DEPOSITIONAL ENVIRONMENTS, SEDIMENTOLOGY, AND STRATIGRAPHY OF THE DOCKUM GROUP (TRIASSIC) IN THE TEXAS PANHANDLE
by BRENT A. MAY, B.S. A THESIS
IN GEOSCIENCE
Submitted to the Graduate Faculty of Texas Tech University in Partial Fulfillment of the Requirements for the Degree of
MASTER OF SCIENCE
Approved
Accepted
August, 1988 ACKNOWLEDGMENTS
I would like to extend my appreciation to Dr. Thomas Lehman for suggesting this project and chairing the thesis committee. Dr. Lehman's advice in the field and during the writing of this study was invaluable. Financial assistance could not have obtained without his help. Dr. Sankar Chatterjee and Dr. Alonzo Jacka offered enlightening discussions and helpful editorial comments for this project.
Invaluable field assistance was given by Dick Record and Kerry "Crockett" Howard. Special thanks goes to Susan Ensenat for her helpful questions and observations in the field. Dr. Necip Guven and Darrell Brownlow gave generous assistance for the interpretation of X-ray diffraction patterns. Thanks goes to my parents for providing food, shelter, and moral support during research in the Palo Duro Canyon area. I would like to thank Edwin and Claudine May for shelter and excellent food while conducting field work in the Tule Canyon area. Appreciation is extended to Dave May for the use of a field vehicle along with computer hardware, software, and advice.
The Texas Parks and Wildlife Department was gracious enough to allow this study to be conducted in the Palo Duro Canyon and Caprock Canyons State Parks. Funding for this project was granted by the American Association of Petroleum Geologists and the Geological Society of America.
This project is dedicated to the memory of my grandfather, David Ross Sherman.
ii CONTENTS
ACKNOWLEDGMENTS ii
ABSTRACT vi
LIST OF FIGURES vii I. INTRODUCTION 1
Objectives and Study Methods 1
Study Area, General Geology, and Stratigraphy 2 Previous Works 10
II. STRATIGRAPHY 13
Introduction 13 Biostratigraphy 14
Tecovas Formation 23
Aeolian unit 24
Channel sandstones 27
Paleosols 29 Multi-colored mudstones 33 Upper mudstone 36
Trujillo Formation 38
Stratigraphic Problems 44 Lithostratigraphic correlation 47
Naming inconsistencies 48 Informal subdivisions 52
m III. SEDIMENTARY FACIES 53 Introduction 53
Aeolian Faci es 54 Dune sub-facies 56
Interdune sub-facies 59
Fluvial Channel Facies 60 Thalweg sub-faci es 61
Channel-fill sub-facies 61
Point bar sub-facies 66
Abandonment sub-facies 78
Small-scale channel deposits 80
Overbank Faci es 85
Levee sub-facies 85 Crevasse splay sub-facies 86
Sheet f 1 ow sub-f aci es 89
Floodplain sub-facies 92
Valley-Fill Facies 93
IV. SANDSTONE PETROGRAPHY 101 Methods 101
Sandstone Mineralogy 101
Sandstone Classification 107 Cements 119
IV V. PALEOCURRENT ANALYSIS 123 Methods 123
Previous Analyses 126
Results 129
VI. DEPOSITIONAL ENVIRONMENTS 138
Introduction 138
Current model 138
An Alternate Depositional Model 140
Stream patterns 140
Depositional history 141
Overbank environments 147
Paleogeomorphology 150
Paleoclimate 152
Problematic Deposits 153
Lacustrine interpretation 154
Fluvial overbank interpretation 160
Source of volcanic ash 162
Fluvial and Deltaic-Lacustrine Comparisons 163
Point bars and sheetflows versus deltas 163
Fluvial overbank versus lacustrine 164
Paleocurrent data versus sandbody geometry 167
Suggestions for Further Research 168
VII. CONCLUSIONS 170
BIBLIOGRAPHY 172 ABSTRACT
The Triassic Dockum Group consists of nonmarine "red beds" exposed around the Southern High Plains of Texas and New Mexico. The Dockum Group is subdivided into two formations: the lower Tecovas Formation and the upper Trujillo Formation. Both formations are comprised of conglomerate, sandstone, siltstone, and mudstone. Sandstones from the two formations are readily distinguished petrographically. Aeolian, fluvial channel, fluvial overbank, and valley-fill facies are present in the Dockum Group. Dune and interdune sub-facies are recognized in the aeolian facies. The fluvial channel facies is divided into thalweg, channel-fill, point bar, and abandonment sub-facies. Levee, crevasse splay, sheetflow, and floodplain sub-facies are identified in the fluvial overbank facies. Valley-fill facies are exposed at three localities in the study area. Extrabasinal conglomerates are associated with the valley-fill deposits. Paleocurrent data yield a grand vector mean of N 19 W for Dockum sandstones. This value compares well with previous paleocurrent analyses. The prevailing depositional model for the Dockum indicates the presence of deltaic, lacustrine, and fluvial environments. A "Dockum lake" is postulated to have formed in the relict Midland Basin. This study proposes a wholly fluvial setting for the Dockum Group. Deposits of low and high sinuosity streams dominate Dockum deposystems. Lacustrine deposits reflect only small local lakes or ponds. No deltaic deposits are present.
VI ^^^^aP^fflSSSH»W^^iM^D^BBB^IB^H®i«!»Sl!e«™iL3K!a'aaiE5^^
FIGURES
1.1 Outcrop areas of the Dockum Group and location of the detailed outcrop studies in this report 3 1.2 Location of detailed outcrop studies in the Palo Duro Canyon area 6 1.3 Location of detailed outcrop studies in the Caprock Canyons area 8 2.1 Correlation of Triassic continental deposits in the United States 15 2.2 Divisions of the Alpine and Germanic sequences and phytosaur zonations in Germany 18 2.3 Phytosaur zonation and correlation of North American Triassic strata 21
2.4 Correlation of measured sections of the Dockum Group .. In Pocket
2.5 The unconformable Permo-Triassic contact 25
2.6 Aeolian sandstone at the base of the Tecovas Formation .... 26
2.7 Typical sandstone from the Tecovas Formation 28
2.8 Typical paleosol unit occurring in a basal aeolian sandstone unit 30 2.9 Manganese nodules occurring in the paleosol unit 31 2.10 Silcrete horizon occurring in a paleosol unit 32 2.11 Capitol Peak in Palo Duro Canyon 34 2.12 Typical red and yellow mudstones in the Tecovas
Formation 35
2.13 The upper sandstone of the Tecovas Formation 37
2.14 Interbedded sandstones and mudstones of the Trujillo Formation 39 vii 2.15 Generalized cross-section of the Dockum Group along the eastern "Caprock" escarpment showing major sandstone sequences; contacts between the Permian, Tecovas, Trujillo, and Tertiary; and underlying Paleozoi c structural features 42
2.16 Comparison of the stratigraphic nomenclature applied to the Dockum Group in Texas 45 2.17 Correlation of the lower Dockum Group with Triassic deposits in New Mexico and the Colorado Plateau 50 3.1 Example of the dune and interdune sub-facies in aeolian deposits of the Dockum Group 57 3.2 Trough cross-beds with re-activation surface 63 3.3 Example of the channel-fill sub-facies, thalweg sub-facies, and floodplain sub-facies 64 3.4 Example of stacked channel-fill sub-facies 67 3.5 Lateral accretion surfaces in point bar sub-facies 69 3.6 Example of point bar sub-facies 71 3.7 Outcrop drawing of a cross-sectional view of a channel
deposit exhibiting point bar sub-facies 73
3.8 Example of a channel deposit with point bar sub-facies ... 75
3.9 Clay plug filling a stream channel 79
3.10 Example of abandonment sub-facies and a capping channel- fill sub-facies 81 3.11 Example of a small-scale channel deposit 83 3.12 Example of levee sub-facies within floodplain sub-facies 87 3.13 Example of crevasse splay sub-facies and floodplain sub-facies 90 3.14 Outcrop drawing of a composite valley-fill sequence 96 3.15 Slumped sandstone boulders at the base of a valley- fill sequence that has eroded into channel abandonment sub-facies 98 viii 3.16 Multiple stream channel facies within a valley-fill sequence 99 4.1 Photomicrograph of polycrystalline quartz grain 102 4.2 Photomicrograph of a typical orthoclase grain which is slightly altered and shows no twinning 104 4.3 Foliated quartz-mica schist rock fragment and muscovite grain 105 4.4 Photomicrograph of a squashed claystone fragment 106 4.5 Photomicrograph of a rounded zircon grain 108 4.6 Photomicrograph of a vertebrate bone fragment in calclithic conglomerate 109 4.7 Mineralogical composition of Dockum sandstones showing compositional groups A, B, and C discussed in text Ill 4.8 Photomicrograph of a typical Tecovas sandstone sample .... 114 4.9 Photomicrograph of a typical Trujillo sandstone sample ... 116 4.10 Photomicrograph of a calclithic conglomerate 118 4.11 Photomicrograph of clay coating surrounding a monocrystalline quartz grain 120 4.12 Photomicrograph of pore-filling kaolinite-dickite booklets 122
5.1 Trough cross-bedding in plan view 124
5.2 Trough cross-bedding in cross-section oblique to paleo-flow 125 5.3 Rose diagrams displaying paleocurrent directions from sandstones in the Trujillo Formation 127 5.4 Paleocurrent data from Boone's "meanderbelt system" in Tule Canyon 130 5.5 Rose diagrams generated from paleocurrent data collected from Trujillo Formation sandstones 133 5.6 Rose diagrams constructed for the Trujillo Formation, Tecovas Formation, and the Dockum Group total 135 ix Sv^ESs^^ssaH
6.1 Generalized cross-section from the sediment source area to the Dockum Basin 142
6.2 Photomicrograph of Tecovas silcrete 145
6.3 X-ray diffraction pattern of a red overbank mudstone sample from the Trujillo Formation 148 6.4 X-ray diffraction patterns of yellow smectite-rich mudstone from a problematic deposit in the Tecovas Formation .... 155 6.5 Possible "lungfish burrows" within smectite-rich mudstones of the Tecovas Formation 158 6.6 Gray mudstone and a rare concretionary carbonate in a pond deposit 161 6.7 Fine-grained epsilon cross-beds 165 CHAPTER I
INTRODUCTION ^
The Triassic Dockum Group is composed of nonmarine "red beds" exposed around the Southern High Plains of Texas and New Mexico. The Dockum crops out beneath the "Caprock" of the High Plains in many places. The abundance and quality of exposures allow an excellent opportunity to study the origin of these deposits. The sedimentology of the Dockum has not been studied in detail until recently, probably because of the lack of economic deposits within the Dockum. The Dockum is known for its vertebrate fossils and hence, early studies tended to concentrate on paleontology. The Dockum Group is comprised of two formations, the lower Tecovas Formation and the upper Trujillo Formation.
Objectives and Study Methods The main objective of this study is to develop a paleoenvironmental model for the Dockum Group in the Panhandle of Texas. Secondary objectives are to:
(1) Delineate the various lithofacies present. (2) Describe the sedimentology of the different lithofacies. (3) Identify and correlate the formations present. (4) Determine the paleocurrent trends and sediment source areas.
(5) Describe the sandstone petrography of the formations. This report is the second in a series of studies at Texas Tech -2 University to document the sedimentology of the Dockum strata around the Texas High Plains (first study - Frelier, 1987).
Suitable outcrops were located to study the lateral relationships of the lithofacies present in the Dockum. Detailed stratigraphic sections were measured using brunton compass and Jacob staff. The sections were used to correlate the two formations and to interpret the sedimentology. Outcrop drawings were produced and used to establish the vertical and lateral relationships between the various lithofacies present. Samples were collected from the different sandstone bodies for examination and petrographic analysis. Samples were analyzed to determine any petrographic differences between the formations. Mineral composition, texture, cements, and diagenetic affects were described in each thin section and source terrains were inferred from the mineral compositions. Paleocurrent data was collected from sedimentary structures in the major sandstone bodies to determine the depositional paleoslope of the Dockum. X-ray diffractometry was used to determine the types of clay minerals present in the mudstones of the Dockum.
Study Area, General Geology, and Stratigraphy The study area comprises an elongate belt oriented north-south in the Texas Panhandle (Fig. 1.1). Triassic sediments crop out along the eastern edge of the High Plains. Streams have eroded into the edge of the High Plains creating steep cliffs and narrow canyons with a relief of several hundred feet. Quality of the exposures generally depends on the steepness of the slopes. Outcrops in the northern part of the study area generally are better than those of the south.
^ 1
1
F1g. 1.1. outcrop area, of the Ooc.u. Group and location^of^the ^^^^^_ detailed outcrop studies in this repori:. ^M. LC ;; '•-T-E..ras;:i.i».s.>iiaiafc:^a;S;»S5^^ BiffiliHirtirf-ir"--rni iinir
OKLAHOMA TEXAS
^P Triassic Outcrop
0 50 miles LOCALITIES PD (2-9) Palo Dure Canyon #2--9 FC-3 Floyd County #3 CTC Ceta Canyon DC-1 Dickens County #1 CC-1 Wayside Crossing #1 DC-2 Dickens County #2 CC-2 Wayside Crossing #2 DC-3 Dickens County #3 TC Tule Canyon DC-4 Dickens County #4 SRC-1 Silverton Roadcut SF-1 Silver Falls CRC (1-2) Caprock Canyons #1-2 CBC-1 Crosby County #1 FC-1 Floyd County #1 CBC-2 Crosby County #2 FC-2 Floyd County #2 CBC-3 Crosby County #3 5 Detailed studies were conducted in Randall, Armstrong, Briscoe, Floyd, Motley, Dickens, and Crosby Counties. Reconnaissance of Triassic outcrops was performed in Garza, Potter, and Oldham Counties. Specific localities were chosen for detailed outcrop studies (Figs. 1.1, 1.2, 1.3) and when possible, outcrops were surveyed between these localities.
Permian, Triassic, Tertiary, and Quaternary sediments are exposed in the study area. Permian sediments are represented by the Quartermaster Formation. The Quartermaster consists of red and orange siltstone and mudstone with interbedded gypsum veins. These beds have been described as shallow marine shoreline deposits (Hood, 1977).
The Triassic Dockum Group overlies the Quartermaster and is composed of continental "red beds" ranging in thickness up to 145 meters. The sediments in the Dockum are conglomerate, sandstone, siltstone, and mudstone. The mudstones generally are red in color while the conglomerates and sandstones are a green gray color. The mudstones and siltstones weather to produce "badlands" while the sandstones are prominent ledge formers. The Dockum sediments are fluvial in origin with coarse-grained sediments representing channel deposits and fine-grained sediments overbank materials.
The Dockum is overlain by the Miocene Ogallala Formation in most parts of the study area. It is composed of calcareous sandstone, conglomerate, and claystone, with a well developed calcrete in the upper part (Hood, 1977). The calcrete is thick and dense and forms a distinct ledge resistant to weathering, known locally as the "Caprock." Fig. 1.2. Location of detailed outcrop studies in the Palo Duro Canyon area. Edo«of escarpment > 2
State Parit miles boundary Fig. 1.3. Location of detailed outcrop studies in the Caprock Canyons area. Motley County
^ 0 12 3 4
Edge of escarpment miles 10 Local Quaternary formations overlie the Dockum in the Tule Canyon and Palo Duro Canyon areas. The Tule Formation (Tule Canyon) is composed of gray siltstones and represents lacustrine deposits. The Cita Formation (Palo Duro Canyon) is also considered to be lacustrine in origin and is composed of silts, clays, and sandstones.
Structural features present during Triassic time influenced Dockum sedimentation (McGowen et al., 1983). Dockum sediments were deposited in the relict Late Paleozoic Palo Duro and Midland basins. A rift that later formed the Gulf of Mexico had opened during Triassic time. The doming associated with that event is attributed to the sedimentation of the Dockum. The Dockum Basin (relict Palo Duro and Midland basins) may be interpreted as an extra-arch basin flanking the rift. The uplifted Ouachita Orogenic Belt supplied sediment for the Dockum streams (McGowen et al., 1983). The Matador Arch separates the Palo Duro and Midland basins and remained a positive feature during Triassic time, but was not a major sediment source.
Previous Works The Dockum Group was named after Dockum Creek in Dickens County, Texas by Cummins (1890). He reported these deposits as Triassic in age. Drake (1891) described the Dockum and subdivided it into three units. Adams (1929) equated the Santa Rosa and Chinle Formation of New Mexico to two of Drake's (1891) units. Gould (1907) upgraded the
Dockum to a group and named two formations, the upper Trujillo and the lower Tecovas. Gould's (1907) nomenclature will be used in this study. The above workers described the Dockum as nonmarine "red beds." 11 Green (1954) was one of the first to attempt to determine paleoenvironments for the Dockum. He postulated aeolian, fluvial, and lacustrine environments. Kiatta (1960) and Cazeau (1960) reiterated Green's (1954) findings and attempted to determine source areas for the Dockum. Both concluded that the Llano region of Texas was the likely source area for the Dockum. Cramer (1973) postulated the Ouachita Uplift as the source area and reinforced the concept of the fluvial nature of the sediments. Hood (1977) presented a good overview of the Permian, Triassic, Tertiary, and Quaternary deposits in Palo Duro Canyon.
Paleontological studies have provided the impetus for sedimentological studies in the Dockum. E.D. Cope and E.C. Case conducted paleontological studies in the Dockum during the early part of the century. Green (1954) studied the vertebrate fossils along with the sedimentology. Murray (1982) conducted a comprehensive study on the biostratigraphy of the Dockum. Recently, Chatterjee (1985, 1986) has conducted several studies on the vertebrate fossils found in the Dockum.
McGowen et al. (1979, 1983) completed the first detailed report on the facies and depositional patterns of the Dockum. These works were compiled from several unpublished theses: Seni (1978), Boone (1979), and Granata (1981). McGowen et al. (1979, 1983) proposed that the Dockum was deposited in a complex series of deltaic, lacustrine, and fluvial environments.
Frelier (1987) conducted a detailed sedimentological study on the Dockum in Garza County. Frelier (1987) has proposed an alternate model 12 to the one by McGowen et al. (1979, 1983). Frelier (1987) postulated that the Dockum is composed exclusively of fluvial sediments and that no deltaic or lacustrine sediments are present.
This study agrees with many of the ideas presented by Frelier (1987). The alternative model proposed by him is independently confirmed in this study. The Dockum is composed of fluvial deposits with the exception of one enigmatic deposit in which no conclusive origin can be postulated in this study (see Chapter VI). This deposit, though, is only a small part of the fluvial deposits of the Dockum.
This study will also attempt to clarify the inconsistent nomenclature and correlation of the stratigraphy of the Dockum. The stratigraphy has been problematic from the early part of the century to the present day. CHAPTER II
STRATIGRAPHY
Introduction The Dockum Group unconformably overlies Permian deposits of the Quartermaster Formation in the study area. The Quartermaster Formation consists of evenly bedded reddish brown siltstone and mudstone with thin gypsum interbeds and is characterized by pronounced local solution collapse deformation (Boone, 1979). Gypsum is not present immediately below the Permo-Triassic contact, and hence the gypsum interbeds cannot be used as an indicator of the contact.
Several formations unconformably overlie the Dockum Group, the major one being the Tertiary (Miocene) Ogallala Formation. In the Tule Canyon area (see Fig. 1.1), the Ogallala Formation is absent and the Quaternary Tule Formation overlies the Triassic (Boone, 1979). The Tule Formation is a thin sequence (up to 30 meters) of fossiliferous silts and clays formed in a Pleistocene pluvial lake environment (Boone, 1979). This formation overlies the Triassic in the upstream section of Tule Canyon, while the Ogallala Formation covers the downstream section. The Quaternary Cita Formation overlies the Dockum Group in limited areas of Palo Duro Canyon (see Fig. 1.1) (Hood, 1977). Hood (1977) described the basal unit of the Cita Formation as a calcareous, caliche-impregnated, quartz sandstone with interbedded claystone, shale, and some diatomite. These Pliocene lacustrine
13 14 deposits rest on an erosional surface incised through the Ogallala and into the Dockum (Hood, 1977). The Dockum Group in Texas is divided into two formations: the Tecovas Formation and the Trujillo Formation (Gould, 1907). These formations have been correlated with other Triassic sediments in the United States including: the Dockum Group in New Mexico, the Chinle Formation in the Colorado Plateau area, the Dolores Formation in southwestern Colorado, the Popo Agie Member of the Chugwater Formation in Wyoming, and the Newark Supergroup along the Atlantic coast of the United States (Fig. 2.1).
Biostratigraphy
Triassic continental sediments in the United States are correlated on the basis of vertebrates, plant megafossils, and palynology. Because the Dockum Group and its formations are correlated with other Triassic sections primarily on the basis of biostratigraphy, a brief overview of proposed correlations is given here. Most of this discussion is based on Chatterjee (1986).
The Triassic type section is located in Germany. It is divided into lower nonmarine (Buntsandstein), middle marine (Muschelkalk), and upper nonmarine units (Keuper). An entirely marine type section (Alpine) has also been designated with subdivisions that are now considered the global standard sequence of stages. In the case of continental sediments, however, the Germanic types are still used.
The Keuper has six divisions: Lettenkeuper, Gipskeuper, Schilfsandstein, Rotewand (Bunte Mergel), Stubensandstein, Fig. 2.1. Correlation of Triassic continental deposits in the United States (modified after Morales, 1987). 16
S.Utah S.W. New Mexico Wyoming W. Atlantic lU N. Arizona Colorado N.W. N. Central Texas Coast 1-210 .'...'.. 1111 I.I .•.y,v,;.^;..l,i.i ,1.1,1.1 i.i.i.i.i.v.i.v.i.'.u'. WIngate ;:$: Wingate Newai1< Super Dolores Chinle Tmjillo Chinle Chinle group Vf^nf^t W9V Popo Agio Santa Tecovas Rosa l.i.i.i.i.i.V
230
9o 2 I MoenkopI Anton Chico 240 Moenkopi or Si 1250 lUUJiUJbUU^ ^uuMi^^j^iu^^y^^^uuuyyi 17 Knollenmergel, and Rhatkeuper (Rhat Sandstein) (Fig. 2.2). The Keuper divisions are equivalent to the Carnian and Norian divisions of the Alpine sequence except for the Lettenkeuper which is equivalent to the Ladinian. The Rhatkeuper is equivalent to the Rhaetian, but many workers include the Rhaetian in the Norian. The Dockum Group in Texas ranges from Schilfsandstein to Knollenmergel (Chatterjee, 1986).
Vertebrate remains found in the Dockum Group include those of fish, amphibians, and reptiles. Most vertebrate paleontologists rely on the reptiles for correlation, particularly the phytosaurs. Fish, amphibians, and reptiles (excluding phytosaurs) are used to supplement correlations based on phytosaurs. The phytosaurs permit more detailed correlation and paleontological subdivisions of the Late Triassic because individual taxa are restricted temporally and have a fairly wide geographic distribution (Chatterjee, 1986).
Chatterjee (1986) recognizes four phytosaur genera in North America: Parasuchus, Angistorhinus, Rutiodon, and Nicrosaurus. In the German section, Parasuchus is exclusively Carnian with Nicrosaurus and Mystriosuchus being Norian (Fig. 2.2). Rutiodon spans both Carnian and Norian stages and hence is not considered useful for zonation. Because Parasuchus is restricted to the Carnian it is used to delineate the "lower faunal zone" of the Dockum (Chatterjee, 1986). Chatterjee (1986) favored Nicrosaurus over Rutiodon, for his "upper faunal zone," because there is no overlap in the ranges of Parasuchus and
Nicrosaurus. Both Parasuchus and Angistorhinus are found in the Tecovas Formation, which is assigned a Carnian age. Based on the Fig. 2.2. Divisions of the Alpine and Germanic sequences and phytosaur zonations in Germany (Chatterjee, 1986). 19
EPOCH AGE Ma GERMANY RANGE OF PHYTOSAURS
LIAS HETTAGIAN 213 RHAETIAN RHATKEUPER . 210 KNOLLEN MERGEL NORIAN UTE t t STTJBEN- TRIASSIC Nlnmnniirim Mytriosiirhim SANOSTEIN 225 STEINMERGE L 4 4 Butlc ROTEWAND KEUPE R ^ J^ CARNIAN SCHILFSANDSTEIN i GPSKEUPER 231 LFTTENKEUPER MIDDl^ TRIASSIC OOINIAN MUSCHE1J Chatterjee (1986) places the Tecovas Formation and the Popo Agie Member of the Chugwater Formation in the Parasuchus zone (Fig. 2.3). The Trujillo Formation, Newark Supergroup, Dockum Group of New Mexico, and Chinle Formation of Arizona are placed in the Nicrosaurus zone. Lucas and Hunt (1987) have placed the Anton Chico Formation (lowermost formation of the Dockum Group in New Mexico) in the Middle Triassic owing to the occurrence of Eocyclotosaurus, a capitosaurid labyrinthodont. The Santa Rosa Formation is considered to be equivalent to the Tecovas Formation based on lithostratigraphic relationships (Lucas et al., 1987). This point is discussed in more detail below. Plant megaflora and palynomorphs each provide zonations different from the vertebrate faunal zones, thus allowing additional biostratigraphic correlations. Ash (1980) divided the upper Triassic of North America into three floral zones: Eoginkgoites, Dinophyton, and an unnamed upper zone. The zone of Eoginkgoites is the lowermost zone. Ash (1980) believes this zone to be of middle Carnian age. He placed the Shinarump Member of the Chinle Formation in the Colorado Plateau region and part of the Newark Supergroup within this zone. The middle zone of Dinophyton is considered to be late Carnian and perhaps Norian in age. Ash (1980) placed the Dolores Formation, the Petrified Forest and Monitor Butte Members of the Chinle Formation, the Trujillo Formation, the Tecovas Formation, the Santa Rosa Formation, and parts Fig. 2.3. Phytosaur zonation and correlation of North American Triassic strata (modified after Chatterjee, 1986). 22 ALPINE EUROPE NORTH AMERICA EPOCH BIOZONE FACIES ATLANTIC WEST NEW GERMANY ARIZONA COAST TEXAS MEXICO WYOMING I CO 5; lO ^£WARK KNOLLEN SUPERGROUP MERGEL o CO g z STUBEN (COOPER) SANDSTEIN § N DOCKUM FM. DOCKUM FM CHINLE FM. ^ ROTEWAND I I TECOVAS M. I POPO AGIE M SCHILFSANDSTEIN 23 of the Newark Supergroup in this zone. The upper zone is poorly known, but probably extends into the Jurassic (Ash, 1980). Dunay and Fisher (1978) described three palynological zones in the European marine Triassic sequence and seven zones in the Germanic Keuper. They correlated species from the Dockum Group to one of their marine zones, which is Carnian in age. Species that are equivalent in age to those found in the Dockum Group are also found in the Chinle Formation of the Colorado Plateau and part of the Newark Supergroup, thus supporting a Carnian age for these formations. The biostratigraphy for Late Triassic deposits is somewhat controversial. Many workers only agree that the majority of North American Triassic deposits are Late Triassic in age and span the Carnian and perhaps Norian ages. A major problem has been that most phytosaur remains are sparse and fragmentary. Workers have had a tendency to name new species and genera that are actually referable to earlier named species. Plant fossils are even more scarce and fragmentary than vertebrate remains. Most workers using plant megafossils and palynology have, however, drawn the same conclusion as the vertebrate workers, the Dockum Group and most Triassic continental deposits in North America are Carnian and/or Norian in age. Tecovas Formation The Tecovas Formation was named for Tecovas Creek in Potter County, Texas, by Gould (1907). Gould did not publish any measured sections from Tecovas Creek, but did provide sections from localities 24 in Palo Duro Canyon, the North Branch of North Canyon Cita Creek in Randall County, and West Amarillo Creek in Potter County. The Tecovas Formation is the lowest Triassic formation present in the study area. It rests unconformably on Permian sediments. The Permo-Triassic contact appears to be gradational in some areas: Wayside Crossing (locality CC-1), Silverton Roadcut (locality SRC-1), Caprock Canyons (localities CRC-1 and CRC-2), and Floyd County (locality FC-2) (Fig. 2.4). The gradational contact is usually associated with an aeolian sand body which lies on top of known Permian deposits yet below more typical Tecovas lithologies. This seemingly gradational contact suggests continuous sedimentation from Permian to Late Triassic time (McGowen et al., 1979). The majority of localities however have an obviously unconformable contact that is sharp and distinct (Fig. 2.5). Aeolian unit. The aeolian unit is the lowermost unit, where present, in the Tecovas Formation. This unit extends from the Wayside Crossing (locality CC-1) in the north to Floyd County (locality FC-2) in the south. It varies in thickness from 3 to 4 meters in Tule Canyon to 18 meters in Caprock Canyons (locality CRC-1). This unit is a red brown to orange, very fine to medium grained, poorly cemented sandstone (sedimentology discussed further in Chapter III). Horizontal bedding and parallel lamination is observed in some places (Fig. 2.6), but the predominant sedimentary structure is large scale trough cross-bedding (80 to 110 centimeters thick) indicative of aeolian deposition. This unit cannot easily be placed in either Triassic or Permian systems. Its coloration is similar to both Quartermaster and Tecovas, but it is 25 Fig. 2.5. The unconformable Permo-Triassic contact. The orange colored siltstones are part of the Permian Quartermaster Formation. Purple, gray, and yellow mudstones belong to the Tecovas Formation. Note the white carbonate nodules in the Triassic mudstones. Location is Palo Duro Canyon (PD-5). Field of view is approximately 8 meters. 26 Fig. 2.6. Aeolian sandstone at the base of the Tecovas Formation. Jacob staff is 2 meters long. Location is the Wayside Crossing (CC-1). 27 petrographically dissimilar to both formations. It is herein tentatively considered as Triassic, however, no fossils are known from this unit. If this unit is indeed Triassic in age, it could be much older than the overlying Dockum Group (?Early or Middle Triassic), because a marked paleosol separates it from the overlying Tecovas Formation. Channel sandstones. Channel sandbodies occur locally along the base of the Tecovas Formation and in places occupy channels incised into underlying Permian strata. These sandbodies range from 0.5 to 48 meters thick. The sandstones are petrographically identical to those occurring higher in the formation and are one of the criteria used to identify the Tecovas Formation (see Chapter IV). These sandstones appear at localities: PD-6, PD-7, CC-2, Tule Canyon, SRC-1, CRC-1, CRC-2, FC-1, FC-2, near DC-2, DC-3, and CBC-1. The sands occur as individual lenticular bodies that are obvious channel-fills or as a series of stacked sandbodies (see Chapter III). The largest of these lenticular bodies was observed 0.5 mile south of locality DC-2 and 0.25 miles north of U.S. highway 82, outside of Dickens, Texas (Fig. 2.7). These sands commonly contain chert granules and pebbles. The thickest sequence of stacked channels occurs in Tule Canyon. The sands are tan brown, coarse grained, quartz-rich, and poorly cemented. The lenticular sandbodies wedge out laterally over meters or tens of meters. The stacked sandstones can be laterally more extensive (localities SRC-1 and Tule Canyon). 28 Fig. 2.7. Typical sandstone from the Tecovas Formation. Field of view is approximately 8 meters. Location is Dickens, Texas (0.5 mile south of DC-2). 29 Paleosols. At localities north of Tule Canyon a prominent paleosol rests on the Permo-Triassic contact. At localities, including Tule Canyon, south to Floyd County (locality FC-1) the paleosol rests stratigraphically higher, above or on the aeolian unit described above (Fig. 2.8 and Fig."2.4). The paleosol was not observed in Motley, Dickens, or Crosby Counties. The thickness of the paleosol varies from 2 to 11 meters. The Silverton Roadcut (locality SRC-1) has the thickest section. The paleosol unit is purple to purple gray in color. It occurs in mudstones, sandstones, or both. In Tule Canyon and at the Silverton Roadcut bright yellow and red mottled zones also occur. The Trujillo Formation rests unconformably on the paleosol at the Silverton Roadcut. The Trujillo Formation rests on an erosional surface incised into the Tecovas Formation, at this locality, partially truncating the paleosol. At locality PD-6, manganese nodules are observed in the paleosol where it is formed in the basal sandstone of the Tecovas (Fig. 2.9). The manganese nodules are observed in thin section as detrital grains surrounded exclusively by manganese oxide (pyrolusite?) cement. A silcrete horizon is observed in the paleosol at three locations: PD-5, CC-1, and SRC-1. The silcrete is massive or laminated, white in color, and is strongly lithified. At location CC-1, a thinly laminated maroon lens is found along with the massive, white variety (see Chapter VI for detailed description of the paleosol and mode of emplacement). The silcrete is 50 centimeters thick and extends laterally, as a continuous unit, for about 0.5 kilometers at locality SRC-1 (Fig. 2.10). At PD-5 and CC-1 the silcrete is a discontinuous bed only 5 to 10 centimeters thick. The unit is probably 30 Fig. 2.8. Typical paleosol unit occurring in a basal aeolian sandstone unit. Note mottled appearance. Location in Floyd County (FC-1). Rock hammer for scale. 31 Fig. 2.9. Manganese nodules occurring in the paleosol unit, Location in Palo Duro Canyon (PD-6). Note hammer for scale. 32 Fig. 2.10. Silcrete horizon (continuous white bed just above Jacob staff) occurring in a paleosol unit. Note mottling in purple mudstone. Sandstone at the top of the photo is from the Trujillo Formation. Location is the Silverton Roadcut (SRC-1). Jacob staff is 2 meters long. 33 exposed in other parts of the field area, but because it is very thin and discontinuous it is difficult to locate. At two localities in Crosby County (CBC-2 and CBC-3), siliceous materials that are similar to the silcrete are found in the same stratigraphic position, but they are not associated with a recognizable paleosol. The siliceous materials are cylindrical and are similar in appearance to root casts. No confident correlation can be drawn, however, between these siliceous materials and the silcrete. Multi-colored mudstones. Located above the paleosol and basal sandstones are what Gould (1907) described as "variegated shales." This unit is made up of maroon, red, red brown, and yellow mudstones (Fig. 2.11). In most areas the reddish mudstones rest on the paleosol followed by a yellow unit and another reddish unit (Fig. 2.12). This is not always the case, however, there may be up to eight colored zones within this mudstone unit. Each colored zone has a gradational contact with the other zones and the colors are transitional between two differently colored zones. The unit, as a whole, varies in thickness from 5 to 32 meters with wide variation in thickness of individual colored zones. This unit is well exposed in the Palo Duro Canyon area. These mudstones thin from Palo Duro southward to Tule Canyon, toward the Matador Arch, then reappear and thicken southward from the arch. The unit is not present at Tule Canyon, the Silverton Roadcut, Caprock Canyons, or in Floyd County, but reappears in Dickens and Crosby Counties. At various stratigraphic positions within the mudstone are carbonate 34 WS" - A Fig. 2.11. Capitol Peak in Palo Duro Canyon (north of PD-5). Orange unit is the Permian Quartermaster Formation, purple gray beds above the Permian are Triassic paleosols, red and yellow beds are mudstones of the Tecovas Formation. 35 Fig. 2.12. Typical red and yellow mudstones in the Tecovas Formation. Location in Dickens County (DC-4). Jacob staff is 2 meters long. 36 nodules ranging in diameter from 1 to 15 centimeters. These nodules are weathered out of the mudstones and litter the surface as rubble. The majority are 1 to 2 cm in diameter, but in Palo Duro Canyon much larger varieties are found. These larger nodules are also different texturally than the smaller ones. Some of these large nodules have trace fossils on the exterior, some are cylindrical, and some are botryoidal (see Chapter VI for further discussion). The cylindrical nodules appear to be biological in origin. Throughout the mudstones in the Tecovas Formation are thin conglomerates (5 to 20 centimeters). These conglomerates are composed of quartz-rich sand and intrabasinal granules and pebbles (see Chapter IV). The conglomerates are massively bedded and have lateral extents of 10 to 60 meters. Occurring locally in Palo Duro Canyon and at the Wayside Crossing (location CC-2) is a white, very clean, poorly cemented, lenticular sandstone. Compositionally it is the same as other sandstones from the Tecovas Formation, but it is finer grained (very fine sand) and bone white in color (Fig. 2.13). It is very poorly cemented and appears in outcrop as unconsolidated sand. Hood (1977) stated that this sandstone was locally referred to as "sugar sand." This sandstone appears only at the locations described above and appears to represent channel-fill (see Chapter III). Upper mudstone. The upper unit of the Tecovas Formation is typically a red or maroon colored mudstone. This unit is present at almost every locality where the Tecovas is exposed and the erosional surface at the base of the Trujillo Formation has not scoured deeply 37 -•—-^-^ i p^^M .•4 %^ .^j^. 0^ Fig. 2.13. The upper sandstone of the Tecovas Formation. This sandstone is locally called "sugar sand." Note the climbing ripple cross-lamination. Location in Palo Duro Canyon (PD-3). Hammer for scale. iflte the T§§gvi§ F§fffiiti§R. This ffludsteRi i^ dlfftfint from ths muditgnil l''vyf.f i„ iiiR §§eti@fl, A1th'_""jli thi cgler 1s -.Imnar to the r%i'. inH fflaregfi§ gf th§ Umr fny^§i@fl§§, it dgf. net vary or grad© Into ytllnwi ifld pufpi'§, hut E'Sfj hiv§ thin grmr^ zenti (;• to 40 ((•ntiffl'-t§r§) div§1gp§d vhrfiiyghgyt 1t§ »hlekni§-.. Carbonate nodules are net n: abundiflt in thi§ unit &§ thiy or'. ln tht lowtr units. Leesl pa1i©§0l% hoy& dtvileptd en th^se muditonis at thi Tecuvai-lrujino eentget (1eeg]1t1f§ PD-a, CC-1). Thiy are limited laterally t.o 2D or 30 m§t§r§ and ar© a fiw fnttars thick. These paleosols are compesid ef mfi+.tlei: purpli mudstone and sandstone. They are not strati graphically equivalent to th© lowor Tecovas paleosol described above, TryjIIHig Formation Gould (1907) naraiedl the T'.viTlo Formation for Trujillo Creek in Oldham County, Texas. Gould (1907) describee the Trujillo Formation as several ledges of isiassiwe Oir cross bs^ied sandstone and conglomerate with interbedded red and gr% ihsles. Ssaild (1907) also included the :;ppermost shales of tine Ztockiuinn lira tltne Trujillo Formation. In this study, tine Triuijii''lo Frri-e-tcn is considered to include the entire Triass'^c sectH'?-.™ iocve tlhie Tecovas ForEstion. The Trujillo Formation is eowiposedt ®f g-i-'r- grsj^ sarndsre-^s and conglomerates with iiiniterbeds of red, nmainQKisiiii^ mii p^eeun grajf mwdstones (Fig. 2.14). The (comglorotfitts w© mM^ ®f limtir^tesiiifiiall gramiusles and pebbles 39 Fig. 2.14. Interbedded sandstones and mudstones of the Trujillo Formation. Note large scour at the bottom of the photo. Location at the Wayside Crossing (CC-1). Fence posts at the top of the photo are approximately 1.4 meters tall. 40 (sedimentary rock fragments), but some extrabasinal (quartzite and chert) conglomerates are also present. The base of the Trujillo Formation is marked by a thick sandstone or conglomerate. The contact between the Tecovas and Trujillo Formations is an unconformity with substantial erosional relief. The basal sandstones of the Trujillo rest on deep scours in the underlying Tecovas. Many workers have placed the contact between the formations at the base of the first massive sandstone unit, but Hood (1977) places the contact at the lowermost sandstone (see below), whether it is a major or minor sandstone body. The author agrees with Hood's (1977) observation. Typical Trujillo sandstones are green gray, very fine to fine grained, micaceous, clayey, and contain abundant rock fragments. Sedimentary structures include massive bedding, parallel lamination, trough cross-bedding, ripple cross lamination, and climbing ripple cross lamination. The conglomerates are green gray in color with clasts ranging from granular to boulder in size. Intrabasinal clasts include carbonate grains, mudclasts, reworked sandstone and siltstone fragments. Extrabasinal clasts include quartzite, vein quartz, and chert clasts with the quartzite clasts the dominant grain type. Intrabasinal conglomerates are most common, but extrabasinal clasts outnumber intrabasinal clasts at a few localities. These localities (PD-5, PD-7, Tule Canyon, FC-3, DC-1, SF-1) have conglomerates that are comprised almost exclusively of quartzite and chert pebbles. In Tule Canyon the extrabasinal conglomerates occupy a large valley-fill that cuts through the Tecovas Formation and into the underlying Permian (see 41 Chapter III). The mudstones interbedded with typical Trujillo sandstones are red, maroon, green gray, or gray in color. They are massively bedded with very few sedimentary structures present. From Palo Duro Canyon in the north, to Caprock Canyons in the south, the basal sandstone of the Trujillo is a very thick sandstone body (10 to 90 meters) that forms a vertical cliff. South of Caprock Canyons this sandstone is not as prominent as in the northern part of the study area. This sandstone unit has been referred to as the "Trujillo sandstone" or even the entire Trujillo Formation. This unit is not as massive or uniform as some workers have suggested. In the Palo Duro Canyon area this unit appears as one massive sandstone body, but can be traced laterally into as many as three different sandstones. Even where it does comprise one massive sandstone body, the sandstone is made up of several sedimentation units with no mudstone separating them. This sandstone is continuous from Palo Duro Canyon southward to Caprock Canyons. Individual sandstone bodies that comprise this unit are, however, not continuous. The unit is hence made of an overlapping "nested" series of different sandstone bodies. Each individual sandstone body wedges in and out of the unit. Owing to their complex interrelationships, correlation between the individual sandstone bodies was not attempted in Fig. 2.4. A generalized cross section of the sandstone bodies is presented (Fig. 2.15). Another problem complicating use of the term "Trujillo sandstone" is that pre-Ogallala erosion has in places removed many of the Trujillo sandstone bodies. This occurs at location CC-2 and several localities south of Caprock Canyons. In the southern part of the study area the .^ o O OJ >, •— <-) u -a to E c c 3 O (O I— .^ 4J (O O to "1. O "0 O x: •c o z v> CO m c CO (D 3 > o CO CO o to o > o o o CO o o o o o o o o o o ^w o o o o o o o o o o (D o en 00 h- Several sandstone bodies also occur above the "massive" sandstone unit, but mudstone increases in abundance upward and predominates over the sandstone. The sandstone thicknesses can vary from a few centimeters to 60 meters. These upper sandstone bodies are even more lenticular than the ones lower in the formation and can wedge out within a kilometer or less. In places these sandstones are as thick as the lower sandstone, however. Although the sandstones in the upper Trujillo, are smaller and less abundant than those of the lower Trujillo, they are petrographically identical to those at the base of the Trujillo Formation. Stratigraphic Problems Most workers have described two or three stratigraphic units for the Dockum Group in the Palo Duro Basin (Fig. 2.16). Drake (1891) recognized three informal units. Gould (1907) proposed two formations: the Tecovas and the Trujillo. Adams (1929) recognized Drake's three units and referred Goulds's Trujillo Formation to the Santa Rosa and Chinle. The lower formation was equated to Gould's Tecovas (Fig. 2.16). Adams' (1929) equivalent of the Chinle Formation is not well exposed in the Palo Duro Basin, but good exposures occur in Garza County and further south in the Midland Basin. Chatterjee (1986) applied the name "Cooper Member" to Fig. 2.16. Comparison of the stratigraphic nomenclature applied to the Dockum Group in Texas. 46 Drake (1891) Adams (1929) McGowen et al. Chatterjee Gould (1907) (1979) (1986) Upper Chinle Upper Cooper unit Formation Dockum Member Trujillo 0. Formation a. 13 Q. 6 o— ) O _> i- oCC oc CC o ^ (i) o CD CO g Middle 2 2 Santa Rosa :5 Trujillo 2 unit Formation ID Member o o o o ^J oQ oQ Q Lower " § Dockum Tecovas Lower Tecovas Basal Member unit Formation shales 47 Adams' "Chinle." Although Frelier (1987) recognized the existence of the Tecovas and Trujillo Formations north of Crosby County, he could not distinguish the two formations in Garza County or in adjacent counties where the Dockum Group crops out. Lithostratigraphic correlation. Chatterjee (1986) formally proposed that the Dockum Group be reduced in rank to a formation. Chatterjee stated that Gould's formational units are local and cannot be traced from one region to another. The present study has shown that the formations may be traced along the eastern escarpment of the High Plains (Fig. 2.4). Gould's subdivisions of the Dockum Group meet the criterion of mappability for formational status. Thus, the Tecovas and Trujillo formations should not be reduced to members and the Dockum Group is retained as a group in this study. The Dockum Group is well exposed in Palo Duro Canyon, with the Tecovas and Trujillo formations being easily distinguishable from each other and from the Permian and Tertiary deposits. For this reason, field work for the present study began in this area and progressed systematically southward. Exposures deteriorate progressively south to Motley County, but good exposures appear again in Dickens and Crosby Counties. As sections were measured in more southerly areas, variation in the formations was noted and the two Dockum formations could be successfully differentiated even in poor exposures. 48 Distinguishable characteristics of the Tecovas formation are: (1) Typical quartz-rich sandstones (described above and in Chapter IV). (2) Purple mottled paleosols, including silcrete horizons. (3) Maroon and yellow smectite-rich mudstones (see Chapter VI about clay content). (4) Large botryoidal and cylindrical carbonate nodules (described above and in Chapter VI). Distinguishable characteristics of the Trujillo formation are: (1) Typical micaceous lithic-rich sandstones (described above and in Chapter IV). (2) The abundance of sandstone units. (3) Maroon and red mudstones with the absence of any yellow or purple mudstones. (4) Mudstones composed of kaolinite, illite, and smectite. The contact between the two formations is placed at the base of the lowest sandstone with typical "Trujillo" lithology. All Triassic deposits above this sandstone are included in the Trujillo Formation. Using the criteria described, the formations can be identified in most exposures, even in Garza County and further south. Naming inconsistencies. The term Santa Rosa Formation should not be used for the basal Trujillo sandstones. Subsurface workers in the Permian Basin incorrectly apply the term to the massive sandstone bodies of the Dockum. Santa Rosa and Trujillo sandstones are distinct petrographically. The Santa Rosa Formation of the Dockum Group in northeastern New Mexico has been correlated with the Tecovas Formation 49 of Texas (Lucas et al., 1987) (Fig. 2.17). Most workers do not recognize an equivalent of the Santa Rosa Formation in outcrops on the Texas side of the High Plains; however, the possibility exists that the basal lenticular quartzose sands of the Tecovas represent thin tongues of the Santa Rosa Formation. McKee (1957) and Chatterjee (1986) considered the use of the term "Chinle" for the upper Trujillo undesirable because the entire Dockum Group is equivalent in age to the Chinle Formation of the Colorado Plateau. Instead a new name, the "Cooper Member", has been coined for the upper unit of the Dockum (Chatterjee, 1986). This distinction is a difficult one to make because Gould's Trujillo Formation is not readily divisible into two formations (Trujillo and "Chinle" or "Cooper"). The only criterion workers have used to split the Trujillo Formation into two separate formations is that the upper part of the Trujillo Formation has a higher percentage of mudstone compared to sandstone. There is, however, no distinct boundary between the upper and lower part of the formation. There are major sandstone bodies within the upper part, some of which are as large and extensive as those in the lower part. The entire formation, independent of geographic location, is lithologically and petrographically consistent. The "Trujillo sandstone" (basal sand of the Trujillo Formation) is prominent in the Palo Duro Basin, but cannot be confidently identified further south. This sandstone body should itself not be considered a separate formation, only part of the Trujillo formation. Gould's (1907) subdivisions of the Dockum Group remain the most useful when applied to outcrops in Texas. Fig. 2.17. Correlation of the lower Dockum Group with Triassic deposits in New Mexico and the Colorado Plateau. Note that the Tecovas Formation is equated with the Santa Rosa Formation and the lower shale member of the Chinle Formation (from Lucas et al., 1987). 51 COLORADO EAST-CENTRAL PALO DURO PLATEAU PER . NEW MEXICO BASIN SU B P . STAG E SONSELA S. CUERVO M. TRUJILLO F. lower PETRIFIED shale m. FOREST M. CHINL E F . TRES ^MOSS LAGUNAS M. ^BACK M. TECOVAS F. LAT E CARNIA N MONITER DOCKU M GROU P BUTTE M. LOS^^B ESTEROS i^ TRIASSI C SHINARUMP TECOLOTITO SANT A ROS F . M. M. :'X*!'X*>:%'X;;>;:;I;I;y;t;X;t>;X;I:;;y;>';i;I:*; "^X*X'>>>I:I:*x'*X:X;>:;Ix>>I;I;.;.;.;. :.;.;•;•;• X<'X%%*>I*'^'vvXx;;;;.;.;.;.;.;.;.;.X';>>I*:%'; *X'XvI;X'X*X'Xx''X%>X''%*!'I*'v'**'^**'**' I'XvX*X'I*X'X'I'X»I'X*X'I'X*.'.'.X">>X'>i HOLBROOK ANTON M. CHICO F. MIDDL E ANISIA N MOQUI M. •!-X*/*"X"X*!'"'X*I*'*X*i'*'*".*'*'''*-'''*' Wf}i 52 Informal subdivisions. In the past decade many workers, especially subsurface workers, have abandoned the use of formation names for the Dockum (Boone, 1977; Seni, 1978; McGowen et al., 1979, 1983; and Johns and Granata, 1987). These workers contend that the formal stratigraphic units have local distributions and cannot be traced, particularly in the subsurface (Seni, 1978). These workers have instead informally subdivided the Dockum Group into "upper" and "lower" Dockum, basically for ease of subsurface work (Fig. 2.16). In outcrop, the "lower Dockum" of the subsurface workers is equivalent to the Tecovas and Trujillo Formations and the upper Dockum to the "Chinle Formation" (Johns et al., 1987). It would produce more consistency and less confusion, however if Gould's (1907) nomeclature were used in the Texas subsurface. CHAPTER III SEDIMENTARY FACIES Introduction Deposits of the Dockum Group are entirely nonmarine in origin. The sedimentary facies present are delineated by size, geometry, sedimentary structures, grain size, mineral composition, sorting, and boundary types. A facies is defined as a distinctive rock that forms under certain conditions of sedimentation, reflecting a particular process or environment (Reading, 1986). A facies sequence is a series of facies vertically gradational with one another (Reading, 1986). Such a vertical sequence of conformably stacked facies is thought to represent sedimentary environments that formerly existed lateral to one another. The facies sequence concept is applied here to Dockum outcrops. Sedimentological and external controls influence these sequences. These relationships are used to gain a better understanding of the different facies present in the Dockum Group. Measured sections, outcrop drawings, and field observations allow the identification of several sedimentary facies. Stratigraphic columns constructed from the measured sections are the primary tool used in identification of the facies, but emphasis is also placed on outcrop drawings and field observations. Stratigraphic columns are used to illustrate the field observations, but in some cases give incomplete information. Hence, other methods are to used in conjunction with measured sections. 53 54 Sedimentary structures are of great use in facies analysis. Different types of structures and varied relationships between them are diagnostic of different facies. The concept of bounding surfaces is used to delineate "packages" of sedimentary structures and their order of occurrence (Miall, 1987). Four orders of bounding surfaces are observed in the Dockum sandbodies. First order surfaces may truncate any lower order surface (or orders of the same rank), but a lower order does not truncate a higher order bounding surface. The base of a channel body represents a first order surface (highest order). Second order bounding surfaces are represented by lateral accretion surfaces. Lateral accretion surfaces record point bar building in fluvial channels. Third order surfaces are the boundaries between cross-bed sets. Fourth order bounding surfaces are re-activation surfaces within sets of cross-beds. Aeolian Facies Aeolian deposits are localized in the study area. They are only found in Briscoe and Floyd Counties (Fig. 2.4). These deposits are comprised of sandstone, siltstone, mudstone, and clay rip-up clast conglomerate. Sandstone is the major lithology in this facies. The sandstones are orange to red brown in color, vary from very fine sand to medium sand, and are well sorted. The grains are subangular and have moderate sphericity. The sandstones are poorly cemented with quartz overgrowths, ferrans, pore-filling authigenic clays, and small amounts of calcite. The upper and lower boundaries of the facies are gradational (discussed in Chapter II). Sedimentary structures present 55 are massive bedding, horizontal parallel lamination, inclined parallel lamination, and large-scale trough cross-beds (set height = 50 - 110 cm). Throughout the vertical sequence the sandstones are interbedded with thin lenses of siltstone, mudstone, and clay rip-up clast conglomerate. The siltstones are red in color and contain horizontal lamination. These units have an average thickness of 60 cm and wedge out laterally within a few meters. Within the siltstones are numerous horizontal lenticular gray reduction zones. These zones are approximately 2 cm thick and 5 to 10 cm long. The mudstones are red in color and show no sedimentary structure (massive bedding). The mudstones are 20 to 40 cm thick and comprise a small percentage of the facies. Pebble-size clay clast zones interrupt the sandstone sequence in places. The clay clasts are red in color and were probably reworked from the mudstone layers described above. These layers are 5 to 10 cm thick and extend laterally approximately 1 to 2 meters. These deposits are postulated to be aeolian in origin because of the characteristics described above and the two distinct sub-facies identified within the unit that are described below. Large-scale cross-bedding, well sorted sand, fine to medium grained sand, high roundness and sphericity of grains, frosted grain surfaces, lack of detrital clays and mica, lack of marine fossils, and rare presence of vertebrate fossils are diagnostic features of aeolian deposits. For confident interpretations of aeolian deposits as many of these criteria as possible should be used because, independently, most can be ambiguous (Collinson, 1986). The sandstones in the aeolian facies of 56 the Dockum meet the above criteria except for roundness, sphericity, grain frosting, and presence of vertebrate fossils. It is believed the roundness, sphericity, and grain frosting criteria are not met because numerous quartz overgrowths have obliterated the original grain shape and surface. The absence of fossils in this unit does not negate its aeolian origin. Aeolian deposits in the Dockum comprise a sheet-like sandbody, not a lenticular sandbody, composed of two sub-facies (dune and interdune). Lenticular sandbodies are relict dune or draa forms and are quite rare because of the special conditions required to preserve them (Collinson, 1986). Dune sub-facies. The dune sub-facies comprises about 95 percent of the aeolian sediments exposed in the Dockum (Fig. 3.1). It is predominantly composed of parallel lamination and large-scale trough cross-bedding though some massive bedding is also apparent. Ahlbrandt and Fryberger (1982) refer to large-scale trough cross-bedding in aeolian deposits as wedge planar stratification. The dune sub-facies is observed to be 4 to 15 meters thick. The dominant structure present is trough cross-bedding with an average thickness of 1 meter. Trough cross-beds represent preserved foresets of migrating dunes with undulose crests. Dune foresets are formed through the alternating processes of grainfall and avalanching (grainflow). Saltating grainfall is postulated to be the predominant process of dune migration in the aeolian deposits in the Dockum Group. Avalanching (grainflow) deposits are not observed in this facies, however, this may be owing to lack of preservation of upper dune deposits. Fig. 3.1. Example of the dune (D) and interdune (ID) sub-facies in aeolian deposits of the Dockum Group. Note the large-scale trough cross-beds and parallel lamination of the dune sub-facies. The interdune sub-facies contains laminated siltstone and clay clast layers. Location is Caprock Canyons (CRC-1). 58 2748" ID lll^TT D K:^^ ^ -<: 'K^ >r^o^ N.^N: \00 O O Q-g-oJ »> -^=•^^^3^?^ ::^v:i.w^i:-:.^L>; 0 IToo% 0%i I -\ m I I I I I ss si cl vf vc gr cb 59 Parallel lamination may represent tractional deposition at wind velocities too high for ripples to form (Collinson, 1986). This type of lamination forms on the stoss side of dunes and is usually rare. It may also represent a transitional facies between the dune and interdune facies. More likely, it represented translational strata produced by wind ripple migration. The latter is probably the type present in the Dockum. The parallel laminated deposits represent a sand sheet that surrounds the dunes and laterally grades into the interdunal areas. In this study the parallel laminated sand sheets are included in the dune sub-facies, because they are intimately interbedded with the trough cross-bedding. Interdune sub-facies. The interdune sub-facies is composed of siltstone, mudstone, and clay clasts (Fig. 3.1). This sub-facies comprises approximately 5 percent of the aeolian sediments exposed in the Dockum. The siltstones and mudstones are interpreted as grainfall deposits between dunes. The clay clasts could be mudstones that were eroded by wind action into small granule and pebble-size clasts. No evidence was observed for the interpretation of lacustrine or wet interdunal areas. The aeolian sediments present in the Dockum Group cover a two county area. This "dune field" was not large relative to most aeolian deposits. From the sedimentary structures exposed it is concluded that only dunes and smaller bedforms were produced and there is no evidence that there were bedforms larger than dunes (i.e., draas). 60 Fluvial Channel Facies The majority of the Dockum sediments, in the study area, are considered to be fluvial in origin. Although all of the different facies identified in outcrop are used to make this determination, the channel facies is considered to be the most diagnostic. Channel deposits in the Dockum have four recognized sub-facies: channel-fill, point bar, thalweg, and abandonment. Channel facies can be recognized by sandbody geometry, erosional bases, grain size, sedimentary structures, paleocurrent distribution, and relationships with adjacent facies. The channel facies is formed from the lateral and vertical accretion of sediment within a stream channel. Fluvial channel deposits have a general fining upward grain-size sequence and display sedimentary structures reflecting an upward decrease in flow regime (in ascending order: upper flow regime flat beds, lower flow regime cross-bedding and cross lamination, and lower flow regime flat beds). Sandbody shape depends largely on channel type and behavior (Collinson, 1986). Basically, channels can be divided into fixed or mobile channels. Fixed channels are restricted laterally and produce highly elongated sandbodies isolated in fine grained sediment (Collinson, 1986). They have steep sides and concave upward bases. They probably represent highly sinuous streams with mixed sediment loads. Mobile channels deposit more laterally extensive sandbodies which are isolated in fine grained sediments or occur in composite sandbodies with little or no fine sediment (Collinson, 1986). These channels can have steep sides and flat bases or gentle 61 concave upward bases. The mobile flat bottom channels are loosely considered to be highly sinuous streams while the mobile concave bottom channels are low sinuosity streams (Collinson, 1986). The various sandbody shapes discussed above are observed in Dockum outcrops. Fixed channel sandbodies are observed only in the Tecovas Formation. Mobile channel bodies with concave upward bases are observed in both Tecovas and Trujillo while the flat bottomed variety is observed only in the Trujillo Formation. Thalweg sub-facies. The thalweg is the lowest point in a stream channel. Sediment transport is as bedload and usually the coarsest material in the channel is observed in thalweg deposits. These coarse grained materials are referred to as channel lag and consist of granule size and larger clasts composed of claystone, reworked sedimentary rock fragments, vertebrate fossils, unionid shells, fossil wood, and reworked carbonate nodules. The majority of materials comprising channel lag in the Dockum are intrabasinal pebbles (carbonate nodules and reworked sedimentary rock clasts). Clay clasts are apparent in some lag material, but are not common, probably because they were destroyed by prolonged exposure to flowing water in the channel. Fossils are rarely found in channel lags, the majority found are coprolites. Thalweg deposits are discontinuous and never very thick. If channel lag is present, it indicates the base of the channel (Reineck, 1980). Thalweg deposits range from 2 to 80 centimeters in thickness. They are present at the base of many, but not all channels. Channel-fill sub-facies. Channel-fill sub-facies in the Dockum are comprised of conglomerate, sandstone, siltstone, and mudstone. The 62 predominant lithology is sandstone. Sedimentary structures are prominent in the channel-fill sub-facies. Sedimentary structures observed in the Dockum are: trough and tabular cross-bedding, parallel lamination (upper flow regime), massive bedding, ripple cross lamination, and climbing ripple cross lamination. Trough-cross beds and parallel lamination are the most abundant sedimentary structures present in this sub-facies. Tabular cross-bedding is rarely seen in the Dockum channel-fill sub-facies. Trough cross-beds with re-activation surfaces are observed in Dockum outcrops, these structures are thought to represent varying flow velocities (Fig. 3.2). These structures occur when one dune overtakes another. Parallel lamination in the Dockum is considered to reflect upper flow regime conditions because of parting lineation observed on bedding planes. The sandstones with parallel lamination are very fine to medium sand grain sizes. The channel-fill sub-facies in the Dockum Group, in general, follow the classic facies models. There is a fining upward sequence, but an easily recognizable decrease in flow regime is not always apparent. Trough cross-bedding and planar-bedding can form at similar flow velocities in different grain sizes, but comparing the entire sequence from the thalweg up there is a slight decrease in flow regime. The trough cross-beds generally decrease in set thickness upward and there is a lack of coarse-grained material above the thalweg subfacies (Fig. 3.3). Channel-fill sub-facies in the Dockum occur as single channels or multiple channels forming thick sandbody sequences. Multiple channel deposits, are similar to the single fixed channel 63 Fig. 3.2. Trough cross-beds with re-activation surface (right-center). Note the clay drapes separating the sandstones. Location is Palo Duro Canyon (PD-4). Jacob staff divisions are 20 cm. Fig. 3.3. Example of the channel-fill sub-facies (Ch), thalweg sub-facies (Th), and floodplain sub-facies (Fp). The channel facies is from a lenticular single channel deposit. Note the fining upwards sequence of the channel facies and for the entire fluvial deposit. Location is Palo Duro Canyon (PD-7). 3171" 65 10 Fp Ch Th n nr~\ vc gr cb 66 deposits except for sandbody geometry. Complex scouring events are recognized in these sandbodies (less frequent in fixed channels) large-scale trough cross-beds (up to several meters) rest on scour surfaces on lower beds forming concave upward third order bounding surfaces. These scouring events complicate the sequences of sedimentary structures present, but the use of the bounding surfaces concept delineates these structures easily. Complex stacked channel sandbodies occur because of the lateral shifting of the stream with little deposition of overbank materials (Fig. 3.4). The top two channel sequences shown in Fig. 3.4 show a true fining upward sequence. Some channel facies show no change in grain size. This is because they have been truncated and the upper portions of the facies is missing. Conglomerates are observed near the top of the channel facies (not in a thalweg) in some cases. This can occur during a flood event when coarser material is transported and deposited higher on the channel flanks. Clay drapes occur at the top of a sequence and are probably deposited as water velocities decrease (subsiding after a flood event) and suspended material can settle. Point bar sub-facies. Lateral accretion surfaces (epsilon cross-beds) are observed in many of the Dockum sandbodies (Fig. 3.5). Epsilon cross-beds dip at angles nearly normal to the current flow. They are generally comprised internally of small-scale structures such as small-scale cross-beds and ripple cross lamination (Collinson, 1986). Epsilon cross-beds are the result of deposition on point bars due to the lateral migration of a "meandering" river during flooding Fig. 3.4. Example of stacked channel-fill sub-facies (Ch). Note the channel lag and fining upward sequences. Location is Caprock Canyons (CRC-2). 68 2838" 10 Ch SSsssiS goo *> »" e_o_±0_ o ^ Ch S^VNS^^:^ ^r:^ r"\ ^ Ch s:=^^vS?«* SB _ ^•"^---ft. n n q o'o P » o " o O o o " o Ch lag fToo% 0%' I 1 mI I I I I rrn ss si cl vf vc gr cb 69 Fig. 3.5. Lateral accretion surfaces in point bar sub-facies. Note the clay drapes delineating the epsilon cross-beds. The sandstone is a green gray color that has been stained by red mudstones above it. Location is the Silverton Roadcut (SRC-1). Jacob staff is 2 meters long. 70 (Reineck, 1980). Epsilon cross-beds have also been reported in low sinuosity streams (Jackson, 1978; Miall, 1987). Point bar deposits show a fining upward sequence (Fig. 3.6). Siltstone or clay drapes may separate the individual sandstone layers, and represent deposition by receding floodwaters and decreasing flow rates. Even though sand grain sizes remain constant throughout the section in Fig. 3.6 there is an upwards cyclic increase in silt content. Sedimentary structures reflect an upward decrease in flow power, from planar bedding to ripple cross lamination. Point bar deposition is a periodic event. Building a point bar is a slow process, but the depositional events are very rapid. Each layer is deposited rapidly during a flood event, but it may be several years between each flood. The channel migrates during point bar deposition. Channel depths (bank-full) correspond to the height of the lateral accretion surfaces, but one must be careful when making this comparison. Most point bar deposits have been truncated, so only minimum depths may be estimated. A point bar exposed in Palo Duro Canyon (Fig. 3.7) is approximately 12 meters thick, thus minimum bank-full depth of the river that produced the point bar is equivalent to this. This point bar is very thick and has numerous epsilon cross-beds present. It displays the "classic" fining upward sequence and the sedimentary structures expected in a point bar deposit (Fig. 3.8). Point bars form in types of channels other than highly sinuous ones. Low sinuosity channels can contain epsilon cross-bedding (Jackson, 1978; Miall, 1987). The deposits of such channels are Fig. 3.6. Example of point bar sub-facies (PB). Note the lateral accretion surfaces, fining upward sequence, and the sedimentary structures reflecting decreasing flow power upward. Location is the Silverton Roadcut (SRC-1). 72 2868' :- 4 0 1 m I I I I I r~n vf vc gr cb o c u O 3 r— C3 •1— •— to o QJ O-r- C 01 (0 c a. j= •*-> r— • (O CU CO C -M CU O O T- •1- z o 4-> ta tJ H- 0) . 1 to in X) 1 a> 3 10 •!— to to o o m CU t. »f- j= U 1 -t-> UD Q. D» T- -O 3 J3 C . O •>- fO ^-^ x: <1^ X to o . Q. o c> > •r- CU t. ta u_ -a tj CJ 74 o Z o CO s o (O o in o ® O CO o CM _ O O CO — o Fig. 3.8. Example of a channel deposit with point bar sub-facies (PB). Note the thickness of the section, its fining upward sequence, and sedimentary structures displaying an upward decrease in flow power. This section closely fits the "classic" point bar model. Location is Palo Duro Canyon (PD-3). 76 3262* PB e5?? 0 ' ' 1 ml I I I I n~i vf vc Qr db 77 coarser grained and can lack the finer materials present in "classic" point bars. The sedimentary structures present are the same as those present in "classic" point bar models. A fining upward sequence, however, may not occur, but there may be a clay drape at the top of the sequence. Coarse-grained point bars can be divided into an upper and lower point bar. The upper point bar has chute bar sediments (Reineck, 1980). Chutes develop across the top surface of point bars during flood events. Chute bars are deposited at the downstream end of the point bar and produce thick sets of tabular cross-beds. Many times upper point bar deposits are not recognized, but this is not unusual because the upper point bar many times is not preserved. Chute bar deposits are not recognized in Dockum sediments. A coarse-grained point bar deposit is observed at the Wayside Crossing (CC-1) (see p. 96). It is composed of inclined planar bedding interpreted as epsilon cross-bedding. A channel lag of carbonate nodules occurs at its base. The lower part of the point bar deposit is a very fine to fine-grained sandstone with small green clay wisps and a slight fining upward sequence. The deposit laterally grades into red and green siltstone that contains inclined parallel lamination and climbing ripple cross lamination. These finer grained sediments represent the toe of the point bar and are probably channel-fill. Nami and Leeder (1978) described similar deposits with epsilon cross-beds laterally grading into fine-grained abandoned channel-fill sediments. The Dockum deposit is approximately 10 meters thick and 120 meters long in outcrop (this may not be a true measurement of the meanderloop 78 radius). This exposure occurs at an elevation of approximately 3031 feet at the measured section CC-1 on the northwest side of Highway 207. A transition zone may also occur in a point bar sequence. The transition zone occurs on the upstream side of a point bar. This is where stream velocity is highest near the inner bank, opposite of the typical point bar. At the transition zone the thalweg is switching from the outer bank of the upstream point bar to the outer bank of the downstream point bar. In the transition zone, grain size change is unclear and may coarsen upwards (Jackson, 1976; Collinson, 1986). Sedimentary structures may not be as well ordered as in the "classic" point bar sequence. Transition zone deposits were not confidently recognized in the Dockum, but it could explain rare coarsening upward packets observed in the sandstones. Point bar sub-facies can form over a wide range of sediment types and may display a wide range of variability within a single point bar (Reineck, 1980). Point bar sub-facies exposed in outcrop do not always show the same characteristics from location to location. Parts of one point bar deposit may not fit the "classic" point bar model. Variations are noted and strict application to the "classic" model is not always possible. Abandonment sub-facies. Abandonment sub-facies represent sediment that has filled stream channels after the stream has abandoned that channel course (Fig. 3.9). Channel abandonment can occur from chute cutoff, neck cutoff, or avulsion (Collinson, 1986). After the channels or portions of a channel are abandoned, overbank material from the stream is deposited in the empty channels. Clay, silt, and organic 79 --^T^JMj. af-Vn- Fig. 3.9. Clay plug (abandonment sub-facies) filling a stream channel. This deposit is in a stacked sequence of channel sands. Note layering of the red and gray mudstones and siltstones. Location is Palo Duro Canyon (PD-3). Hammer for scale (left-center). 80 matter are the filling sediments. These sequences are sometimes referred to as clay plugs. Clay and silt are most abundant in the center of the abandoned channel. Sandy plugs occur at the end of the channel fill deposit (Reineck, 1980). The abandonment sub-facies observed in the Dockum closely resemble the facies described above. They fill channels and are composed of laminated siltstone and mudstone with rare sandstone beds (Fig. 3.10). Sedimentary structures are parallel lamination, ripple cross lamination, and climbing ripple cross lamination. Sandstone, when present, is thinly bedded (5 - 20 cm) and laminated. Many of the clay plugs are covered or truncated by incision of other channels into them. Small-scale channel deposits. There are numerous channel deposits that are smaller in size than the channels described above. These small channel facies have many of the same characteristics of the larger channel facies. They show a fining upwards sequence, similar sedimentary structures, grain size, channel lag, and channel shape (Fig. 3.11). The small-scale channel deposits in the Dockum contain trough cross-beds, planar bedding, and rare ripple cross lamination. Grain sizes are, for the most part, slightly coarser (medium size sand) than in the larger channels. Channel shapes are similar, with concave upward bases, but the width to depth ratios are larger. The small channel deposits probably represent fixed channels that formed in floodplain materials and are not found in close proximity to stacked channel bodies. The streams that deposited these sediments were probably small tributaries feeding the larger streams (Frelier, 1987). Fig. 3.10. Example of abandonment sub-facies (Ab) and a capping channel-fill sub-facies (Ch). Note the fining upward sequence and the sedimentary structures present. Location is Palo Duro Canyon (PD-3). 82 3192' Ch Ab \ Ch fToo%" o%i « 1 ml I I I I r-m ss si cl vf vc gr cb Fig. 3.11. Example of a small-scale channel deposit (SCh). This deposit is very similar to the larger channel facies. Note the thalweg sub-facies (Th) and the floodplain sub-facies (Fp). Location is Crosby County (CBC-3). 84 2582' Fp SCh Th TH rnr-i vf vc gr cb 85 These smaller streams were probably eroding material from the floodplains of earlier deposits, thus many of the sandstones are more mature mineralogically and texturally than sandstones of larger channel deposits. Overbank Facies Overbank deposits comprise the majority of the sediments in the Dockum Group. Overbank sediments are deposited on a river floodplain as fluvial channels migrate laterally. They are also deposited between channels in a multi-channel system. Overbank sediments are deposited through the process of floodwaters spilling over or breaching channel banks and settling out of suspension when the waters subside. Overbank deposits can be classified as proximal or distal. Proximal overbank deposits in the Dockum are divided into three sub-facies: levee, crevasse splay, and sheetflow. Distal overbank deposits in the Dockum comprise one sub-facies, the floodplain sub-facies. Levee sub-facies. Levees are deposited on the flanks of a stream during flood events. After the waters overtop the banks the velocity drops and suspended materials are deposited near the banks. Levees are not only the result of the fall-out of coarse suspended load material, but also may be partly the result of the accretion of crevasse splay deposits (Collinson, 1986). Levees slope away from the channel onto the floodplain. Levees fine upward and laterally away from the channel. Typical sedimentary structures present in levees are ripple cross lamination, climbing ripple cross lamination, and parallel lamination. Generally a sand layer is overlain by a mud layer in 86 alternation (Reineck, 1980). Soils are sometimes recognized in levee deposits, because sedimentation only occurs during the flood stages of a stream. Levee sub-facies are rare in Dockum outcrops. They are only recognized in the Tecovas Formation. They occur proximal to the channel facies identified in the Tecovas. The levee sub-facies is composed of poorly cemented sandstone, silty sandstone, sandy siltstone, siltstone, and mudstone. Levee deposits in the Dockum show a fining upward sequence (Fig. 3.12). Sedimentary structures present are climbing ripple cross lamination and parallel lamination. Mud drapes are present and dip away from the channel facies as do the laminated sands and silts. Structures are difficult to recognize in the sands and silty sands because of poor cementation. Sediments in the levee facies have coloration associated with soil forming processes (orange brown, red, yellow, purple gray). The rare occurrence of levee deposits in the Dockum is not uncommon for fluvial environments. Proximal sand-rich levee deposits are of low preservation potential because of their susceptibility to erosion by channel migration (Collinson, 1986). Preserved levee deposits are associated with the fixed and sinuous type of channel. This type of stream does not migrate laterally widely compared to the other stream types. Thus, levees have a higher preservation potential than streams that migrate widely. Crevasse splay sub-facies. During floods the levee may be breached. When this occurs, crevasses are formed. A crevasse is a pathway through the levee and crevasse splays are tongues of sediment Fig. 3.12. Example of levee sub-facies (L) within floodplain sub-facies (Fp). Note the inclined lamination, fining upward sequence, sedimentary structures, and alternating sand and mud layers. Location is Palo Duro Canyon (PD-8). 88 3085' Fp Fp rToo% 0%' ' 1 r-T—I ss si cl vc gr cb 89 deposited on the floodplain. Crevasse splays may be sites of re-occurring deposition. Once the levee is breached it may stay open through successive floods allowing sediment through each time. Crevasse splay deposits are generally coarser grained than levee deposits (Reineck, 1980). Ripple cross lamination, climbing ripple cross lamination, and parallel lamination can be present in the crevasse splay sub-facies. Drifted plant material and fossil remains may be concentrated in crevasse splay deposits (Reineck, 1980). Crevasse splay sub-facies are recognized in the Dockum. These deposits are usually thin (1-2 meters) and are located within floodplain sequences. They are composed of sandstone, intrabasinal granules and pebbles, siltstone, and mudstone. Sedimentary structures recognized are small-scale trough cross-beds, massive bedding, ripple cross lamination, and parallel lamination (Fig. 3.13). Many of these deposits contain abundant plant debris. The plant debris consists of small branches, 5 to 10 cm long. Some of the plant debris is carbonized and well preserved, but some is mineralized with sulfur, sulfate, and iron oxides leaving only partial plant remains recognizable. Sheet flow sub-facies. Where a channel is unable to contain a major flood discharge, sheet flows may cover wide areas flanking the channel (Collinson, 1986). Sheet flow deposits are comprised of sand and some coarser sized material. They may be up to two meters thick, but are usually thinner. Sedimentary structures present are parallel lamination, climbing ripple cross lamination, and tabular and trough Fig. 3.13. Example of crevasse splay sub-facies (CSp) and floodplain sub-facies (Fp). Note the diverse lithologies, sedimentary structures, and the abundance of plant remains. Location is Palo Duro Canyon (PD-5). 91 3179" Fp ^•^•t^ CSp • .• . o .' ' o _ o ' • 'Q . • Fp •100% 0%l « im I I I I I I I I ss si cl vf vc gr cb 92 cross-beds. Sheet flow deposits can be laterally wide spread and can reflect the paleogeomorphology at the time of deposition. Sheet flows in the Dockum are common. They are comprised of fine to coarse sand with some amounts of intrabasinal granules and pebbles. These deposits are represented by the calclithic conglomerates (see Chapter IV). Sedimentary structures observed are massive bedding, parallel lamination, and small-scale trough cross-bedding. Many of these deposits accumulated under upper flow regime conditions as shown by the structures present. Thicknesses range from a few centimeters up to 25 centimeters. They are laterally extensive and may be up to 60 meters wide. Some of the sheet flow deposits contain fossil debris. The sheet flow deposits of the Dockum are believed to reflect the paleogeomorphology of the Dockum (Frelier, 1987). Many are horizontal, but some dip up to several tens of degrees. They can be concave upward, suggesting the blanketing of a small stream cut or even a small enclosed depression (discussed further in Chapter VI). Floodplain sub-facies. Floodplain sub-facies occur in the distal portion of overbank deposits. Floodplain deposits are produced by the sedimentation of suspended materials from flood waters that spilled over stream banks. Floodplain deposits are vertically accreted, and consist of clay, silt, and sand sized material with clay being dominant. Sedimentary structures present are parallel lamination, climbing ripple cross lamination, and small-scale trough cross-bedding. Depending on the climate, different types of floodplain deposits may occur. Backswamps and lakes are dominant in humid regions while 93 oxidized sediments and playa lakes may dominate in semi-arid regions (Reineck, 1980; Collinson, 1986). In humid regions, overbank sediments contain large quantities of organic matter and peats. Organic lacustrine deposits can occur along with coals of the backswamp area. Semi-arid regions produce red sediments produced from oxidation. Playa lakes may form, depositing evaporites. Deltaic deposits may be recognized in the lacustrine facies of both settings. Floodplain sub-facies observed in the Dockum are red or maroon in color, composed of mudstone or siltstone, and rarely display sedimentary structures. Massive mudstone comprises the majority of the sediments (Figs. 3.3, 3.11, 3.13). Within these mudstones are thin silty sand layers (1-2 cm) and locally abundant green and yellow reduction spots (discussed further in Chapter VI). Carbonate nodules are observed in many of the red mudstones. These nodules vary from 1 to 5 centimeters in diameter. They usually litter the weathered surface of the mudstone slopes, but may be observed in situ by digging into the mudstone. The nodules are scattered throughout the mudstone, and do not occur as a distinct bed or mass. These nodules are probably of pedogenic origin (discussed further in Chapter VI). Valley-Fill Facies Valley-fill deposits accumulate when streams erode into their channel creating a valley, and then deposit sediments, thus filling the valley. 94 Valley-fill deposits in the Dockum are unusual. They are confidently observed in three localities: Tule Canyon, Wayside Crossing (CC-1), and Palo Duro Canyon (near PD-5 and PD-7). These valley-fill deposits have concave upward bases and steeply dipping sides. The valley-fill sequence in Tule Canyon is the largest reported in the Dockum Group. It is 1.3 km wide and 64 m thick (Boone, 1979). The top portion of this deposit is in the Trujillo Formation, thus the entire deposit is considered Trujillo. It truncates the Tecovas Formation and cuts into the underlying Permian. The valley-fill is comprised of three units each with a distinct rock type (Boone, 1979). The lower unit is composed of granule to small cobble sized quartzite, quartz, and chert conglomerate. The base contains boulder size clasts which represent slump from the valley walls. The middle unit is a conglomeratic sandstone of the same composition as the lower unit (Boone, 1979). The upper unit is predominantly granular subchertarenite (Boone, 1979). Sedimentary structures observed in valley-fill deposits are trough cross-beds, tabular cross-beds, and planar bedding. A slight fining upward sequence was recognized by Boone (1979). Boone (1979) classified the valley-fill deposits as confined braided stream facies (refer to Boone, 1979 for a detailed description). In the present study, the deposits are classified as a confined low sinuosity stream after Jackson's (1978) nomenclature. Boone (1979) interpreted the valley incision and fill as a response to changing base level. He stated that the base level changes 95 are the result of a low-stand and a high-stand of the "Dockum lake" (see Chapter VI for further discussion). There are three different valley cut and fill sequences present at the Wayside Crossing (Fig. 3.14). First order bounding surfaces are observed at the base of each sequence. Sequences 2 and 3 incise into 1. Whether 3 cuts into 1 or vice-versa is impossible to determine because the top of the outcrop has been eroded. Lithologies of the three sequences at Wayside Crossing are similar. Medium grained sandstone, intrabasinal pebble-sized conglomerate, siltstone, and red mudstone are present. Sedimentary structures present are planar bedding, trough cross-beds, massive bedding, and epsilon cross-beds. Sequences 2 and 3 display epsilon cross-beds. Lateral accretion surfaces are recognized in sequence 2 (Figs. 2.14 and 3.14). At the base of sequence 3 are boulder size blocks of sandstone (Fig. 3.15). These blocks slumped into the channel from the valley walls. This indicates the valley flank sediments were lithified before the stream incised the valley. The valley-fill deposits are not simple sedimentation events, they are internally complex. Several different sandstone bodies and mudstones are present. The stream that deposited these sediments did not occupy the entire width of the valley, but probably migrated across the valley floor producing a series of scours and fills (Fig. 3.16). Seni (1978) described two valley-fill sequences in Palo Duro Canyon. He described them as sandstone and chert pebble conglomerate. One is located approximately .25 mile south of locality PD-5 and the other is located approximately 1.5 miles northeast of locality PD-7. CU (U JC O JZ o c +J (O CU LU 3 X cr o • CU ^— CU to CU u J3 c c f- *—^ r™ i- o I—( •1— 3 to 1 l*- to 1— CJ 1 ns o S". c ^.^ CU o to r— •r— CU o> ^ •4-J •r— c (O CU O •r- > S- o Z to m CD CD o CO 98 Fig. 3.15. Slumped sandstone boulders at the base of a valley-fill sequence that has eroded into channel abandonment sub-facies (mudstone in lower right). Note the sedimentary structures present in the slumped boulders. Location is the Wayside Crossing (CC-1). Hammer for scale. 99 Fig. 3.16. Multiple stream channel facies within a valley-fill sequence. The sandstones represent channel facies and the mudstone within the valley-fill is a clay plug. Note the concave upward base. Location is the Wayside Crossing (CC-1). Cedar trees at the top of the photo are approximately 2 meters tall. 100 The PD-5 valley-fill is 34 m thick while the PD-7 deposit is 25 m thick (Seni, 1978). Three other possible valley-fill deposits are present in the study area. They are extrabasinal conglomerates located at Silver Falls SF-1), Dickens County (DC-1), and Floyd County (FC-3). The conglomerates are identical compositionally and texturally to the conglomerates in valley-fills at Palo Duro Canyon and Tule Canyon. These conglomerates are poorly exposed, thus their geometry cannot be identified. The deposit in Floyd County (FC-3) is an abandoned gravel pit, thus the geometry, sedimentary structures, and any other characteristics are destroyed. Because the extrabasinal conglomerates and valley-fill deposits are so unusual, it is probable that they are related. The valley-fill sequences (except Wayside Crossing) and gravel deposits are roughly equivalent stratigraphically. They are located near the base of the Trujillo Formation. It is possible that the valley-fill sequence at Tule Canyon represents a major channel with the other valley-fills and gravel deposits being smaller tributaries feeding the main stream. The valley-fill sequences at the Wayside Crossing are not included in this hypothesis because they contain little extrabasinal conglomerate and are stratigraphically higher than the other deposits (located stratigraphically in the mid-section of the Trujillo). CHAPTER IV SANDSTONE PETROGRAPHY Methods Forty six sandstone samples were collected from Dockum outcrops. Specimens were taken from fresh exposures, cataloged, and prepared for petrographic analysis. The samples were taken from localities representative geographically and stratigraphically of the study area (see Figs. 1.1 - 1.3). Many of the samples were poorly cemented and had to be impregnated with epoxy before thin sections could be made. Thin sections were stained with Alizarin Red S solution to distinguish calcite and dolomite cements. Sandstone Mineralogy The predominant detrital grains in Dockum Group sandstones are quartz, feldspar, metamorphic rock fragments, sedimentary rock fragments, and muscovite. Less common constituents are chlorite, heavy minerals (zircon and tourmaline), and vertebrate bone fragments. The majority of the Dockum sandstones are comprised of quartz grains (up to 97 percent). Most of the sandstones range from 64 to 97 percent quartz. The quartz grains are of both monocrystal!ine and polycrystalline type. The monocrystalline grains are common "plutonic" quartz with straight to slightly undulose extinction. The polycrystalline varieties (Fig. 4.1) are of metamorphic and "stretched" metamorphic quartz. Individual crystals in metamorphic 101 102 Fig. 4.1. Photomicrograph of polycrystalline quartz grain. Note pink calcite and purple ferroan calcite. Field of view is 0.5 mm. Crossed polarizers. 103 polycrystalline quartz grains have straight to slightly undulose extinction, while those in stretched metamorphic quartz grains have strong undulose extinction. Feldspars comprise up to 10 percent of the detrital grains. In half of the samples feldspars total 1 percent or less of the framework grains. The abundance of feldspars generally increases from north to south in the study area. The greatest amounts are present in samples from Crosby County. Orthoclase, plagioclase, and microcline are the feldspar varieties present. Almost all of the orthoclase is untwinned; hence, features such as alteration, cleavage, and relief were used to distinguish orthoclase from quartz (Fig. 4.2). Metamorphic rock fragments comprise up to 28 percent of detrital grains, but average around 4 percent. Almost all of the metamorphic rock fragments are of schists. The schist fragments are composed of quartz crystals encompassing mica flakes with obvious foliation (Fig. 4.3). Sedimentary rock fragments consist of carbonate, claystone, and chert grains. In some samples, calcium carbonate rock fragments comprise up to 96 percent of the detrital grains. These grains are reworked from calcium carbonate nodules formed penecontemporaneously in the surrounding Dockum mudstones. Claystone fragments comprise up to 19 percent of the framwork grains. These grains were very ductile during transport and deposition. Many are squashed between the more rigid grains (Fig. 4.4). Chert grains comprise from 0 to 36 percent of the detrital grains, with the higher percentages coming from the Tule 104 Fig. 4.2. Photomicrograph of a typical orthoclase grain which is slightly altered and shows no twinning. Note the surrounding celestite and/or barite cement (yellow). Field of view is 1 mm. Crossed polarizers. 105 Fig. 4.3. Foliated quartz-mica schist rock fragment (center grain) and muscovite grain (upper left). Note the pink calcite cement. Field of view is 0.5 mm. Crossed polarizers. 106 Fig. 4.4. Photomicrograph of a squashed claystone fragment. Field of view is 0.5 mm. Crossed polarizers. 107 Canyon and Silver Falls areas. The chert grains are typically very well rounded. Miscellaneous (non-essential) constituents, where present, occur in trace amounts (single grains) and up to 3 percent of detrital grains. These grains include muscovite, chlorite, zircon, tourmaline, and vertebrate bone fragments. Muscovite is observed in almost all of the Trujillo sandstone samples with some sandstones containing 2 or 3 percent muscovite. Chlorite grains usually occur in trace amounts up to around 1 percent. The heavy minerals, zircon and tourmaline, are present in trace amounts in the Tecovas sandstones. Zircon was observed in about half of the Tecovas samples (Fig. 4.5) and tourmaline was identified in only one sample (Sample CBC-3B). These two ultra-stable heavy minerals are indicative of mature sandstones (Krumbein, 1963). Vertebrate bone fragments (Fig. 4.6) were observed in six samples. Sandstone Classification The classification scheme proposed by Folk (1974) was used to group the sandstones. Folk's standard point counting procedure was used with a minimum of one hundred counts of essential framework constituents per slide (Table 4.1). Half of the samples fell in the sublitharenite group while most of the other half fell in the litharenite group (Fig. 4.7). When all the rock fragments from the sublitharenites and the litharenites are recalculated to 100 percent and placed in the rock fragment daughter triangle all of the samples plot on the sedimentary and metamorphic 108 Fig. 4.5. Photomicrograph of a rounded zircon grain (center). Dark area between the grains is porosity. Note the chert grain in the lower right corner. Field of is 0.5 mm. Crossed polarizers. 109 Fig. 4.6. Photomicrograph of a vertebrate bone fragment (tan fragment in the center) in calclithic conglomerate. Field of view is 5 mm. Plane light. 110 Table 4.1. Point Count Data for essential constituents of Dockum Group Sandstones Saapit Formation Qtz Feld MRF CRF Clyst Chert P0-2C Tecovas 94 1 3 0 0 2 PD-2D Trujillo 75 0 1 24 0 0 PD-2E* Trujillo 32 0 3 56 9 0 P0-2F Trujillo 77 6 U 3 3 0 PD-3A* Tecovas 32 1 0 59 8 0 PD-3B Tecovas 94 1 0 0 5 0 PD-3C Trujillo 88 3 7 0 0 2 PD-30 Trujillo 78 3 0 0 19 0 P0-3E Trujillo 77 1 5 5 12 0 PD-3F* Trujillo 0 0 0 91 9 0 PD-3G* Trujillo 1 0 0 96 3 0 PD-3H* Tecovas 76 0 3 18 0 3 PD-4A* Tecovas 80 0 0 19 0 1 P0-4B Trujillo 76 0 0 15 9 0 P0-4C Trujillo 72 3 4 10 7 4 PD-4D Trujillo 83 2 10 3 0 2 PD-SB* Trujillo 41 0 0 58 0 1 PO-SC Trujillo 73 1 11 5 7 3 PD-5D* Trujillo 1 0 2 94 3 0 PD-6A Tecovas 92 0 3 0 0 5 PD-6B* Trujillo 1 0 0 99 0 0 P0-7A Trujillo 68 3 28 0 0 1 PD-7E Tecovas 88 1 0 0 0 U PD-8B* Tecovas 4 0 0 96 0 0 PD-8C Tecovas 97 1 1 0 0 1 CC-IA Trujillo 88 4 5 2 0 1 CC-IB* Trujillo 37 2 1 55 4 1 CC-IC Trujillo 68 4 1 22 3 2 CC-10 Trujillo 76 3 12 7 0 2 SRC-1A Tecovas 79 0 4 0 0 17 SRC-IC Tecovas 80 1 9 1 0 9 SRC-ID* Trujillo 5 0 0 95 0 0 SRC-1E* Trujillo 5 0 1 92 1 1 SRC-IF* Trujillo 1 0 0 99 0 0 FC-IB Trujillo 72 5 9 4 5 5 FC-2A Trujillo 64 2 2 18 0 14 TC-3A Tecovas 86 2 1 0 1 10 TC-4A Trujillo 64 0 0 0 0 36 TC-5A Tecovas 89 1 1 0 0 9 TC-5B Tecovas 92 4 0 0 0 4 SF-IA Trujillo 67 1 1 9 8 14 CBC-IA Tecovas 89 3 1 0 0 7 CBC-2B Trujillo 73 10 8 2 6 1 CBC-20 Tecovas 88 7 3 0 0 2 CBC-3A Trujillo 69 8 8 13 0 2 CBC-3B Trujillo 80 6 8 2 2 2 * Saiyles classified as calclithic conglooerates. Fig. 4.7. Mineralogical composition of Dockum sandstones showing compositional groups A, B, and C discussed in text. Classification scheme after Folk (1974). 112 SRF Q S*. Sh 27 points 8 point* VRF MRF RF-tr1angl« F/R ratio o Tecovas Formation (Group A) • Trujillo Formation (Group B) + Calclithic conglomerate (Group C) 1. Quartz arenite 2. Subarkose 3. Sublitharenite 4. Arkose 5. Lithic arkose 6. Feldspathic litharenite 7. Litharenite 113 sides showing that no volcanic rock grains are present. When the sedimentary rock grains are recalculated to 100 percent and plotted on the sedimentary daughter triangle a large scatter is produced, but most samples plot in the carbonate apex of the triangle. Three distinct groups can be discerned on Folk's ternary diagram. These three groups allow sandstones of the Tecovas and Trujillo formations to be separated petrographically. The first group (Group A) ranges from quartz percentages of 86 to 97 and includes sublitharenites (subchertarenites, subphyllarenites, and subclaystone-arenites), a quartz arenite, and a subarkose. This group is almost exclusively comprised of Tecovas sandstones (Fig. 4.8). Grain sizes average from 0.26 to 0.48 mm (medium sand). Grain sizes were calculated from the Friedman equation (1958) where equivalent sieve grain sizes are converted from those measured in thin section. Roundness is usually subangular to rounded, and the sandstones are well sorted. Almost all of these samples are classified as supermature. Relative maturity was calculated on the basis of clay matrix content, sorting, and roundness (as described by Folk, 1974). These sandstones are very different mineralogically from the other two groups (described below) in that they have more quartz, fewer feldspars, and usually high percentages of chert as rock fragments. Many of the samples also have trace amounts of zircon and tourmaline that are readily recognized in thin section. These heavy minerals are typically found in very mature sandstones (Krumbein, 1963). The Tecovas sandstones can be qualitatively distinguished from the Trujillo sandstones in hand samples. 114 Fig. 4.8. Photomicrograph of a typical Tecovas sandstone sample (Group A). Note the predominance of quartz grains and quartz overgrowths. Field of view is 1 mm. Crossed polarizers. 115 The Tecovas sandstones are quartz-rich, usually coarser grained, cleaner-looking, poorly cemented, and tannish brown in color. The second group (Group B) is composed almost exclusively of Trujillo sandstones (Fig. 4.9). Quartz percentages range from 64 to 83 and these sandstones are classified as sublitharenites, litharenites (phyllarenites, calclithites, chertarenites, and claystone-arenites), and a feldspathic litharenite. Roundness varies from subangular to subrounded with most of the samples being moderate to well sorted. Grain sizes average from 0.11 to 0.26 mm (very fine to fine sand), generally finer than the Tecovas sandstones. Sixty five percent of the Trujillo sandstones are submature and almost all of the remainder are mature. One of the most obvious differences between Groups A and B is that the Trujillo sandstones (Group B) have abundant calcium carbonate grains whereas the Tecovas sandstones contain none (with the exception of the calclithic conglomerates discussed below). Muscovite grains occur in trace amounts or abundantly in the samples of group B, but the Tecovas sandstones (Group A) generally lack muscovite. The Trujillo sandstones also have greater amounts of metamorphic rock fragments and less chert than those of the Tecovas. In hand samples, the Trujillo sandstones are finer grained, not as quartzose, clayey, better cemented, and green-gray in color. Five points plotted within the overlap between Groups A and B. The two points from Group A were very quartzose like the other Tecovas sandstones, but had enough chert grains to move them down from the quartz apex. Both are from the Silverton Roadcut locality. The three points from Group B are similar to the Trujillo sandstones, but are 116 Fig. 4.9. Photomicrograph of a typical Trujillo sandstone sample (Group B). Note the calcium carbonate grain in the upper middle part of the photo and the pink poikilotopic calcite cement engulfing the grains. Field of view is 5 mm. Crossed polarizers. 117 more mature. These samples are thought to represent sands reworked from earlier floodplain materials (see Chapter III). The last group (Group C) that is distinguishable on Folk's classification triangle is the calclithic conglomerate group (Fig. 4.10). This series of samples are comprised of granular sandstones to pebble conglomerates from both formations. They differ from those in the other two groups in consisting entirely of calcium carbonate granules and pebbles surrounded by a sandy matrix or calcite cement. The sandy matrix is similar in mineralogy, maturity, texture, and grain size to the other sandstone groups depending on the formation from which the conglomerate was derived. The calcium carbonate pebbles and granules are well rounded and range from 2 to 6 mm in diameter (very coarse sand to pebble size). The maturity ranges from submature to mature with many classified as bimodal mature (according to nomenclature of Folk, 1974). The granules and pebbles are exclusively calcium carbonate grains. Seni (1978) and Frelier (1987) describe these grains as being morphologically similar to larger calcite nodules seen in the several paleocaliche horizons in the Dockum. Frelier (1987) reported these grains as occurring in his fourth and fifth order channel bodies in quantities over 50 percent. These same statements can be applied to the calclithic conglomerates in this study area. It is concluded that these calcium carbonate grains are reworked nodules from paleosols formed penecontemporaneously in overbank materials. 118 Fig. 4.10. Photomicrograph of a calclithic conglomerate (Group C). Note the rounded calcium carbonate grains in a sandy matrix. Field of view is 5 mm. Crossed polarizers. 119 Cements Quartz overgrowths, clay coatings, calcite, ferroan calcite, dolomite, celestite and/or barite, manganese, and authigenic clays make up the cements recognized in the Dockum sandstones. Quartz overgrowths and calcite appear in almost all of the samples, while clay coatings, ferroan calcite, dolomite, and authigenic clays are less abundant. Celestite and/or barite and manganese cements each appear in one slide. Calcite is the most prominent cement. It has filled the primary porosities completely in many samples, while in others there are only small scattered patches of calcite cement. Most of the calcite is poikilotopic in nature, though minor amounts of other fabrics (schalenohedral and blocky equant) are observed in the calclithic conglomerates. These latter calcite cement fabrics indicate meteoric environments (Asquith, 1979). Quartz overgrowths are present in almost all of the samples. It is a minor cementing agent in most sandstones, however in some of the Tecovas samples it is the only cementing agent or the major one. These overgrowths are epitaxial on single grains or appear as several grains growing together. Dolomite and clay coatings both appear in the majority of samples, but only in minor amounts. Dolomite is observed replacing calcite and in some cases ferroan calcite. Clay coatings surround quartz grains (Fig. 4.11). Clay coatings are clays (pedogenic in nature) forming early in the diagenetic sequence, whereas authigenic clays (mentioned earlier) refer to kaolinite-dickite and illite pore filling cements occurring later in the diagenetic sequence. 120 Fig. 4.11. Photomicrograph of clay coating surrounding a monocrystalline quartz grain. Field of view is 0.5 mm. Crossed polarizers. 121 Ferroan calcite, kaolinite-dickite, and illite appear in fewer slides than dolomite. Ferroan calcite, where present, comprises up to 50 percent of the cement. Kaolinite-dickite occurs as booklets (Fig. 4.12) and illite as fibers. Kaolinite-dickite occurs principally as a primary void filling and generally is associated with feldspar grains showing some dissolution (Frelier, 1987). Many muscovite grains are observed disintegrating into illite fibers. Celestite and/or barite occurs in only one sample (Sample CBC-2C), a Tecovas sandstone, where it makes up 25 percent of the cement present, it has a poikilotopic fabric. Manganese also occurs in one sample (Sample PD-6A), a Tecovas sandstone. It has obliterated all of the pore space and is the only cement present in the slide. The diagenetic sequence of the Dockum sandstones is fairly simple and consistent throughout the study area and stratigraphic column, with only minor deviations. The general diagenetic sequence in order of occurrence is as follows: quartz overgrowths and clay coatings on detrital grains, calcite cement filling primary pore space, minor replacement of the calcite cement with dolomite cement, and varying amounts of dissolution of calcium carbonate cements. In several slides, ferroan calcite replaces the calcite before dolomite. The authigenic clays also occur last in the sequence, following calcite. A few of the Tecovas sandstones have only quartz overgrowths and clay coatings followed by authigenic clays. 122 Fig. 4.12. Photomicrograph of pore-filling kaolinite-dickite booklets. Field of view is 0.25 mm. Crossed polarizers. CHAPTER V PALEOCURRENT ANALYSIS Methods Paleocurrent measurements for this study were taken exclusively from large trough cross-beds (set height = greater than 20 cm). Large trough cross-beds were used because of their high value in indicating primary current flow directions (Miall, 1974; Potter and Petti John, 1977). Horizontal and gently-inclined parallel lamination is the major sedimentary structure present in Dockum sandstones, but these structures do not allow paleocurrent determination. There are several localities where numerous sets of large trough cross-beds are exposed and can easily be measured for directional qualities. Many sandstones, however, are poorly cemented and no sedimentary structures can be recognized. Ripple cross lamination and parting lineation are rare. Trough cross-beds exposed in 3 dimensions are used in this study. Trough cross-beds were studied in 3 dimensions to determine the actual direction of flow. Few of the sedimentary structures are exposed in the horizontal plane (Fig. 5.1), the majority are exposed in cross-section (Fig. 5.2). Not all 3-dimensional trough cross-beds exposed in cross-section were incorporated into the study. A true direction could not be discerned from many of the sedimentary structures because orientation in exposures did not allow a confident direction measurement. Current directions from horizontally 123 124 Fig. 5.1. Trough cross-bedding in plan view. Current direction is from the bottom to the top of the photo. Location in Palo Duro Canyon (PD-3). Hammer for scale. 125 Fig. 5.2. Trough cross-bedding in cross-section oblique to paleo-flow. Location in Crosby County (CBC-2). Jacob staff divisions are 20 cm. 126 exposed cross-beds were ascertained from the downdip direction of the beds (Fig. 5.1). Correction for structural dip is not needed. The Dockum beds are relatively horizontal with only about 4 ft. per mile slope to the south. Eighty-six measurements were taken and processed using the computer program "Direx." "Direx" was developed by Michael McLane as IBM compatible software. The program processes entered data and calculates a vector mean (x), length of the resultant vector (R), and the percent length of the resultant vector (L). Equations used in the program follow procedure explained by Potter and Petti John (1977). Rose diagrams can be constructed by the program and grouped into any number of classes. The performance of the program was satisfactory with calculations taking only a few hours. Previous Analyses Cazeau (1960) measured paleocurrent directions from cross-beds in Garza County and surrounding areas. Cazeau (1960) took 278 measurements from the Trujillo and Tecovas sandstones, but only his Trujillo measurements will be discussed here, because other workers measured only Trujillo sandstones. Cazeau (1960) calculated a vector mean of N 40 W (320 degrees) (Fig. 5.3). Kiatta (1960) measured paleocurrent directions from cross-beds in Texas. He reported a mean of N 61 W (299 degrees) from 90 measurements. Cramer (1973) measured paleocurrent directions from trough cross-beds in Trujillo outcrops from Oldham County, in the north, to Mitchell County, in the south. Cramer incorporated some of Cazeau's (1960) data into his 235 Fig. 5.3. Rose diagrams displaying paleocurrent directions from sandstones in the Trujillo Formation. Works from Cazeau (1960), Cram (1973), and Frelier (1987) are incorporated in this figure. Note the unimodal direction, the rough 180 degree spread, and the symmetry of the diagrams. 128 Cazeau (1960) x= N40W n = 200+ Cramer (1973) X = N 29 W n = 235 Frelier (1987) x = N2E n = 158 N —I 20 interval percent of n 129 measurements, but it is not clear how much of Cazeau's work was used. Cramer calculated a vector mean of N 29 W (331 degrees) (Fig. 5.3). Boone (1979) took 141 measurements from trough cross-beds exposed in the thick basal sandstones (his "meanderbelt system") of the Trujillo Formation located in the Tule Canyon area. He did not take any measurements from sandstone bodies higher in the section because he felt they were not representative of true current directions (see Chapter VI). Boone (1979) did not report a grand vector mean, but did graphically display his results (Fig. 5.4) showing a northeasterly paleocurrent direction. Frelier (1987) measured current directions (n = 158) from large trough cross-beds in a major sandstone body ("Macy Ranch sandstone") of the Trujillo Formation. He calculated a vector mean of N 2 E (2 degrees) (Fig. 5.3). He also calculated a vector mean from other major sandstone bodies in the Trujillo as N 9 W (351 degrees). Frelier's (1987) measurements came from exposures in Garza County. Results A total of 86 directional values for trough cross-beds were entered into the "Direx" program. The measurements were gathered from the sandstones of the Tecovas and Trujillo formations. The measurements from the Tecovas are very few (6) because the channel sandstone bodies are poorly cemented. Sedimentary structures are destroyed by weathering and cannot be detected in these friable sandstones. eu = JZ E -M CU •4-» OJ to 4-> >1 o to z 4-> ^—CU c•: jQ o s- >> (O O) CU E r— 3 - f- C7^ C r^ •1— (T> »-l on ) to •r- •C U •«-(O> c E o i- o o CQ IL. E o o r" i- r-" c.: *f- •1™ o •"-J •r- (O 3 ••-> 4J L. U «1— CU •o u L. •r- 4-> CU T3 c S CU O l~ >> i. ^— t- 3 CU (U O J= -»J O -t-> to (U la f— o (U (O .+-> J= Q. +j 4-> s_ c: o CU c: • r— ^ (O in• — L. 3 CU C7.> cr c •r- CU CU Li ^—' en 131 132 Rose diagrams are presented for Trujillo sandstones at 8 localities (Fig. 5.5). The diagrams are grouped into 12 azimuth classes (30 degrees each). Figure 5.5 shows vector means varying from northwest to northeast, but a general paleoslope to the north is apparent. The paleocurrent data collected at Tule Canyon is not from the same units that Boone (1979) measured. This explains the discrepancy between his results and those presented here. Paleocurrent directions were determined from sandstone units above the one Boone (1979) studied. The present study shows a general paleocurrent direction to the northwest while Boone (1979) reported a northeasterly direction. Measurements from the 8 localities are grouped in a rose diagram representing a total for the Trujillo (Fig. 5.6). A vector mean of N 15 W (345 degrees) was calculated along with a length value of 54 percent for the resultant vector. These results agree closely with earlier paleocurrent studies. There is some variance in this unimodal diagram, but the variance is symmetrical about the vector mean suggesting that the Dockum fluvial systems had sinuous channel patterns. Paleocurrent data from Tecovas sandstones give a vector mean quite different from the Trujillo sandstones (Fig. 5.6). The vector mean calculated for the Tecovas sandstones is S 33 W (213 degrees). This value differs 132 degrees from the Trujillo sandstones. The small number of measurements, however, make these findings tenuous. More measurements are needed before some confidence in the Tecovas paleocurrent direction can be assured. Measurements from both Dockum Fig. 5.5. Rose diagrams generated from paleocurrent data collected from Trujillo Formation sandstones. Note the general northerly direction. 134 ARMSTRONG RANDALL BRISCOE SWISHER MOTLEY DICKENS 30 miles ^ Triassic Outcrop 1. Palo Duro Canyon 5. Caprock Canyons 2. Ceta Canyon 6. Floyd County 3. Wayside Crossing 7. Silver Falls 4. Tule Canyon 8. Crosby County Fig. 5.6. Rose diagrams constructed for the Trujillo Formation, Tecovas Formation, and the Dockum Group total. Note the unimodal directions and synmetry of the Trujillo and Dockum total. 136 Trujillo sandstones x = N 15 W R = 43 L = 54% n = 80 Tecovas sandstones X = S 33 W R = 4 L = 68% n = 6 Total Dockum x = N 19 W R = 41 L = 47% N n = 86 —I 20 Interval percent of n 137 formations are incorporated into a single rose diagram and a grand vector mean was calculated (Fig. 5.6). All of the studies cited have determined a north to northwest regional paleo-slope for the Dockum and suggest that the fluvial systems had sinuous channel patterns. The north to northwest regional Dockum paleo-slope is nearly opposite to the present regional slope of the Dockum sandstones to the south at 4 ft. per mile (Fig. 2.15). This regional slope opposite to paleocurrent directions is probably due to post-Triassic deformation. The Amarillo Mountains were uplifted approximately 150 m in the Late Miocene (Budnik, 1987) and the Dockum Basin could have been subsiding after the close of the Triassic. CHAPTER VI DEPOSITIONAL ENVIRONMENTS Introduction Interpretation of sedimentary facies and their interrelationships is essential to understanding the environments in which the Dockum accumulated. Facies interpretations are, however, not the only factors used in determining ancient depositional environments. Tectonism, fluctuations of sea level, variation of climate, variation in source rock terrain or provenance, evolutionary patterns of organisms, variation in preservation within the geologic record, and diagenetic effects contribute to the identification of depositional environments and complicate the interpretations (Scholle and Spearing, 1982). Current model. For the last decade the accepted depositional model for the Dockum Group was that proposed by McGowen et al. (1979, 1983). McGowen et al. (1979, 1983) proposed a variety of depositional environments for the Dockum: braided and meandering streams, alluvial fans and fan deltas, distributary-type lacustrine deltas, ephemeral and long-lived lakes, and mud flats. Outcrop and subsurface data from the Texas Panhandle and eastern New Mexico was used to determine these depositional environments. A "Dockum lake" was postulated to have formed in the relict Midland Basin. This "lake" was the controlling factor for the deposition in the Dockum. Stream systems from the Ouachita uplift, Wichita Mountains, Amarillo Mountains, Glass Mountains, Sacramento Uplift, Sierra Grande Arch, and Sangre De Cristo 138 139 Uplift supplied sediments for the peripheral filling of the lake. The streams built lobate deltas into the lake. Alluvial fans and fan deltas prograded into the lake from the Glass Mountains and the Amarillo-Wichita mountain systems. Alternating periods of wet and dry climate caused cyclic sedimentation in the Dockum Group (McGowen et al., 1983). The alternation of climate caused lake levels to fluctuate and the geographic extent of the lake to contract and expand. When the lake contracted during dry climates, deltas prograded into the lake, streams incised into their channels, lake salinities increased, and mud flats appeared around the lake edges. During wet climates the lake expanded, transgressed over the deltas and river valleys, streams deposited sediments in their channels, lake salinities dropped, and mud flats were inundated. Dockum sandbodies are thought to represent delta distributary channels, delta foresets and topsets, and fluvial channels. Fine-grained Dockum sediments represent lacustrine deposits, mudflats, and minor overbank sediments. McGowen et al. (1979, 1983) described coarsening-upward sequences as representing deltaic environments. Gamma-ray well logs were used to identify facies in the subsurface sediments of the Midland Basin. Fine-grained elastics in the subsurface were interpreted as lacustrine while coarse-grained elastics are thought to represent fluvio-deltiac deposits. For further discussion of the "McGowen model" refer to McGowen et al. (1979, 1983) and Johns and Granata (1987). 140 An Alternate Depositional Model An alternate interpretation of a fluvial depositional setting for the Dockum is derived from the facies described in earlier chapters. Channel, point bar, valley-fill, abandonment, levee, crevasse splay, sheetflow, and floodplain facies indicate a fluvial environment for the Dockum. Stream patterns. Cross-sectional sandbody shapes were discussed in Chapter III, but not inferred stream patterns. Simple classification of streams into braided or meandering is no longer satisfactory (Jackson, 1978; Miall, 1987). Meandering streams were originally thought to carry fine-grained sediments, have prominent epsilon cross-beds, clay plugs, and levees. In contrast, braided streams carry coarse-grained material, have abundant scours, lack epsilon cross-beds, clay plugs, and levees. Jackson (1978) has shown that many of these "distinct" characteristics of each stream type appear together in "braided" and "meandering" streams. In ancient stream channels a distinct stream pattern cannot in many cases be determined using the "classic" criteria for braided and meandering streams. Many streams display braided characteristics at low discharge, but during flood stages (bank full) a meandering pattern is attained. The terms "braided" and "meandering" are not used in in this study to describe inferred channel patterns in the Dockum. High sinuosity and low sinuosity are more suitable general terms for describing channel patterns in the Dockum. Many workers (Asquith and Cramer, 1975; Seni, 1978; Boone, 1979; McGowen et al., 1979, 1983) have described "braided" and "meandering" 141 channels in the Dockum. Characteristics of these types of streams are observed in the sandstones of the Dockum, but in many cases features of both are observed in the same channel deposit. The majority of the channel deposits observed in the Trujillo Formation were probably deposited by highly sinuous mixed load streams. Streams of the Tecovas were probably deposited by bedload low sinuosity streams. Well developed point bar deposits, lateral accretion surfaces, widely distributed and symmetrical paleocurrent data, and thick overbank deposits suggest a high sinuosity for the Trujillo streams. Trujillo streams were probably of mixed load type with high amounts of fine-grained suspended sediment carried in the channels. Tecovas channels are interpreted as bedload streams because of the abundance of coarse-grained materials in the channel deposits and the absence of mud and clay. Confined and unconfined stream channel deposits are evident in both formations. Depositional history. In Late Triassic time the Dockum Basin (a relict of the Permian, Midland, and Palo Duro Basins) subsided in response to doming along the flank of the rift that was forming the incipient Gulf of Mexico (Salvador, 1987; McGowen et al., 1979, 1983; Wood and Walper, 1974). In response to this uplift, northwesterly flowing fluvial systems developed (Fig. 6.1). These fluvial systems flowed from the Ouachita Uplift area through the Dockum Basin in a northwesterly direction. The rivers may have continued on to the Triassic sea located to the northwest. Deposition of Dockum sediments began with the development of these fluvial systems. 4J O M- +-> -r- S- lO CU CU CU CD u c s- o 3 I— O (O to c: C (O CU ^ E t. •I- CU •a -«-> OJ to o . 0) sz m JZ CLCO O l-H E E O (O •> I- 4J • H- •rto- CU QJ •4-> to o I O C to CaU. CUu oxX: 2S: t- 4-C>U - 0 s_ CU CU T3 -IJ -tJ 01 o t- N z: (O •I— I— TJ (0 ' CU I- c: •!- (U ••- 4- CU lO -o CD aa o E-E « 3 i-H J>i£ t. .(JO) UJ o -a Q I— • 3 cn CU o •r- sz sz 143 O o u. O CD H CL o t I] o Q. Zi < u. O b o < O 5 144 The fluvial systems migrated laterally and deposited sediments across the entire basin. Smaller tributaries fed the major trunk streams. The major streams carried sediments from the exposed metamorphic interior zone of the Ouachita Uplift, while the tributaries eroded local Permian and Dockum sediments. As the streams laterally migrated the floodplain accreted vertically. At the beginning of the Late Triassic, channel sands were deposited in both confined and unconfined streams along with overbank deposits. In the Tule Canyon area the stream channel sands have an unconfined character resulting in a thick multistory sandbody. Later during Tecovas deposition equilibrium between sediment accumulation and erosion on the floodplain resulted in the formation of a paleosol. The paleosol is marked by a mottled purple mudstone and the local appearance of a thin and discontinuous silcrete (Fig. 2.10). This paleosol is similar to one described by Martins and Pfefferkorn (1988) in the Lower Triassic of Germany and that described as the "purple mottled unit" by Dubiel (1987). Silcretes are zones of silicification in pedological environments (Thirty and Millot, 1986). The Tecovas silcrete is present in a smectite-rich mudstone that is probably a product of devitrification of volcanic ash (clay discussed below). Much of the silica may have been derived from the ash. The silcrete is composed of detrital quartz grains cemented by microquartz and chalcedony. Some samples are exclusively composed of microquartz cement. Microquartz, chalcedony, and megaquartz are observed as pore-filling cements (Fig. 6.2). Silcretes can form in either humid or arid environments (Meyer 145 Fig. 6.2. Photomicrograph of Tecovas silcrete. Note the detrital quartz grains, microquartz cement, chalcedony, megaquartz, and iron oxide stain. Chalcedony and megaquartz are filling pores. 146 and Dos Reis, 1983; Ross and Chiarenzelli, 1985). Humid region silcretes show evidence of low pH conditions, such as the presence of authigenic anatase. Arid environment silcretes form from alkaline groundwaters with a high pH. They are characterized by vug-filling quartzine and silica (Ross and Charenzelli, 1985). The Dockum silcretes probably formed in an arid or semi-arid climate because of the large amounts of pore-filling silica and the absence of anatase. Deposition through the end of Tecovas time involved quartz-rich sands carried by the low sinuosity bedload streams and the deposition of large quantities of overbank material. Dockum and Permian sediments were exposed to erosion, as indicated by the presence of Permian siltstone clasts and Dockum materials (carbonate nodules, clay clasts, sandstone clasts) within some of the channels. Source areas for Tecovas sediments are thought to be in the Ouachita Uplift, but sparse paleocurrent data allow some question to remain. There is a change of sedimentation at the beginning of the deposition of the Trujillo. Channels change from the predominantly fixed variety of the Tecovas to more mobile type. Sandstone composition changes from quartzose in the Tecovas to more lithic-rich in the Trujillo. The majority of the lithic fragments are of metamorphic rocks suggesting that metamorphic terrains were exposed in the Ouachita Uplift during Trujillo time. Sedimentary rock terrains had been exposed during Tecovas time. At the beginning of Trujillo deposition, streams were unconfined and thick multistory sandstone bodies were deposited ("Trujillo sandstone"). Later the streams became more confined and produced large quantities of overbank materials. 147 Overbank environments. The majority of overbank sediments in the Dockum are "red beds." Colors of green, gray, yellow, and brown are also present. The red color of the mudstones probably resulted from early post-depositional modification. The red stain is observed coating detrital grains and in clay clasts found in channel deposits. This suggests that the colors had formed earlier on the alluvial plain, prior to later erosion of the floodplain by streams. The red color is probably due to the oxidation of detrital iron compounds (McBride, 1974). Oxidation presumably occurred during dry seasons when the water table was lowered sufficiently to permit decomposition of organic matter and oxidation of ferrous iron in a vadose environment (McBride, 1974). X-ray diffractometry was used to determine the types of clay minerals present in the red overbank mudstone. A bulk sample, oriented clay, and glycolated clay slides were analyzed. Clay minerals present in the overbank sediment are smectite, illite and kaolinite (Fig. 6.3). Other minerals present are quartz, calcite, dolomite, feldspar, and hematite. The smectite, kaolinite, and illite could be authigenic or detrital. The quartz and feldspar are detrital minerals while the calcite, dolomite, and hematite are authigenic. Green mudstones and siltstones are present as thin beds (2 - 20 cm) and small lenses (1 - 20 cm long). Some of the green mudstones may have been colored prior to deposition because the coloration does not cross bedding planes. The majority of green rocks have post-depositional coloring since the coloring crosses bedding planes. Green beds can form as thin layers (2 - 10 cm thick) under channel Fig.6.3. X-ray diffraction pattern of a red overbank mudstone sample from the Trujillo Formation. Note the types of clay minerals and , I detrital grains identified. 149 28 2 5 10 15 20 25 30 oriented sample glycolated sample 20 2 bulk sample 150 sands while others are small lenses and ellipsoids (reduction spots) within red mudstones. The green beds under channels are the result of the reduction of ferric iron in red beds by water seepage from open channels at the surface (McBride, 1974). Reduction spots and lenses along bedding planes are also the result of the reduction of ferric iron to ferrous iron. Carbonaceous materials are thought to be responsible for the reduction process (Mykura and Hampton, 1984; McBride, 1974). A major well-differentiated paleosol in the Tecovas Formation has been discussed above, but there are numerous minor paleosol horizons exposed in the Dockum sediments. These paleosol horizons are evident as zones of carbonate nodules in the mudstones. The carbonate occurs as discrete individual nodules, not as beds, but they do form in distinct horizons within the mudstones. These nodules are similar to calcretes described by Allen (1986) though they are not as well developed. Calcretes can be common in overbank deposits. When little deposition or erosion is occurring on the floodplain soils can develop. The carbonate nodules are present throughout both formations in the Dockum, but the Tecovas has more prominent horizons. These types of paleosols are formed in association with high pH soil waters of semi-arid and arid climates (Allen, 1986; Collinson, 1986). Paleogeomorphology. Sheetflow deposits from the Dockum streams mantled the alluvial plain and preserved a record of the paleogeomorphology of the Dockum (Frelier, 1987). Floodplain topography can be produced during periods when streams erode the alluvial plain. The scouring of the floodplains is accomplished by 151 base-level changes. When the base-level of a major stream drops, the tributaries and gullies in the drainage basin are affected. Scouring occurs within the drainage basin and a dissected landscape can be produced (Kraus and Middleton, 1987). During flood events, floodwaters spill over the stream banks and sheetflow deposits bury the surrounding floodplain. The sheetflows can be produced from the main trunk stream, its tributaries, or gullies that were filled during floods. Such sheetflow deposits mirror the topography that existed at the time of deposition. These scouring and sheetflow events can occur numerous times with younger scours cutting into older sheetflows and scours. At some point base-level may rise and sediments are deposited preserving the paleo-topography. The floodplain deposits of the Dockum record such scouring and filling events. Sheetflow deposits can be observed as flat beds, inclined beds, or bowl-shaped beds in cross-section. Some scours do not have sheetflow deposits preserved within them, but fine-grained sediments of a different color or consistency producing a contrast between the different sediments. Much of the ancient floodplain topography in the Dockum is preserved by sheetflow deposits. Many of the sheetflow deposits display concave upward slopes, thus reflecting the topography of a gully, small valley, or a small enclosed depression. Approximately one mile north of locality CBC-3 a bowl-shaped sheetflow deposit is exposed. Relief on the deposit is several meters and in map view it covers an area of approximately 0.25 square kilometer. From field observations it was determined that the 152 sheetflow deposit had blanketed a small enclosed depression. This depression was probably dry except during flood events when the sheetflow was deposited. Above the sheetflow deposits are massive red mudstones which represent filling of the depression during subsequent flood events. The flood events that deposited the mudstones were probably of less magnitude than those which resulted in sheetflow deposits because of the lack of coarse-grained material in the mudstone. Water did not stand in the depression long because lacustrine-type deposits are not present. Such enclosed depressions may not be related to base-level changes of the Dockum streams. An alternate mechanism for the formation of enclosed depressions could be dissolution of underlying Permian salts by groundwater during Triassic time. Today dissolution of underlying Permian deposits is responsible for the creation of many playa lakes in the Texas Panhandle (Reeves and Temple, 1986). The valley-fill deposits give excellent evidence for base-level changes in the Dockum streams. The size of many of these features (up to 60 m thick) indicates a large change in base-level. Base-level changes can occur from climatic processes, tectonic processes, and eustatic sea level changes. It is unknown which of these processes had the greatest impact on Dockum sedimentation. Paleoclimate. The Dockum climate was probably semi-arid to sub-humid (McGowen et al., 1979, 1983; Frelier, 1987). The paleosols identified in the Dockum are characteristic of these types of climates. The red coloration of overbank materials suggest alternating wet and dry periods (McBride, 1974). The development of carbonate 153 nodules suggest the early phases of calichification or formation of calcrete. Calcretes result in areas where precipitation is too low for an organic layer to form on the soil and waters have a high pH (Collinson, 1986). This combined with smaller volumes of percolating water, means that solution rates are relatively low and that water, rather than escaping into the groundwater system, is retained in the soil profile and eventually lost through evaporation. This results in precipitation of material from solution within the profile, under alkaline conditions (Collinson, 1986). Hence, the Dockum climate was probably not arid or arid long enough for the formation of a well developed calcrete. An arid climate for the Dockum is refuted by paleontological evidence and by lack of primary evaporite deposits (Frelier, 1987). Rare gypsum crystals are observed in some Dockum mudstones, but these could have been produced later diagenetically. Underlying Permian strata contain abundant gypsum veins that could have been a source for the gypsum found in the Dockum, through groundwater movement. Problematic Deposits Most of the sediments in the Dockum Group have been described and grouped into specific facies and indentified with depositional environments. One unit of the Dockum, however, does not display enough diagnostic features to confidently identify its environment of deposition. This unit is located in the Tecovas Formation and was described as "multicolored mudstone" in Chapter II (Figs. 2.11, 2.12). The unit at its thickest point is 32 m and is composed of 154 alternating bands of the colored mudstone (see Chapter II for a description). This unit, from field observations, was originally thought to be an overbank deposit. Samples of the mudstone were collected from localities in Palo Duro Canyon (PD-7) and Dickens County (DC-4) and analyzed by X-ray diffraction to determine the clay minerals present. Bulk sample, oriented clay, and glycolated clay slides were analyzed. The analysis show that the clay present is almost pure sodium-rich smectite (Fig. 6.4). Other interesting minerals identified are analcime and ankerite. The analyses for samples from Palo Duro Canyon are very similar to the ones from Dickens County. The smectite and zeolite are thought to be products of devitrification of volcanic ash (Eslinger and Pevear, 1988; Hay, 1978). Many workers have described similar units (High and Picard, 1965; Reynolds, 1970; Hay, 1978; Blakely and Gubitosa, 1983). The Petrified Forest Member of the Chinle Formation and the Popo Agie Member of the Chugwater Formation (both Late Triassic continental deposits) have similarly described units. Stewart et al. (1986) cited McGowen (1977) as proclaiming similarly described beds in the Dockum as having a volcanic origin. This unit displays features of both lacustrine and fluvial overbank environments. Neither environment is favored over the other in this study because of the lack of conclusive evidence for either. Both interpretations are presented in the discussion below. Lacustrine interpretation. Seni (1978) described these mudstones as being lacustrine in origin. Deposits containing devitrified volcanic ash and zeolites have been interpreted as lacustrine I n^ Fig. 6.4. X-ray diffraction patterns of yellow smectite-rich mudstone from a problematic deposit in the Tecovas Formation. Note the presence of smectite, analcime, and the relative size of the smectite peak. Smectite is the only clay mineral identified in this analysis. 156 t5 29 2 10 15 20 25 30 35 40 oriented sample glycolated sample 20 2 5 10 15 20 25 30 35 40 45 50 55 60 bulk sample 157 sediments. He interpreted the massive mudstones lacking carbonate nodules as lake bottom sediments and the carbonate nodule-rich mudstones as mudflats surrounding the lake. He postulated that the nodules represented calcrete horizons. In the Palo Duro Canyon area unusual carbonate concretions occur within the mudstones. They are composed of micritic and sparry calcite and dolomite. These concretions are observed with three different external shapes: cylindrical, botryoidal, and massive. They are completely unlike the carbonate nodules described elsewhere in this study as calcrete. The cylindrical concretions are from 3 to 8 cm in diameter and can be up to 40 cm long (Fig. 6.5). They appear to be similar to features described as "lungfish" burrows in the Late Triassic Dolores and Chinle Formations (Dubiel, Blodgett, and Bown,1987). McAllister (1988) disputed the indentification of these features as lungfish burrows, however, and claimed that lungfish fossil remains must be present in the burrows before confident identification can be ascertained. No fossil remains are observed in the "burrows" of the Dockum, thus they cannot be confidently identified as lungfish burrows. Many of the "burrows" are horizontal and merge together laterally. This is another feature that is not indicative of lungfish burrows. Whatever their origin, they were probably biogenically produced. The botryoidal concretions are from 5 to 15 cm in diameter, and resemble clusters of large grapes. The interiors of the concretions are composed of micritic calcite and sparry dolomite rhombs. These 158 Fig. 6.5. Possible "lungfish burrows" within smectite-rich mudstones of the Tecovas Formation. Note the vertical and horizontal orientations. Location is Palo Duro Canyon (PD-6). Field of view is approximately 2 meters. 159 concretions are not generally associated geographically with the other two types, but occur in approximately the same stratigraphic position. The massive concretions are the most abundant of the three. They measure approximately 5 to 20 cm in diameter. These concretions are massive and are composed of micritic calcite and dolomite. The exteriors are covered with burrow fillings approximately 1 cm in diameter. The burrow fillings consist of the same material as the concretion. Seni (1978) identified these burrows as the trace fossils Scoyenia and Teichichnus. Seni (1978) postulated that the "Dockum lake" that produced these deposits would expand and contact in accordance with wet and dry periods. This would explain the interbedding of "mudflat" sediments and "lake bottom" sediments. The smectite is sodium-rich, and thus saline conditions probably prevailed during its precipitation. The saline conditions could have occurred during the dry periods when the lake waters would be concentrated by evaporition; however, no evaporite deposits are preserved. The sodium could also have been supplied diagenetically by dissolution of underlying halite-rich Permian sediments. Sodium-rich groundwaters could have interacted with the clay. Sedimentary structures are absent from this unit. Lacustrine deltas and beach deposits are not present. No gray organic-rich mudstones characteristic of typical Triassic lacustrine deposits are present (see Gore, 1988). One of the major identifying characteristics of lacustrine deposits is also absent--fossils of freshwater organisms. 160 These beds are exposed in Potter, Randall, Armstrong, Dickens, and Crosby Counties and possibly surrounding counties outside the study area. If the sediments do indeed represent lacustrine deposits they probably record a series of shallow lakes covering these counties. Fluvial overbank interpretation. The same features described above, but interpreted differently can be applied to a fluvial overbank model. The mudstones could represent floodplain deposits with calcrete soils forming during periods of no erosion or deposition. Hay (1978) described zeolite-rich soils forming from volcanic ash. Smectite-rich soils can also form from these deposits (Eslinger and Pevear, 1988). The anomalous concretions could have formed pedogenically during these soil-forming periods. The cylindrical and massive concretions could have formed near the soil-air interface with the botryoidal type forming further down in the profile. At a locality in Crosby County (CBC-2) a definite lacustrine or • "pond" deposit is observed within the smectite-rich beds. The pond deposit consists of massive dark gray mudstone (Fig. 6.6). Viewed from atop adjacent hills the semi-circular shape of the deposit is apparent. Smectite and illite are present in these mudstones, but they differ mineralogically from the smectite-rich mudstones they are found within. The pond deposit contains large quantities of coprolites, vertebrate fossils, and fossil logs. Murray (1982, 1987) reported fossil remains of freshwater sharks in these deposits. Thin fossiliferous sandstones are also intercalated with the deposit. These sandstones probably represent deposits of small streams feeding the pond. Such a pond deposit is easily explained in a fluvial overbank 161 ^ / ^^ ' '^*/^'I ^''^^r--, ^: ,^ . ; - *»>'. .•^v> IL Fig. 6.6. Gray mudstone and a rare concretary carbonate in a pond deposit. The concretion is structureless and composed of dolomite. Location is Crosby County (CBC-2). Hammer for scale. 162 environment, however, it is difficult to postulate how pond sediments of different composition could be deposited within adjacent lacustrine sediments. It is possible that the main lake contracted during a dry period and that the pond sediments accumulated in depressions on the dry lake floor. However, pond sediments should be the same composition as the sediments deposited in the main lake if this occurs, but they are not. Source of volcanic ash. The question of which environment these sediments were deposited in remains open. A question also arises regarding the origin of the volcanic ash. These sediments may represent ash falls directly in the basin area or ash reworked and transported by streams. The source of the ash can only be postulated at this time. Two possible sources are: (1) the rift associated with the opening of the Gulf of Mexico and, (2) the postulated Triassic Magmatic Arc in western Mexico and the western United States (Stewart et al., 1987). The rift of the proto-Gulf of Mexico may be discounted because of the type of igneous rocks associated with it. The Eagle Mills Formation of Late Triassic and Jurassic is present in the rift. This formation is composed of diabase and basalt dikes and sills within red beds (Salvador, 1987). Basalts and diabase are not usually associated with large amounts of volcanic ash, but bimodal basalt-rhyolite associations have been reported in recent rift-related volcanism (Hall, 1987). No volcanic ash beds have been reported in the Eagle Mills, but little research has been conducted on this formation. Thus the rift, as a possible source, should not be completely discounted. The Triassic Magmatic Arc is the only other possible 163 source. However, this source is approximately 1000 kilometers away from the Dockum Basin and very little is known about it. Large volcanic eruptions must have occurred during Late Triassic time because of large volumes of devitrified volcanic ash beds reported in the literature from the western United States. These deposits are reported in the Chinle of the Colorado Plateau region, the Popo Agie of Wyoming, and the Dockum of Texas. Because of the thickness of these deposits it is likely that the majority of the volcanic ash was carried into the Dockum Basin by fluvial systems. The ash probably fell in the headlands of the stream systems. Some ash may have fallen within the study area, but probably much smaller amounts. Fluvial and Deltiac-Lacustrine Comparisons Dockum sediments have been interpreted as fluvial in origin in this study. In contrast, McGowen et al. (1979, 1983) describe them as deltaic and lacustrine in origin. Various features characteristic of the two paleoenvironmental models presented are examined and compared here. Reasons why these features are described, in this study, as fluvial instead of deltaic-lacustrine are also presented. Point bars and sheetflows versus deltas. Deltas are recognized by coarsening upward grain-size sequences, sedimentary structures showing increasing flow power upward, gradational bases, delta foreset bedding, and soft sediment deformation. Point bar deposits are recognized by fining upward grain-size sequences, evidence for decreasing flow power upward, erosional bases, lateral accretion surfaces, and sedimentary 164 structures oriented near normal to the dip of the epsilon cross-beds. Sheetflow deposits are thin sandstone units usually found in floodplain sediments. Point bar lateral accretion surfaces and sheetflow deposits in the Dockum are interpreted as delta foreset deposits by McGowen and others. Some of the deposits in the Dockum do not show significant grain-size changes within vertical sequences, but they do display erosional bases, evidence for decreasing flow power upward, and sedimentary structures oriented normal to the dip of the inclined beds (Fig. 6.7). Sheetflow deposits are sometimes deposited in succession, thus displaying a stacked appearance. Frelier (1987) described the "delta foresets" of McGowen et al. (1979, 1983) as sheetflow deposits and was able to trace them to a nearby channel sandbody. Many of the sheetflow deposits in this study are proximal to channel sandbodies. Many of the sheetflow units also truncate older sheetflows showing dipping beds changing strike within a few meters. These features are not described as delta foreset bedding in this report. Fluvial overbank versus lacustrine. Lacustrine deposits are recognized by fine-grained organic-rich sediments, carbonate layers, parallel laminations, wave ripples, beach deposits, lacustrine deltas, evaporite minerals, and most importantly--fossils of freshwater organisms (Picard and High, 1972). The characteristics of fluvial overbank deposits are fine-grained sediments, close proximity to fluvial channels, soil profiles, and small isolated pond deposits (Collinson, 1986). The fine-grained sediments of the Dockum more closely fit the criteria of fluvial overbank deposits (except the problematic sediments described above). The majority of these CU l_ +J (/) o c (O c c o XJ -r- I— IB r— (O -(-> to lO (O CU •!- Q.I— c c :$ CU I— C o 4-> CU S- cm (u O r— C Cf- ^ (O •r- O I— CU j_ (O 0) u 4-> o z to c CU CO T3 IB x: to u CU C h- c ra J3 -r- (U o -u I E > Crt (/» (O • CU o ••- 0 (/I XI o c o CU O I I— 3 c 3 o u to CU o E •«-> CU u o r- O >> l/t I CJ (- Q. Q. <+- s R3 CU CL. +J 0) a.tz (O •I- o I— > «/> E •a I. •— (U •!- X C T3 Crt X) (/) CI o •r- C Q. O (O o t- n) (O CU t. E •r" Q. s- a. ••-> CL o> to o) CU 1.1— o 0) C X] to • i/i o a. •4- -a c -o • $- -1- (U CU o (U r-~ o "o I— XJ •r" ^—' 1 -M x1> • to (O to U3x: CU to l/> E 1/1 cnx: •a o t- o • 3 -)-> CU t. o 4. cn o X3 U U_ CJ •r- I. o 166 167 sediments are red in color, not the gray of organic-rich sediments associated with most lake deposits. Many lacustrine deposits of gray organic-rich mudstone are reported in the Newark Supergroup (Late Triassic) of the Atlantic Coast (Gore, 1988). Some small and localized gray mudstones are observed in the Dockum and probably do indeed represent "pond" deposits. However, the only carbonate deposits observed are pedogenic carbonate nodules (see, however, description of problematic types). Sedimentary structures in the mudstones have been obliterated and the deposits are massive. No wave ripple cross lamination is observed in the sandstones. Beach and delta deposits are not observed in the Dockum. Evaporite minerals are rare in the Dockum, yet in saline lakes abundant evaporite minerals should be preserved. Fossils in the Dockum are predominantly of terrestrial or semi-aquatic vertebrates (reptiles and amphibians). The only fossils of freshwater organisms are unionid clam shells, fish, and freshwater sharks. These organisms are indicative of either fluvial or lacustrine settings. Paleocurrent data versus sandbody geometry. McGowen et al. (1979, 1983) and Johns and Granata (1987) constructed sandbody geometry maps from subsurface well logs. Sandstone percentages were used to construct these maps. McGowen et al. (1979, 1983) felt sedimentary structures are not good paleocurrent indicators and instead orientations of the sandbodies were used to determine paleoslopes. Their maps show sandstones (which represent channels) filling the Dockum Basin peripherally, thus supporting a closed basin and a lacustrine setting for the Dockum. In contrast, several paleocurrent studies (Chapter V), using sedimentary structures, in outcrop have 168 shown a general northwesterly paleoslope direction, and infer an open basin with streams flowing through to an exit in the northwest. Several workers agree on the validity of sedimentary structures as paleocurrent indicators (Potter and Pettijohn, 1977; Miall, 1974). The sandbody geometry maps constructed by McGowen et al. (1979, 1983) are interpretations and thus could be considered subjective. The maps could also be drawn showing sandbodies flowing through the basin to the northwest with flanking tributaries feeding the main streams. Suggestions for Future Research There is much need for additional research in the Dockum. The anomalous smectite-rich mudstones of the Tecovas need to be studied in greater detail to determine their depositional environment, source areas of the volcanic ash, and formation of the unusual carbonate concretions within them. The extrabasinal conglomerates and the valley-fill deposits associated with them need further study. The dimensions of these deposits are not known and their origin is poorly understood. The well-differentiated Tecovas paleosols were not studied in detail in this report. Apart from their use in stratigraphic correlation, inferring climates, and aiding in environmental interpretation, the origin of the paleosol needs further study. Geochemical studies of the paleosols should prove useful in better understanding the paleoclimate of the Dockum. Determination of a correct age for the aeolian deposits is also necessary. Its stratigraphic affinity is poorly understood. A more clearly delineated paleoslope for the Tecovas Formation is also needed. It may be 169 possible to find representative paleocurrent data at Tecovas outcrops outside the study area. The origin of base-level changes in the Trujillo Formation is uncertain at this time. A single mechanism or a combination of mechanisms could be responsible. Hence, much work remains to be done in order to comprehend the Dockum. CHAPTER VII CONCLUSIONS The study of the sedimentary facies present in the Dockum is useful in interpreting depositional environments. The presence of channel-fill, thalweg, point bar, levee, crevasse splay, and floodplain sub-facies indicate a fluvial setting. Valley-fill, sheetflow, and pond deposits, along with paleocurrent data support the fluvial model. Valley-fill deposits give evidence for flucuating base-levels in the Dockum. Depositional styles vary from the Tecovas to the Trujillo. The Tecovas stream channels were predominantly low sinuosity bedload streams. High sinuosity streams with mixed loads dominated in the Trujillo. The Trujillo has incised into the Tecovas creating an erosional contact. Sandstone compositions vary between the two formations with the Tecovas being more quartz-rich and the Trujillo more enriched in lithic grains. The Ouachita Orogenic Belt was probably the source area for Dockum sediments based on paleocurrent directions (N 19 W) and sandstone mineralogy, but sparce paleocurrent data for the Tecovas suggest that it could have had a different area. In Tecovas time, fluvial systems were filling the relict Midland and Palo Duro Basins. In Trujillo time, these relict basins subsided forming the Dockum Basin. This deformation was in response to doming along the flank of the rift forming the incipient Gulf of Mexico. Trujillo fluvial systems flowed northwest from the relict Ouachita 170 171 Uplift region through the Dockum Basin and probably onto the Triassic sea located in the northwest. The major streams carried sediment from the exposed metamorphic interior of the relict Ouachita Uplift, while the tributaries eroded local Permian and Triassic sediments. The Tecovas and Trujillo formations can be delineated throughout the study area and meet the criteria supporting their formational status. The term "group" should be applied to the Dockum and the Tecovas and Trujillo should retain formational status. The nomenclature proposed by Gould (1907) is the best that can be applied to the Dockum of Texas. This study disputes the theory that deltaic and lacustrine deposits dominate the Dockum. With the exception of the problematic deposit discussed in Chapter VI, the Dockum is entirely fluvial in origin. Deltaic deposits are not observed in the Dockum Group and lacustrine deposits (pond deposits) are small, localized, and associated with fluvial deposits. BIBLIOGRAPHY /Adams, A.E., MacMenzie, W.S., and Guilford, C, 1984, Atlas of sedimentary rocks under the microscope: J. Wiley and Sons, New York, 104 p. Adams, J.E., 1929, Triassic of West Texas: Amer. Assoc. Petr. Geol. Bull., V. 13, p. 1045-1055. Ahlbrandt, T.S. and Fryberger, S.G., 1982, Eolian deposits, in Scholle, P.A. and Spearing, D., eds.. Sandstone depositional environments: Amer. Assoc. Petr. Geol. Memior 31, Tulsa, p. 11-48. Allen, J.R.L., 1984, Sedimentary structures, their character and physical basis, in developments in sedimentology, v. 30: Elseveir, Amsterdam, 1256 p. Allen, J.R.L., 1986, Pedogenic calcretes in the Old Red Sandstone facies (Late Siluian - Early Carboniferous) of the Anglo-Welsh area. Southern Britain, in Wright, V.P., ed., Paleosols: Princeton Univ. Press, Princeton, p. 58-82. Allen, P.A., 1986, Lakes, in Reading, H.G., ed.. Sedimentary .environments and facies: Blackwell Scientific Publ., Oxford, p. 63-94. Ash, S.R., 1972, Upper Triassic Dockum flora of eastern New Mexico and Texas: New Mex. Geol. Soc. Guidebook, 23th Conf., p. 124-128. Asquith, G.B., 1979, Subsurface carbonate depositional models: a concise review: PennWell Books, Tulsa, 121 p. Asquith, G.B., and Cramer, S.L., 1975, Source and depositional environments of the Upper Triassic of the Texas High Plains: Geol. Soc. Amer., South Central Sect., 9th Annual Meeting. Asquith, G.B. and Cramer, S.L., 1975, Transverse braid bars in the Upper Triassic Trujillo Sandstone of the Texas Panhandle: J. Geology, v. 83, p. 657-661. Ayers, W.B., 1986, Lacustrine and Fluvial-Deltaic depositional systems. Fort Union Formation (Paleocene), Powder River Basin, Wyoming and Montana: Am. Assoc. Petr. Geol. Bull., v. 70, p. 1651-1673. Barnes, V.E. and Eifler, G.K., 1967, Geologic atlas of Texas, Lubbock sheet: Univ. Texas Bur. Econ. Geol. 172 173 Barnes, V.E. and Eifler, G.K., 1968, Geologic atlas of Texas, Plainview sheet: Univ. Texas Bur. Econ. Geol. Barnes, V.E. and Eifler, G.K., 1969, Geologic atlas of Texas, Amarillo sheet: Univ. Texas Bur. Econ. Geol. Blakley, R.C. and Gubitosa, R., 1983, Late Triassic paleogeography and depositional history of the Chinle Formation, southern Utah and northern Arizona, in Reynolds, M.W. and Dolly, E.D., eds., Mesozoic paleogeography of west-central United States: Soc. Econ. Paleont. Mineral., Rocky Mtn. Sect., p. 57-77. /^Boone, J.L., 1979, Lake margin depositional systems of the Dockum Group (Upper Triassic) in Tule Canyon, Texas Panhandle: Unpubl. Master's thesis Univ. Texas, 147 p. Bown, T.M. and Kraus, M.J., 1987, Integration of channel and floodplain suites, I. Developmental sequence and lateral relations of alluvial paleosols: J. Sed. Pet., v. 57, p. 587-601. Broadhead, R.F., 1984, Subsurface petroleum geology of Santa Rosa Sandstone (Triassic), northeast New Mexico: New Mex. Bur. Mines, Cir. 193, 23 p. Budnik, R.T., 1987, Late Miocene reactivation of Ancestral Rocky Mountain structures in the Texas Panhandle: A response to Basin and Range extension: Geology, v. 15, p. 163-166. Cant, D.J., 1982, Fluvial facies models, in Scholle, P.A. and Spearing, D., eds.. Sandstone depositional environments: Amer. Assoc. Petr. Geol. Memior 31, Tulsa, p. 115-138. Cazeau, C.J., 1960, Cross-bedding directions in Upper Triassic sandstones of West Texas: J. Sed. Pet., v. 30, p. 459-465. Cazeau, C.J., 1962, Upper Triassic deposits of West Texas and Northeastern New Mexico: Unpubl. Ph.D. dissertation, Univ. North Carolina, 94 p. Chatterjee, S., 1985, Postosuchus, a new thecodontian reptile from the Triassic of Texas and the origin of Tyrannosaurs: Phil. Trans. R. Soc. Lond., V. 309, p. 395-460. Chatterjee, S., 1986, The Late Triassic vertebrates: their stratigraphic and paleobiogeographical significance, in Padian, K., ed., The beginning of the age of dinosaurs: Cambridge Univ. Press, p. 139-150. Clemmensen, L.B., 1978, Lacustrine facies and stromatolites from the Middle Triassic of East Greenland: J. Sed. Pet., v. 48, p. 1111-1128. 174 Collinson, J.D., 1978, Vertical sequence and sand body shape in alluvial sequences, in Miall, A.D., ed.. Fluvial sedimentology: Can. Soc. Petr. Geol. Memoir 5, Calgary, p. 577- 586. Collinson, J.D., 1986, Alluvial sediments, in Reading, H.G., ed.. Sedimentary environments and facies: Blackwell Scientfic Publ., Oxford, p. 20-62. Collinson, J.D., 1986, Deserts, in Reading, H.G., ed.. Sedimentary environments and facies: Blackwell Scientific Publ., Oxford, p. 95-154. Cramer, S.L., 1973, Paleocurrent study of the Upper Triassic Sandstone, Texas High Plains: Unpubl. Masters thesis. West Texas State Univ., 31 p. DeLuca, J.L. and Eriksson, K.A., 1986, Synchronous ephemeral and perennial stream sedimentation in the Middle Chinle formation, northeastern New Mexico: Geol. Soc. Amer., Abstracts with Programs, v. 18, p. 582. Donahue, J.J., 1985, Topographic inversion of badland buttes in Montana: Geol. Soc. Amer. Bull., v. 96, p. 911-917. -^Drake, N.F., 1891, Stratigraphy of the Triassic formation of northwest Texas: Texas Geol. Surv. 3th Annual Rept., p. 227-247. Dubiel, R.F., 1987, Sedimentology of the Upper Triassic Chinle Formation, southeastern Utah: Paleoclimatic implications: J. Arizona-Nevada Academy of Sciences, v. 22, p. 35-45. Dubiel, R.F., Blodgett, R.H., and Bown, T.M., 1987, Lungfish burrows in the Upper Triassic Chinle and Dolores formations, Colorado Plateau: J. Sed. Pet., v. 57, p.512-521. Dubiel, R.F., Blodgett, R.H., and Bown,T.M., 1988, Lungfish burrows in the Upper Triassic Chinle and Dolores Formations, Colorado Plateau - Reply: J. Sed. Pet., v. 58, p. 367-369. Dunay, R.E. and Fisher, M.J., 1974, Late Triassic palynofloras of North America and their European correlatives: Review of Palaeobotany and Palynology, v. 17, p. 179-186. Dunay, R.E. and Fisher, M.J., 1978, The Karnian palynofloral succession in the northern Calcareous Alps, Lunz-Am-See, Austria: Pollen et Spores, V. 20, p. 177-187. y Dunay, R.E. and Fisher, M.J., 1979, Palynology of the Dockum Group (Upper Triassic) Texas, U.S.A.: Review of Paleobotany and Palynology, v. 28, p. 61-92. 175 Edler, R.L., 1987, Taphonomy and paleoecology of the Dockum Group, Howard County, Texas: J. Arizona-Nevada Academy of Sciences, v. 22, p. 85-94. Eslinger, E. and Pevear, D., 1988, Clay minerals for petroleum geologists and engineers: Soc. Econ. Paleont. Mineral. Short Course Notes no. 22, 413 p. Folk, R.L., 1968, Petrology of sedimentary rocks: Hemphill's Book Store, Austin, TX, 170 p. Fouch, T.D. and Dean, W.E., 1982, Lacustrine environments, in Scholle, P.A. and Spearing, D., eds.. Sandstone depositional environments: Amer. Assoc. Petr. Geol. Memoir 31, p. 87-114. ^Frelier, A.P., 1987, Sedimentology, fluvial paleohydrology, and paleogeomorphology of the Dockum formation (Triassic), West Texas: Unpbubl. Master's thesis, Texas Tech Univ., 198 p. Gore, P.J.W., 1988, Paleoecology and sedimentology of a Late Triassic lake, Culpeper Basin, Virginia, U.S.A.: Paleogeo., Paleoclim., Paleoecol., v. 62, p. 593-608. Gould, C.N., 1907, The geology and water resources of the western portion of the Panhandle of Texas: U.S.G.S. Water Supply Paper, v. 191, p. 21-29. Granata, G.E., 1981, Regional sedimentology of the Late Triassic Dockum Group, West Texas and eastern New Mexico: Unpubl. Master's thesis, Univ. Texas, 198 p. /Green, F.E., 1954, The Triassic deposits of northwestern Texas: Unpubl. Ph.D. dissertation, Texas Tech College, 196 p. Gregory, J.T., 1972, Vertebrate faunas of the Dockum Group, Triassic, eastern New Mexico and West Texas: New Mex. Geol. Soc. Guidebook, 23th Conf., p. 120-124. Hay, R.L., 1978, Geologic occurrence of zeolites, in Sand, L.B. and Mumpton, F.A., eds.. Natural zeolites - Occurrence, properties, use: Pergamon Press, New York, p. 135-143. Hall, A., 1987, Igneous petrology: J. Wiley and Sons, New York, p. 1-23. High, L.R. and Picard, M.D., 1965, Sedimentary petrology and origin of analcime-rich Popo Agie Member, Chugwater (Triassic) Formation, west-central Wyoming: J. Sed. Pet., v. 35, p. 49-70. Hood, C.H., 1977, Geology of Fortress Cliff Quadrangle, Randall County, Texas: Unpubl. Master's thesis. West Texas State Univ., 123 p. 176 Hubert, J.F. and Hyde, M.G., 1983, Sheet-flow deposits of graded beds and mudstone on a alluvial sandflat-playa system: Upper Triassic Blomidon redbeds, St. Mary's Bay, Nova Scotia: Sedimentology, v. 29, p. 457-474. Jackson, R.G., 1976, Depositional model of point bars in the lower Wabash River: J. Sed. Pet., v. 46, p. 579-594. Jackson, R.G., 1978, Preliminary evaluation of lithofacies models for meandering alluvial streams, in Miall, A.D., ed.. Fluvial sedimentology: Can. Soc. Petr. Geol. Memoir 5, Calgary, p. 543-576. /Johns, D.A., 1986, Role of structure in the distribution of sandstone in the Dockum Group, Texas Panhandle: Geol. Soc. Amer., Abstracts with Programs, v. 18, p. 646. i/ Johns, D.A. and Granata, G.E., 1987, Regional tectonic setting and depositional trends of the Dockum Group, West Texas and eastern New Mexico: J. Arizona-Nevada Academy of Sciences, v. 22, p. 53-72. Kelley, V.C., 1972, Triassic rocks of the Santa Rosa country, in New Mex. Geol. Soc. Guidebook, 23th Conf., p. 84-90. Kiatta, H.W., 1960, A provenance study of the Triassic deposits of northwestern Texas: Unpubl. Masters thesis, Texas Tech College, 63 p. Kraus, M.J., 1987, Integration of channel and floodplain suites, II. Vertical relations of alluvial paleosols: J. Sed. Pet., v. 57, p. 602-612. Kraus, M.J. and Bown, T.M., 1986, Paleosols and time resolution in alluvial stratigraphy, in Wright, V.P., ed., Paleosols: Princeton Univ. Press, Princeton, p. 180-207. Kraus, M.J. and Middleton, L.T., 1987, Dissected paleotopography and base-level changes in a Triassic fluvial sequence: Geology, v. 15, p. 18-21. Krumbein, W.C. and Sloss, L.L., 1963, Stratigraphy and sedimentation: Freeman and Co., San Francisco, 660 p. Kuenzi, W.D., Horst, O.H., and McGehee, R.V., 1979, Effect of volcanic activity on fluvial-deltaic sedimentation in a modern arc-trench gap, southwestern Guatemala: Geol. Soc. Amer. Bull., v. 90, p. 827-838. Leopold, L.B. and Miller, J.P., 1964, Fluvial processes in geomorphology: Freeman and Co., San Francisco. 177 Lucas, S.G. and Hunt, A.P., 1987, The age of the Anton Chico and Santa Rosa Formations of north-central New Mexico: J. Arizona-Nevada Academy of Science, v. 22, p. 21-33. Lucas, S.G., Hunt, A.P., and Hayden, S.N., 1985, The Triassic system in the Dry Cimarron Valley, New Mexico, Colorado and Oklahoma: New Mex. Geol. Soc. Guidebook 36th Conf., p. 97-115. Lucas, S.G., Hunt, A.P., and Morales, M., 1985, Stratigraphic nomenclature and correlation of Triassic rocks of East-Central New Mexico: A preliminary report: New Mex. Geol. Soc. Guidebook 36th Conf., p. 171-196. Lucas, S.G., Oakes, W., and Froehlich, J.W., 1985, Triassic microvertebrate locality, Chinle Formation, East-Central New Mexico: New Mex. Geol. Soc. Guidebook 36th Conf., p. 205-212. Martins, U.P. and Pfefferkorn, H.W., 1988, Genetic interpretation of a Lower Triassic paleosol complex based on soil micromorphology: Paleogeo., Paleoclim., Paleoecol., v. 64, p. 1-14. Matthews, W.A., 1969, The geologic story of Palo Duro Canyon: Univ. Texas Bur Econ Geol., Guidebook 8, 49 p. McAllister, J.A., 1988, Lungfish burrows in the Upper Triassic Chinle and Dolores Formations, Colorado Plateau - Comments on the recognition criteria of fossil lungfish burrows: J. Sed. Pet., v. 58, p. 365-366. McBride, E.F., 1974, Significance of color in red, green, purple, olive, brown, and gray beds of Difunta Group, northeastern New Mexico: J. Sed. Pet., v. 44, p. 760-773. ^ McGowen, J.H., Granata, G.E., and Seni, S.J., 1979, Depositional framework of the lower Dockum Group (Triassic), Texas Panhandle: Texas Bur. Econ. Geol. Rept. Inv. 97, 60 p. I McGowen, J.H., Granata, G.E., and Seni, S.J., 1983, Depositional setting of the Triassic Dockum Group, Texas Panhandle and eastern New Mexico, in Reynolds, M.W. and Dolly, E.D., eds., Mesozoic paleography of west-central United States: Soc. Econ. Paleont. Mineral., Rocky Mtn. Sect., p. 13-38. Meyer, R. and Pena Dos Reis, R.B., 1985, Paleosols and alunite in continental Cenozoic of western Portugal: J. Sed. Pet., v. 55, p. 76-85. Miall, A.D., 1974, Paleocurrent analysis of alluvial sediments: A discussion of directional variance and vector magnitude: J. Sed. Pet., V. 44, p. 1174-1185. 178 Miall, A.D., 1987, Recent developments in the study of facies models, in Ethridge, F.G., Flores, R.M., and Harvey, M.D., eds.. Recent developments in fluvial sedimentology: Soc. Econ. Paleont. Mineral. Spec. Publ. 39, Tulsa, p. 1-9. /Murray, P., 1982, Biostratigraphy and paleoecology of the Dockum Group (Triassic) of Texas: Unpubl. Ph.D. dissertation. Southern Methodist Univ., 459 p. /Murray, P., 1987, Notes on the stratigraphy and paleontology of the Upper Triassic Dockum Group: J. Arizona-Nevada Academy of Sciences, v. 22, p. 73-84. Mykura, H. and Hampton, B.P., 1984, On the mechanism of formation of reduction spots in the Carboniferous/Permian red beds of Warwickshire: Geol. Mag., v. 121, p. 71-74. Nami, M. and Leeder, M.R., 1978, Changing channel morphology and magnitude in the Scalby Formation (M. Jurrassic) of Yorkshire, England, in Miall, A.D., ed.. Fluvial sedimentology: Can. Soc. Petr. Geol. Memoir 5, Calgary, p. 431-440. VPicard, M.D., 1965, Iron oxides and fine-grained rocks of Red Peak and Crow Mountain Sandstone Members, Chugwater (Triassic) Formation, Wyoming: J. Sed. Pet., v. 35, p. 464-479. Picard, M.D. and High, L.R., 1972, Criteria for recognizing lacustrine rocks, in Rigby, J.K. and Hamblin, W.K., eds.. Recognition of ancient sedimentary environments: Soc. Econ. Paleont. Mineral. Spec. Publ. no. 16, p. 108-145. Potter, P.E. and Pettijohn, F.J,, 1977, Paleocurrent and basin analysis: Springer-Verlag, Berlin, 296 p. Reading, H.G., 1986, Facies, in Reading, H.G., ed.. Sedimentary environments and facies: Blackwell Scientific Publ., Oxford, p. 4-19. Reeside, J.B., 1957, Correlation of the Triassic formations of North America exclusive of Canada: Geol. Soc. Amer. Bull., v. 68, p. 1451-1514. Reeve, S.C, 1975, Paleomagnetic studies of sedimentary rocks of Cambrian and Triassic age: Unpubl. Ph.D. dissertation, Univ. Texas at Dallas, 278 p. Reeves, CC, 1976, Caliche: Estacado Books, Lubbock, 233 p. Reeves, CC and Temple, M., 1986, Permian salt dissolution, alkaline lake basins, and nuclear waste storage. Southern High Plains, Texas and New Mexico: Geology, v. 14, p. 939-942. 179 Reineck, J.B. and Singh, I.B., 1980, Depositional sedimentary environments: Springer-Verlag, New York, p. 262-314. Ross, G.M. and Chiarenzelli, J.R., 1985, Paleoclimatic significance of widespread Proterozoic silcretes in the Bear and Churchill provinces of northwestern Canadian Shield: J. Sed. Pet., v. 55, p. 196-204. Roth, R., 1943, Origin of siliceous Dockum conglomerates: Am. Assoc. Petr. Geol. Bull., v. 27, p. 622-631. Salvador, A., 1987, Late Triaasic-Jurassic paleogeography and origin of Gulf of Mexico Basin: Amer. Assoc. Petr. Geol. Bull., v. 71, p. 419-451. /Scholle, P.A., 1979, A color illustrated guide to constituents, textures, cements, and porosities of sandstones and associated rocks: Am. Assoc. Petr. Geol. Memoir 28, 201 p. Scholle, P.A. and Spearing, eds., 1982, Sandstone depositional environments: Amer. Assoc. Petr. Geol. Memior 31, Tulsa, 410 p. Schultz, L.G., 1963, Clay minerals in Triassic rocks of the Colorado Plateau: U.S.G.S. Bull. 1147-C, 71 p. Schumm, S.A., 1972, Fluvial paleochannels, in Rigby, J.K. and Hamblin, W.K., eds.. Recognition of ancient sedimentary environments: Soc. Econ. Paleont. Mineral. Spec. Publ. no. 16, p. 98-107. Sellards, E.H., Adkins, W.S., and Plummer, F.B., 1932, The geology of Texas, Volume I: Stratigraphy: Univ. Texas Bull., no. 3232, p. 242-253. Sellards, E.H. and Baker, C.L., 1934, The geology of Texas, Volume II: Structural and economic geology: Univ. Texas Bull., no. 3401, p. 390-399. Selley, R.C, 1976, An introduction to sedimentology: Academic Press, New York, 408 p. /Seni, S.J., 1978, Genetic stratigraphy of the Dockum Group (Triassic), Palo Duro Canyon, Panhandle, Texas: Unpubl. Master's thesis, Univ. Texas, 157 p. Shepard, R.G., 1987, Lateral accretion surfaces in ephemeral-stream point bars, Rio Puerto, New Mexico, in Ethridge, F.G., Flores, R.M., and Harvey, M.D., eds.. Recent developments in fluvial sedimentology: Soc. Econ. Paleont. Mineral. Spec. Publ. 39, Tulsa, p. 93-98. 180 /sidwell, R., 1945, Triassic sediments in West Texas and Eastern New Mexico: J. Sed. Pet., v. 15, p. 50-54. Spiegel, Z., 1972, Problems of the Triassic stratigraphy in the Canadian River Basin, Quay, San Miguel, and Guadalupe Counties, New Mexico: New Mex. Geol. Soc. Guidebook, 23th Conf., p. 79-83. Stewart, J.H., J.H., Poole, F.G., and Wilson, R.F., 1972, Stratigraphy and origin of the Chinle Formation and related Upper Triassic strata in the Colorado Plateau region: U.S.G.S. Prof. Paper 690, 336 p. Stewart, J.H., Anderson, T.H., Haxel, G.B., Silver, L.T., and Wright, J.E., 1986, Late Triassic paleogeography of the southern Cordillera: The problem of a source for voluminous volcanic detritus in the Chinle Formation of the Colorado Plateau region: Geology, v. 14, p. 567-570. Summerfield, M.A., 1984, Isovolumetric weathering and silcrete formation. Southern Cape Province, South Africa: Earth Surface Proc. and Landforms, v. 9, p. 135-141. vThiry, M. and Millot, G., 1987, Mineralogical forms of silica and their sequence of formation in silcretes: J. Sed. Pet., v. 57, p. 343-352. Wood, M.L. and Walper, J.L., 1974, The evolution of the interior Mesozoic Basin and the Gulf of Mexico: Gulf Coast Assoc. Geol. Soc. Trans., v. 24, p. 31-41. I PERMISSION TO COPY In presenting this thesis in partial fulfillment of the requirements for a master's ciegree at Texas Tech University, I agree that the Library and my major department shall make it freely avail able for research purposes. Permission to copy this thesis for scholarly purposes may be granted by the Director of the Library or my major professor. It is understood that any copying or publication of this thesis for financial gain shall not be allowed without my further written permission and that any user may be liable for copy right infringement. Disagree (Permission not granted) Agree (Permission granted) Student's signature Student's signature ' p-z-n Date Date Q. o (O tl) • O ^ O) to '"s era t—I to •!- to c -a o •I- ns o c i_ U- OJ tJ 3 o, 43