1

1 Temporal variations in fluvial architecture, paleohydrology, and detrital

2 composition of the Upper Triassic Chinle Formation in southwestern Utah, U.S.A.

4 K. Shibataa,*, Billy G. Adhiperdanab, M. Itoc, A.R.C. Milnerd, M.G. Lockleye Deleted: .

a
6 Yokosuka City Museum, 95 Fukadadai Yokosuka, Kanagawa 238-0016 Japan

b
7 Department of Geology, Padjadjaran University, Jatinangor 45363 Indonesia

c
8 Department of Earth Sciences, Chiba University, Chiba 263-8522 Japan

d
9 St. George Dinosaur Discovery Site at Johnson Farm, 2180 East Riverside Drive, St.

10 George, UT 84790 U.S.A.

e
11 Dinosaur Trackers Research Group, University of Colorado Denver, PO Box 173364,

12 Denver, CO, 80217-3364 U.S.A.

13

14 * Corresponding author.

15 E-mail address: kenichirou-shibata@city.yokosuka.kanagawa.jp (K. Shibata)

16

17 Keywords: fluvial architecture, paleohydrology, clay mineral, climate, Triassic, Chinle

18 Formation, U.S.A.

1
 2

20 Running head: Fluvial architecture, paleohydrology, and detrital composition of the

21 Chinle Formation

22

23 ABSTRACT

24 Late Triassic humid climate of the western Pangaea that was affected by

25 monsoonal circulation is interpreted to have collapsed after 215 Ma. However, detailed

26 analyses of Late Triassic depositional systems and paleoclimates have been conducted

27 mainly based on geologic data from the Four Corners region in western U.S.A. The

28 present study analyzed spatial and temporal variations in fluvial architecture,

29 paleohydrological features, and detrital composition of the Chinle Formation in

30 southwestern Utah, which consists of the Shinarump Member, Cameron Member, and

31 “purple pedogenic beds” in ascending order. The studied formation is correlated to the

32 strata in the Four Corners region and is older than 215 Ma. Architectural analyses of the

33 Chinle Formation fluvial deposits revealed that the fluvial style of the Shinarump

34 Member was represented by braided channels, and that of the Cameron Member and

35 “purple pedogenic beds” by single thread, stable and fixed channels with wide

36 floodplains, respectively, and the channel density decreased upsection. Paleohydrological

37 features of the Chinle Formation fluvial deposits exhibit upward decrease in channel

2
 3

38 cross-sectional sizes and bankfull discharges. Vertical changes in sandstone framework

39 composition and detrital clay mineral composition in the formation indicate a decline in

40 the intensity of chemical weathering in the source mountains from the Shinarump

41 Member to Cameron Member and “purple pedogenic beds”. The combination of these

42 datasets suggests that a shift from humid to wet-dry seasonal climatic conditions occurred

43 before 215 Ma in southwestern Utah. In addition, temporal variation in architecture and

44 stacking pattern of the Chinle Formation fluvial deposits are interpreted to have been

45 controlled by the climate shift rather than by tectonism.

46

47 1. Introduction

48

49 Fluvial systems are sensitive to allogenic controls, such as tectonic and climatic

50 changes in sedimentary basins and their hinterlands (Blair and Bilodeau, 1988; Miall,

51 1996; Blum and Törnqvist, 2000). Spatial and temporal variations in longitudinal profiles

52 and accommodation spaces in the fluvial systems are considered to be influenced

53 primarily by the rate of basin subsidence (Bridge and Leeder, 1979; Martinsen et al.,

54 1999). In addition, precipitation controls hydrological features of the fluvial systems,

55 such as water and sediment discharges, denudation rates, and vegetation (Knox, 1983;

3
 4

56 Ohmori, 1983). These fluvial hydrological features are also considered to control spatial

57 and temporal variations in fluvial architecture and stacking patterns of fluvial deposits

58 (Blum et al., 2000; Ray and Chakraborty, 2002; Paredes et al., 2007, 2018; Hillier et al.,

59 2007; Colombera et al., 2017). Although the interaction between climatic and tectonic

60 fluctuations has been considered to be responsible for spatial and temporal variations in

61 fluvial systems, relative contribution of these allogenic controls still remains

62 controversial (Dickinson et al., 1994; Cleveland et al., 2007; Varban and Plint, 2008;

63 Antia and Fielding, 2011; Dubiel and Hasiotis, 2011; Trendell et al., 2012; Miall, 2014).

64 During the Late Triassic, the Pangaean monsoonal circulation is considered to

65 have been at maximum strength, and western Pangaea changed from wet to wet-dry

66 seasonal climatic conditions (Parrish, 1993). This climatic change, together with the

67 development of a volcanic arc system along the western margin of the Pangaea, is

68 interpreted to have been documented in fluvial and lacustrine systems formed in the

69 present western U.S.A. (Stewart et al., 1972; Blakey and Gubitosa, 1984; Dubiel and

70 Hasiotis, 2011; Nordt et al., 2015). These Late Triassic terrestrial depositional systems

71 are widely distributed around the Colorado Plateau. (Fig. 1). However, detailed analyses

72 of fluvial architecture, paleoclimates, and geologic ages of the fluvial systems and related

73 depositional systems have mainly been conducted in the Upper Triassic successions that

4
 5

74 are developed in the Four Corners region (a junction area of Arizona, Utah, Colorado, and

75 New Mexico that includes Petrified Forest National Park in northern Arizona) (Blakey

76 and Gubitosa, 1984; Kraus and Middleton, 1987; Dubiel, 1991; Riggs et al., 2003; Dubiel

77 and Hasiotis, 2011; Trendell et al., 2012, 2013; Atchley et al., 2013; Howell and Blakey,

78 2013; Nordt et al., 2015; Jin et al., 2018; Rasmussen et al., 2020). In contrast, the Upper

79 Triassic successions in southwestern Utah have not yet been focused upon in terms of

80 climatic and tectonic changes in western Pangaea (Martz et al., 2017), although geologic

81 studies in southwestern Utah should have a potential to fill a niche for a better

82 understanding of climatic and tectonic fluctuations in the western Pangaea during the

83 Late Triassic. One of the potential stratigraphic successions for these studies in

84 southwestern Utah is the Upper Triassic Chinle Formation.

85 This study focuses on the Chinle Formation that is distributed in and around Zion

86 National Park on the western margin of the Colorado Plateau of southwestern Utah (Fig.

87 2). The main purpose of this study is to analyze spatial and temporal variations in fluvial

88 architecture, stacking patterns, and paleohydrological features of the Chinle Formation,

89 together with analyses of vertical changes in detrital compositions of sandstone

90 framework grains and clay minerals for elucidating the relationships between fluvial

91 sedimentation, and climatic and tectonic fluctuations in an Upper Triassic sedimentary

5
 6

92 basin that formed in the western Pangaea.

93

94 2. Geological setting

95

96 2.1. Tectonic setting and provenance

97

98 The Chinle Formation and its age-equivalent strata are widely distributed in the

99 Colorado Plateau and adjacent areas (Fig. 1). The formation is characterized by fluvial

100 and lacustrine deposits, and is interpreted to have developed in a continental backarc

101 region in western Pangaea (Blakey and Gubitosa, 1984; Dubiel et al., 1991; Tanner, 2000;

102 Trendell et al., 2012; Martz et al., 2017). The Chinle Formation developed in sedimentary

103 basins that were surrounded by Cordilleran volcanic arc in California, the Ancestral

104 Uncompahgre and Front Range uplifts in Colorado, the Mogollon Slope in Arizona

105 (Dubiel, 1992), Amarillo–Wichita Uplift in northwestern Texas, and Ouachita–Marathon

106 Orogenic Belt in Texas (Fig. 1). The major sediment transport direction in the Colorado

107 Plateau region is interpreted to have been northwest during the Late Triassic (Stewart et

108 al., 1972; Dubiel et al., 1991; Blakey and Ranney, 2008; Dubiel and Hasiotis, 2011; Martz

109 et al., 2017), and the major provenance of the detrital fragments of the Chinle Formation

110 in southwestern Utah is interpreted to have been from the Mogollon Slope in southern

6
 7

111 Arizona (Stewart et al., 1972; Bilodeau, 1986; Dubiel, 1992; Nordt et al, 2015; Martz et

112 al., 2017; Jin et al., 2018). The Mogollon Slope consists of the Proterozoic sedimentary,

113 volcanic, granitic, and metamorphic rocks, and the Paleozoic sedimentary rocks, such as

114 conglomerates, sandstones, siltstones, shale, limestone, dolostone, and chert (Stewart et

115 al., 1972; Parker et al., 2005). In addition, the Late Triassic Cordilleran volcanic arc is

116 also considered to have supplied some volcanic detritus to the Chinle Formation

117 sedimentary basins (Riggs et al., 2003, 2013, 2016). The Chinle Formation in

118 southwestern Utah is interpreted to have deposited within the Vermillion Cliffs

119 paleovalley (Martz et al., 2017).

120

121 2.2. Lithostratigraphy and age

122

123 The Chinle Formation in southern Utah is subdivided into the Shinarump,

124 Monitor Butte/Cameron, Moss Back, Petrified Forest, Owl Rock (partially the Kane

125 Springs Beds), and Church Rock members in ascending order (Martz et al., 2017) (Fig.

126 3A). In the study area, the Chinle Formation overlies the lower Middle Triassic coastal

127 plain deposits of Upper Red Member of the Moenkopi Formation. The uppermost part of

128 the Moenkopi Formation locally includes paleosol mottling with reddish brown and light

7
 8

129 greenish gray colors, which was informally called “mottled strata”, and was included in

130 the basal part of the Chinle Formation (Stewart et al., 1972). The Chinle Formation is

131 subdivided, in ascending order, into the Shinarump (~40 m thick) and Cameron (~95 m

132 thick) members, and the “purple pedogenic beds” (~25 m thick) (Martz et al., 2017).

133 Although the upper part of the formation (i.e., the Cameron Member and “purple

134 pedogenic beds”) was previously correlated to the Petrified Forest Member (Stewart et al.,

135 1972; Hintze, 1997), recent studies of vertebrate biostratigraphy indicate that the upper

136 part correlates to the Cameron Member rather than to the Petrified Forest Member on the

137 basis of the occurrences of phytosaur skull in the lower part of the Cameron Member,

138 which is likely a basal members of the clade Leptosuchomorpha (Martz et al., 2017). The

139 formation is unconformably overlain by the Upper Triassic–Lower Jurassic Moenave

140 Formation (Kirkland and Milner, 2006; Lucas and Tanner, 2007; Kirkland et al., 2014)

141 (Fig. 3B). The unconformity at the Chinle–Moenave contact (or J-0 unconformity of

142 previous authors) is overlain regionally by an approximately one meter thick poorly

143 cemented chert and anhydrite pebble conglomerate (Kirkland and Milner, 2006; Kirkland

144 et al., 2014; Martz et al., 2017).

145 The radioisotopic ages of the Chinle Formation have been investigated using

146 samples mainly from the Petrified Forest National Park in northern Arizona (Riggs et al.,

8
 9

147 2003; Ramezani et al., 2011; Atchley et al., 2013; Nordt et al., 2015; Rasmussen et al.,

148 2020). The Chinle Formation in northern Arizona is subdivided into the Mesa Redondo,

149 Blue Mesa, Sonsela, Petrified Forest, and Owl Rock members in ascending order, and the

150 radioisotopic dating indicates that the Mesa Redondo, Blue Mesa, Sonsela, and Petrified

151 Forest members are correlative to the Norian, while the Owl Rock Member is Rhaetian in

152 age (Fig. 3A). Although the radioisotopic age of the Chinle Formation in southwestern

153 Utah has not yet been conducted, the Shinarump Member is correlative to the Mesa

154 Redondo Member in northern Arizona, and the Cameron Member and “purple pedogenic

155 beds” are considered to be correlative to the Monitor Butte Member in southeastern Utah

156 and the Blue Mesa Member in northern Arizona (Martz et al., 2017) (Fig. 3).

157 Consequently, the age of the Chinle Formation in the study area is interpreted to be the

158 Norian (Fig. 3B).

159

160 2.3. Paleoclimates

161

162 Fluctuations in paleoclimates during the deposition of the Chinle Formation have

163 been studied on the basis of sedimentological and paleontological evidence (Dubiel et al.,

164 1991), fossil plant assemblages (Lindström et al., 2016; Baranyi et al., 2018), paleosol

9
 10

165 types (Dubiel and Hasiotis, 2011; Lucas and Tanner, 2018), geochemical features of

166 paleosol (Prochnow et al., 2006; Atchley et al., 2013; Nordt et al., 2015), and

167 petrographic features (Jin et al., 2018) of the formation. These studies indicate that the

168 lower part of the Chinle Formation was deposited under unusually humid climatic

169 condition with seasonality that may have been caused by “megamonsoon” circulation

170 (Dubiel et al., 1991; Dubiel and Hasiotis, 2011; Lucas and Tanner, 2018). In addition, the

171 climates are considered to have shifted to more arid conditions after 215 Ma (Atchley et

172 al., 2013; Nordt et al., 2015; Baranyi et al., 2018; Jin et al., 2018). This climatic change is

173 interpreted to have been caused by the northward drift of the Chinle Formation

174 sedimentary basins (Dubiel and Hasiotis, 2011) or the development of a rain shadow in

175 response to the development of the Cordilleran volcanic arc in the western margin of the

176 Pangaea (Atchley et al., 2013; Nordt et al., 2015).

177

178 3. Data set and methods

179

180 3.1. Fluvial architecture

181

182 Three-dimensional analyses of lithofacies organization and architecture of the

183 Chinle Formation fluvial deposits were conducted using large outcrops in seven major

10
 11

184 locations (Fig. 2). This study mainly used photographic mosaics and measured sections

185 for the present analyses. We employed the basic concept of the hierarchical

186 bounding-surface and architectural element analyses (Miall, 1985, 1988; 1996; Holbrook,

187 2001; Ford and Pyles, 2014) for the definition of fluvial forms in the outcrops.

188 Paleocurrent directions were measured from dip directions of the axial part of trough or

189 planar cross-bedding.

190

191 3.2. Paleohydrology

192

193 Paleochannel depths have been estimated on the basis of the empirical

194 relationships between the thickness of bar deposits and bankfull channel depth (Leeder,

195 1973; Bridge and Diemer, 1983; Alexander et al., 2020). The present study measured

196 thickness of well-preserved bar deposits from the photographic mosaics and stratigraphic

197 sections. Bankfull channel depth has also been estimated using cross-set thickness in the

198 bar deposits using empirical equations by Leclair and Bridge (2001) and Bridge (2003).

199 Mean bankfull channel depth (= hydraulic radius at a bankfull water stage), bankfull

200 channel width, and bankfull water discharge were also calculated using empirical

201 equations that were established from modern fluvial channels (e.g., Williams, 1984;

11
 12

202 Bridge and Mackey, 1993; Adhiperdana et al., 2018; Shibata et al., 2018).

203

204 3.3. Detrital composition

205

206 Pebble samples were collected from conglomerates and pebbly sandstones of the

207 Shinarump and Cameron members at six sites for each member (Routes 2, 4, and 5 for the

208 Shinarump Member, and Routes 3, 4, and 6 for the Cameron Member in Fig. 2). The

209 numbers of pebble samples from each site are 110–122 for the Shinarump Member, and

210 122–135 for the Cameron Member. Thin sections of 22 selected samples were observed

211 using a polarizing microscope to identify rock types of the pebbles and the other samples

212 were visually examined to identify their rock types.

213 For identifying framework composition of the sandstones, a total of 28 and 25

214 medium- to very coarse-grained sandstone samples were collected from the Shinarump

215 and Cameron members, respectively. Samples were collected mainly from Routes 3, 4,

216 and 6 in Figure 2. The framework composition was determined by point counting more

217 than 500 sand-sized grains in a thin section using the Gazzi-Dickinson method (Ingersoll

218 et al., 1984). Framework grains were classified following the grain categories of Winkler

219 (1984). The thin sections were stained to facilitate the identification of K-feldspar using a

12
 13

220 method of Bailey and Stevens (1960).

221 A total of 76 samples from mudstones and 52 samples from sandstones were also

222 taken from the Shinarump and Cameron members and “purple pedogenic beds” for the

223 X-ray diffraction (XRD) analysis. These samples were collected mainly from Routes 3, 4,

224 and 6 in Figure 2. About 3 g mudstone fragments and 15 g sandstone fragments of each

225 sample were grinded using a mortar. Clay-size fractions less than 2 m were separated

226 from the grinded samples using motionless gravity settling of suspension (Stokes’ law).

227 Clay suspension was dropped on slide glasses, and was dried at 60 ℃ in an oven to

228 prepare oriented clay slides. Ethylene glycol was added to the dried oriented clay slides

229 for the identification of smectite. The XRD analysis was conducted using RIGAKU

230 RAD-X System in the Department Earth Sciences, Chiba University. This X-ray unit uses

231 Ni-filtered Cu-radiation (Cu Kα) target setting of 40 kV and 20 mA. The relative

232 composition of major clay minerals is estimated semi-quantitatively, following the peak

233 area method (Biscaye, 1965). The relative proportions of chlorite and kaolinite were

234 calculated from the ratio of peak heights (3.53 and 3.58 Å, respectively). If this ratio is 1,

235 the amount of chlorite is assumed to be twice that of kaolinite (Suresh et al., 2004). A total

236 of four samples of mudstone and two samples of sandstone were also observed by a

237 scanning electron microscope (SEM) using HITACHI S-2400 in the Department Earth

13
 14

238 Sciences, Chiba University to distinguish detrital and authigenic grains.

239

240 4. Fluvial architecture

241

242 The present study identified twelve lithofacies types, nine architectural element

243 types, and six major bounding-surface (BS) types (1st- to 6th-order in increasing scale

244 order) in the Chinle Formation fluvial deposits. Lithofacies types were identified only by

245 the combination of lithologic types and physical and biological sedimentary structures

246 (Ford and Pyles, 2014). In contrast, architectural element types were mainly classified in

247 terms of external forms. The hierarchical classification of the bounding-surface types was

248 defined by cross-cutting relationships of the surface types, in association with the

249 combination between lateral continuity, geometry, lithofacies features, and paleocurrent

250 directions of the encased fluvial deposits.

251

252 4.1. Lithofacies

253

254 The Chinle Formation consists of very coarse- to fine-grained, dull yellow

255 orange-colored pebbly sandstones (Lithofacies St/Sp, Sh, Sc, and Sg/Sig), with local

14
 15

256 associations of pebble-sized conglomerates (Lithofacies Gc, and Gt/Gp), and varicolored

257 (e.g., gray, yellowish gray, brownish gray, reddish gray, grayish red, and reddish brown)

258 mudstones (Lithofacies Fm, Pm, Fr, and Fl). Minor massive limestone (Lithofacies Lm)

259 is also locally intercalated in the formation (Fig. 4). Descriptions and interpretations of

260 lithofacies types are summarized in Table 1.

261

262 4.2. Architectural element

263

264 Descriptions and interpretations of architectural element types are summarized in

265 Table 2 and their schematic illustrations are given in Figure 5.

266

267 4.2.1. Channel deposits

268 Description. Distinct erosional surfaces in the base of sandstones are locally

269 overlain by thin and discontinuous, pebble-sized, clast-supported conglomerates

270 (Lithofacies Gc, Gt/Gp, and Element SG, Fig. 6A). The erosional surfaces and the basal

271 conglomerates are overlain by moderately sorted, very coarse- to fine-grained pebbly

272 sandstones with trough or planar (Lithofacies St/Sp) cross-stratification, in association

273 with minor horizontal planar stratification (Lithofacies Sh). Scattered pebble-sized gravel

15
 16

274 is also locally present along the stratification. Some cross-set boundaries indicate minor

275 truncation of underlying sandy deposits. These cross-set boundaries dip towards

276 downstream directions (Element DA, Figs. 6B and C), and are also locally incline

277 orthogonal to the downstream direction (Element LA, Fig. 6A). The distinction between

278 Elements DA and LA was made using an angle of 60° between channel flow directions

279 and orientations of the inclined boundaries, following Miall (1996). The cross-set

280 boundaries are also locally horizontal (Element SD, Fig. 6C). Sets of cross-stratified

281 sandstones commonly represent upward-fining trends, and often include fossil plants.

282 Locally, moderately sorted, medium- to fine-grained, trough cross-stratified pebbly

283 sandstones (Lithofacies St) with lenticular geometry and erosional basal surfaces

284 (Element SCF, Fig. 7A), which are up to 1.0 m thick and 5.2 m wide, occurs on the

285 cross-stratified sandstones with upward-fining trends. Large-scale lenticular element (~8

286 m thick and 40 m wide) is locally filled with alternating beds of mudstones (Lithofacies

287 Fm, Fl, and Fr) and sandstones, which consist of the cross-stratified or horizontally

288 stratified sandstones (Lithofacies St/Sp, Sh) that are covered by current-ripple

289 cross-laminated sandstones with local development of convolute laminations (Lithofacies

290 Sr and Sc), and show upward-fining trends (Element LCF) (Fig. 7B). The stacked

291 elements of DA, LA, SD, SCF, and LCF are locally overlain by inclined alternating beds

16
 17

292 of sandstones (Lithofacies St/Sp, Sh, Sr) and mudstones (Lithofacies Fm, Fr, and Fl)

293 (Element UB, Fig. 7C).

294 Interpretation. The sandstones with erosional basal surfaces are considered to

295 represent deposits in fluvial channels (Miall, 1988; Hampson et al., 2013; Ford and Pyles,

296 2014). Discontinuous pebble-sized, clast-supported conglomerates (Element SG) are

297 interpreted as lag deposits on the erosional surfaces in the base of channels (Jo, 2003;

298 Hampson et al., 2013). Trough or planar cross-stratified, very coarse- to fine-grained

299 sandstones (Lithofacies St/Sp) and pebble-sized conglomerates (Lithofacies Gt/Gp)

300 indicate deposition from straight- or sinuous-crested dunes that migrated in the

301 downstream direction (Allen, 1982; Miall, 1996; Hampson et al., 2013). The inclined

302 stratification in conglomerates and sandstones (Lithofacies Gt/Gp, St/Sp, Sh, and Sr),

303 which shows inclined direction similar to paleocurrents and is represented by boundaries

304 with minor erosional surfaces, indicates downstream accretion (Element DA) of braid bar

305 deposits (Thomas et al., 1987; Miall, 1988; Lunt et al., 2004; Ford and Pyles, 2014). In

306 contrast, the inclined stratification in conglomerates and sandstones (Lithofacies Gt/Gp,

307 St /Sp, Sh, and Sr), which shows inclined directions orthogonal to paleocurrents, is

308 considered to indicate lateral accretion (Element LA) of braid bar and point bar deposits

309 (Thomas et al., 1987; Miall, 1988; Ford and Pyles, 2014). In addition, sandstones with

17
 18

310 horizontal cross-set boundaries (Element SD) are interpreted to represent aggradation of

311 dunes on low-relief bars (Miall, 1985). The upward-fining successions of stacked

312 sandstones (Elements DA, LA, SD) indicate the reduction of bed shear stress from the

313 base to the top of the channel or bar (Allen, 1970), and the successions are interpreted to

314 be channel and component bar deposits (Allen, 1970; Leeder, 1973; Lorenz et al., 1985;

315 Lunt et al., 2004; Adams and Bhattacharya, 2005). Inclined alternating beds of

316 sandstones and mudstones that developed over stacked sandstones (Element UB) indicate

317 upper bar deposits (Bridge et al., 2000; Bridge, 2003). Small-scale lenticular sandstones

318 (Lithofacies St and Element SCF) developed on the top of upward-fining successions

319 (Elements DA, LA, SD) as well as alternating beds of mudstones and sandstones

320 (Lithofacies Fm, Fl, Fr, St, Sp, Sh, and Sr) with large-scale erosional base (Element LCF)

321 are interpreted as infills of small channels on bars (Allen, 1983; Jo, 2003) and abandoned

322 channels (Miall, 1985, 1996; Bridge, 1993; Ford and Pyles, 2014), respectively.

323 Alternating beds of mudstones and sandstones in Element LCF indicate repetitive

324 depositions from suspension and traction loads, respectively. The cross-stratified or

325 horizontally stratified sandstones (Lithofacies St/Sp, Sh), which are overlain by

326 current-ripple cross-laminated sandstones (Lithofacies Sr) with upward-fining trends are

327 considered to document deposition during a waning flow stage in flood events (Ray,

18
 19

328 1976).

329

330 4.2.2. Floodplain deposits

331 Description. The couplets of sandstones with basal erosional surfaces are encased

332 in, or overlain by varicolored sets of massive, thick mudstone (~25 m thick) (Lithofacies

333 Pm) or gray-colored interbeds of mudstones and very fine-grained sandstone (Lithofacies

334 Fl, Fr, and Fm) (Element FF, Fig. 7D) with local association of climbing ripple

335 cross-laminations (Lithofacies Fr) (Fig. 4D). Varicolored mudstones (Lithofacies Pm)

336 locally show distinct slickenside and cross-cutting structure. Massive limestone

337 (Lithofacies Lm), which lacks body fossils, locally occurs in association with the

338 varicolored mudstones (Lithofacies Pm). The thick mudstones contain moderately sorted,

339 very fine- to medium-grained sandstones (~2.6 m thick) (Element FS, Fig. 7D). These

340 sandstones typically contain inverse grading (Lithofacies Sig), planar parallel- or

341 cross-stratification (Lithofacies Sh, St, and Sr), and/or normal grading (Lithofacies Sg),

342 from the base to top of a bed (Fig. 7D). Sandstones of Element FS typically show

343 upward-fining trends, and do not have distinct erosional surfaces and are represented

344 mainly by basal flat surfaces. They also show laterally thinning trends, and locally pass

345 into mudstones or very fine-grained muddy sandstones (Lithofacies Fm, Pm, Fr, Fl,

19
 20

346 Element FF).

347 Interpretation. Massive (Lithofacies Fm) and laminated (Lithofacies Fr, Fl)

348 mudstones represents accumulation from suspension and traction loads, respectively.

349 Development of climbing ripple cross-laminations in the intercalated very fine-grained

350 sandstone beds (Lithofacies Fr) indicates dominance of suspension loads relative to

351 traction loads in association with very high deposition rate (Allen, 1982). Varicolored

352 mudstones with distinct slickenside and cross-cutting structures (Lithofacies Pm) are

353 interpreted as pedogenetic modification of muddy deposits that were accumulated from

354 suspended loads (Cleveland et al., 2007; Dubiel and Hasiotis, 2011; Trendell et al., 2013).

355 Massive limestone (Lithofacies Lm) indicates chemical precipitation of inorganic

356 calcareous deposits, and is interpreted to have formed in small carbonate lakes or ponds

357 (Tanner, 2000). These mudstones (Element FF) are interpreted as floodplain muddy

358 deposits accumulated mainly by suspended loads with local association with tractional

359 structures (Miall, 1985, 1988; Ford and Pyles, 2004).

360 Normally or inversely graded sandstones (Lithofacies Sg/Sig) and planar parallel-

361 or cross-stratified sandstones (Lithofacies Sh, St, and Sr) represent depositions from

362 suspension and traction loads, respectively. The inversely graded sandy deposits have

363 been described from modern flood-induced deposits in an overbank environment, and are

20
 21

364 interpreted to represent temporal increase in discharge and sediment concentration in

365 flood water (Iseya, 1989). Planar parallel- or cross-stratified sandstones (Lithofacies Sh,

366 St, and Sr) and normally graded sandstones (Lithofacies Sg) that show overall

367 upward-fining trends suggest the deposition from waning flow during flood events (Ray,

368 1976). A depositional model of crevasse-splay deposits by Burns et al. (2017) suggests

369 that their average thicknesses in proximal, medial, and distal parts are 2.1 m, 1.5 m, and

370 0.8 m, respectively. The lateral variation in thickness of Element FS of the present study

371 exhibits a pattern similar to that in the model. Therefore, the sandstones of Element FS

372 are interpreted to be overbank sandy deposit generated by flood events.

373

374 4.3. Bounding-surface (BS) types and architectural hierarchy

375 An erosional surface type, which is commonly overlain by laterally discontinuous

376 pebble-sized conglomerates (Element SG), corresponds to a channel base (4th-order BS),

377 and a set of Element SG and overlying sandstones (Element SG, DA, LA, SD, and UB)

378 indicates sandy channel deposits ranging from 3 m to 13 m in thickness (Figs. 8A and 10).

379 A basal surface of abandoned channel deposits (Element LCF) is also represented by this

380 order surface type (Fig. 7B). Top surfaces of the channel deposits that are overlain by

381 thick mudstones (Element FF) represent bar-top surfaces (3rd-order BS) (Fig. 8A). The

21
 22

382 basal- and top-surfaces of the floodplain sandy deposits (Element FS) encased in the

383 floodplain muddy deposits (Element FF) are also equivalent to this order surface

384 (3rd-order BS) (Fig. 7D). Within the channel and bar deposits, gently inclined

385 stratification that is bounded by minor erosional surfaces (Elements DA and LA) and

386 horizontal stratification (Element SD) records accretion and vertical aggradation of bars

387 (2nd-order BS) (Figs. 6 and 7), respectively. In addition, the constituent cross-set

388 boundaries (1st-order BS) represent migration of dunes and ripples.

389 Vertically and laterally stacked sets of channel deposits are bounded by

390 higher-scale, compound basal erosional surfaces (5th-order BS) (Fig. 8A). A set of

391 channelized deposits that is bounded by this bounding-surface type is termed, herein, as a

392 channel complex (up to 25 m thick). A compound basal erosional surface (5th-order BS)

393 commonly incises into underlying finer-grained deposits (Elements UB and FF) (Fig. 8A).

394 In addition, a basal surface of the stacked channel complexes corresponds to a sequence

395 boundary (6th-order BS) (Fig. 8, see Discussion).

396

397 4.4. Stratigraphic change

398

399 4.4.1. Shinarump Member

22
 23

400 In the Shinarump Member, cross-stratified sandstones with inclined accretion

401 surfaces (Elements DA, LA, and UB) and with horizontal stratification (Element SD)

402 constitute 74–99 % of the total thickness, with a small proportion of conglomerates of

403 Element SG (<5%) (Figs. 8A and 10), and the component cross-sets are up to 135 cm

404 thick. Cross-stratified sandstones with downstream-directed accretion (Element DA) are

405 dominant throughout the member (Fig. 6B), although laterally accreting, cross-stratified

406 sandstones (Element LA) also locally occur in the upper part of the member (Fig. 6A).

407 These sandstones and conglomerates are mainly channel deposits, which are bounded by

408 erosional basal surfaces of channel deposits (4th-order BS) or channel complexes

409 (5th-order BS) (Fig. 8A). The upper part of an upward-fining trend in channel deposits is

410 characterized locally by inclined alternating beds of sandstones and mudstones (Element

411 UB) (Fig. 7C). A channel complex is overlain locally by alternating beds of mudstones

412 and very fine-grained sandstones (Element FF), which are truncated by distinct erosional

413 surfaces (5th-order BS) and are overlain by a younger channel complex (Fig. 8A). These

414 mudstones and very fine-grained sandstones constitute a small proportion (1–10 %) in the

415 total thickness of the member. Overall, the Shinarump Member consists of vertically

416 stacked two or three cycles of channel complexes, which contain 1–4 cycles of channel

417 deposits (Fig. 10). Trough cross-stratification and current-ripple cross-laminations in

23
 24

418 sandstones (Elements DA, LA, and SD) are characterized by unidirectional paleocurrents

419 that show a dominant flow direction toward the north, northwest, and/or west (Figs. 6–8,

420 and 10). An exceptional flow direction is in the south-southeast (Route 5 in Figs. 2 and

421 10), which are observed in some laterally accreting deposits (Element LA) (Figs. 6A).

422

423 4.4.2. Cameron Member and “purple pedogenic beds”

424 The stacked channelized sandstones of the Shinarump Member are abruptly

425 replaced by varicolored and massive mudstones of the Cameron Member and “purple

426 pedogenic beds” (Element FF), which were formed mainly in floodplains and comprise

427 72–95 % of the total thickness (Fig. 10). Thick varicolored mudstones contain channel

428 deposits and channel complexes with distinct erosional basal surfaces (4th- and 5th-order

429 BS) (Figs. 9 and 10) in the Cameron Member. Cross-stratified conglomerates and

430 sandstones of the channel deposits are minor components of the Cameron Member and

431 “purple pedogenic beds” in total thickness (<2 % and 6–20 %, respectively). Cross-sets in

432 the channel deposits are up to 90 cm in thickness, and the cross-set boundaries are

433 commonly horizontal (Element SD), and locally dip obliquely to paleocurrent directions

434 (Element DA) (Fig. 6C). The channel deposits lack alternating beds of sandstones and

435 mudstones (Element UB) in their uppermost part, and are covered only by muddy

24
 25

436 deposits (Element FF). Isolated channel complexes consist of 1–3 cycles of the channel

437 deposits, and are encased in varicolored floodplain muddy deposits. The largest channel

438 complex in the Cameron Member (Route 3 in Fig. 2) is about 20 m in thickness and about

439 450 m in width in an outcrop that trends nearly orthogonal to a mean paleocurrent

440 direction (Fig. 9) with a width/thickness ratio of 22.5. Sandstone interbeds of floodplain

441 sandy deposits (Elements FS) (~2.6 m thick) are also intercalated in the thick varicolored

442 mudstones (Fig. 7D), and are also minor components of the member (0–7 % in the total

443 thickness). Paleocurrent data of the channel deposits and the floodplain sandy deposits

444 show flow directions mainly towards the north and northwest (Fig. 10).

445

446 5. Paleohydrology

447

448 Stratigraphic thicknesses of bar deposits (h) are interpreted to be slightly less

449 (90%) than bankfull channel depth (d) (Bridge and Diemer, 1983; Bridge and Mackey

450 1993), and Ethridge and Schumm (1978) and Lorenz et al. (1985) used a 10 %

451 compaction factor to correct stratigraphic thickness for the estimation of paleochannel

452 depths from sandy bar deposits. The present study also adopted the 10% compaction

453 factor, and we used corrected h for the calculation of d as follows:

25
 26

454

455 d = h  10/9  10/9 (Eq. 1)

456

457 The present study also calculated the mean cross-set thickness (Smean) and

458 standard deviation (Ssd), and estimated the mean dune height (Hmean), using the following

459 equations (Leclair and Bridge, 2001; Bridge, 2003):

460

461 Hmean = 5.3β (Eq. 2)

462 β ≈ Smean/1.8 (Eq. 3)

463

464 In this method, Ssd/Smean should be approximately equal to 0.88 (± 0.3). Mean dune height

465 increases with the increase in flow depth, and the flow depth is interpreted to have a range

466 from 6 to 10 times of the mean height of river dunes (Bridge, 2003).

467 Empirical equations between d and mean bankfull channel depth (= hydraulic

468 radius at a bankfull water stage) (dm), between dm and bankfull channel width (Wb), and

469 between Wb and water discharge (Q) have been established using data from modern

470 fluvial systems at various tectonic and climatic settings (e.g., Williams, 1984; Bridge and

471 Mackey, 1993; Adhiperdana et al., 2018; Shibata et al., 2018). However, the empirical

472 equations should be applied carefully considering tectonic and climatic conditions,

26
 27

473 because these conditions affect valley slope, rate of uplift, inputs of volcaniclastic

474 materials from active volcanic hinterlands, and precipitation and evaporation, each of

475 which controls geomorphological and hydrological features of fluvial systems, such as

476 bed material size, proportion of bedload relative to suspended load, and channel forms

477 (Williams, 1984; Adhiperdana et al., 2018; Shibata et al., 2018). The present study

478 adopted the empirical equations from modern Indonesian fluvial systems (Adhiperdana et

479 al., 2018) for the estimation of dm, Wb, and bankfull discharge (Qb), because the empirical

480 equations were developed from fluvial systems formed in a volcanic arc setting that is

481 influenced by a monsoon climate. These tectonic and climatic settings are considered to

482 have been similar to those for the development of the Chinle Formation. The empirical

483 equations that were used by the present study are as follows:

484

485 dm = 0.65 d (Eq. 4) Formatted: German (Germany)

486 Wb = 52.88 dm0.7 (Eq. 5)

487 Qb = 1.098 Wb1.42 (Eq. 6)

488

489 For the reconstruction of paleohydrological features of the Chinle Formation, the

490 present study used outcrop data from well-exposed eight channel complexes of the

27
 28

491 Shinarump and Cameron members (CHC I–VIII in Fig. 10). Although some bar deposits

492 are truncated by overlying younger bar or channel-fill deposits, the thickest values of bar

493 deposits can be used to estimate the maximum geomorphological and hydrological

494 parameters in each channel complex (Shibata et al., 2018; Alexander et al., 2020).

495 Thickness of the thickest bar deposits with upward-fining trend (h), which are defined by

496 the 4th- (basal surface of channel deposits) and 3rd- (top surface of channel deposits)

497 order bounding surfaces, range from 7.2 to 9.9 m in the Shinarump Member (CHC I–V).

498 On the basis of the empirical equation between h and d (Eq. 1), the d value can be

499 estimated at 8.9–12.2 m. Following the similar procedures, the h and d values of the

500 fluvial channels of the Cameron Member (CHC VI–VIII) can be estimated at 4.6–5.4 m

501 and 5.7–6.7 m, respectively.

502 The Smean values of the channel complexes in the Shinarump and Cameron

503 members are 42.1–58.2 cm and 30.9–38.6 cm, respectively, and the Hmean values in the

504 members were estimated at 124.1–171.3 cm and 90.9–113.7 cm, respectively, with the

505 values of Ssd/Smean ranging from 0.58 to 0.84 (Eqs. 2 and 3). These results indicate that

506 paleo-water depths of the fluvial channels were 7.4–17.1 m for the Shinarump Member,

507 and 5.5–11.4 m for the Cameron Member, respectively.

508 Using the empirical relationships between d and dm (Eq. 4), and between Wb and

28
 29

509 dm (Eq. 5), the Wb values were estimated at 159.5–285.7 m for channel complexes of the

510 Shinarump Member, and 128.3–214.5 m for those of the Cameron Member. Similarly, the

511 relationship between Wb and Qb (Eq. 6) indicate that the Qb values can be estimated,

512 respectively, at 1473.7–3372.8 m3s-1 and 1081.9–2245.0 m3s-1 for the channel complexes

513 of the Shinarump and Cameron members. These estimated values are summarized in

514 Figure 11.

515 The estimated values indicate that paleochannels of the Shinarump Member were

516 generally larger than those of the Cameron Member. The Wb and Qb values of the

517 Shinarump Member fluvial channels are almost equal to those of the modern Yellowstone

518 River near Miles City, Montana. In contrast, the estimated values of the Cameron

519 Member fluvial channels are approximately equivalent to or slightly larger than those

520 obtained from the modern Green River near Green River, Utah and from the modern

521 Colorado River near Hite, Utah (Table 4).

522

523 6. Detrital composition

524

525 6.1. Clast composition of conglomerates

526

29
 30

527 Chert and quartzite are dominant clast types of conglomerates in both the

528 Shinarump and Cameron members, although quartzite and chert clasts are most common

529 clast types in the Shinarump and Cameron members, respectively (Figs. 12A, B, and 13).

530 Quartz fragments are also common in the Shinarump Member, while mudstone and

531 sandstone clasts are minor components in both members (Figs. 12C, D and 13). Mudstone

532 pebbles are represented commonly by black hard mudstone, and a brownish laminated

533 mudstone pebble was identified only in the Cameron Member. Furthermore, three types

534 of pebble-sized volcanic rock fragments were identified. The first type is acidic

535 (rhyolitic) volcanic rock fragments with felsitic texture, and includes partially devitrified

536 volcanic grass and embayed monocrystalline quartz (Fig. 12E). Only one clast of this type

537 is identified in the Cameron Member. The second type is well devitrified tuff including Commented [BGA1]: In this case only: As opposed to
the use of ''partially devitrified'' at above, could we
538 clots of quartz and/or feldspar with a dull appearance (Fig. 12F), and was identified in just use here ''almost fully devitrified'' instead of ''well
devitrified''? This is only my tentative opinion,
539 both the Shinarump and Cameron members. The third type is partially devitrified tuff, decision is yours. You don't have to change anything
here if you want.
540 which is characterized by fine-grained groundmass and embayed monocrystalline quartz.

541 This type was found only in the Cameron Member.

542

543 6.2. Sandstone framework composition

544

30
 31

545 Framework composition of the Chinle Formation sandstones is generally

546 represented by quartzose grains in association with minor feldspar grains and lithic

547 fragments (Figs. 14 and 15; Appendix I). The Shinarump Member is characterized by

548 more contents of quartzose grains than the Cameron Member, which has more feldspar

549 grains and lithic fragments than the Shinarump Member (Figs. 14 and 15; Appendix I).

550 Monocrystalline quartz grains show straight or undulose extinction (Fig. 14A and

551 B), and some of them indicate embayed shape with rounded corner (Fig. 14C). Overall,

552 monocrystalline quartz grains are dominant throughout the formation, and the other

553 quartz grains, such as the polycrystalline quartz, quartz-tectonite, and chert (Fig. 14D),

554 are minor components. K-feldspar grains with perthitic textures (Fig. 14A) were Commented [BGA2]: I think this microcline does not
show a typical perthite texture, which may consists of
555 commonly altered and replaced by clay minerals and calcite (Fig. 14E). Plagioclase paralel exsolution lamellae (I attached for you to see
perthitic texture in your sample on a separate PPT
556 feldspar is also represented by polysynthetic twinning and commonly sericitized (Fig. file). This microcline instead, shows cross-hatched
twinning or texture, which is a combination of albite
557 14F). K-feldspar is more abundant than plagioclase feldspar in the two members. and pericline twinning.
Cross-hatched twinning in microcline also is referred
558 Volcanic lithic fragments are acidic, and are classified into two types. One is intersertal to as tartan or gridiron twin.

559 (felted) volcanic lithic fragments, which show completely devitrified volcanic glass with

560 a dull appearance (Fig. 14G), and the other type shows felsitic texture (Fig. 14H). Small

561 amounts of sedimentary lithic fragment (Fig. 14A) and quartz-mica tectonite grains (Fig.

562 14I) also occur in some samples of the Shinarump and Cameron members.

31
 32

563

564 6.3. Clay minerals

565

566 Clay mineral composition of the rock samples from the Shinarump Member,

567 Cameron Member, and “purple pedogenic beds” that was analyzed by XRD is

568 summarized in Figures 16–19 and Appendix II. Sandstone samples from the Shinarump

569 Member contain abundant kaolinite followed by illite, smectite and chlorite (Figs. 17 and

570 18), although chlorite was identified only in six sandstone samples. In contrast, mudstone

571 samples of the Shinarump Member are represented by nearly equal amounts of illite and

572 kaolinite, and less amounts of chlorite (Figs. 17 and 18) (Appendix II). Mudstone samples

573 from the Shinarump Member do not have smectite at all. SEM images of both sandstone

574 and mudstone samples from the Shinarump Member do not show authigenic features

575 such as booklet structure and fan-like morphology, which are interpreted to characterize

576 authigenic kaolinite (Do Campo et al., 2010; Huyghe et al., 2011) (Fig. 20A).

577 Sandstone samples from the Cameron Member also have abundant kaolinite,

578 followed by minor illite and smectite in association with non-clay feldspar, although

579 chlorite was identified only in one sandstone sample (Figs. 16, 18 and 19). In contrast,

580 mudstone samples of the Cameron Member and “purple pedogenic beds” are

32
 33

581 characterized by the dominance of kaolinite and smectite, and by minor illite and chlorite

582 (Figs. 18 and 19). The relative abundance of kaolinite and smectite also shows temporal

583 variation in the Cameron Member and “purple pedogenic beds”. The amounts of kaolinite

584 are inversely proportional to those of smectite, and their relative abundance shows

585 temporal sequential zigzag patterns. Smectite in some mudstone samples are randomly

586 interstratified with illite (Fig. 16A). Either kaolinite with booklet structure and fan-like

587 morphology, or smectite with honeycomb structure and rose-like texture, which are

588 common structure of authigenic clays (Do Campo et al., 2010; Huyghe et al., 2011), were

589 not observed in the samples from the Cameron Member and “purple pedogenic beds”

590 using SEM images (Fig. 20B–C).

591

592 7. Discussion

593

594 7.1. Fluvial styles

595

596 Recent studies on modern fluvial systems indicate that plan view channel forms

597 are highly variable, and transitional channel forms between end-member types, such as

598 braided, meandering, and straight, are often observed (Schumm, 2005; Ethridge, 2011).

33
 34

599 In addition, some modern rivers develop channel belts that include braided, meandering,

600 and anastomosing segments within a same valley reach (Lunt et al., 2004; Schumm,

601 2005). Thus, Ethridge (2011) criticized the prediction and classification of ancient

602 channel planform types in terms of straight, meandering, anastomosed, and braided

603 channels, on the basis only of facies and architectural analyses of stratigraphic

604 successions. Nevertheless, the Chinle Formation fluvial deposits document temporal

605 changes in architectural element types from Element DA dominant stacked channel

606 complexes (Fig. 8A) to isolated channel complexes with Element SD and DA that are

607 encased in floodplain muddy deposits (Element FF) (Fig. 9). This suggests that a distinct

608 change in fluvial styles occurred from the Shinarump Member to the Cameron Member

609 and “purple pedogenic beds”.

610 The fluvial style of the Shinarump Member is interpreted to have been low

611 sinuous and braided types (Fig. 21). In general, downstream-accreting bar deposits

612 (Element DA) have been interpreted to be a characteristic component of downstream

613 parts of braid bar deposits (Miall, 1985; Lunt et al., 2004; Adams and Bhattacharya, 2005;

614 Ford and Pyles, 2014). Small-scale channel fills (Element SCF) are indicative of

615 cross-bar channels, and suggest the channel pattern was multi-threaded. Fine-grained fills

616 with erosional bases, which are equivalent to Element LCF of the present study, have also

34
 35

617 been documented from ancient braided channel deposits (Ford and Pyles, 2014).

618 Paleocurrent data, which show flow directions toward the north, northwest, or west (Fig.

619 10), also suggest the existence of low sinuous fluvial channels, although a channel

620 complex (CHC-IV in Fig. 10) are locally composed of laterally accreting, cross-stratified

621 sandstone (Element LA) that exhibit the southeastward-directed paleocurrents (Fig. 6A).

622 This channel complex possibly indicates a local development of a point bar in response to

623 bending of paleochannels.

624 The channel complexes of the Cameron Member are generally isolated within

625 floodplain muddy deposits and do not show lateral continuity between the studied

626 sections (width/thickness ratio ≤ 30). In addition, they consist mainly of horizontally

627 stratified sandstones (Element SD) in association with minor downstream-accreting,

628 cross-stratified sandstones (Element DA) (Figs. 6C and 9). These lithofacies and

629 architectural features are interpreted that the Cameron Member fluvial system was

630 formed mainly by vertical aggradation in a stable and fixed channel (Nadon, 1994).

631 Although the Elements SD and DA generally represents upward-fining trends, the

632 channel deposits of the Cameron Member lack upper bar deposits (Element UB), which

633 are represented by alternating beds of sandstone and mudstone on Elements SD and DA,

634 and are abruptly covered by floodplain muddy deposits (Element FF). These architectural

35
 36

635 features are consistent with a facies model of anastomosed river by Nadon (1994).

636 Although anastomosed fluvial deposits are interpreted to typically contain coal (Flores

637 and Pillmore, 1987), some modern anastomosed river systems even in tropical climatic

638 settings do not develop peat (Smith, 1986). Makaske (2001) defined that anastomosed

639 rivers are composed of two or more interconnected channels that enclose floodbasins.

640 However, the Cameron Member channel complexes do not document two or more

641 interconnected channels, which are represented by sparsely distributed channel

642 complexes in a stratigraphic cross-section (Fig. 10). Because paleocurrent data show the

643 northward-, northwestward-, and/or westward-directed flows (Fig. 10), the Cameron

644 Member fluvial system may document relatively low sinuous rivers. Consequently, the

645 fluvial system of the Cameron Member and “purple pedogenic beds” is interpreted to

646 have deposited as single thread, stable, and fixed channels and wide floodplains (Fig. 21).

647 The size of channel cross-sectional area and bankfull discharge (Qb) are controlled

648 by flood magnitudes that are associated with large rainfalls, except for snowmelt runoff

649 (Knox, 1985; Carson et al., 2007). In addition, the Qb value is proportional to the area size

650 of the drainage basins in case of fluvial systems that are characterized by high

651 precipitation and perennial streams as represented by modern Japanese fluvial systems

652 (Sakaguchi et al., 1986). The estimated geomorphological and hydrological values

36
 37

653 indicate that the Chinle Formation was formed by a fluvial system, which had geometric

654 and hydrological features larger than those of the modern Virgin River that flows near the

655 study area (Table 4). This comparison implies that the Chinle Formation drainage basin

656 may have been larger than that of the modern Virgin River, and climatic condition of the

657 Chinle depositional basin and its source area may have been more humid than that of the

658 drainage basin of the Virgin River. Consequently, the Chinle Formation fluvial channels

659 are considered to have been almost equivalent to modern large rivers in the Colorado and

660 Missouri river basins. The main climate types of the Colorado and Missouri river basins

661 are cold or arid (Peel et al., 2007), and the precipitation of the Chinle Formation drainage

662 basin seems to have been higher than that of the modern Colorado and Missouri river

663 basins. Therefore, the drainage basins of the Chinle Formation fluvial systems in the

664 study area may have been smaller than those of the modern Colorado and Missouri river

665 basins. The main source area of the Chinle Formation was not changed during deposition,

666 so that the temporal variation in the area size of the Chinle Formation drainage basin is

667 also interpreted to have been insignificant. In addition, the effect of a snowmelt runoff in

668 the Chinle Formation drainage basin can be ignored because of its low latitude climatic

669 condition. Therefore, the upward decrease in the estimated Qb values for the Chinle

670 Formation fluvial channels (Table 4) possibly indicates temporal decrease in both flood

37
 38

671 magnitude and precipitation in the Chinle Formation depositional basin as well as in its

672 upper reaches.

673

674 7.2. Detrital composition

675

676 Clast compositions show a distinct change from the Shinarump to Cameron

677 members, which is characterized by upward decrease in the relative amounts of quartzite

678 and quartz, and by upward increase in those of chert and acidic volcanic rock fragments

679 (Fig. 13). The stratigraphic change in the clast composition is interpreted to reflect a

680 temporal change in the source areas, and/or a successive removal of quartzite and parent

681 rocks of quartz by denudation, which may have resulted in an expansion of the exposure

682 of chert and acidic volcanic rocks in the source areas. Three clast types of volcanic rock

683 fragments imply that they were derived from different volcanic terranes in terms of

684 locations and/or ages.

685 Some embayed monocrystalline quartz grains of the Cameron Member (Fig. 14C)

686 imply the supply of volcaniclastic fragments in response to active volcanism during the

687 sedimentation of the member, and is interpreted to have been formed in acid volcanic

688 and/or sub-volcanic rocks (Donaldson and Henderson, 1988). Partially dissolved

38
 39

689 feldspars, which are surrounded and/or replaced by authigenic clay minerals (Fig. 14E)

690 suggests that these grains were affected by chemical weathering after the deposition. Two

691 types of volcanic lithic fragment suggest that they were derived from different volcanic

692 terranes in different locations or at different ages. Commented [BGA3]: To me, most of the volcanic rock
fragments seemed not to have been affected to much
693 The stratigraphic variations in the framework compositions of sandstones (Fig.15) by weathering-related alteration, such as oxidation
and hydrolysis. Because both their mineralogy and
694 are also interpreted to document spatial and/or temporal changes in major rock types in their textures, appeared relatively unmodified, if
compared to the weathered volcanic fragments. This
695 the source areas. Alternatively, they are also considered to have been controlled by a feature can be attributed to direct influx of coeval
volcanic eruption. As another option, those relatively
696 change in the intensity of chemical weathering in the provenance terranes (Suttner and unaltered grains were derived from dissected volcanic
arc which was not affected by intensive chemical
697 Dutta, 1986; Garzanti et al., 2013). For example, an intense chemical weathering weathering (hot and humid climate) after being
exposed. The later explanation may confirm the
698 condition under hot and humid climates may have produced quartzose sands in the climate shift from humid to less humid condition.

699 provenance (Garzanti et al., 2013). A temporal change in the intensity of volcanic activity

700 in the hinterlands is also a possible explanation for the upward increase in feldspars and

701 volcanic lithic fragments in association with the occurrences of embayed monocrystalline

702 quartz grains and volcanic lithic fragments with felsitic texture. However, relative

703 abundances of volcanic lithic fragments are limited as a framework composition in

704 sandstones of both the Shinarump and Cameron members (Appendix I), and the effect of

705 the temporal change in volcanic activity in the source terranes can be excluded as a major

706 cause in the vertical change in the sandstone framework composition. Consequently, we

39
 40

707 interpret that the temporal variation in the sandstone framework composition was

708 controlled by the changes in the basement rock types in the source mountains and/or the

709 temporal changes in the intensity of chemical weathering of the source rocks.

710 Although smectite in some mudstone samples are randomly interstratified with

711 illite (Fig. 16A), high percentages of smectite and kaolinite in some sandstone and

712 mudstone samples indicate that the Chinle Formation fluvial deposits have been

713 subjected only to the effect of an early stage in diagenesis. SEM images, which do not

714 show authigenic structures (Fig. 20), also suggest that diagenetic modification of

715 minerals after the deposition were insignificant. Thus, clay minerals from Chinle

716 Formation fluvial deposit samples are mainly interpreted to be detrital in origin, although

717 some diagenetic modification of detrital minerals are observed in partially dissolved

718 K-feldspar that is surrounded and replaced by clay minerals (Fig. 14E). Detrital illite and

719 chlorite, therefore, are considered to have been mechanically ground clay minerals

720 (Singer, 1984; Do Campo et al., 2010; Huyghe et al., 2011; Ghosh et al., 2019). Illite is

721 interpreted to have been derived primarily from granitic, metamorphic, and/or

722 sedimentary rocks in the source areas. Chlorite may have been derived from the igneous

723 and/or sedimentary rocks in the source areas, or from mafic volcanic input as suggested

724 by Jin et al. (2018), who analyzed sandstone samples from the Chinle Formation in the

40
 41

725 Petrified Forest National Park, Arizona. Because illite and chlorite are unstable under

726 warm and humid climatic conditions (Singer, 1984, Clift et al., 2014; Ghosh et al., 2019),

727 some of these clay minerals found in the Chinle Formation are considered to have been

728 derived from bedrocks distributed near its depositional basin. For example, the

729 underlying Moenkopi Formation in southwestern Utah includes abundant illite and

730 chlorite (Schultz, 1963), and it may have been a source of illite and chlorite during Chinle

731 Formation fluvial depositions. Relative abundance of detrital kaolinite and smectite in the

732 mudstone samples from the Cameron Member and “purple pedogenic beds” shows

733 vertical variation with a zigzag pattern (Figs. 18 and 19), which is interpreted to have

734 caused mainly by temporal fluctuation in the intensity of chemical weathering of the

735 basement rocks in the source areas, rather than by the change in rock types of the

736 basement rocks. The occurrence of smectite may also indicate an effect of the input of

737 volcaniclastic sediments from the source terraces (Singer, 1984; Do Campo et al., 2010).

738 However, the effect of the input of volcaniclastic sediments is interpreted to have been

739 excluded for the supply of detrital smectite to the Chinle Formation, because of a

740 relatively small proportion of volcanic rock fragments as sandstone framework grains

741 (Appendix I).

742 The channel sandstones and floodplain mudstones show very high contrast

41
 42

743 between clay mineral compositions in each member (Figs. 17–19). The differences in the

744 clay mineral compositions between channel sandstones and floodplain mudstones may

745 have been caused by segregation of clay particles during the transportation and deposition

746 in the fluvial systems (Gibbs, 1977). Larger clay particles (i.e., kaolinite) seem to have

747 selectively deposited in fluvial channels, while smaller clay particles, such as smectite,

748 illite, and chlorite, may have been concentrated in suspended load from which floodplain

749 muddy deposits formed. In contrast, clay mineral composition of mudstone samples from

750 the Cameron Member and “purple pedogenic beds” document distinct vertical variation

751 in the relative abundance of kaolinite and smectite. This vertical variation is considered to

752 have been caused by fluctuation in the intensity of chemical weathering in the source

753 areas in response to temporal variation in climatic conditions. The occurrence of

754 abundant kaolinite is generally linked to a humid climate, and the increase in the relative

755 abundance of smectite is interpreted as a signal of a climate with contrasting seasonality

756 that may have been associated with a pronounced dry season in a continental setting

757 (Singer, 1984; Chamley, 1989; Thiry, 2000; Net et al., 2002; Do Campo et al., 2007,

758 Huyghe et al., 2011; Wan et al., 2007; Clift et al., 2014). Smectite and kaolinite in the

759 mudstone samples of the Cameron Member and “purple pedogenic beds”, whose relative

760 abundances show inversely proportional zigzag patterns (Figs.18 and 19), are interpreted

42
 43

761 to reflect periodic changes of humid and wet-dry seasonal conditions in the source areas.

762 The development of vertisol (Martz et al., 2017) also suggests wet-dry seasonal

763 conditions during the deposition of the Cameron Member. The mudstone samples of the

764 Shinarump Member include kaolinite, yet do not contain smectite (Figs. 17 and 18).

765 These clay mineral compositions imply that wet-dry seasonal condition had not

766 developed, and the climate was probably humid during the deposition of the Shinarump

767 Member.

768

769 7.3. Climatic shift

770

771 The paleohydrological data of the Chinle Formation indicate a temporal decrease

772 in flood discharge (Fig. 11). In addition, detrital compositions of sandstone framework

773 grains (Fig. 15) and clay minerals (Figs. 17–19) also suggest a temporal reduction in the

774 intensity of chemical weathering. These data imply that climatic conditions for the

775 development of the Chinle Formation may have shifted from humid to wet-dry seasonal

776 conditions with time. The stable and fixed channels of the Cameron Member (Fig. 21)

777 also suggest a seasonal change in discharge, because the stable and fixed anastomosed

778 fluvial systems have commonly been documented from modern fluvial systems, which

43
 44

779 have a seasonal peak in the annual hydrograph that is caused by meltwater from snow and

780 glacier during spring and summer or by seasonal intensification of rainfall in adjacent

781 mountains (Nadon, 1994). Paleoenvironmental interpretation of the Chinle Formation in

782 the northern Arizona indicates that humid monsoonal condition is considered to have

783 collapsed at about 215 Ma (Nordt et al., 2015; Baranyi et al., 2018; Jin et al., 2018), and

784 the age of the climatic shift is correlated to the depositional age of the Sonsela Member

785 (Nordt et al., 2015). Dubiel and Hasiotis (2011) and Lucas and Tanner (2018) described

786 vertisols from the Petrified Forest Member, which is distributed mainly in the Four

787 Corners region, and is younger than the Sonsela Member, and interpreted a wet-dry

788 seasonal climate from the paleosols. These wet-dry seasonal climatic data are consistent

789 with those from the Cameron Member and “purple pedogenic beds” in southwestern Utah,

790 which also contain vertisols (Martz et al., 2017) and is characterized by the occurrence of

791 smectite (Figs. 16, 18, and 19). Although the radioisotopic age of the Chinle Formation in

792 southwestern Utah has not yet been determined, the Cameron Member and “purple

793 pedogenic beds” in southwestern Utah can be correlated to the Monitor Butte Member

794 (225–220 Ma?) in northern Arizona on the basis of vertebrate biostratigraphy (Martz et

795 al., 2017). This correlation indicates that the development of wet-dry seasonal climatic

796 condition may have started in the Chinle Formation depositional basin in southwestern

44
 45

797 Utah before 215 Ma. This earlier development of the wet-dry climate in the southwestern

798 Utah may have responded to a local development of rain shadow of the Cordilleran

799 volcanic arc (Atchley et al., 2013; Nordt et al., 2015). A part of the Cordilleran volcanic

800 arc, such as the Scheelite pluton (232–219 Ma, Riggs et al., 2016), which was active

801 during the depositional periods of the Shinarump Member, Cameron Member and “purple

802 pedogenic beds”, is considered to have blocked a monsoonal circulation to southwestern

803 Utah as a rain shadow.

804

805 7.4. Basin evolution

806

807 Several sequence boundaries have been defined in the Chinle Formation on the

808 basis of identification of distinct erosional surfaces that are interpreted to have formed in

809 response to the development of fluvial paleovalleys and interfluve paleosols (Dubiel,

810 1991; Demko et al., 1998; Cleveland et al., 2007; Dubiel and Hasiotis, 2011). Mottled

811 mudstones of the uppermost part of the Moenkopi Formation, which was formerly

812 included in the basal part of the Shinarump Member by Stewart et al. (1972), is

813 interpreted as pedogenically modified deposits (Fig. 8C) that document long term

814 subaerial exposure of sediments and basement rocks. These pedogenically modified

45
 46

815 deposits have been interpreted to be accompanied by valley incision during degradation

816 periods (Wright and Marriott, 1993; Shanley and McCabe, 1994). Because the basal

817 surface of a channel complex of the Shinarump Member is a composite erosional surface

818 of channel deposits and channel complexes (Fig. 10), the composite basal erosional

819 surface is interpreted to have formed in response to alternation of incision and deposition,

820 and is equivalent to a composite scour surfaces of Holbrook and Bhattacharya (2012) and

821 Blum et al. (2013). Composite scour surfaces in fluvial successions are not consistent

822 with the concept of the traditional sequence boundary, which is interpreted to represent a

823 single falling stage topographic surface of erosion (i.e., unconformity) (Posamentier et al.,

824 1988; Van Wagoner et al., 1988). Holbrook and Bhattacharya (2012) proposed an

825 extension of the conceptual definition of the sequence boundary to include traits of

826 regional composite scour surfaces. Following this extended definition of sequence

827 boundary, the basal composite scour surfaces of the channel complexes of the Shinarump

828 Member is interpreted to be a sequence boundary (6th-order BS, Figs. 8A and 10).

829 A factor that controlled architecture and stacking pattern of the Chinle Formation

830 fluvial system has been controversial. Dubiel and Hasiotis (2011) and Miall (2014)

831 interpreted that aggradation-and-degradation cycles of the Chinle Formation may have

832 been controlled primarily by climatic fluctuation, while Blakey and Gubitosa (1984),

46
 47

833 Cleveland et al. (2007), and Trendell et al. (2012) emphasized that tectonism was the

834 prime controlling factor for the development of the formation. The Chinle Formation

835 depositional basin is interpreted to have developed in a backarc region of the Cordilleran

836 volcanic arc, and also have been surrounded by the Cordilleran volcanic arc to the west,

837 highlands of the Ancestral Rocky Mountains to the east, and a broad and gently sloping

838 elevated area that is called the Mogollon Slope to the south (Fig. 1). Therefore, the basin

839 is considered to have been affected by differential subsidence (Kraus and Middleton,

840 1987; Dickinson and Gehrels, 2008). However, in southwestern Utah, stratigraphic

841 thicknesses of the Chinle Formation fluvial deposits do not exhibit distinct spatial

842 variation, and the paleocurrent directions also do not show distinct spatial and temporal

843 variations (Fig. 10). Furthermore, the effects of faulting and folding within the basin are

844 not recognized during the deposition of the Chinle Formation. Consequently, the tectonic

845 movements are not interpreted to have influenced the architecture and stacking patterns

846 of the Chinle Formation fluvial system in southwestern Utah. Alternatively, the reduction

847 of water discharge and sediment load is interpreted to cause an upstream shift of a distal,

848 fine-grained fluvial facies belt in association with development of isolated channels,

849 regardless of low rate of basin subsidence (Strong et al., 2005). Geomorphological and

850 paleohydrological analyses of the Chinle Formation fluvial channels indicate upward

47
 48

851 decreases in the sizes of paleochannels and bankfull discharges (Fig. 11). Sediment load

852 discharged by a fluvial system under a wet-dry seasonal climate is generally greater than

853 those observed under a more humid monsoonal climate as a response to restriction of

854 vegetation under a wet-dry seasonal climate (Cecil, 1990). The locally abundant

855 occurrence of fossil plants including a variety of ferns, conifers, horsetails, and

856 bennettitaleans from the channel deposits of both Shinarump and Cameron members

857 (A.R.C. Milner, unpublished data), implies that the uplands of the Chinle Formation

858 depositional basin were well forested. In addition, the studies of fossil plant assemblages

859 of the Chinle Formation in the Four Corners region and the adjacent areas suggest that the

860 uplands, lowlands, and floodplains of the Chinle Formation depositional basin may have

861 been well vegetated and forested during the Norian time (Litwin et al., 1991; Lindström et

862 al., 2016; Baranyi et al., 2018). Consequently, sediment loads in the Chinle Formation

863 fluvial system is considered not to have increased as a result of thick vegetation cover

864 under the wet-dry seasonal climate. The upward decrease in channel density in the Chinle

865 Formation, therefore, is interpreted to have responded to the upward decrease in bankfull

866 discharge (Fig. 11) in harmony with the climatic shift (Figs. 15, 17–19). Consequently,

867 the stacking patterns of the Chinle Formation channel deposits in southwestern Utah is

868 interpreted to have been controlled mainly by the shift from humid to wet-dry seasonal

48
 49

869 climatic condition.

870

871 8. Conclusions

872

873 This study investigated spatial and temporal variations in fluvial architecture,

874 stacking patterns, paleohydrological features, and detrital composition of the Upper

875 Triassic (Norian) Chinle Formation, which consists of the Shinarump Member, Cameron

876 Member, and “purple pedogenic beds” in ascending order in the southwestern Utah, for

877 elucidating the relationships between fluvial sedimentation and climatic fluctuation in a

878 sedimentary basin that was formed in western Pangaea. The main conclusions of the

879 study are as follows:

880 (1) The Shinarump Member consists of amalgamated channel complexes, which

881 represent multiple-thread, braided fluvial style. In contrast, the overlying the Cameron

882 Member and “purple pedogenic beds” is represented by isolated channel complexes and

883 component single thread, stable, and fixed channel deposits, which were encased in

884 floodplain muddy deposits.

885 (2) Geomorphological and paleohydrological analyses of the channel deposits

886 represent upward decrease in channel cross-sectional sizes and bankfull discharges of the

887 Chinle Formation fluvial system. Paleohydrological features of the Chinle Formation

49
 50

888 fluvial channels are almost equivalent to those of some modern rivers in the Colorado and

889 Missouri river basins, such as Yellowstone River near Miles City, Montana for the

890 Shinarump Member paleochannels, and the Green River near Green River, Utah for the

891 Cameron Member paleochannels, respectively.

892 (3) Sandstone samples from the Shinarump Member are represented by more

893 quartzose framework composition than those from the Cameron Member. Mudstone

894 samples from the Cameron Member and “purple pedogenic beds” are characterized by

895 abundant contents of kaolinite and smectite, while those from the Shinarump Member

896 lack smectite. These stratigraphical changes in detrital compositions suggest that the

897 intensity in chemical weathering of source rocks in the provenance terranes temporally

898 decreased.

899 (4) The temporal variations in the paleohydrological features and detrital

900 compositions documented in the Chinle Formation are interpreted to have been in

901 harmony with a temporal shift from humid to wet-dry seasonal climatic conditions in

902 southwestern Utah during the Norian time. This shift suggests a local development of a

903 wet-dry seasonal climate before 215 Ma in the Chinle Formation depositional basin.

904 (5) The stratigraphical change in the fluvial architecture from the amalgamated

905 channel complexes of multiple-thread, braided fluvial system to the isolated channel

50
 51

906 complexes of stable and fixed channels and thick floodplain muddy deposits of the Chinle

907 Formation is interpreted to have been controlled primarily by the shift from humid to

908 wet-dry seasonal climatic conditions rather than by tectonic movements in the

909 sedimentary basin and its hinterland.

910

911 Acknowledgments

912

913 This research was conducted under permit issued by the United States Department

914 of the Interior National Park Service (Permit numbers: ZION-2005-SCI-0022,

915 ZION-2006-SCI-0019, and ZION-2013-SCI-0015) and Bureau of Land Management

916 (BLM) in Utah (Tracking/Agreement number: 5-17). We are grateful to Drs. Jerry D.

917 Harris (Dixie State University), M. Matsukawa (Tokyo Gakugei University), A. Inoue

918 and N. Furukawa (Chiba University) for valuable discussions and suggestions. This

919 research has been supported in part by the Japan Society for the Promotion of Science

920 KAKENHI Grant Number JP17K12968.

921

922 References

923

924 Adams, M.M., Bhattacharya, J.P., 2005. No change in fluvial style across a sequence

925 boundary, Cretaceous Blackhawk and Castlegate Formations of central Utah, U.S.A.

926 Journal of Sedimentary Research 75, 1038–1051.

51
 52

927 Adhiperdana, B.G., Hendarmawan, Shibata, K., Ito, M., 2018. Relationships between

928 discharge parameters and cross-sectional channel dimensions of rivers in an active

929 margin influenced by tropical climate: The case of modern fluvial systems in the

930 Indonesian islands. Catena 171, 645–680.

931 Alexander, J.S., McElroy, B.J., Huzurbazar, S., Murr, M.L., 2020. Elevation gaps in

932 fluvial sandbar deposition and their implications for paleodepth estimation.

933 Geology 48, 718–722, https://doi.org/10.1130/G47521.1

934 Allen, J.R.L., 1970. Studies in fluviatile sedimentation: a comparison of fining-upwards

935 cyclothems, with special reference to coarse member composition and

936 interpretation. Journal of Sedimentary Petrology 40, 298–323.

937 Allen, J.R.L., 1982. Sedimentary structures: Their character and physical basis Volume I.

938 Developments in Sedimentology 30A. Elsevier, Amsterdam (593p).

939 Allen, J.R.L., 1983. Studies in fluviatile sedimentation: bars, bar complexes and

940 sandstone sheets (low-sinuosity braided streams) in the Brownstones (L. Devonian),

941 Welsh Borders. Sedimentary Geology 33, 237–293.

942 Antia, J., Fielding, C.R., 2011. Sequence stratigraphy of a condensed

943 low-accommodation succession: lower upper Cretaceous Dakota Sandstone, henry

944 Mountains, southeastern Utah. The American Association of Petroleum Geologists

52
 53

945 Bulletin, 95, 413–447.

946 Atchley, S.C., Nordt, L.C., Dworkin, S.I., Ramezani, J., Parker, W.G., Ash, S.R., Bowring,

947 S.A., 2013. A linkage among Pangean tectonism, cyclic alluviation, climate change,

948 and biologic turnover in the Late Triassic: the record from the Chinle Formation,

949 southwestern United States. Journal of Sedimentary Research 84, 1147–1161.

950 Bailey, E.H., Stevens, R.E., 1960. Selective staining of K-feldspar and plagioclase on

951 rock slabs and thin sections. The American Mineralogist 45, 1020–1025.

952 Baranyi, V., Reichgelt, T., Olsen, P.E., Parker, W.G., Kürschner, W.M., 2018. Norian

953 vegetation history and related environmental changes: New data from the Chinle

954 Formation, Petrified Forest National Park (Arizona, SW USA). GSA Bulletin 130,

955 775–795.

956 Berwick, V.K., 1962. Floods in Utah, Magnitude and Frequency. Geological Survey

957 Circular 457. U.S. Geological Survey (24p).

958 Biscaye, P.E., 1965. Mineralogy and Sedimentation of resent deep-sea clay in the Atlantic

959 Ocean and adjacent seas and oceans. Geological Society of America Bulletin 76,

960 803–832.

961 Bilodeau, W.L., 1986. The Mesozoic Mogollon Highlands, an Early Cretaceous rift

962 shoulder. Journal of Geology 94, 724–735.

53
 54

963 Blair, T.C., Bilodeau, W.L., 1988. Development of tectonic cyclothems in rift, pull-apart,

964 and foreland basins: sedimentary response to episodic tectonism. Geology 16,

965 517–520.

966 Blakey, R.C., Gubitosa, R., 1984. Controls of sandstone body geometry and architecture

967 in the Chinle Formation (Upper Triassic), Colorado Plateau. Sedimentary Geology

968 38, 51–86.

969 Blakey, R.C., Ranney, W., 2008. Ancient landscapes of the Colorado Plateau. Grand

970 Canyon Association (156p).

971 Blum, M.D., Guccione, M.J., Wysocki, D.A., Robnett, P.C., 2000. Late Pleistocene

972 evolution of the Central Mississippi Valley, Southern Missouri to Arkansas.

973 Geological Society of America Bulletin 112, 221–235.

974 Blum, M.D., Martin, J., Milliken, K., Garvin, M., 2013. Paleovalley systems: Insights

975 from Quaternary analogs and experiments. Earth-Science Reviews 116, 128–169.

976 Blum, M.D., Törnqvist, T.E., 2000. Fluvial responses to climate and sea-level change: a

977 review and look forward. Sedimentology 47 (Suppl. 1), 2–48.

978 Bridge, J.S., 1993. Description and interpretation of fluvial deposits: a critical perspective.

979 Sedimentology 40, 801–810.

980 Bridge, J.S., 2003. Rivers and floodplains: forms, processes, and sedimentary record.

54
 55

981 Blackwell publishing, Malden (491p).

982 Bridge, J.S., Diemer J.A., 1983. Quantitative interpretation of an evolving ancient river

983 system. Sedimentology 30, 599–623.

984 Bridge, J.S., Leeder, M.R., 1979. A simulation model of alluvial stratigraphy.

985 Sedimentology 26, 617–644.

986 Bridge, J.S., Mackey, S.D., 1993. A theoretical study of fluvial sandstone body

987 dimensions. In: Flint, S.S., Bryant, I.D. (eds.), Geological Modeling of

988 Hydrocarbon Reservoirs. International Association of Sedimentologists, Special

989 Publication 15, pp. 213–236.

990 Bridge, J.S., Jalfin, G.A., Georgieff, S.M., 2000. Geometry, lithofacies, and spatial

991 distribution of Cretaceous fluvial sandstone bodies, San Jorge Basin, Argentina:

992 outcrop analog for the hydrocarbon-bearing Chubut Group. Journal of Sedimentary

993 Research 70, 341–359.

994 Burns, C.E., Mountney, N.P., Hodgson, D.M., Colombera, L., 2017. Anatomy and

995 dimensions of fluvial crevasse-splay deposits: Examples from the Cretaceous

996 Castlegate Sandstone and Neslen Formation, Utah, U.S.A. Sedimentary Geology

997 351, 21–35.

998 Carson, E.C., Knox, J. C., Mickelson, D.M., 2007. Response of bankfull flood

55
 56

999 magnitudes to Holocene climate change, Uinta Mountains, northeastern Utah.

1000 Geological Society of America Bulletin 119, 1066–1078.

1001 Cecil, C.B., 1990. Paleoclimate controls on stratigraphic repetition of chemical and

1002 siliciclastic rocks. Geology 18, 533–536.

1003 Chamley, H., 1989. Clay Sedimentology. Springer-Verlag, New York (623p).

1004 Cleveland, D.M., Atchley, S.C., Nordt, L.C., 2007. Continental sequence stratigraphy of

1005 the Upper Triassic (Norian-Rhaetian) Chinle strata, northern New Mexico, U.S.A.:

1006 allocyclic and autocyclic origins of paleosol-bearing alluvial successions. Journal

1007 of Sedimentary Research 77, 909–924.

1008 Clift, P.D., Wan, S., Blusztajn, J., 2014. Reconstructing chemical weathering, physical

1009 erosion and monsoon intensity since 25 Ma in the northern South China Sea: A

1010 review of competing proxies. Earth-Science Reviews 130, 86–102.

1011 Colombera, L., Arévalo, O.J., Mountney, N.P., 2017. Fluvial-system response to climate

1012 change: The Paleocene-Eocene Tremp Group, Pyrenees, Spain. Global and

1013 Planetary Change 157, 1–17.

1014 Demko, T.M., Dubiel, R.F., Parrish, J.T., 1998. Plant taphonomy in incised valleys:

1015 Implications for interpreting paleoclimate from fossil plants. Geology 26,

1016 1119–1122.

56
 57

1017 Dickinson, W.R., Gehrels, G.E., 2008. U-Pb ages of detrital zircons in relation to

1018 paleogeography: Triassic paleodrainage networks and sediment dispersal across

1019 southwest Laurentia. Journal of Sedimentary Research 78, 745–764.

1020 Dickinson, W.R., Soreghan, G.S., Giles, K.A., 1994. Glacio-eustatic origin of

1021 Permo-Carboniferous stratigraphic cycles: evidence from the southern Cordilleran

1022 foreland region. In Dennison, J.M., Ettensohn, F.R. (eds.), Tectonic and eustatic

1023 controls on sedimentary cycles. Concepts in Sedimentology and Paleontology 4, pp.

1024 25–34. SEPM.

1025 Do Campo, M, del Papa, C., Jiménez-Millán, J., Nieto, F., 2007. Clay mineral

1026 assemblages and analcime formation in a Palaeogene fluvial-lacustrine sequence

1027 (Maíz Gordo Formation Palaeogen) from northwestern Argentina. Sedimentary

1028 Geology 201, 56–74.

1029 Do Campo, M., del Papa, C., Nieto, F., Hongn, F., Petrinovic, I., 2010. Integrated analysis

1030 for constraining paleoclimatic and volcanic influences on clay-mineral assemblages

1031 in orogenic basins (Paleogene Andean foreland, Northwestern Argentina).

1032 Sedimentary Geology 228, 98–112.

1033 Donaldson, C.H., Henderson, C. M. B., 1988. A new interpretation of round embayments

1034 in quartz crystals. Mineralogical Magazine 52, 27–33.

57
 58

1035 Dubiel, R.F., 1991. Architectural-facies analysis of nonmarine depositional systems in

1036 the Upper Triassic Chinle Formation, southeastern Utah. In: Miall, A.D., Tyler, N.

1037 (eds.), The three-dimensional facies architecture of terrigenous clastic sediments

1038 and its implications for hydrocarbon discovery and recovery. Concepts in

1039 Sedimentology and Paleontology 3, pp. 103–110. SEPM.

1040 Dubiel, R.F., 1992. Sedimentology and depositional history of the Upper Triassic Chinle

1041 Formation in the Uinta, Piceance, and Eagle basins, northwestern Colorado and

1042 northeastern Utah. United States Geological Survey Bulletin 1787-W, 1–25.

1043 Dubiel, R.F., Hasiotis, S.T., 2011. Deposystems, paleosols, and climatic variability in a

1044 continental system: the Upper Triassic Chinle Formation, Colorado Plateau, U.S.A.

1045 In: Davidson, S.K., Leleu, S., North, C.P. (eds.), From River to rock record: the

1046 preservation of fluvial sediments and their subsequent interpretation. SEPM Special

1047 Publication 97, pp. 393–421.

1048 Dubiel, R.F., Parrish, J.T., Parrish, J.M., Good, S C., 1991. The Pangaean Megamonsoon

1049 -Evidence from the Upper Triassic Chinle Formation, Colorado Plateau. PALAIOS

1050 6, 347–370.

1051 Ethridge, F.G., 2011. Interpretation of ancient fluvial channel deposits: review and

1052 recommendations. In: Davidson, S.K., Leleu, S., North, C.P. (eds.), From River to

58
 59

1053 rock record: the preservation of fluvial sediments and their subsequent

1054 interpretation. SEPM Special Publication 97, pp. 9–35.

1055 Ethridge, F.G., Schumm, S.A., 1978. Reconstructing paleochannel morphologic and flow

1056 characteristics: methodology, limitations, and assessment. In: Miall, A. D. (ed.),

1057 Fluvial sedimentology. Canadian Society Petroleum Geologists Memoir 5, pp.

1058 703–721.

1059 Flores, R.M., Pillmore, C.L., 1987. Tectonic control on alluvial paleoarchitecture of the

1060 Cretaceous and Tertiary Raton Basin, Colorado and New Mexico. In: Ethridge, F.G.,

1061 Flores, R.M., Harvey, M.D. (eds.), Recent Development in Fluvial Sedimentology.

1062 SEPM Special Publication 39, pp. 311–320.

1063 Ford, G.L., Pyles, D.R., 2014. A hierarchical approach for evaluating fluvial systems:

1064 Architectural analysis and sequential evolution of the high net-sand content, middle

1065 Wasatch Formation, Uinta Basin, Utah. The American Association of Petroleum

1066 Geologists Bulletin 98, 1273–1304.

1067 Garzanti, E., Padoan, M., Andò, S., Resentini, A., Vezzoli, G., Lustrino, M., 2013.

1068 Weathering and relative durability of detrital minerals in equatorial climate: sand

1069 petrology and geochemistry in the East African Rift. The Journal of Geology 121,

1070 547–580.

59
 60

1071 Gibbs, R.J., 1977. Clay mineral segregation in the marine environment. Journal of

1072 Sedimentary Petrology 47, 237–243.

1073 Ghosh, S., Mukhopadhyay, J., Chakraborty, A., 2019. Clay mineral and geochemical

1074 proxies for intense climate change in the Permian Gondwana Rock Record from

1075 eastern India. AAAS Research 2019, https://doi.org/10.34133/2019/8974075.

1076 Hampson, G.J., Jewell, T. O., Irfan, N., Gani, M. R., Bracken, B., 2013. Modest change in

1077 fluvial style with varying accommodation in regressive alluvial-to-coastal-plain

1078 wedge: Upper Cretaceous Blackhawk Formation, Wasatch Plateau, central Utah,

1079 U.S.A. Journal of Sedimentary Research 83, 145–169.

1080 Hillier, R.D., Marriott, S.B., Williams, B.P.J., Wright, V.P., 2007. Possible climate

1081 variability in the Lower Old Red Sandstone Conigar Pit Sandstone Member (early

1082 Devonian), South Wales, UK. Sedimentary Geology 202, 35–57.

1083 Hintze, L.F. 1997. Geologic highway map Utah with a topographic map that shows

1084 western national parks. Brigham Young University Geology Studies–Special

1085 Publication 3.

1086 Holbrook, J.M., 2001. Origin, genetic interrelationships, and stratigraphy over the

1087 continuum of fluvial channel-form bounding surfaces: an illustration from middle

1088 Cretaceous strata, south-eastern Colorado. Sedimentary Geology 144, 179–222.

60
 61

1089 Holbrook, J.M., Bhattacharya, J.P., 2012. Reappraisal of the sequence boundary in time

1090 and space: Case and considerations for an SU (subaerial unconformity) that is not a

1091 sediment bypass surface, a time barrier, or an unconformity. Earth-Science Reviews

1092 113, 271–302.

1093 Howell, E.R., Blakey, R.C., 2013. Sedimentological constraints on the evolution of the

1094 Cordilleran arc: New insights from the Sonsela Member, Upper Triassic Chinle

1095 Formation, Petrified Forest National Park (Arizona, USA). Geological Society of

1096 America Bulletin 125, 1349–1368.

1097 Huyghe, P., Guilbaud, R., Bernet, M., Galy, A., Gajurel, A.P., 2011. Significance of the

1098 clay mineral distribution in fluvial sediments of the Neogene to Recent Himalayan

1099 Foreland Basin (west-central Nepal). Basin Research 23, 332–345.

1100 Ingersoll, R.V., Bullard, T.F., Ford, R.L., Grimm, J.P., Pickle, J.D., Sares, S.W., 1984. The

1101 effect of grain size on detrital modes: a test of the Gazzi-Dickinson point-counting

1102 method. Journal of Sedimentary Petrology 54, 103–116.

1103 Iseya F. 1989. Mechanism of inverse grading of suspended load deposits. In Taira, A. and

1104 Masuda, F. eds., Sedimentary Facies in the Active Plate Martin, 113–129. Terra

1105 Scientific Publishing Company, Tokyo.

1106 Jin, C., Dworkin, S., Atchley, S., Nordt, L., 2018. Eogenetic diagenesis of Chinle

61
 62

1107 sandstones, Petrified Forest National Park (Arizona, USA): A record of Late

1108 Triassic climate change. Sedimentology, 65, 1277–1300, doi: 10.111/sed.12421

1109 Jo, H.R., 2003. Depositional environments, architecture, and controls of Early Cretaceous

1110 non-marine successions in the northwestern part of Kyongsang Basin, Korea.

1111 Sedimentary Geology 161, 269–294.

1112 Kirkland, J.I., Milner, A.R.C., 2006. The Moenave Formation at the St. George Dinosaur

1113 Discovery Site at Johnson Farm, St. George, southwestern Utah. In: Harris, J.D.,

1114 Lucas, S.G., Spielmann, J.A., Lockley, M.G., Milner, A.R.C., and Kirkland, J.I.

1115 (eds.), The Triassic-Jurassic Terrestrial Transition, pp. 289–309. New Mexico

1116 Museum of Natural History & Science Bulletin 37.

1117 Kirkland, J.I., Milner, A.R.C., Olsen, P.E., Hargrave, J.E., 2014. The Whitmore Point

1118 Member of the Moenave Formation in its type area in northern Arizona and its age

1119 and correlation with the section in St. George, Utah: Evidence for two major

1120 lacustrine sequences. In: MacLean, J.S., Biek, R.F., Huntoon, J.E. (eds.), Geology

1121 of Utah’s Far South, 321–355. Utah Geological Association Publication 43.

1122 Knox, J.C., 1983. Responses of river systems to Holocene climates. In: Wright, H.E.Jr.

1123 (ed.), Late-Quaternary Environments of the United States Volume 2 The Holocene,

1124 pp. 26-41. University of Minnesota Press, Minneapolis.

62
 63

1125 Knox, J.C., 1985. Responses of floods to Holocene climatic change in the Upper

1126 Mississippi Valley. Quaternary Research 23, 287–300.

1127 Kraus, M.J., Middleton, L.T. 1987. Dissected paleotopography and base-level changes in

1128 a Triassic fluvial sequence. Geology 15, 18–21.

1129 Leeder, M. R., 1973. Fluviatile fining-upwards cycles and the magnitude of

1130 palaeochannels. Geological Magazine 110, 265–276.

1131 Leclair, S.F., Bridge, J.S., 2001. Quantitative interpretation of sedimentary structures

1132 formed by river dunes. Journal of Sedimentary Research 71, 713–716.

1133 Lindström, S., Irmis, R.B., Whiteside, J.H., Smith, N.D., Nesbitt, S.J., Turner, A.H., 2016.

1134 Palynology of the upper Chinle Formation in northern New Mexico, U.S.A.:

1135 Implications for biostratigraphy and terrestrial ecosystem change during the Late

1136 Triassic (Norian–Rhaetian). Review of Palaeobotany and Palynology 225,

1137 106–131.

1138 Litwin, R.J., Traverse, A., Ash, S.R., 1991. Preliminary palynological zonation of the

1139 Chinle Formation, southwestern U.S.A., and its correlation to the Newark

1140 Supergroup (eastern U.S.A.). Review of Palaeobotany and Palynology 68,

1141 269–287.

1142 Lorenz, J.C., Heinze, D.M., Clark, J.A., Searls, C. A., 1985. Determination of widths of

63
 64

1143 meander-belt sandstone reservoirs from vertical downhole data, Mesaverde Group,

1144 Piceance Creek Basin, Colorado. The American Association of Petroleum

1145 Geologists Bulletin 69, 710–721.

1146 Lucas, S.G., Tanner, L.H., 2007. Tetrapod biostratigraphy and biochronology of the

1147 Triassic–Jurassic transition on the southern Colorado Plateau, USA. Palaeogeogr.

1148 Palaeoclimatol. Palaeoecol. 244, 247–256.

1149 Lucas, S.G., Tanner, L.H., 2018. Record of the Carnian wet episode in strata of the Chinle

1150 Group, western USA. Journal of Geological Society 175, 1004–1011.

1151 Lunt, I.A., Bridge, J.S., Tye, R.S., 2004. A quantitative, three-dimensional depositional

1152 model of gravelly braided rivers. Sedimentology 51, 377–414.

1153 Makaske, B., 2001. Anastomosing rivers: a review of their classification, origin and

1154 sedimentary products. Earth-Science Reviews 53, 149–196.

1155 Martinsen, O.J., Ryseth, A., Helland-Hansen, W., Flesche, H., Torkildsen, G. Idil, S.,

1156 1999. Stratigraphic base level and fluvial architecture: Ericson Sandstone

1157 (Campanian), Rock Springs Uplift, SW Wyoming, USA. Sedimentology 46,

1158 235–259.

1159 Martz, J.W., Kirkland, J.I., Milner, A.R.C., Parker W.G., Santucci, V.L., 2017. Upper

1160 Triassic lithostratigraphy, depositional systems, and vertebrate paleontology across

64
 65

1161 southern Utah. Geology of the Intermountain West, 4, 99–180.

1162 Miall, A.D., 1985. Architectural-element analysis: a new method of facies analysis

1163 applied to fluvial deposits. Earth Science Reviews 22, 261–308.

1164 Miall, A.D., 1988. Reservoir heterogeneities in fluvial sandstones: Lessons from outcrop

1165 studies. The American Association of Petroleum Geologists Bulletin 72, 682–697.

1166 Miall, A.D., 1996. The Geology of Fluvial Deposits. Springer-Verlag, Heidelberg (582p).

1167 Miall, A.D., 2014. Fluvial depositional systems. Springer International Publishing,

1168 Switzerland (316p).

1169 Nadon, G.C., 1994. The genesis and recognition of anastomosed fluvial deposits: data

1170 from the St. Mary River Formation, southwestern Alberta, Canada. Journal of

1171 Sedimentary Research B64, 451–463.

1172 Net, L.I., Alonso, M.S., Limarino, C.O., 2002. Source rock and environmental control on

1173 clay mineral associations, Lower Section of Paganzo Group (Carboniferous),

1174 Northwest Argentina. Sedimentary Geology 152, 183–199.

1175 Nordt, L., Atchley S., Dworkin S., 2015. Collapse of the Late Triassic megamonsoon in

1176 western equatorial Pangea, present-day American Southwest. Geological Society of

1177 America Bulletin 127, 1798–1815.

1178 Ohmori, H., 1983. Characteristics of the erosion rate in the Japanese mountains from the

65
 66

1179 viewpoint of climatic geomorphology. Zeitschrift für Geomorphologie, Formatted: German (Germany)

1180 Supplement band 46, 1–14.

1181 Osterkamp, W.R., Hedman, E.R., 1982. Perennial-streamflow characteristics related to

1182 channel geometry and sediment in Missouri River Basin. U. S. Geological Survey

1183 Professional Paper 1242 (37p).

1184 Paredes, J.M., Foix, N., Piñol, F.C., Nillni, A., Allard J.O., Marquillas, R.A. 2007.

1185 Volcanic and climatic controls on fluvial style in a high-energy system: The Lower

1186 Cretaceous Matasiete Formation, Golfo San Jorge Basin, Argentina. Sedimentary

1187 Geology 202, 96–123.

1188 Paredes, J.M., Foix, N., Allard J.O., Valle, M.N., Giordano, S.R. 2018. Complex alluvial

1189 architecture, paleohydraulics and controls of a multichannel fluvial system: Bajo

1190 Barreal Formation (Upper Cretaceous) in the Cerro Ballena anticline, Golfo San

1191 Jorge Basin, Patagonia. Journal of South American Earth Sciences 85, 168–190.

1192 Parker, J.T.C., Steinkampf, W.C., Flynn, M.E., 2005, Hydrogeology of the Mogollon

1193 Highlands, central Arizona. U.S. Geological Survey Scientific Investigations

1194 Report 2004–5294 (87 p).

1195 Parrish, J.T., 1993. Climate of the supercontinent Pangea. The Journal of Geology 101,

1196 215–233.

66
 67

1197 Peel, M.C., Finlayson, B.L., McMahon, T.A., 2007. Updated world map of the

1198 Köppen-Geiger climate classification. Hydrology and Earth System Sciences 11,

1199 1633–1644.

1200 Posamentier, H.W., Jervey, M.T., Vail, P.R., 1988. Eustatic controls on clastic deposition

1201 I–conceptual framework. In Wilgus C.K., Hastings, B.S., Kendall, C.G.St.C.,

1202 Posamentier, H.W., Ross, C.A., Van Wagoner, J.C. (eds.), Sea-level research: an

1203 integrated approach. SEPM Special Publication 42, pp. 198–124.

1204 Prochnow, S.J., Nordt, L.C., Atchley, S.C., Hudec, M.R., 2006. Multi-proxy paleosol

1205 evidence for Middle and Late Triassic Climate Trends in eastern Utah. Palaeogeogr.

1206 Palaeocl. Palaeoecol. 232, 53–72.

1207 Ramezani, J., Hoke, G.D., Fastovsky, D.E., Bowring, S.A., Therrien, F., Dworkin, S.I.,

1208 Atchley, S.C., Nordt, L.C., 2011. High-precision U-Pb zircon geochronology of the

1209 Late Triassic Chinle Formation, Petrified Forest National Park (Arizona, USA):

1210 Temporal constraints on the early evolution of dinosaurs. Geological Society of

1211 America Bulletin 123, 2142–2159.

1212 Rasmussen, C., Mundil, R., Irmis, R.B., Geisler, D., Gehrels, G.E., Olsen, P.E., Kent,

1213 D.V., Lepre, C., Kinney, S.T., Geissman, J.W., Parker, W.G., 2020. U-Pb zircon

1214 geochronology and depositional age models for the Upper Triassic Chinle

67
 68

1215 Formation (Petrified Forest National Park, Arizona, USA): Implications for Late

1216 Triassic paleoecological and paleoenvironmental change. GSA Bulletin, doi: Formatted: Swedish (Sweden)

1217 https://doi.org/10.1130/B35485.1.

1218 Ray, P. K., 1976. Structure and sedimentological history of the overbank deposits of a

1219 Mississippi River point bar. Journal of Sedimentary Petrology 46, 788–801.

1220 Ray, S., Chakraborty, T., 2002. Lower Gondwana fluvial succession of the Pench-Kanhan

1221 valley, India: stratigraphic architecture and depositional controls. Sedimentary

1222 Geology 151, 243–271.

1223 Riggs, N.R., Ash, S.R., Barth, A.P., Gehrels, G.E., Wooden, J.L., 2003. Isotopic age of the

1224 Black Forest bed, Petrified Forest Member, Chinle Formation, Arizona: An

1225 example of dating a continental sandstone. Geological Society of America Bulletin

1226 115, 1315–1323.

1227 Riggs, N.R., Oberling, Z.A., Howell, E.R., Parker, W.G., Barth, A.P., Cecil, M.R., Martz,

1228 J.W., 2016. Sources of volcanic detritus in the basal Chinle Formation,

1229 southwestern Laurentia, and implications for the Early Mesozoic magmatic arc.

1230 Geosphere 12, 439–463.

1231 Riggs, N.R., Reynolds, S.J., Lindner, P.J., Howell, E.R., Barth, A.P., Parker, W.G., Walker,

1232 J.D., 2013. The Early Mesozoic Cordilleran arc and Late Triassic paleotopography:

68
 69

1233 The detrital record in Upper Triassic sedimentary successions on and off the

1234 Colorado Plateau. Geosphere 9, 602–613.

1235 Sakaguchi, Y., Takahashi, Y., Ohmori, H., 1986. Nature in Japan Part 3: Rivers of Japan.

1236 Iwanami Shoten, Tokyo (248p). (in Japanese)

1237 Schultz, L.G., 1963. Clay minerals in Triassic rocks of the Colorado Plateau. U.S.

1238 Geological Survey Bulletin 1147-C. United States Government Printing Office

1239 (71p).

1240 Schumm, S.A., 2005. River variability and complexity. Cambridge University Press

1241 (220p).

1242 Shanley, K.W., McCabe, P.J., 1994. Perspectives on the sequence stratigraphy of

1243 continental strata. The American Association of Petroleum Geologists Bulletin 78,

1244 544–568.

1245 Shibata, K., Adhiperdana, B.G., Ito, M. 2018. Quantitative reconstruction of

1246 cross-sectional dimensions and hydrological parameters of gravelly fluvial

1247 channels developed in a forearc basin setting under a temperate climatic condition,

1248 central Japan. Sedimentary Geology 363, 69–82.

1249 Singer, A., 1984. The paleoclimatic interpretation of clay minerals in sediments–a review.

1250 Earth-Science Reviews 21, 251–293.

69
 70

1251 Smith, D.G., 1986. Anastomosing river deposits, sedimentation rates and basin

1252 subsidence, Magdalena River, northwestern Colombia, South America.

1253 Sedimentary Geology 46, 177–196.

1254 Stewart, J.H., Poole, F.G., Wilson, R.F., 1972. Stratigraphy and origin of the Chinle

1255 Formation and related Upper Triassic strata in the Colorado Plateau region. U.S.

1256 Geological Survey Professional Paper 690 (336p).

1257 Strong N., Sheets, B., Hickson T., and Paola C. 2005. A mass-balance framework for

1258 quantifying downstream changes in fluvial architecture. In Blum M. D., Marriott S.

1259 B., Leclair S. F. eds. Fluvial sedimentology VII, vol. 35. International Association

1260 of Sedimentologists, Special Publication: 243-253.

1261 Suresh, N., Ghosh, S.K., Kumar, R., Sangode, S.J., 2004. Clay-mineral distribution

1262 pattern in late Neogene fluvial sediments of the Subathu sub-basin, central sector of

1263 Himalayan foreland basin: implications for provenance and climate. Sedimentary

1264 Geology 163, 265–278.

1265 Suttner, L.J., Dutta, P.K., 1986. Alluvial sandstone composition and paleoclimate, I.

1266 Framework mineralogy. Journal of Sedimentary Petrology 56, 329–345.

1267 Tanner, L.H., 2000. Palustrine-lacustrine alluvial facies of the (Norian) Owl Rock

1268 Formation (Chinle Group), four corners region, southwestern U.S.A.: implications

70
 71

1269 for Late Triassic paleoclimate. Journal of Sedimentary Research 70, 1280–1289.

1270 Thiry, M., 2000. Palaeoclimatic interpretation of clay minerals in marine deposits: an

1271 outlook from the continental origin. Earth-Science Reviews 49, 201–221.

1272 Thomas, R.G., Smith, D.G., Wood, J.M., Visser, J., Calverley-Range, E.A., Koster, E.H.,

1273 1987. Inclined heterolithic stratification--Terminology, description, interpretation

1274 and significance. Sedimentary Geology 53, 123–179.

1275 Trendell, A.M., Atchley, S.C., Nordt, L. C. 2012. Depositional and diagenetic controls on Formatted: Norwegian (Bokmål)

1276 reservoir attributes within a fluvial outcrop analog: Upper Triassic Sonsela member

1277 of the Chinle Formation, Petrified Forest National Park, Arizona. The American

1278 Association of Petroleum Geologists Bulletin 96, 679–707.

1279 Trendell, A.M., Atchley, S.C., Nordt, L.C., 2013. Facies analysis of a probable

1280 large-fluvial-fan depositional system: the Upper Triassic Chinle Formation at

1281 Petrified Forest National Park, Arizona, U.S.A. Journal of Sedimentary Research

1282 83, 873–895.

1283 Varban, B.L. and Plint, A.G., 2008. Sequence stacking patterns in the Western Canada

1284 foredeep: influence of tectonics, sediment loading and eustasy on deposition of the

1285 Upper Cretaceous Kaskapau and Cardium formations. Sedimentology 55: 395–421.

1286 Van Wagoner, J.C., Posamentier, H.W., Mitchum, R.M., Vail, P.R., Sarg, J.F., Loutit, T.S.,

71
 72

1287 Hardenbol, J., 1988. An overview of the fundamentals of sequence stratigraphy and

1288 key definitions. In Wilgus C.K., Hastings, B.S., Kendall, C.G.St.C., Posamentier,

1289 H.W., Ross, C.A., Van Wagoner, J.C. (eds.), Sea-level research: an integrated

1290 approach. SEPM Special Publication 42, pp. 39–45.

1291 Wan, S., Li, A., Clift, P.D., Stuut, J.-B.W., 2007. Development of the East Asian

1292 monsoon: mineralogical and sedimentologic records in the northern South China

1293 Sea since 20 Ma. Palaeogeogr. Palaeoclimatol. Palaeoecol. 254, 561–582.

1294 Williams, G.P., 1984. Paleohydrological methods and some examples from Swedish

1295 fluvial environments II–River meanders. Geografiska Annaler 66A, 89–102.

1296 Winker, W., 1984. Paleocurrents and petrography of the Gurnigel-Schliren Flysh: A basin

1297 analysis. Sedimentary Geology 40, 169–189.

1298 Wright, V.P., Marriott, S.B., 1993. The sequence stratigraphy of fluvial depositional

1299 systems: the role of floodplain sediment storage. Sedimentary Geology 86,

1300 203–210.

1301

1302

1303 Figure and Table Captions

1304 Fig. 1. Late Triassic paleogeography of the southwestern United States and distribution of

72
 73

1305 the Upper Triassic terrestrial strata (modified from Martz et al., 2017). Inset map in

1306 the upper right indicates the distribution of the Pangaea during Triassic (modified

1307 from Trendell et al., 2012).

1308 Fig. 2. Geologic map of the study area (simplified from Hintze, 1997). Circled numbers

1309 indicate the investigated routes; 1: Anasazi Trail, 2: Long Valley Recreation Area, 3:

1310 East Reef, 4: Hurricane Mesa, 5: Rockville, 6: Smithsonian Butte, 7: Cedar City.

1311 Fig. 3. A: Stratigraphy of the Upper Triassic in northeastern Arizona and southern Utah

1312 with an interpretation of climatic change. Data from Nordt et al. (2015) and Martz et

1313 al. (2017). B: Stratigraphy of the Middle Triassic–Lower Jurassic strata in

1314 southwestern Utah indicating the studied interval. Modified from Lucas and Tanner

1315 (2007) and Martz et al. (2017).

1316 Fig. 4. Lithofacies types of the Chinle Formation. A: Clast-supported conglomerates (Gc)

1317 and trough cross-stratified sandstones (St) in the Shinarump Member at Route 5

1318 (37°09'53.3"N, 113°01'33.5"W). B: Surface view of trough cross-stratified

1319 sandstones (St) in the Shinarump Member at Route 5 (37°10'25.4"N,

1320 113°02'10.7"W). C: Horizontally stratified sandstones (Sh), sandstones with

1321 convoluted laminations (Sc), and massive mudstones (Fm) in the Shinarump

1322 Member at Route 4 (37°14'20.8"N, 113°12'30.1"W). D: Current ripple

73
 74

1323 cross-laminated (Fr), and horizontally laminated (Fl), very fine-grained sandstones

1324 in the Shinarump Member at Route 4 (37°14'18.3"N, 113°12'30.1"W). E:

1325 Varicolored massive mudstones (Pm) in the Cameron Member at Route 6

1326 (37°07'34.0"N, 113°06'28.4"W). F: Massive limestone (Lm) in the “purple

1327 pedogenic beds” at Route 7 (37°40'27.3"N, 113°02'31.3"W). See Table 1 for

1328 descriptions and interpretations of the lithofacies types. See Fig. 2 for the

1329 investigated routes.

1330 Fig. 5. Schematic illustration of architectural element types. Element SG: Scoured

1331 surface with gravel lags. Element DA: Downstream accreting, cross-stratified

1332 sandstones. Element LA: Laterally accreting, cross-stratified sandstones. Element

1333 SD: Sandy bedforms. Element SCF: Small-scale channel-fill deposits. Element LCF:

1334 Large-scale channel-fill deposits. Element UB: Upper bar deposits. Element FF:

1335 Floodplain muddy deposits. Element FS: Floodplain sandy deposits. Circled

1336 numbers indicate hierarchy of bounding-surface (BS) types. See Table 2 for

1337 descriptions and interpretations of architectural element types.

1338 Fig. 6. A: Scoured surface with clast-supported gravel lags (SG) and laterally accreting,

1339 cross-stratified sandstones (LA) in the Shinarump Member at Route 5 (37°10'18.0"N,

1340 113°02'00.6"W). B: Downstream accreting, cross-stratified sandstones (DA) in the

74
 75

1341 Shinarump Member at Route 6 (37°08'06.8"N, 113°06'12.2"W). C: DA and sandy

1342 bedforms (SD) in the Cameron Member at Route 3 (37°12'07.6"N, 113°21'14.8"W).

1343 N in rose diagrams indicates the number of measurements. See Fig. 2 for the

1344 investigated routes.

1345 Fig. 7. A: Small-scale channel-fill deposits (SCF) in the Shinarump Member at Route 4

1346 (37°14'19.9"N, 113°12'29.8"W). B: Large-scale channel-fill deposits (LCF) in the

1347 Shinarump Member at Route 4 (37°14'20.8"N, 113°12'30.1"W). C: Upper bar

1348 deposits (UB) in the Shinarump Member at Route 4 (37°14'17.0"N 113°12'30.3"W).

1349 D: Floodplain muddy deposits (FF) and floodplain sandy deposits (FS) in the

1350 Cameron Member at Route 4 (37°14'33.6"N, 113°11'44.2"W). N in rose diagrams

1351 indicates the number of measurements. See Fig. 2 for the investigated routes.

1352 Fig. 8. A: Panorama photograph of the Shinarump Member at Route 4 (37°14'17.1"N,

1353 113°12'18.7"W). N in rose diagram indicates the number of measurements. B: Basal

1354 unconformity of the Chinle Formation on which a channel complex of the Shinarump

1355 Member overlies unsoiled sandstones and mudstones of the Moenkopi Formation.

1356 Route 4 (37°14'11.9"N, 113°12'31.0"W). C: Basal unconformity of the Chinle

1357 Formation on the pedogenically modified mudstones of the uppermost part of the

1358 Moenkopi Formation. Route 5 (37°09'48.9"N, 113°01'29.3"W). See Fig. 2 for the

75
 76

1359 investigated routes.

1360 Fig. 9. Panorama photograph of the Cameron Member and “purple pedogenic beds” at

1361 Route 3 (37°12'11.5"N, 113°21'13.4"W). FF indicates floodplain muddy deposits. N

1362 in rose diagram indicates the number of measurements. See Fig. 2 for the

1363 investigated route.

1364 Fig. 10. Stratigraphic cross-section of the Chinle Formation. Circled numbers indicate the

1365 investigated routes shown in Fig. 2. Paleocurrent directions are average of 5–10

1366 measurements. CHC I–VIII indicates channel complexes in which paleohydrological

1367 analyses were conducted. See Fig.11 for the results. A–F indicates sampling horizons

1368 for clast composition (Fig. 13). Friable sandstone, gray beds, and “middle

1369 sandstones” in Route 5 are after Martz et al. (2017).

1370 Fig. 11. Results of geomorphological and hydrological analyses of the Chinle Formation

1371 paleochannels. CHC: Channel complex, N: the number of measurements for

1372 cross-set thickness.

1373 Fig. 12. Cross-polarized microscopic photographs of thin sections of conglomerate clasts

1374 from the Chinle Formation. A: Chert from the Cameron Member at Route 3. B:

1375 Quartzite from the Shinarump Member at Route 4. C: Polycrystalline quartz from the

1376 Shinarump Member at Route 4. D: Mudstone from the Cameron Member at Route 3.

76
 77

1377 E: Acidic (rhyolitic) volcanic rock fragment showing felsitic texture including

1378 embayed monocrystalline quartz (Qm) from the Cameron Member at Route 3 F:

1379 Acidic vitric tuff with clots of quartz and/or feldspar (devitrification feature) from the

1380 Shinarump Member at Route 4. See Fig. 2 for the investigated routes.

1381 Fig. 13. Clast composition of conglomerates in the Chinle Formation. See Fig. 10 for the

1382 sampling locations and horizons.

1383 Fig. 14. Cross-polarized microscopic photographs of thin-sections of sandstone samples

1384 from the Chinle Formation. A: Monocrystalline quartz (Qm), K-feldspar (K), and

1385 sedimentary lithic fragment (Ls), the Cameron Member at Route 3, ER-Cs13. B: Qm

1386 with undulose extinction, the Shinarump Member at Route 6, SB-Ss16. C: Embayed

1387 Qm with rounded outline, the Cameron Member at Route 6, SB-Cs01. D: Chert (Ch),

1388 the Shinarump Member at Route 6, SB-Ss16. E: Partially dissolved K surrounded

1389 and replaced by authigenic clay minerals, the Shinarump Member at Route 6,

1390 SB-Ss12. F: Plagioclase feldspar (P), the Cameron Member at Route 3, ER-Cs05. G:

1391 Intersertal (felted) volcanic lithic fragment (Lv) with dull appearance, the Cameron

1392 Member at Route 3, ER-Cs05. H: Lv with felsitic texture, the Cameron Member at

1393 Route 3, ER-Cs13. I: Quartz-mica tectonite (QMT), the Shinarump Member at Route

1394 6, SB-Ss06. See Fig. 2 for the investigated routes.

77
 78

1395 Fig. 15. Triangular QFL plot of the Chinle Formation sandstones. Q is total quartzose

1396 grains including monocrystalline quartz (Qm), polycrystalline quartz (Qp)

1397 (sub-grains < 0.063 mm), tectonic quartz (QT) (sub-grains < 0.063 mm), and

1398 sedimentary chert (Ch). F is total feldspar grains including plagioclase (P) and

1399 K-feldspar (K). L is total unstable lithic fragments including volcanic lithic fragment

1400 (Lv), sedimentary lithic fragment (Ls), and quartz-mica tectonite (QMT).

1401 Fig. 16. Representative X-ray diffraction patterns (ethylene-glycol solvated) of sandstone

1402 and mudstone samples from the Chinle Formation. A: Mudstone sample from the

1403 "purple pedogenic beds" at Route 3, ER-Cm29. B: Mudstone sample from the

1404 Cameron Member at Route 3, ER-Cm04. C: Sandstone sample from the Cameron

1405 Member at Route 3, ER-Cs06. D: Mudstone sample from the Shinarump Member at

1406 Route 4, HM-Sm04. E: Sandstone sample from the Shinarump Member at Route 6,

1407 SB-Ss10. Ka: kaolinite, Chl: chlorite, I: illite, Sm: smectite. See Fig. 2 for the

1408 investigated routes.

1409 Fig. 17. Stratigraphic variations in clay mineral composition and sandstone framework

1410 composition of the Chinle Formation at Route 4 in Fig. 2. Relative abundance of four

1411 major clay minerals was calculated following the method of Biscaye (1965). See Fig.

1412 10 for legend of section and Fig. 15 for the Q, F, and L categories of framework grain

78
 79

1413 types of sandstones.

1414 Fig. 18. Stratigraphic variations in clay mineral composition and sandstone framework

1415 composition of the Chinle Formation at Route 6 in Fig. 2. See Fig. 17 for legend of

1416 clay mineral composition and of framework grain types of sandstones, Fig. 10 for

1417 legend of section, and Fig. 15 for the Q, F, and L categories of framework grain types

1418 of sandstones.

1419 Fig. 19. Stratigraphic variations in clay mineral composition and sandstone framework

1420 composition of the Chinle Formation at Route 3 in Fig. 2. See Fig. 17 for legend of

1421 clay mineral composition and of framework grain types of sandstones, Fig. 10 for

1422 legend of section, and Fig. 15 for the Q, F, and L categories of framework grain types

1423 of sandstones.

1424 Fig. 20. Scanning electron microscopy (SEM) images of sandstone and mudstone

1425 samples from the Chinle Formation. A: Mudstone from the Shinarump Member at

1426 Route 4, HM-Sm09. B: Kaolinite-rich sandstone from the Cameron Member at

1427 Route 3, ER-Cs06. C: Kaolinite-rich mudstone from the Cameron Member at Route

1428 3, ER-Cm04. D: Smectite-rich mudstone from the "purple pedogenic beds" at Route

1429 3, ER-Cm29. See Fig. 2 for the investigated routes.

1430 Fig. 21. Schematic illustrations of fluvial systems in the Shinarump Member, and the

79
 80

1431 Cameron Member and “purple pedogenic beds” of the Chinle Formation showing

1432 paleohydrological parameters.

1433 Table 1. Lithofacies classification of the Chinle Formation.

1434 Table 2. Architectural element types of the Chinle Formation.

1435 Table 3. Hierarchy of bounding-surface (BS) types used in this study.

1436 Table 4. Hydrological features of some modern fluvial channels. Wb: Bankfull channel

1437 width (m). Qb: Bankfull discharge (m3s-1).

1438

1439 Appendix I. Framework composition of sandstones from the Chinle Formation. See Fig.

1440 15 for the framework grain types of sandstones.

1441 Appendix II. Clay mineral composition of sandstone and mudstone samples from the

1442 Chinle Formation. Ka: kaolinite, Chl: chlorite, I: illite, Sm: smectite, Sds: sandstone

1443 sample, Mds: mudstone sample.

80