gravel because of cementation; slight con­ of the bluff near Matanuska are fed by ground- solidation of the lower gravel has been report­ water flow from the north. One of these ed in well logs (table 2). One well (305} springs supplied the community of Matanuska on tract 132 obtained water after being deep­ (now mostly abandoned). In 1949 the Alaska ened by drilling in 1951. Railroad obtained 8,000 to 9,000 gpd from this spring. Well 363, drilled for the city of Palmer and completed in January 1952, yielded 118 Hydrologic conditions in' the area be­ gallons of water per minute with 35 feet of tween Palmer and the experiment station are drawdown; the nonpumping level is 20 feet poorly known. The water table slopes south­ below the surface. This well obtains it.s ward. In the Blunck well (378) the water water from sand and sandy gravel beneath the stands 119 feet below the surface. No other buried till. The well was finished with a wells have been drilled west or southwest of 20-foot section of well screen. Similar quan­ this one, in this physiographic unit, except tities of water should be available at many several near Wasilla Creek and near the other localities in this area if wells are experiment station. The Blunck well and sev­ constructed with screens; larger quantities eral wells near the experiment station obtain might be available if larger pumps are used. water from gravel beneath a buried till stra­ tum which may be the same as that found north­ Some other wells in this part of the west of Palmer. Water in the Blunck well is physiographic unit obtain water from the not under artesian pressure; the till stratum younger gravel. In most places this gravel is probably interrupted to the northwest. lies upon the buried till, but in some places Artesian conditions are present in some wells it is likely that gravel-filled channels west of the experiment station. indent the upper surface of the till or even

24 Little Susitna-Goose Bay Morainic outwash gravel in alluvial-plain deposits, Area (Unit 6) pitted deposits, and thin deposits which locally mantle hills of till. Alluvial-plain The Little Susitna-Go9se Bay physiogr^ph- gravel at the gravel pit 1^ miles west of ic unit is characterized by extensive areas Pittman is 20 feet thick or more; it rests of ground moraine separated by gravel-flo.bred on relatively impermeable till. Elsewhere in valleys. The topography shows a conspicuous this vicinity crevasse fillings of till which southwestward trend, and drainage is toward protrude through gravel show that the gravel the southwest. / is relatively thin (probably not thicker than about 10 to 20 feet). Gravel in some of the Three large areas in which the su/rface alluvial plains is locally openwork cobble is ground moraine are shown on plate 1. One gravel. extends from the vicinity of the expeyriment station and Wasilla to Goose Bay; another is The gravel in all these deposits is ex­ north and west of Wasilla; the third; lies cellent water-bearing material wherever it is north and west of Pittman. In each/of these saturated. Springs along the walls of the areas till forms the surface over aixtensive conspicuous valley 2 miles southeast of tracts or is mantled by thin depos/lts of Wasilla, and on the hillside east of the gravel. Many small valleys and closed depres­ mouth of Pish Creek, and probably those on sions contain somewhat thicker gravel depos­ the Pleckenstein property 1-|- miles southwest its which are the most promising sources of of Wasilla, are contact springs; they derive near-surface ground water in areas of ground their water from gravel lying on till and moraine. It is emphasized thatf wells in till discharge at the outcrop of the contact be­ are likely to obtain water* hftx»ai' ooly BHMPtl tween the gravel and the till. qnnnt.ltioa aro nfrtniniafrlo "from layers of sand or gravel in the till, "if gravel depos­ Little is known of ground water in the its are present nearby, shallow wells in gravel-covered terraces north of Knik Arm. till should not be constructed. The writer believes that the gravel is thin in most localities (as in wells 622 and 623), Five extensive areas of gravel deposits and that the quantities of water available are present in this physiographic unit. are small. Quantities of water sufficient Gravel-covered terraces indent the ground for household use should be readily available moraine above the bluff north of Knik Arm. in valleys on these terraces, however. The topography of the eroded surface of till beneath the gravel is irregular, and locally Pew wells have been drilled in the pitted hills of till protrude through the gravel dep-osits northeast-of Wasilla, but it appears (pi. 1). At the bluff overlooking'Knik Arm that water is readily available between the near Lucy Lake till is exposed at the land lakes. Deposits beside (to the north and surface; in well 631, about 1 mile southwest, south of) the lakes are more likely to be the gravel, as reported by R. Lathrop (per­ underlain by till. Till was found in wells sonal communication, 1950), is at least 60 502 and 506, beside Wasilla Lake. feet thick. Elsewhere, however, shallow wells reach till (for example, well 623, in Wasilla, situated on gravel deposits which till is 15 feet beneath the surface), between Wasilla Lake and Lake Lucile, undoubt­ and it is likely that till is near the sur­ edly has large available supplies of ground face over much of the area of these terraces. water. Till is known to be present as a thin Poor surface drainage on some of the terraces layer beneath part of the community (p. 18). is probably due to the presence of till near Till found in the Bergman well (531) may be a the surface. part of this layer or of a thicker body. Because the till beneath Wasilla may be part An extensive area of pitted gravel depos­ of a thin layer, it may reasonably be excava­ its lies northeast of Wasilla. Locally the ted for a depth of several feet in the hope gravel is thick (46 feet in well 470), but that the underlying gravel will be reached. elsewhere the gravel is thin. The old King However, this is not true on the hills north well (463) south of Kings Lake reached till and south of Wasilla (and north and south of 10 feet beneath the land surface. the lakes to the east and west); under these hills the till is probably much thicker. Stream-laid gravel deposits border the series of lakes extending northeast of Wasilla. Ground water should be readily available To the southwest these deposits merge to form almost anywhere in the alluvial-plain deposits a conspicuous narrow alluvial plain. Locally, west and north of Wasilla. The quantity of as at Wasilla, the gravel is pitted. Small water available is limited, however, where islands of till protrude through the gravel till is near the surface (that is, where the west of Wasilla; narrow valleys joining the saturated gravel is thin). Where till does alluvial plain are gravel floored. The gravel, not crop out, poor drainage of gravel deposits particularly in the smaller valleys, probably may be a sign of its presence not far beneath is generally thin. the surface. Another alluvial plain extends south- North and west of Pittman even the most westward from the vicinity of Kings Lake; it extensive gravel deposits appear to be thin. merges southwest of Pittman with the alluvial Water should be readily available to these plain west of Wasilla. The thickness of the deposits, however, even in closed depressions, gravel in these deposits is not known. for the surrounding higher ground is under­ lain by relatively impermeable till. House­ Gravel deposits associated with ground hold or farm water supplies are probably moraine north and west of Pittman have not available from gravel in many of the shallow been differentiated (pi. 1). They consist of basins in this part of the area, but larger 25 quantities of water will be more difficult organic pollution, Is high In water from some to obtain. wells (2, 494, 123, 145, 347, and 502). All but ore (494) of the samples of high-nitrate The occurrence of ground water In gravel water are from shallow dug wells, which are beneath a stratum of burled till, near Palmer particularly susceptible to pollution. and near the experiment station, has been described In preceding pages. The writer be­ Water of the chemical character of any lieves that the burled till In these locali­ of these samples would be satisfactory chemi­ ties and the till at or near the surface west cally for use in irrigation. of Wasilla are the same. About 3 miles east of Wasilla and 2 to 3 miles west and north­ Water from gravel beneath the buried till west of the experiment station, four wells northwest of Palmer does not seem to differ (490, 491, 577, and 578) obtain water from significantly from water from wells that ob­ gravel beneath till. In well 490 the water tain their water above the till (compare stands 45 feet above the base of the till, and analyses of water from wells 363 and 386-). 100 feet In well 578. The till layer is probably discontinuous locally, so that water above and beneath it The presence of water-bearing gravel Is in a single ground-water body. beneath till In three areas suggests that the older gravel underlies the till in much of the Water in small perched bodies in wind­ agricultural area west of the Matanuska River blown sand (well 275) is less concentrated in and east of Wasilla. Undoubtedly the till is most constituents than other ground-water interrupted locally. Geologic. evidence samples analyzed. This may be due to the re­ suggests that this same till is at or near the charge of these ground-water bodies from local surface over the greater part of the agricul­ precipitation; the water has not traveled tural area west of Wasilla. At Goose Bay long distances through the sediments, as ground this till lies upon older gravel. The writer water represented by many other samples prob­ believes it likely that the older gravel under­ ably has. Furthermore, the windblown sand lies the till elsewhere in the area wejt of probably is thoroughly leached and contains Wasilla. If the gravel is so extensive, it little soluble material. It is likely also may offer the possibility of the development that the water in the small perched bodies of larger ground-water supplies than are is continually renewed by rainfall, and dis­ probably available from surface gravel west charged by plants and by leakage through the of Wasilla. floors of the small basins, so that the water does not remain long in the sediments. QUALITY OF WATER During the early days of the agricultur­ al colony several wells were drilled that Water containing less than 500 parts per obtained highly mineralized water. There is million (ppm) of dissolved solids is general­ listed in the files of the Alaska Rural Re­ ly satisfactory for domestic use unless it Is habilitation Corp. an abandoned well in exceptionally hard or contains objectionable Palmer that obtained "sulphur water" in bed­ amounts of iron. Hardness in excess of 150 rock at a depth of 121 feet. Salt water was ppm is noticeable in ordinary use and may found In the old slaughterhouse (144} and cause the formation of scales in boilers and hospital (135) wells and in two other wells. heating units. According to reports, analysis of water from the hospital well showed a carbonate hardness Chemical analyses of water samples (table of 4,300 ppm, a chloride content of 3,520 3) suggest that ground water In the agricul­ ppm, and a pH of 6.0. The writer believes tural area generally contains less than 300 that this salt water is old and unrelated to ppm of dissolved solids. The hardness is that in Knik Arm (p. 23). generally 100 to 200 ppm and Is due largely to calcium and magnesium bicarbonate. Water from well 145, which Is seated on bedrock, is Water-Level Fluctuations very hard (490 ppm) and has a relatively high content of dissolved solids (652 ppm). Water Reports of well owners indicate that from wells that penetrate bedrock (wells 70 the fluctuation of ground-water levels between and 123) also has a higher mineral content wet and dry seasons is as much as several than water from most wells in unconsolidated feet. Seasonal fluctuations, the water levels sediments. Well 494, which is in gravel, being lower in winter and early spring, have obtains hard water having a relatively high also been reported in several wells. Since content of dissolved solids (638 ppm). The 1949 the Geological Survey has carried on reason for the high mineral /content of this periodic observation of selected wells. water Is not evident. Chemical softening of this water has' been necessary for satisfactory Climatologic data for 1949 are incom­ household use (Richards, Harold, personal plete but available information indicates communication, 1950). The iron in water from that 1949 was an average year in precipita­ well 462 makes this water objectionable for tion. The year 1950 was abnormally dry; 1951 household consumption or use by livestock. probably approached the average. Because the The iron content of water from several other ground is usually frozen during the melting wells (123, 60, and 347) is sufficient to of most or all of the snow in this area, cause staining of clothing laundered in It. and because wind usually removes much of the snow cover, recharge of ground water seems Most of the water samples listed are to be chiefly from rainfall. chemically suitable for human consumption. The nitrate content, a possible indicator of 26 Table 3. Analyses of ground water from the Matanuska Valley agricultural area, Alaska [Results expressed in parts per million] Well no. and Alaska laboratory no* 70 123 145 38 60 101 187 315 347 363 306 157 215 158 309 2Q252/ 213 155 20242/ 1071 Silica(SiOp) 19 16 16 8.2 9.1 10 7.0 13 14 13 Total iron(Fe) .03 .06 .02 .02 .06 .08 .02 .02 .05 .04 Calcium(Ca) 75 147 51 76 41 46 55 37 20 30 5.6 9.4 4.6 5.9 20 9.4 Sodium (Na) and Potassium(K) 34 33 5.5 9.2 5.5 2.5 11 5.7 Bi carbonate (HC03 ) 194 266 371 143 168 236 102 160 172 159 Sulfate(S04) 120 35 73 36 27 38 40 8.7. 58 9.6 Chloride ( CD 4 42 74 4.5 8 4.8 5.0 1.8 3 1.8 Fluorlde(P) .2 .1 .0 .1 .0 .1 .0 .1 .0 Nitrate(NOs) .. .3 35 96 .8 2.7 9.4 1.6 2.3 46 .2 Dissolved solids - . . 388 652 182 ... 273 155 159 292 155 Hardness as CaCOg -- 244 269 490 150 162 228 122 140 219 131 pH- H 8.0 6.8 6.8 7.7 7.9 7.0 7.3 7.6 7-13-50 8-26-49 8-30-49 8-27-49 6-15-50 10-4-48 8-29-49 8-22-49 Oct!l948 1-31-52 Brazil 462 494 2 502 522 660 75 Spring Spring Dinklei/ 159 153 217 151 216 324 1060 2023 £*!/ 2143/ Silica(Si02 ) . .. 23 26 20 21 15 28 15 16 18 Total ironTFe) 7.2 .02 .02 .03 .02 1.5 .02 .05 .02 /

±/I/ 2-1/2 miles southeast of Wasllla, 2/ Salt LakeLs City~ - - laboratory- - number. 3/ Spring. Table 4.--Data from mechanical analysis of sediments from the Matanuaka Valley agricultural area [Expressed in percent by weight ] Finer Finer Finer Type sediment Sample 4-2 2-1 1-t *-* i-vb than 1/8-1/16 than 1/16-1/256 than Locality nun mm nun. nun mm 1/8 nun 1/16 mm 1/256 mm mm mm Till ...... 2A 3 3 6 10 q 66 3 north of Palmer. Do ...... 21 4 6 12 20 18 38 2 one-half mile southwest of ex­ periment station. Sand (layer in till) S40 2 59 30 6 3 East bluff of Matanuaka River about one -half mile north of Matanuska River highway bridge Do ...... S39 3 q 71 11 6 Do. Fine fraction of S27 29 34 24 7 3 3 of Wasilla. Outwaah sand .... SI 1 2 37 45 5 10 Modern flood plain at Matanuska River highway bridge. Do ...... S5 2 31 57 q 1 Waailla Do ...... S18 2 7 6 TO 31 42 of Four Corner a. Outwaah silt . . . . 2F 1 6 10 80 3 Modern flood plain at Matanuaka '5 River highway bridge. B2 9, 15 31 33 7 7 Dune on Matanuaka bluff about 2 mi lea north of Palmer. 0 6 40 52 2 northwest edge of Palmer. Figure 6 shows graphically the water- been unsuccessful in this area, so that lin­ level fluctuations observed in four wells. ings of pipe or barrels, as well as those of Each well shows a marked decline in water blocks or poured concrete, must be construc­ level during 1950, interpreted as reflecting ted from the bottom up. In gravel wood crib­ lack of recharge from rainfall. Water-level bing must be used and later removed from the recovery Is most complete in the "Woods well} bottom as the permanent lining is constructed. this well is on the extensive terrace south These other types of lining are more permanent of Palmer, where the water table is controlled than, wood and, if properly constructed, more by the Matanuska River. Data for the Linn sanitary. They do not, however, permit much well, and possibly for the Tryck well, suggest inflow of water from water-bearing layers slight recovery during the late spring of higher than the bottom of the well, as may be 1950 bsfore further decline. Each well shows desirable in wells dug in till. partial recovery during the summer of 1951. The LaRose well, which is on a hillside, shows Although dug wells are less satisfactory the greatest decline and the most sluggish than drilled wells for many reasons, their recovery. The Tryck well is on a hillside lesser cost and the fact that they can be northeast of Lake Lucile. The Linn well is constructed by individuals account for the on a hill but the ground-water level Is prob­ large number of dug wells. Dug wells general­ ably controlled by the water table beneath ly can be constructed only a few feet deeper the alluvial flat west of Matanuska; hence than the water table, and some wells dug the water-level fluctuation in the Linn well, during seasons of high water table have gone like that in the Woods well, is relatively dry during drier seasons. Some of these slight. wells have been deepened by additional digging or by drilling. If recovery of ground-water levels con­ tinues during 1952, the declines shown by For sanitary reasons, few of the dug figure 6 probably may be taken as representa­ wells in this area are satisfactory. Many tive of those to be expected after an unusual­ are close to barns or privies} the wood crib­ ly dry year. bing does not prevent entrance of near-surface water into the wells; and few are adequately The rapid decline and slow recovery of sealed at the surface. ground-water levels, as shown by these data, suggest that annual additions to the ground- water body represent only a small proportion Driven Wells of the annual precipitation received in the agricultural area. Conditions suitable for the construction of driven wells seldom persist to sufficient depth in this area to permit wells of this Construction of Wells type to reach water. The Clay Johnson (35) and Hocca (650) wells obtain water from clean Dug wells sand at depths less than 20 feet. The pres­ ence of gravel or till at or near the land Most of the wells constructed by individ­ surface in most of the agricultural area uals in the Matanuska Valley agricultural makes the success of driven wells unlikely. area are hand-dug wells. They are generally less than 50 feet deep but a few are much deeper. The Collier well (272) is 105 feet Drilled Wells deep, and the Owen Moffitt well (305) was 95 feet deep before it was deepened by drilling. Since establishment of the agricultural Commonly dug wells in this area are square or colony most of the wells constructed, Includ­ rectangular, and 3 to 4 feet on a side. ing all but a few of those deeper than 50 feet, have been drilled. For many years the The walls of wells in till stand after Alaska Rural Rehabilitation Corp. operated a excavation, but in most places gravel must cable-tool drilling machine} more recently be supported during the digging. Pea gravel this machine has been purchased and operated is particularly difficult to excavate because by private drillers. Several other private walls in it commonly slump before they can be drillers have been active in this area In supported. Wood cribbing is most commonly recent years. used to line dug wells. Poles or rough boards have been used, extending vertically and fas­ Most of the wells drilled In this tened to cross supports, but they are diffi­ are lined with 4-lnch steel casing; a few are cult to put in place and to support satisfac­ 6-inch wells, relined with 4-lnch casing torily. More generally short notched logs or which extends to greater depth. In all but 2- by 6-inch boards are used; the notched a few of the drilled wells water passes Into sections are added to the bottom of the crib­ the casing through the open lower end and bing, parallel to the bottom of the hole, as through %- to 4-inch slots cut longitudinally digging proceeds. Several types of wood have in the lower few feet of the casing. In only been found satisfactory for well cribbing} one well in this area has a screen been em­ spruce is considered excellent, but cotton- ployed. Open-end and .slotted casing has prov­ wood is said (Wilson, T., personal communica­ ed satisfactory In most wells supplying water tion, 1949) to give the water a taste of for home or farm use, but it Is unlikely that organic decomposition. Other materials that any of these wells has been pumped at a rate have been used for lining dug wells include exceeding 10 gpm, or that many have been concrete blocks, poured concrete, concrete pumped continuously at any rate for periods pipe, and metal oil barrels. Attempts to longer than a few hours. Exceptions are add pipe or barrels at the top and follow wells 37, 57, and 363 (see well records). the digging with the lining have generally 29 Thick sand was penetrated In a few wells. string of tools Is more satisfactory than a Quicksand was penetrated between depths of smaller one. 47 and 185 feet In the Hosier well (213), and difficulty was experienced In drilling. It The availability of electric power on was necessary to drill the well to a depth of farms in this district has resulted in the 187 feet to reach gravel which could be widespread use of electric pumps. Lift is by developed with the open-end casing, but the jet pump in nearly half the wellsj small cen­ water level In the completed well stood at 47 trifugal pumps are used in a few wells. Lift- feet. type hand pumps are still widely employed, although many are equipped with motor and Of the several types of drilling machines pump jack. In only a small proportion of the the cable-tool Is the one best suited to con­ wells is the lift by bucket and windlass. ditions In this area. Cobbles and occasional The electric pumps in use are equipped with boulders are found In both gravel and till. motors in the £- to if-horsepower rangej at The tough silty matrix of till and the coarse the levels for which these pumps are used the cobbles and boulders of all deposits yield pumping rates are generally less than 10 gpm. best to cutting and crushing by a heavy drill Most of the electrically driven pumps are bit. Because of its greater weight, a 6-lnch connected with pressure systems and are in-

40 53

. 42 -5 54 of"

UJ Q

44

46 56 1949 1950 1951 1962 LINN WELL (449)

?48 24

50 -* 25

52

\ 54 27 1949 1950 1991 1952 1949 1950 1951 1952 LA ROSE WELL (314) TRYCK WELL (533)

uj 72 UJUJ '* u-o

a: ofco UJ _ i 73

tn 74 1949 1950 1951 1952 WOODS WELL (185)

Figure 6. Graphs showing water-level fluctuations in wells, stalled in cellars, in pump houses, or in REFERENCES CITED covered pits above the wells. 'Water pipes leading from wells to buildings or passing Am. Soc. Testing Materials, 1950, Procedures between buildings in this region must be for testing soils, p. 76-93, July. buried 7 to 8 feet beneath the surface, below the depth of winter freezing. Black, R. P., 1951, Eolian deposits of Alaska: Arctic, v. 4, p. 89-111. Utilization of Ground Water Capps, S. R., 1940, Geology of the Alaska Railroad region: U. S. Geol. Survey Public Supply Bull. 907. Palmer is the' only community in the Dachnowski-Stokes, A. P., 1941, Peat resources Matanuska Valley 'agricultural area in which of Alaska: U. S. Dept. Agr. Tech. Bull. residents do not generally use individual 769. wells. Initial drilling in the town site during establishment of the agricultural Flint, R. P., 1947, Glacial geology and the colony was/riuint unsuccessful. For many Pleistocene epoch^ 589 p»f New York, John years the Matanuska Valley Farmers Cooperat­ Wiley & Sons, Inc. ing AssociationiSupplied water to its cream­ ery and other establishments by means of a Flint, R. P., and Deevey, E. S.. Jr., 1951, 22,000-foot conduit of wooden-stave pipe from Radiocarbon dating of late-Pleistocene Brazil Springs, about 3 miles northwest of events: Am. Jour. Sci., v. 249, p. 257-300. , Palmer. Excess water was sold to the resi­ dents of Palmer at monthly rates. After the Kellog/Y C. E., and Nygard, I. J., 1951, Ex- / dry season of 1950, the spring flow was pltftatory study of the principal soil insufficient to meet needs. In 1951 the groups of Alaska: U. S. Dept. igr. Mon. 7. existing pipeline was extended, and water is now obtained from Carnegie Creek about a mile Kulp, J. L., 1952, The carbon 14 method of northwest of the springs. age determination: Sci. Monthly, v. 75, p. 259-267. A test well (363) was completed for the city of Palmer in January 1952. Kulp, J. L., Tryon, L. E., Eckelman, W. R., and Snell, W. A., [1952], Lamont natural A community well (515) was dug in Wasilla radiocarbon measurements, II: Science, several years ago, but it is not in use. At v. 116, p. 409-414. present individual wells supply water for all inhabitants. Landes, K. K., 1927, Geology of the Knik- Matanuska district: U. S. Geol. Survey Bull. 792-B. Domestic and Farm Supplies Lutz, H. J., and Griswold, P. S., 1939, In­ During most seasons individual wells fluence of tree roots on soil morphology: are capable of providing sufficient water for Am. Jour. Sci., v. 237, p. 389-400. domestic and farm use throughout most of the agricultural area. Many farmhouses have Martin, G. C., and Katz, F. J., 1912, Geology plumbing and pressure water systems, and the and coal fields of the lower Matanuska water use includes supplying livestock and Valley Alaska: U. S. Geol. Survey Bull. cooling milk. Well water is used for water­ 500. ing gardens on a very small scale. Martin, Paul, 1942, Surface features of the Matanuska Valley, Alaska: Assoc. Am. CONSTRUCTION MATERIALS Geographers Annals, v. 37, p. 127-128. Sand and Gravel Meinzer, 0. E., 1923, The occurrence of ground water in the United States, with a discus­ The wide distribution of deposits of sion of principles: U. S. Geol. Survey sand and gravel in the agricultural area Water-Supply Paper 489. facilitates construction of roads. Pit-run gravel is used for fill in the initial con­ Rivers, V. C., 1950, Report on engineer's struction of all types of roads, and in investigation and general design for finishing and repair of secondary roads. For water system, Palmer, Alaska: Engineer­ construction of paved highways during 1950, ing Study 13, prepared for Alaska Dept. screened gravel was used in part, and crushed Health. gravel was used in the hard-surfacing. Rookie, W. A., 1942, A picture of Matanuska: Sand and gravel are used locally in Geog. Rev., v. 32, p. 353-371. making concrete. A local plant makes most of the concrete blocks used in the agricul­ tural area. Rockie, W. A., 1946, Physical land conditions United States Department of Agriculture, 1941, in the Matanuska Valley, Alaska: U. S. Climate and man: Yearbook of Agriculture. Dept. Agr., Soil Conserv. Service, Phys. Land Survey 41. United States "Weather Bureau, 1939-48, Clima- tological data, Alaska: Annual Summary, Smith, P. S., 1939, Areal geology of Alaska: U. S. Geol. Survey Prof. Paper 192. Washburn, Bradford, 1935, Morainic bandings of Malaspina and other Alaskan glaciers: Stone, K. H., 1930, The Matanuska Valley Geol. Soc. America Bull., v. 46, p. 1879- colony: U. S. Dept. Interior, Bur. Land 1890. Management, 95 p. Wenzel, L. K., 1942, Methods for determining Tuck, Ralph, 1938, The loess of the Matanuska permeability of water-bearing materials: Valley, Alaska: Jour. Geology, v. 46, U. S. Geol. fater-Supply Paper 887. p. 647-653.

UNPUBLISHED REPORTS Bourne, T. B., Associates, Inc., 1952, Report Karlstrom, T. N. V., 1950, Preliminary geolog­ on Palmer city test well, 5 p. fTypewritten ic map of lowlands, Anchorage area, Alaska: report for city of Palmer. ] U. S. Geol. Survey open-file report.

32 RECORDS OP "WELLS

33 Table 5. Records of wells ifc the Matanuska Valley agricultural area, Alaska Qjocation of wells is shown on plate I] Type of well; B, bored with soil auger; D, dug; Dn, driven; Dr, drilled; J, jetted. Method of lift: Power: E, electric motor; G, gasoline motor, H, hand. Depth of well: Depths of wells and of water levels recorded to the nearest foot , Pump: C, centrifugal; J, jet; L, lift; P, pitcher; W, windlass. are reported values; depths to nearest tenth or hundredth of a foot are ,' Use of water: D, domestic; N, not in use; 0, observation well; S, stock (may include measured. cooling milk).

£ ' g i a 3 t> o> I |_ |" 0 Well Location Owner or Driller 3§ S 3 h 11 O 1 Remarks no-. name !*"§ "1 i 0 iH fll "o ^ g o ^C- a * 1tn a o -P o a p a S) P.!0, § i g 1 ID £ 3 01 g B d " £ $ 0> , S 1-foot sand layer in tin yields 100 gpd. 20 Sept. 1950 15 NE& sec. 34 Ray Ferrin do. 1951 Base of . . D 11 36 do. 10.5 Aug. 6, 1951 EJ D Water from sand layer in till. Measuring point hill (MP) top of wood cribbing, 5 ft below land sur­ face. 20 NWt se~, 35 Horace DuFour do. . . Vaney 260 D 6 . . Gravel 0 July 9, 1949 N ), S 21 SEt sec. 35 M. Kopperud George Venne Base of D 9 HL D hUl 22 SEt sec. 35 George Venne do. Hniside * D Tin HL D Soil, 0 to l£ ft; till, l£ to 22 ft; gravel, 22 to 23 ft; bedrock at 23 ft. Water in gravel at . base of tin. T. 17 Nv R. 2 E. 24 MSi sec. 2 M. B. Martin Terrace 176 D 42 Gravel 21.5 July 13, 1949 HW D 25 NEt sec. 2 A. Brager Former owner 1950 do. 183 D 27^0 30 do. 24.46 Aug. 28, 1950 HW D MP top of wood cribbing, 3,10 ft above land surface* 25 Do. L. W. Bizby Ferber Bailey 1951 do. 163 D 22 48 do. 19 Nov. 1951 .. D 27 SEj sec. 2 F. R. Sims T. Moffitt 1949 do. 160 Dr 17 4 do. 8.7 July 13, 1949 EJ D 28 Do. H. Mackie do. 1949 do. 163 Dr 17 4 do. 10 Apr. 1949 HP D 29 Do. Mrs. M. Hosier - 1950 do. 156 D 17 36 do. 14 May 1950 EJ D Greenhouse. 30 ICt seo. 11 James Childers Owner Base of 151 D 7 do. 4 1949 N terrace 31 Do. do. 1948 Terrace 163 Dr 28 4 do. EC D Deepened from 24 ft after going dry in April 1949, 35 SSi seo. 11 day Johnson Owner 1949 do. 143 Dn 18 1* Sand 10 HP D,S Screened drive point. 37 NH% sec, 14 USGS test USGS 1950 do. 124 J 31 3 Gra-vel 2.03 Aug. 8, 1950 N Yielded 44 gpm with 5.6 ft of drawdown after 4| hrs. wen 1 38 SEi sec. 14 dyde King, Jr. Owner , . do. ns D 12 36 do. 10 1949 EJ D 39 Do. Clyde King do. 1950 do. 115 D 18.2 48 do. 17.1 June 13, 1950 EJ D 42 NSj- sec. 23 Wallace Brown do. 1950 do. 107 ),Dn 38 . . do. 30 May 1950 HL D Dug to 28 ft; dry in 1951; probably perched ground water. 43 Do. John Rippy do. 1950 do. 105 D 31 24 do. 27.42 Aug. 18, 1951 HW D Probably perched ground water. 44 SEi sec. 23 C. L. Bastian Ferber Bailey 1951 do. 102 D 38 48 do. 35 Spring 1951 . . . . Do. 45 NWi SEt sec. 23 H. H. Kirk A. and J. D. 1951 do. 100 Dr 68 4 do. 58 1951 . . D Frey T. 17 N., R. 2 E. NWi SEi sec- Z3 H. H. Kixk A. and J. D. 1951 Terrace . . Dr 63 4 Gravel 56 1951 HL D Frey SEi sec. 23 Peter Gallager do. 1951 do. Dr 59 4 do. 53 Sept. 1951 D Driller reports several perched water bodies as much as 3 ft thick; one is 34 ft beneath surface. SBi SBi f seo. 23 W. B, Barahardl do. 1951 do. i 9 Dr 58 4 do. 52 1951 D NBi seo. 26 "The Butte" Ferber Bailey 1951 do. . . D 57 48 do. 54 1951 D Roadhouse. SEi NEi : seo. 26 J. D. Frey A. and J. D. 1951 do. 85 Dr 53 4 do. 43 1951 D Frey

NEi SEi seo « 26 USGS test USGS 1950 do. 83 J 52.0 4 do. 37.0 Aug. 1950 » 0 well 2

T. 16 N., R. 2 E. NJDj f, sea. - 2 Thomas Lepak do. 61 D 22.0 30 do. 17.20 June 15, 1950 H D MP top of wood cribbing, 2.20 ft above land sur­ face. Do. Russell Dow Owner . . do. 55 D. 17.0 42 do. 11.64 July 26, 1950 H D MP top of wood cribbing, 1.5 ft above land surface. T. 17 N.. R. 2 E. SWi se°« 14 Victor Falk, Alaska Rural 1936 Hillside 152 Dr 110 6 Bedrock 62(7) June 1936 HL ),S Jr. Rehabilita­ tion Corp. (ARRC) Mti seo. 23 Charles A. and J. D. 1951 Terrace , . Dr 47 4 Gravel 38 1951 D Weidner Frey SWi SEi : seo. 15 Victor Falk ARRC 1936 do. 156 Dr 80 4 do. 30 May 1936 EJ ), S Has watered as many as 50 cows. 35 June 1948 seo. 22 Darrell Frank do. t do. 136 Dr 44 4 \ c SW; : seo. 22 Arthur Holbrook do. . do. 129 D 32 42 do. 29 1947 EJ >, S Estimated use, about 400 gpd. swi sw ; sec. 22 Lee MoKinley do. . do. 108 Dr 52 4 do. 38.2 Aug. 18, 1949 D NWi UW; : seo. 27 E. Wineok do. . do. 102 Dr 55 4 do. n Well formerly used for construction camp. SEi NWi seo. 27 Mrs. G. do. . do. 97 Dr 56 4 N Dreghorn I o. do. On terrace 79 D 40 do. 35 T) scarp SWi TWi seo. 27 do. ARRC m 0 Terrace 58 Dr 24 4 do. NEi SW seo. 27 do. do. . , do. 54 Dr 18.5 4 do. 11.58 Aug» 27, 1949 N MP well house floor, 1 ft above land surface. Do. Joseph Lentz do. , . do. 54 Dr 26 4 do. 12 \ Q NWir NWi : seo. 34 Harry Eendrick do. . . do. 54 Dr 22 4 do. 21 HTi SW-^ NW- seo. 34 V. K. Moonier T. Moffitt 1949 do. 51 Dr 22 4 do. 20 1949 EJ D NWi: SW; : seo. 34 W. B. Barnhardt ARRC . . do. 51 Dr 22 4 do. 16 HT, NWir Nffi ; sec. 34 Olaf Nelson do. , . do. 64 Dr 22 4 20 -\ o NEi; ME; ; sec. 34 William Casler do. , . do. 64 Dr 30 4 do. 21 SBi; SE - sec. 27 Donald Parks do. . . do. 67 Dr 33 4 do. 24 Formerly watered 1,000 sheep. SW| SE r sec. 27 Alfred Church do. . . do. 75 Dr 35 4 do. BIT n

' § IDiameterofwell H HO p |L Dateofmeasure­ Yearcompleted 3 (inches) Water-*earing H Well Location Owner or Driller Topographio 1 -ti "S ment Remarks H go material name situation U>U>UJ>U> ui io U) ui tN) tN) |N) tN) H H PS jSfj O 00 lO004kU> CNJI-1 Ol Ul H O tO 00 -J Ul 4k IN) Ul IN) H

*pI *£i IHHIfF'IS *pi ff-i en a a 3 3 g 1=3 S S 2 ? S £p £|H *piip *p tl *p a a a a ae> a a a a mo a o (t (t a gg oo oo o o o o o °? 8S gg £888 8

g t1 o £ P 2 vi B 4 ° ' J ll P g !-* w gs.cr

HHH H HH»H »H H H H H *** u> Year oompleted ggg * ifeis' S ' S O> O* CO O> 5 Ul M b-« w M M m . W 5 PL P ft P P p. P , ip>pi pi p. PI 4 H O 4 PI .H3»SHei e'ei gg'»j g'|_ PI P-pi H f> 0 p 0 O P 5 S « 0 P P ffl I Topographic * * § § f " S ' § situation

Altitude above ft t ft' * ' CN) O Ul O Ul Ul -J Ul O sea level (feet) BB B Type of well

Depth of water"" ~ IS S§ fcgtfis 8 below land surface (feet) Diameter of well fe IN) 4» £ (inches)

pjAjQ) AJ QiMOtOt && fij Ol fij QiD> & & 0*B> * * OjQjQjQj D) (0 M ( Water-bearing Ho5

ro ro 4> o»aoio *ui H* 4k* 4k«»* -J 4k w R o> a> un ui tO ?0 S t2 Water level (feet below land surface) o « >=g ^ ^ y * o « o- H H H * Date of measure­ ui ui ro K ment HH H * i? 10 §§ 1 II . PK Method of lift

0 G Use of water 03 03 CO CO CO

*? H s H-HP- fc o-» pi H' H-JS O CTi H- w r" ^* ^1*^ L ~ ocf SJ» n H^*o - Q i cj lift P B S 4k V u> ri- 5> T. 18 N., R. 2 E. 344 SWt seo. 28 Carl Stoekle F. Bailey 1950 Terrace 442 D 27 42 Gravel 25 1950 N

345 Do. Mrs. G. do. 1950 do. D 14 42 do. 12 1950 N Edmunds 346 Do- Richard Demming S. Kbzloski 1952 do. 435 Dr 122 6 D 347 Do. C. E. Albreoht do. D 35 48 D 348 Do. A. Trowbridge A. and J. D. 1951 do. Dr 97 Water from gravel beneath till(?). Trey 350 NEi SEi seo. 29 D. L. Xrwin 1949 do. D 30.0 36 Gravel 21.10 July 8, 1949 EJ D 352 SEi SEi seo. 29 Howard Estelle ARRC 1935 do. 422 Dr 62 4 do. 48 Dec. 28, 1935 EL D,S 353 Do. do. do. 1935 do. . . Dr 119 4 do. 91 Nov. 26, 1935 . N Till, 43 to 115 ft. 355 NEi swi 3eo« 29 Clarence do. 1936 HiU-top 438 Dr 63 4 43 Till, 43 to 61 ft. Hoffman 356 SEir SWi^ seo. 29 Leonard Moffitt do. 1936 Hillside 415 Dr 80 4 Gravel 53 July 27, 1936 EJ D,S Till, 50 to 70 ft. 359 SB;: SWi: sec. 30 L. Z. Scott do. . . Hilltop 431 Dr 120 4 Till 63 Sept. 1947 EJ D Till below 55 ft; originally dug to 55 ft. 360 NWi NEi- seo. 31 L. Wiederkehr Owner 194S Hillside 374 D 18 48 Gravel 11.9 July 7, 1949 EC D 362 HBir NEir seo. 31 A. Thompson ARRC 1947 Level 374 Dr 110 4 50 D o Till, 28 to 84 ft; inadequate supply of water in surface gravel, on till, at 28 ft. 363 SEi NEi 3eo» 31 Palmer Test J. Qirrie, 1951 Hilltop 375 Dr 165 6 Gravel 20 Dec. 1951 . . . Till, 90 to 114 ftj well finished with 20-foot Well E, Young, screen; drawdown, 35 ft after 16 hrs pumping at S. Eozloskl 118 gpm. 364 SWi NWi seo. 32 Mrs. Irene ARRC 1936 do. Dr 88 4 do. 26 May 23, 1936 HL S Till, 65 to 86 ft. Season 365 SEi NWi seo. 32 Oscar Beylund do. 1936 do. 365 Dr 79 4 do. HT, N Till, 58 to 76 ft. 366 NfiJ NWi 3eo. 32 Frank do. do. -. . Dr 67 4 do. 30 1942 EJ D, S Estimated use 800 gpd. MoAUister 369 NWi NBi seo* 32 Robert ELem do. 1936 Terrace 321 Dr 72 4 Sand 50 Nov. 26, 1936 . . N No till reported; -water reported struck at 70 ft. 370 Do. do. do. do. 321 D 15 42 14 Till at 15 ft; estimated uae 300 gpd. 371 NEi NBi seo> 32 Nell Miller do. do. 309 D 24 22 In dug well, nearly 16 ft deep, till was tapped at 15 ft; 2 ft of water in overlying gravel. 372 SEi NEi sec. 32 Virgil Eokert do. » do. 309 D 18 16 Water in gravel overlying till. 375 NEi SEi sec. 32 Eino Wirtanen do. . . D 25 15 Till reported, 8 to 25 ft; water in gravel layer in till, 16 to 21 ft.

GJ T. 17 N., R. 2 E. «0 376 NJ&i seo. o John Mehan ARRC Hillside- 239 Dr 34 4 Bedrock 15 Spring 1949 EJ D Casing set in rook 10 ft below ground surface; well yields 30 to 40 gal between periods of recovery. 377 Do. Leo Lucas do. 265 D 28 24 D O 378 NEi NWi sec. 8 H. H. Blunok ARRC Terrace 247 Dr 126 4 do. 119 Summer 1948 EL D,S Till, 41 to 76 ft; well yields 20 to 30 gal between periods of recovery. 380 NEi NWi 3eo « 5 D. L. Irwin do. Hilltop 266 Dr 74 do. 66 381 NWi NWi seo. 5 M, Perkins do. f . do. 277 Dr 83 4 Sand 55 Aug. 1951 EJ D 382 NEi NWi seo » 5 G. E. 'Murphy J. Cebula 1949 Hillside 259 D 44 48 Gravel 42 Spring 1949 EC D Till below 42 ft. (?) 383 NWi NWi seo. 5 Mrs. P. Harder A, Moffitt 1948 Hilltop 266 Dr 90 4 Gravel 60 1948 . . 385 NWi NEi seo. 6 Vernon France ARRC . , do. . . Dr 144 4 do. Till, 60 to 130 ft. 386 NEi NWi «>o. 6 Mrs. G. France 1947 Depres 284 E 30 48 do. 27 D sion 388 NWi N»i B«°» 6 Carl Mielke Hillside 314 Dr 63 4 do. 55 D T. 18 N,, R* 2 E. 395 SWt Slit seo. 32 A. W. Pear 3 on ARRC 1947 HiTLtop 272 D,Dr 100 T«1 55 F,T Well ends in till; top of till probably near 80 ftj originally dug to 52 ft. 396 SEi SWi seo. 32 Henry Harris on 1914 Hillside D 14 36 13 HT, 397 SMx SWi sec. 32 Victor Johnson ARRC 1936 Hintop 279 D,Dr 82 KT D o Till, 55 to 64 ft; originally dug to 44 ft. 398 NWi SWf seo. 32 P. J. Heamer do. 1936 Level 286 Dr 61 4 Sand 6 1949 EJ D,S Till, 28 to 59 ft; yielded 5.65 gpm, drawdown 7 ft, surface in 23-hour test. 399 SWi seo « 31 Glenn Harris on Owner 1949 Hillside D 28 48 Gravel 27.10 July 5, 1949 HW D Ground surface built about 3 ft higher than origi­ (?) nal surface; till below about 27 ft. T. 18 N.. R. 1 E. 401 SEi SE; seo. 36 William Roark ARRC Hillside * Dr 102 82 T!J D,S 405 SEi SE : seo. 35 George Black Hilltop 383 D,Dr 86 4 Gravel 61 Feb. 1947 EL D Originally dug to 60 ft. 407 SEi SWi: seo. 35 0. A. Berg Hillside , , D 22 Sand HT, Cased after barrel was placed in bottom of well. 410 Nfi r seo, 35 Eugene Reid Owner . . Hilltop 399 D 58.0 42 Gravel 56.0 July 15, 1949 EJ D 411 SWi SE ' sec* 27 Mrs. A, ARRC 1936 do. 366 D 23 36 do. 19.0 July 15, 1949 EJ D,S Havemister to -j ex ui S ro H o ui

H ggS g 3 &a&<&& & # 1. J. co * co co s g

'ro H Hi HHHHHH H H HH U i ro ui o>

i" ScpS S M s!" S. c 2" EM **

g-g-g-g-g1 &

Year completed

Hfp p VS^SVSi J l_l H j « P P OH P-HP- '§ § § Topographic fc* C8 O cf O &&§1 81 35f!:ap.SfSff&& f S t pi p. » o » « » 5 o a g a a H- B H- o situation ' $2 HL-, PItf ° * PJtf rt " P O » 5 & !»

A AOCXCDCD CX ~J Altitude above 3 : :: S H HWWOO CD 05 sea level (feet) O O U O !? Type of well Depth of §P o SH *. below land sur­ face (feet) Diameter of well A 0»« Jk Jk Ji. A A A fe I (inches) Water-bearing 'g-g1 f"! * g ^ & § § ^ P* pt p OOO p O <| material

48' 8 Water level (feet below land i! 8 surface)

* Date of measure­ ment tpH toH *

ill,ft;tocasi42126 H 4 O O Experimentricultural grsandwaterfromand (£ H> 8 *' ° ' O O O O O O 12 I f &-SIO » s.t** p ^

IP3S p.S on. g-8- a?; layertill.in lledtoback68 H ro < B. w . T. 18 N.. R. 1 W. 468 V

481 NWi seo. 35 do. do. 1950 T) 55 Gravel 53 1990 N T. 17 N., R. 1 E. 490 SEt SBt seo. 6 Anthony ARRC Hilltop * Dr 106.06 4 Gravel 36 3.946 HL D,S Till, 30 to 81 ft. Viokaryous

491 SWi SEi seo. 6 do. do. Gentle Dr 109.09 4 HT, N Till, 42 to 107 ft. slope 4P? MEi NWfc seo. 7 42 6 . 38 June 1, 1951 EJ D A 78-ft well on this property penetrated till(?), 42 to 55 ft with sand and gravel beneath. 494 St& SW| seo. 6 do. nr 40 Children's Home, about 50 people; hard water (see Richards analysis). 495 do. T) 16.0 4 15.5 Aug. 27, 1951 N bottom

T. 17 H., R, 1 W. Tpp o 500 SWt SW£ seo. 1 Fred Hard Owner * Level 329 D 14 12 D surface near lake

501 D. W. Roth do. 14 do. «* EC D Greenaores Roadhouse. fnl v TO 1 QAQ 502 SEf NWj seo. 11 Martin Olson Owner Shore of D 22 islo* Till below 12 ft; well may obtain water from lake gravel above till.

505 SW£ seo. 2 E. L. Peek do. Hillside D 18 42 15 above lake

506 Do. do. do. do. D 32 42 do. Till below 28 ft. no MEi NWfc seo. 10 Mrs. W. 16.8 do. 11.77 Aug. 31, 1949 HL D MP, well ourb, 2.0 ft above land surface. Wilson 511 MWi NWi seo. 10 Tin 30 do. 28 51? 348 26 do. 22.0 July 25, 1949 EC D Lined with concrete pipe. 513 do. 34R 23 do. 19 Till, 20 to 21 ft. 514 KW: : HWfr seo. 10 Wasilla Cafe T. Moffitt do. 348 Dr 28 4 do. Cafe. 515 KWirHWf seo. 10 Was ilia Jack Fabian Hilltop 350 D 32 42 do. N Community Well 516 HEl NWi seo. 10 E. Gustafson Owner Hillside 345 D 17 36 do. 15 1945 . D 517 SEt NWt seo. 10 do. do. do. 345 D 21 36 do* 18 1946 EC D 51fl Do. do. 345 N Office 519 Do. Wasilla Hotel Jaok Fabian do. 338 D 22 . . Gravel 18 1949 EJ D Hotel and restaurant; till, 3-foot layer with top at 15(?) ft. 520 Do. do. 342 16.6 . 0 do. 13.7 July 25, 1949 HW D Till layer reported. 521 SWi NWfc seo. 10 Te eland' s Jaok Fabian * * do. 338 D 16 36 do. 13 April 1949 E D Till below 15 ft. Shopping Center 522 Sl£ NWi seo. 10 U. S. Dept. Level 333 21 do. "FT Well lined with concrete pipe; on low divide be­ Interior, surface tween Wasilla Lake and Lake Lucile. Alaska Rail­ road Table 5. Records of wells i the Matanuska Valley agrioultural area Continued

8 nJ Water-level (feet)levelsea wellDiameterof (feetlandbelow surface) Dateofmeasure­ ; to S Altitudeabove H (inches) Water-bearing H Driller completedYear H ; rd 43 Remarks WeU Location Owner or Topographic material

6?4 SWi ^i seo. 30 J. Mazy do. D 11 HT, 625 do D 11 do. 7 HT,

T. 16 N.. R. 2 W. 630 t Level 42 25 HW D,S Water may be derived from till; till is exposed in surface nearby bluff. 6T1 NWi S^ seo. 2 do. Hillside 48 T. 17 N., R. 2 W. 635 NWi seo. 25 A. H. Tober Owner Hillside D 13 42 Till 10.64 Aug. 18, 1950 HW D Water apparently derived from sand layer in till; MP is well cover, 1.8 ft above ground surface.

T. 16 N., R. 2 W. 640 SWi seo. 8 Stanley Owner Hilltop D 30 42 Till 28 1950 Small inflow of water reported at 20 and 26 ft; Coll ins water at 30 ft may be from gravel layer. 641 SWi seo. 18 Raymond do. 1950 Terrace D 28 36 Gravel 25.95 Aug. 18, 1950 HL D Till at 15 ft in hill nearby. Redington

T. 16 N., B. 3 W. 642 SWi seo. 24 Chester Burden Owner Terrace 88 D 30 Gravel 28 1950 HL D 64T NBi seo. 26 do. 28.0 Aug. 26, 1950 HT, ground surface. 650 14 do. FT, 651 do. D 17 36 15 HT, 65? Nwf seo. 28 do. do. D do. 14 HT, T) T. 17 N., R. 2 W. 660 1945 28£ Dr TM 1 1 If \ FT, Interior, slope Alaska toil- road (pittman)