Evaluation of paleohydrologic models for terrestrial inverted channels: Implications for application to martian sinuous ridges
Introduction
Relief inversion is an aspect of landscape evolution on both terrestrial (Pain and Ollier, 1995) and martian (e.g., Williams and Edgett, 2005, Williams et al., 2005, Pain et al., 2007) denuded surfaces, where materials deposited in topographic lows were or became more resistant to erosion than the surrounding terrain. Exhumation of martian terrains was identified in some of the earliest orbital images (Sharp, 1973, Soderblom et al., 1973) and is now recognized as a widespread process on Mars (Malin and Edgett, 2001). Inverted topography of fluvial landforms results when differential erosion preferentially strips away the less resistant valley or channel walls, leaving the original floor as a local topographic high. (We differentiate channels, the conduit through which water flows, from the larger-scale valley.) Several processes can lead to relief inversion, including cementation of the channel floor, armoring of the channel floor by coarse grains, and infilling by a more resistant material (commonly a lava flow).
Cemented, positive-relief channels are good environments for preservation of fluvial sediments, with exposures in three dimensions and retention of many attributes of the original channel form. The preservation of these sedimentary attributes, including grain size distribution and sedimentary structures, provides a record of the fluvial conditions during active flow and is useful for evaluating paleohydrological models. In contrast, lava-capped inverted valleys are the result of lava infilling of the valley, a process that obscures sedimentary structures in plan view. A lava flow may occupy both a fluvial channel and its surrounding valley floor such that morphometric observations of the inverted topography do not necessarily reflect the channel dimensions and formative discharge.
The motivation for this investigation stems from the recent recognition that fluvial landforms on Mars have a variety of preservation states (Williams and Edgett, 2005). In addition to the negative-relief valley networks first identified in Mariner 9 images (McCauley et al., 1972), some valleys are buried (e.g., Mangold et al., 2004) and others are preserved as ridge forms and positive-relief branching networks. In this report, the nongenetic term “sinuous ridges” will be used to refer to these landforms on Mars, although other terms have been adopted in the literature including “raised curvilinear features” (RCFs) (Burr et al., 2006, Williams, 2007). Over 200 sinuous ridge sites have been identified in meter- and decameter-scale images around the martian globe, covering areas ranging from tens to hundreds of square kilometers (Pain et al., 2007, Williams, 2007). In contrast to valley networks, which are primarily found on the ancient cratered highlands, sinuous ridges are found on terrain with ages that span the entire geological history of the planet, from the Noachian to the Amazonian Periods, expanding the record of past water on Mars (Williams, 2007).
All published studies to date have interpreted martian sinuous ridges as fluvial in origin, but authors have differed on the water source. Based on their curvilinear and bifurcating appearance, continuity relationships with negative-relief valley networks, and similarity to terrestrial fluvial landforms, the sinuous ridges are interpreted to be the remnants of former fluvial channels formed by continually flowing water, now expressed in inverted relief (Moore and Howard, 2005, Williams and Edgett, 2005, Pain et al., 2007). Sedimentary structures that support this interpretation are evident in some of the martian sinuous ridges. For example, scroll bars are present in the putative deltaic deposit defined by cross-cutting sinuous ridges within Eberswalde crater (Malin and Edgett, 2003, Moore et al., 2003, Bhattacharya et al., 2005, Wood, 2006). In addition to precipitation-fed surface runoff (e.g., Moore et al., 2003, Moore and Howard, 2005), other scenarios for generating fluids have been proposed in the formation of these landforms, including impact-generated melt from ground-ice (Jerolmack et al., 2004), glacial meltwater or subglacial streams (eskers) (Howard, 1981, Kargel and Strom, 1992, Nussbaumer et al., 2003, McMenamin and McGill, 2005, Burr et al., 2006). The majority of the sinuous ridges have morphology consistent with open-channel flow, as examined in this study. Very few sinuous ridges, confined to specific regions on Mars (e.g., Argyre Planitia and Aeolis/Zephyria Plana; Kargel and Strom (1992) and Burr et al. (2008), respectively), are interpreted to be eskers primarily because they cross pre-existing topographic divides. Under this interpretation, these sinuous ridges would have formed in a closed and potentially pressured environment that is inconsistent with the terrestrial analog site investigated in this paper.
The diverse suite of sinuous ridges preserved on Mars documents a range of paleofluvial environments and attests to the complex fluvial history of the planet. The channel systems as observed today are the products of combined burial, exhumation, and degradation. Several studies (e.g., Moore et al., 2003, Fassett and Head, 2005, Irwin et al., 2005) have applied hydraulic models derived from the study of active fluvial systems on Earth to one or more of these inferred inverted martian landforms, even though few terrestrial studies have documented the applicability of such models to inverted channels. In this study, we evaluate the applicability of both empirically and theoretically derived paleohydrologic models for modern streams to cemented inverted channels based on analysis of field data for inverted paleochannels in east central Utah. Future phases of this research will investigate other agents of inverted relief in fluvial systems, including lava infilling. Results of this study can be used to assess the magnitude of fluvial activity in inferred inverted channels on Mars.
Section snippets
Study region
Multiple exhumed paleochannels are found in parts of the Colorado Plateau. Following uplift of the region in middle to late Cenozoic time, erosion, dominantly by the Colorado River and its tributaries, stripped away younger strata, revealed Late Jurassic and Early Cretaceous paleochannel sediments at several sites in east central Utah (e.g., Williams et al., 2007). The present-day arid climate has inhibited the development of thick soil horizons and pervasive vegetative cover, leaving these
Measurements
The primary objective of field work was to obtain input parameters for the paleohydrologic models (described in Section 3.2). Secondarily, the suite of field measurements was compared to data derived from aerial photographs, as an appraisal of measurements obtained from remotely sensed data. The following sections describe the methods employed in this study.
The topographic survey obtained direct measurements of the present-day width and slope of inverted channels termed “B” and “D” by Harris
Input parameters
Accurate determination of the paleohydraulic model input parameters, such as flow cross-sectional area, former energy gradients, and maximum particle size, is complicated by multiple geologic processes that have operated on the fluvial deposits since the channels were active. Here we discuss how input values were determined based on field observations. Maizels, 1983, Maizels, 1987 discusses in additional detail the sources of error and problems inherent in reconstruction of paleochannel
Discussion
The approach adopted in this study yields an envelope of discharge values, an indication of the plausible range as verified through field observations. The overall mean discharge value tended toward the maximum value in the micro approach and an intermediate value for the macro approach. When evaluating the paleohydraulics of the system, considering both approaches is desirable and important to mitigate uncertainties in each variable and take full advantage of the available data. None of the
Conclusions
Cemented paleochannel deposits, now exposed as ridge forms because of landscape inversion, preserve key attributes of the former fluvial environment. A record of the fluvial conditions during active flow is preserved, such as original channel shape (preserved in some sections), flow direction, and sediment transport capability. Paleochannel dimensions, slope, and grain size distribution can be used to estimate paleoflow conditions using paleohydraulic models. The 14 paleohydraulic models
Acknowledgements
The authors are grateful to A. Johnston (Smithsonian Institution) for technical advice, T. Chidsey (Utah Geologial Survey) and Brian S. Curie (Miami University-Ohio) for information regarding the study site, and Doug Cox for development of USAPhotoMaps. This manuscript was improved by comments of V. Baker, G. Komatsu, A. Howard and two anonymous reviewers. This research was supported by a NASA Mars Fundamental Research Grant #NNX06AB21G and a research grant from the Becker Endowment from the
References (113)
Evolution of calcrete in paleodrainages of the Lake Napperby Area, central Australia
Palaeogeography, Palaeoclimatology, Palaeoecology
(1986)- et al.
Regional paleoclimatic and stratigraphic implications of paleosols and fluvial/overbank architecture in the Morrison Formation (Upper Jurassic), Western Interior, USA
Journal of Sedimentary Geology
(2004) Occurrence of phetic dolocrete within Tertiary clastic deposits of Kuwait, Arabian Gulf
Sedimentary Geology
(1990)Comparisons of the hydraulics of water flows in martian outflow channels with flows of similar scale on Earth
Icarus
(1979)Preliminary Mariner 9 report on the geology of Mars
Icarus
(1972)- et al.
Inversion of relief—a component of landscape evolution
Geomorphology
(1995) - et al.
Inversion of relief on Mars
Icarus
(2007) A classification of natural rivers
Catena
(1994)- et al.
Topography of valley networks on Mars from Mars Express High Resolution Stereo Camera digital elevation models
Journal of Geophysical Research
(2008) - et al.
Guide for Selecting Manning's Roughness Coefficients for Natural Channels and Flood Plains—Metric Version