The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 1 The Decorah structure, northeastern Iowa: Geology and evidence for formation by meteorite impact Bevan M. French1,†, Robert M. McKay2,§, Huaibao P. Liu2,†, Derek E.G. Briggs3,†, and Brian J. Witzke4,† 1Department of Paleobiology, Smithsonian Institution, Washington, D.C. 20013-7012, USA 2Iowa Geological Survey, IIHR—Hydroscience & Engineering, University of Iowa, 340 Trowbridge Hall, Iowa City, Iowa, 52242, USA 3 Department of Geology and Geophysics, and Yale Peabody Museum of Natural History, Yale University, New Haven, Connecticut 06520, USA 4Department of Earth and Environmental Sciences, University of Iowa, 121 Trowbridge Hall, Iowa City, Iowa 52242, USA ABSTRACT The Decorah structure, recently discov- ered in northeastern Iowa, now appears as an almost entirely subsurface, deeply eroded circular basin 5.6 km in diameter and ~200 m deep, that truncates a near-horizontal series of Upper Cambrian to Lower Ordovician platform sediments. Initial analysis of geo- logical and well-drilling data indicated char- acteristics suggestive of meteorite impact: a circular outline, a shallow basin shape, discordance with the surrounding geology, and a filling of anomalous sediments: (1) the organic-rich Winneshiek Shale, which hosts a distinctive fossil Lagerstätte, (2) an under- lying breccia composed of fragments from the surrounding lithologies, and (3) a poorly known series of sediments that includes shale and possible breccia. Quartz grains in drill samples of the breccia unit contain abundant distinctive shock-deformation features in ~1% of the individual quartz grains, chiefly planar fractures (cleavage) and planar defor- mation features (PDFs). These features pro- vide convincing evidence that the Decorah structure originated by meteorite impact, and current models of meteorite crater for- mation indicate that it formed as a complex impact crater originally ~6  km in diameter. The subsurface characteristics of the lower portion of the structure are not well known; in particular, there is no evidence for the ex- istence of a central uplift, a feature generally observed in impact structures of comparable size. The current estimated age of the Deco- rah structure (460–483 Ma) suggests that it may be associated with a group of Middle Ordovician impact craters (a terrestrial “im- pact spike”) triggered by collisions in the as- teroid belt at ca. 470 Ma. INTRODUCTION During the past few decades, impacts of large extraterrestrial objects onto the Earth’s surface have become recognized and generally accepted as an important geological process (Grieve, 1991, 1997, 1998, 2001; French, 1998, 2004; Lowman, 2002; Jourdan and Reimold, 2012; Osinski and Pierazzo, 2013). It has also been recognized that these rare but highly energetic events can produce major and often widespread geological effects, including the near-instanta- neous formation of large geological structures, the generation of large igneous bodies, the creation of economic mineral and petroleum deposits, the deposition of regional and even global ejecta layers, and (in at least one case) the production of a major biological extinction (at the Cretaceous-Paleogene boundary, 66 Ma; Schulte et al., 2010). This recognition of terrestrial impact craters and their effects has been based mainly on dis- tinctive and permanent petrographic and geo- chemical effects produced in target rocks and minerals by the extreme and highly transient conditions of pressure, temperature, stress, and strain generated by the intense shock waves uniquely created by hypervelocity impact events (French and Short, 1968; French, 1998; French and Koeberl, 2010; Koeberl, 2014). In practice, one of the most widespread and widely used cri- teria for the identification of shock waves and meteorite impact structures has been the multi- ple sets of narrow, closely spaced lamellae (pla- nar deformation features [PDFs]) developed in quartz (see papers in French and Short [1968]; also Stöffler and Langenhorst, 1994; Grieve et al., 1996; Ferrière et al., 2009), although a small number of other criteria (e.g., megascopic shat- ter cones and unique geochemical signatures of the impacting projectile) can also provide un- ambiguous identifications of impact structures (Tagle and Hecht, 2006; French and Koeberl, 2010; Koeberl, 2014). Since the 1960s, the number of established terrestrial meteorite structures has increased steadily at a rate of a few new structures per year (Grieve, 1998), chiefly because the recognition of these shock-metamorphic features (generally shatter cones and PDFs in quartz) has made it possible to identify impact structures that are old, poorly exposed, deeply eroded, buried, or even subjected to post-impact metamorphism. At present, more than 190 preserved impact structures have been definitely identified (Earth Impact Database, 2016), and model calculations suggest that at least several hundred preserved impact structures remain to be discovered on the land areas of the earth (Trefil and Raup, 1990; Grieve, 1991; Stewart, 2011; Hergarten and Kenkmann, 2015). This growing population of recognized im- pact structures, and the increasing diversity and complexity displayed by individual structures, have made it possible for current research ef- forts to expand beyond the simple identification of new structures and to explore more general impact-related problems: the mechanics and complexities of large crater formation, the na- ture of impact-produced rock deformation, the establishment of new criteria for impact events, the effects of the target geologic setting on im- pact crater development, and the wider geologi- cal and environmental consequences of impact events, However, despite the large number of presently known impact structures and the so- phisticated state of current impact research, the field continues to evolve rapidly and unpredict- ably, and the identification of new impact struc- tures remains an essential source of new data and of new and often unexpected questions. In this paper, we describe the geology and origin of the Decorah structure, now present GSA Bulletin; Month/Month 2018; v. 130; no. X/X; p. 1–25; https://doi.org/10.1130/B31925.1; 18 figures; 2 tables.; published online XX Month 2018. †frenchb@si.edu, huaibao-liu@uiowa.edu, derek .briggs@yale.edu, brian-witzke@uiowa.edu. §Corresponding author: rjmckayic@gmail.com. For permission to copy, contact editing@geosociety.org © 2018 Geological Society of America French et al. 2 Geological Society of America Bulletin, v. 130, no. XX/XX as a small, isolated circular basin (diameter ~5.6 km) in virtually undeformed Upper Cam- brian and Lower Ordovician cratonic strata in northeastern Iowa (Liu et al., 2009; McKay et al., 2010, 2011). This structure, recognized during the investigation of a new Konservat- Lagerstätte, the Winneshiek Lagerstätte (Liu et  al., 2006, 2007a), is almost entirely sub- surface, appearing as an area of localized and anomalous sedimentary strata and structural deformation. The basin contains strata unlike those of the surrounding Cambrian through Ordovician rocks (McKay et al., 2010, 2011; Witzke et al., 2011; Wolter et al., 2011). At present, two distinct lithologies are well rec- ognized in the upper portion of the basin- filling sediments: (1) the Winneshiek Shale (Wolter et al., 2011), a greenish-gray to black shale, and (2) an underlying polymict brec- cia apparently derived from local lithologies (McKay et al., 2010, 2011). A third series of rock types, shale, and possible sandstone brec- cia, are poorly known from the deeper por- tions of the basin. The areal extent of the basin is presently defined by the known distribution of the Winneshiek Shale, the uppermost basin- fill unit. In particular, we report here the results of petrographic and petrofabric examinations of quartz grains from small samples of the sub- surface polymict breccia underlying the Win- neshiek Shale. We provide evidence that mi- croscopic deformation features in the quartz grains are shock-produced planar fractures (PFs) and PDFs, establishing that the Decorah structure was formed by an extraterrestrial im- pact. We use hereafter the name “Decorah im- pact structure,” after the city of Decorah, which overlies 40% of the structure. The name was originally introduced by Kass et al. (2013a, 2013b), following their initial analysis of air- borne geophysical survey data collected for the study of Precambrian basement geology in a coincident but much larger study area. Although generally similar to many known impact structures of comparable size, the Dec- orah structure reveals some unusual character- istics: (1) it is one of only a small number of impact structures developed entirely in a target of layered sediments, with apparently no in- volvement of the underlying crystalline base- ment; (2) it displays no evidence of a central uplift, a structural feature generally present in impact structures of comparable size; and (3) it has apparently undergone post-impact erosion of ~300–500  m of originally overlying pre- impact sediments, together with such original impact-produced features as an uplifted crater rim and an ejecta layer that surrounded the original structure. GEOLOGY OF THE DECORAH STRUCTURE Regional Geology, Target Rocks, and Age The Decorah structure is located in Win- neshiek County, northeastern Iowa (Fig. 1); its center is located at latitude 43°18′ 49″ and lon- gitude 91°46′19″. The structure lies within the Paleozoic Plateau landform region of Iowa, an area of relatively high topographic relief and thin Quaternary deposits lying upon Paleozoic bedrock (Prior, 1991). The upper surface of the structure, currently best defined by the mostly subsurface distribution of the Winneshiek Shale, spans a circular area of ~24.8 km2. About 10.1 km2 of the structure underlies the town of Decorah. The bedrock-entrenched Upper Iowa River meanders across the southern half of the structure where, on average, 20 m of Quaternary alluvium overlies bedrock. The depth to the top of structure varies from 0  m at the river level outcrop of the Winneshiek Shale near the struc- ture’s eastern edge to 111 m along the north- western margin on the upland above the river valley. Paleozoic bedrock in the region is a gently tilted sequence of Upper Cambrian through Upper Ordovician cratonic sedimentary strata (Table  1), chiefly quartzose and fine-grained feldspathic arenites, carbonate rocks, and lesser shale (Witzke and McKay, 1987; Witzke and Glenister, 1987; McKay, 1988, 1993; Lud- vigson and Bunker, 2005; Runkel et al., 1998, 2007, 2008; Wolter et al., 2011). The maximum thickness of Paleozoic strata is estimated from Winneshiek County Study Area Iowa 43º15´00´´N 91º37´30´´W Decorah Quad Freeport Quad Decorah incorporated boundary N km 0 3 6 91º52´30´´W 91º45´00´´W 43º22´30´´N Upper Iowa Ri ve r Decorah impact structure Figure  1. Location map showing the Decorah impact structure in Winneshiek County, northeastern Iowa. Inset maps show details of the 385 km2 study area, the Decorah impact structure, the incorporated limits of the city of Decorah, Iowa, the Upper Iowa River, and the U.S. Geological Survey 7.5 min Decorah and Freeport quadrangles. The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 3 drill hole information to be 620 m. Formations dip uniformly to the southwest at an average of 7.2 m/km (<0.5°), a subdued structural atti- tude typical of the region. Compared to much of Iowa, bedrock exposures, especially along valley walls, are common, but Quaternary col- luvium, loess, and patchy glacial till up to a few meters thick mantle large portions of the valley walls and uplands. The Paleozoic strata are underlain by a Pre- cambrian (Mesoproterozoic) mafic to ultramafic complex thought to intrude Yavapai province (1.8–1.72  Ga) metagabbro and felsic plutons (Drenth et al., 2015). This basement complex is estimated to be present at depths of 490–620 m below the land surface. Paleozoic formations in the region have pri- marily been dated using biostratigraphy, but ra- diometric age dates from three K-bentonites in the overlying Galena Group (Kolata et al., 1996) provide additional age constraints on the struc- ture. Of the three bentonites, the most accurately dated is the Millbrig K-bentonite, 1  m above the base of the Decorah Formation and ~37 m above the top of the Winneshiek Shale. 40Ar/39Ar single crystal laser fusion experiments on sani- dine phenocrysts from the Millbrig yielded ages of 454–449 Ma (Chetel et al., 2004, 2005; Smith et al., 2011), a Late Ordovician age straddling the Sandbian–Katian stage boundary (Cohen et al., 2013). This figure provides a minimum absolute age for the Decorah structure, but con- odonts from both the crater-filling Winneshiek Shale, and the overlying crater-capping St. Pe- ter Sandstone, are likely middle-late Darriwil- ian (Liu et al., 2017), indicating a minimum age of 460–465 Ma for the Decorah structure. A maximum age is provided by distinctive con- odont faunas from the Shakopee Formation, the youngest rock unit currently preserved that is disturbed by the structure. These faunas, part of the North American Midcontinent Province fauna (NAMP), contain elements of Ibexian fauna D (Smith and Clark, 1996), and are rep- resentative of the middle Tremadocian. Thus the Decorah structure was formed between the middle Tremadocian and the middle Darriwil- ian, 460–483  Ma (Cohen et al., 2013). (A re- cently published study [Bergström et al., 2018] has used δ13Corg chemostratigraphy to estimate the age of the Winneshiek Shale and the Deco- rah impact structure as 464–467  Ma, values that are consistent with the other estimates given here.) General Features of the Decorah Structure The approximate areal extent of the Deco- rah structure is defined chiefly by the occur- rence of the unusual Winneshiek Shale (Liu et al., 2006), which is the uppermost basin-filling unit. Occurrences of the shale, identified in one small surface outcrop, water-well drill-cut- tings, drillers’ logs, and drill cores (McKay et al., 2010, 2011) define a roughly circular area with a diameter of ~5.6 km (Fig. 2). More than 478 study area well records and 85 outcrops were examined in support of this delineation. Of those records, 35 encounter Winneshiek Shale and 23 sub-Winneshiek breccia. Subse- quent airborne transient electromagnetic data identified and mapped the Winneshiek Shale as a circular conductor aligned nearly per- fectly with the distribution mapped from the drill hole and outcrop data (Kass et al., 2013a, 2013b). The structure truncates the Cambrian Lone Rock, St. Lawrence, and Jordan forma- tions, and the Ordovician Oneota and Shako- pee formations (Fig. 3 and Table 1). The basin, and the rocks that fill it, are almost completely in the subsurface. About 94% (23.4 km2) of the structure is disconformably overlain by the St. Peter Sandstone, and ~6% (1.49 km2) is uncon- formably overlain by Quaternary alluvium of the Upper Iowa River. The uppermost basin fill (Winneshiek Shale) and deformed pre-impact target rocks are exposed at the present land surface in two separate small outcrops totaling less than 20 m2. The distribution of Winneshiek Shale (Fig. 2), and the singular occurrence of deformed sedi- mentary rocks, define a roughly circular struc- ture. Data from well cuttings and drill core samples indicate that this structure is a basin ~210 m in maximum depth, filled completely by a series of unique local lithologies not found in the surrounding sedimentary succession. Data on the substructure of the central portion of the basin are currently confined to a single set of well cuttings (Fig. 3: Well 53572). This infor- mation is not sufficient to determine the exact nature of the floor of the structure, the location of the boundary between breccia and underly- ing bedrock, or the degree of deformation and displacement of the pre-basin sedimentary bed- rock units. In particular, there is no indication as to whether or not the Decorah impact structure exhibits a central uplift of the subcrater rocks, a feature typical for impact structures of com- parable size (e.g., Grieve, 1991; Grieve and Pilkington, 1996). (See discussion below under “Apparent Absence of a Central Uplift”). TABLE 1. STRATIGRAPHY OF PALEOZOIC UNITS IN THE VICINITY OF THE DECORAH IMPACT STRUCTURE System Group/formation Thickness (m) Lithology Upper Ordovician Maquoketa Formation 35 Limestone and shale Galena Group Dubuque Formation 10 Limestone and minor shale Wise Lake Formation 21 Limestone, dolomitic, bioturbated Dunleith Formation 42 Limestone, dolomitic, cherty Decorah Formation 13–14 Shale and limestone, two K-bentonites Platteville Formation 8 Limestone Glenwood Shale 3 Shale and minor siltstone Middle Ordovician St. Peter Sandstone 17–31 Fine- to medium-grained quartzose sandstone Lower Ordovician Major unconformity Prairie du Chien Group Shakopee Formation* 19–42 Dolomite, sandy dolomite, and fine- to coarse- grained quartzose sandstone Unconformity Oneota Formation* 59–62 Dolomite, cherty Upper Cambrian Unconformity Jordan Sandstone* 28–31 Very-fine to coarse-grained feldspathic to quartzose sandstone St. Lawrence Formation* 37 Dolomite and dolomitic siltstone, glauconitic Lone Rock Formation* 43 Very fine- to fine-grained feldspathic sandstone, shale, minor dolomite, glauconitic Wonewoc Formation 40–44 Fine- to coarse-grained quartzose sandstone Eau Claire Formation 41–50 Very fine-to fine-grained feldspathic sandstone and shale, minor dolomite, glauconitic Mt. Simon Formation 140–180† Fine- to very coarse-grained quartzose sandstone, minor shale Note: Thickness values derived from measured sections and well data in Iowa Geological Survey files and publications cited in text. *Unit is absent from the area within the Decorah impact structure, and is replaced by the Winneshiek Shale and sub-Winneshiek breccia, sandstone and shale. †Estimated. French et al. 4 Geological Society of America Bulletin, v. 130, no. XX/XX km 0 3 6 43º15´00´´NN 91º37´30´´W 91º52´30´´W 91º45´00´´W 43º22´30´´N Decorah impact structure Upper Iowa River Upper Iowa River well encounters Prairie du Chien Group below the St. Peter Fm., or the first bedrock in well is Prairie du Chien Group well encounters Winneshiek Shale below the St. Peter Fm., or below Quaternary alluvium/colluvium exposure of Prairie du Chien Group exposure of Winneshiek Shale and site of H2 core shale exposure subcrop of Winneshiek Shale beneath Quaternary alluvium or colluvium Figure 2. Map showing the locations and types of geologic data (well boreholes, outcrops, subcrop, etc.) used to define the distribution of the Winneshiek Shale. This unit occurs only within a circle ~5.6 km in diameter, labeled “Decorah impact structure”; outside this circle the Winneshiek is absent and the normal stratigraphy is present. The distribution of the Winneshiek Shale is therefore taken as a proxy for the area occupied by the currently preserved basin of the Decorah impact structure. The structure includes locations where the Winneshiek Shale is present at depth but overlain by younger units, such as the St. Peter Sandstone, as well as areas where the Win- neshiek Shale is present at the bedrock surface but overlain by Quaternary alluvium, i.e., in the southeast quad- rant of the structure. The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 5 58 95 3 31 58 7 25 31 2 & 1 78 78 53 57 2 58 85 & 6 20 35 57 71 3 12 4 60 6540 41 1 31 57 9 31 07 6 32 85 6 & 6 41 55 59 37 0 28 07 7 60 81 9 58 57 8 31 07 5 53 72 3 0 0 1 1 2 2 3 3 4 4 5 5 6 6 987 miles km U ni on P ra iri e C em et ar y U pp er Io w a R iv er C an oe C re ek D ry R un QUATERNARY ORDOVICIAN CAMBRIAN PRECAMBRIAN undifferentiated glacial till, loess, colluvium & alluvium Om Maquoketa Formation Odw Osp Cj PCu ? Csl Clr Cw Cec Cms Od Os Odpg Oo Dubuque & Wise Lake formations St. Peter Sandstone Jordan Sandstone undifferentiated igneous & metamorphic uncertain St. Lawrence Formation Lone Rock Formation Wonewoc Formation Eau Claire Formation Mt. Simon Formation Dunleith Formation Shakopee Formation Decorah, Platteville & Glenwood formations Oneota Formation Ows Winneshiek Shale Oubcs unnamed breccia, conglomerate, sandstone & shale ? well penetration 31076 IGS well number fault downup Om Odw Odw Od Od Odpg Odpg Osp Osp Os Os Oo Oo Oo Cj Cj Cj Csl Csl Csl Clr Clr Cw Cw Cec Cec Ows Oubcs PCu PCu Cms Cms South North Vertical Exaggeration = 24x 1200 1100 1000 900 800 700 500 600 400 300 200 150 200 250 300 350 100 100 50 00 -100 -100 -150 -200 -250 -50 -200 -400 -600 -500 -700 -300 -300 -350 -800 -1000 -900 -1100 Elevation FTM 43º15´00´´N 91º37´30´´W N km0 3 6 Upper Io wa Riv er Cross section and well location map line of cross section Decorah Impact Structure well Figure 3. Geologic cross-section across the study area and through the western portion of the Decorah impact structure. (Orientation approximately north-south [see inset]; cross section hung on elevation; vertical exaggeration 24×.) The uniformly gentle southwest- erly dips of the regionally widespread Upper Cambrian and Lower Ordovician formations are disrupted by the impact basin (center) and its filling of breccia, conglomerate, and sandstone (Oubcs), which is in turn overlain by a cap of Winneshiek Shale (Ows). Much of the impact basin is disconformably overlain by the St. Peter Sandstone, whose thickness varies considerably because of its discon- formable relationship to the Winneshiek Shale and its unconformable relationship to the underlying Shakopee Formation outside the structure. Cuttings from Well 5885, located on the south side of the structure, are interpreted to indicate that the well encountered a down-dropped block of Oneota, Jordan, and St. Lawrence formations below the present crater floor. Cuttings sample logs for all numbered wells are available on the GeoSam section of the Iowa Geological Survey’s (IGS) website, https://www.iihr.uiowa.edu/igs/ geosam/home (accessed April 2018). Minor inconsistencies between this figure and a more detailed cross-section (Fig. 18) are due to: different locations of the two sections within the crater (i.e., this section trends approximately N-S through the western margin of the structure, while the section in Figure 18 trends approximately NW-SE through the center); the use of different drill holes for the different sections; and different presentations of the vertical dimension in the two sections (i.e., as elevation [this figure] and as depth [Fig. 18]). Considering the schematic character of both cross-sections, these inconsistencies are not considered significant. French et al. 6 Geological Society of America Bulletin, v. 130, no. XX/XX The Decorah structure is expressed by features of its internal stratigraphy (Figs. 3 and 4): (1) an absence of units within the normal stratigraphy of the area (Lone Rock through Shakopee forma- tions); (2) indications of deformation and down- dropping of the normal sedimentary section; (3) the circular distribution, at the top of the ba- sin, of the Winneshiek Shale, a distinctive green- ish-gray to black shale which contains a striking fossil Lagerstätte and is unknown outside the structure; and (4) the presence of an unusual and thick breccia unit, composed of large and small fragments of several units in the normal stratigra- phy, underlying the Winneshiek Shale. Surface Exposures and Access to Subsurface Crater-Fill Units Surface exposures of the units involved in the Decorah structure are limited to one small river- bank exposure of the Winneshiek Shale and one demonstrably deformed and brecciated outcrop of Shakopee Formation. Both exposures occur along the Upper Iowa River at low elevations near the up-dip eastern side of the structure. The Winneshiek Shale exposure is the locality where the Winneshiek fauna has been collected (Liu, et al., 2006, 2007a, 2007b, 2009, 2013, 2017; Lamsdell et al. 2015a, 2015b; Nowak et al., 2017, 2018; Briggs et al., 2016; Hawkins et al., 2018). The H2 core (Figs. 5 and 6) was drilled at this locality above the outcrop. A single exposure of deformed and brecci- ated Shakopee dolomite (Fig. 7) was discovered on the grounds of the Oneota Country Club after the basin boundary was mapped. This Shakopee outcrop is located ~50  m east of the bound- ary, which in this area is buried beneath Upper Iowa River alluvium. The outcrop apparently represents pre-impact target rock just outside the crater boundary, and the dolomite exhibits dips that are steep and irregular; values >10° and possibly up to 55° were noted, in contrast to the dips of <1° that are normal in the region. The outcrop also shows significant deformation (Fig. 7) compared to exposures of the same unit further outside the basin: general and pervasive closely spaced fracturing, multiple subparal- lel fracture sets, local in-place autobrecciation, possible tight overturned folding, and possible faulting and displacement. No exotic breccias (e.g., intrusive polymict breccia dikes) were ob- served. Shatter cones, which are features diag- nostic of meteorite impact (French and Koeberl, 2010, p. 129) and are often well developed in fine-grained carbonate target rocks, were also not found. Twenty-three of the 35 drill records that en- counter Winneshiek Shale are deep enough to indicate the presence of sub-Winneshiek brec- cia, but only 18 of those include satisfactory samples of both Winneshiek Shale and sub- Winneshiek breccia. Of those 18 holes, two were cored; the other 16 were drilled as water wells, and drill-cutting chips were routinely col- lected and saved as samples. Four of the water wells are interpreted to penetrate the entire ba- sin fill; three of these are located near the basin margin (Wells 124, 5885, and 25454) and one near the basin center (Well 53572); the remain- ing holes penetrate the sub-Winneshiek breccia from 16 to 108 m. All well numbers have been assigned by the Iowa Geological Survey, and all collected samples are permanently stored in the Rock Library. Glenwood Shale St. Peter Sandstone Decorah Fm. Dunleith Fm. Platteville Fm. Winneshiek Shale sandstone - breccia ? sandstone - breccia ? Wonewoc Fm. Jordan Ss. Oneota Fm. Shakopee Fm. St. Lawrence & Lone Rock fms. Eau Claire Fm. Mt. Simon Fm. shale D ec or ah im pa ct s tru ct ur e - cr at er b as in fi ll Houston County, MN Composite IGS W53572 Skyline Quarry Winneshiek County, IA increasing gamma, counts per second increasing gamma, counts per second depth 10 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 1550 1600 1650 1700 1800 1750 30 50 70 90 110 130 150 170 190 210 230 250 270 290 310 330 350 370 390 410 430 450 470 490 510 530 550 m ft0 0 Figure  4. Gamma-ray inten- sity logs of stratigraphic units within and outside of the Dec- orah impact structure. Left- hand side of the diagram shows a composite natural gamma log for the normal lower Paleozoic formations (fms.) outside the Decorah impact structure, as measured in adjacent Houston County, Minnesota (modified from Runkel, 1996; uses Minne- sota Geological Survey gamma logs LaCrescent #4, Zibrowski, and Spring Grove Creamery). Right-hand side of diagram shows the natural gamma log from the deep Skyline Quarry well (53572), located 0.34  km north of the center of the Deco- rah impact structure. The two logs show relative gamma-ray intensities and are compared using the top of the St. Peter Sandstone (Ss.) as a datum. Note the distinctively higher relative gamma readings of the Winneshiek Shale compared to the Shakopee Formation, and the lower and significantly more uniform relative gamma readings of the sub-Winneshiek breccia compared to the lower Jordan, St. Lawrence, and Lone Rock formations. Rela- tive gamma signatures of the Wonewoc, Eau Claire, and Mt. Simon formations in Well 53572 log are somewhat noisier compared to Houston County. A supporting cuttings sample log for Well 53572 is available on the GeoSam section of the Iowa Geological Survey’s website, https:// www .iihr .uiowa .edu /igs /geosam /home (accessed April 2018). The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 7 The basin fill (Figs. 3, 4, 5, and 6) includes two units of particular significance. A lower (un- named) breccia unit (Figs. 5 and 6) is composed of large and small, poorly sorted fragments of locally derived carbonates and sandstones; this unit may be partly or completely formed by sedimentary processes, It includes numerous rounded single quartz grains ≤2 mm in size, of which approximately ≤1% contain planar mi- crostructures (both PFs and PDFs) produced by shock waves and described below. The overly- ing, clearly sedimentary, Winneshiek Shale var- ies from ~17–27 m in thickness across the basin, contains the unusual Lagerstätte deposit (Liu et al., 2006), and is itself overlain disconformably by the Ordovician St. Peter Sandstone and Qua- ternary alluvium. Crater-Fill Units: The Crater-Fill Breccia A presently unnamed and complicated series of possible sediments and/or breccias, so far ac- cessible only from drill cores and drill cuttings, underlies the Winneshiek Shale and apparently fills the remainder of the basin down to a ten- tatively identified floor of Paleozoic sedimen- tary bedrock (Figs. 3 and 4). Because the few available coherent samples, obtained from two drill cores, show a highly fragmental and poorly sorted texture (see discussion of Sample H2-1-2, below), this unit is provisionally referred to as “crater-fill breccia,” without considering the de- tails of its origin and deposition. The maximum thickness of this unit is estimated to be 184 m. This value, however, is based on cross-section constructions, which are highly dependent on only four wells with apparent full penetration of the unit (Fig. 3). In addition, only one of the full-penetration holes (Well 53572; Fig. 4) is lo- cated near the structure’s center; the others are proximal to the basin edge. Studies of drill hole cuttings indicate that the crater-fill breccia is complex and lithologi- cally variable (Fig. 4). Three units can be dis- tinguished below the Winneshiek Shale in these cuttings, particularly from Well 53572. From top to bottom, they are: (1) a series of lithologi- cally variable breccias containing sandstone, dolomite, and minor shale (thickness ~84  m); (2) red-brown to gray-green shale (~24 m); and (3) sandstone breccia and loose sand (~52 m). The diversity of inferred rock types, and par- ticularly the occurrence of a shale layer between 30 25 20 15 10 5 0 m TD = 33.4 m fin in g up w ar d se qu en ce W in ne sh ie k S ha le S t. P et er S s. un na m ed b re cc ia breccia shale sandy sandstone bioturbated/burrows faint cross laminae weathered shale distorted laminae lenticular structure snail conodont eurypterid phyllocarid Symbol Legend Figure  5. Graphic core log of the H2 core located 0.32 km inside the mapped eastern basin edge (see Fig. 2), and drilled above the only known outcrop of Winneshiek Shale. The core makes a full penetration of the Winneshiek Shale and a partial penetration of the sub-Winneshiek breccia. The upper portion of the breccia grades upward from breccia into very coarse sandstone, then into finer sandstone with shale interbeds, and finally into the fossiliferous Winneshiek Shale. The contact between the breccia and the overlying Winneshiek shale is placed at a depth of ~18.5 m, at the first appearance of shale; this coincides with the presence of minor burrow traces. The overlying discon- formable contact between the Winneshiek Shale and the St. Peter Sandstone (Ss.), at ~2  m depth, is very sharp and displays a truncated and oxidized weathered appear- ance. A concentration of heavy minerals dominated by ilmenite and zircon, i.e., a heavy mineral lag deposit, was recovered at this contact from the adjacent excavated outcrop. TD—total depth. Figure  6. Macroscopic view of a segment of the H2 core (depth 33.23–33.38 m) from which sample H2-1-2 was cut at a depth of 33.36 m. The sample is a matrix-supported breccia containing larger (cm-sized) angu- lar to subangular clasts that range in color from light gray to medium gray. These larger clasts are dominantly dolomite with lesser sandstone; white clasts are chert. The matrix is a mixture of mm- to sub-mm-sized quartz grains, lesser amounts of feldspar and glauconite grains, and dolomite grains ranging in size from silt to sand. French et al. 8 Geological Society of America Bulletin, v. 130, no. XX/XX two breccia units, suggests a basin-filling history that is complex and prolonged, perhaps reflect- ing the post-impact sedimentary complexities that occur in impact structures formed in marine targets involving a significant depth of overlying water (e.g., Ormö and Lindström, 2000; Dypvik and Jansa, 2003). The most informative samples of the breccia unit come from partial penetrations of the up- per 16–26 m of the breccia in two cores within the structure, 62035 (CS1) and 80238 (H2) (see Figs. 3 and 5); all other breccia samples are rep- resented by drill cuttings. The breccia is com- posed of angular to subangular clasts of dolo- mite, sandstone, chert, and less shale in a matrix of very fine to coarse sand-sized dolomite and quartz, with subordinate grains of feldspar and glauconite. Clasts, in particular dolomite clasts, range up to 25 cm in length; Figure 6 shows a representative core sample of breccia that con- tains smaller clasts. In addition to the two cores, 16 logged wells yielded relatively reliable drill cuttings of sub- Winneshiek material. Prior to the recovery of the first core (H2), we suspected that several sub-Winneshiek drill cuttings sample sets repre- sented penetration of breccia or conglomeratic strata (due to the unusual mix of rock and grain types), with occasionally observed fine matrix coating on what appeared to be small (<1 cm) clasts in a few samples. Uncertainty about the actual rock fabric prevailed, however, until we were able to directly observe the breccia fabric in core samples. This observational uncertainty must have also puzzled earlier geological sur- vey well loggers, because historic logs, dating between 1939 and the early 1990s, noted an ab- normal mix of rock and grain types, as well as apparently abnormal formational thicknesses, even though the loggers applied normal strati- graphic calls to the logs. Upon retrieval of the two cores, which verified the presence of brec- cia below the shale, we reinterpreted all the his- toric logs and reexamined some of those sample sets. However, uncertainty concerning several aspects of both the older and more recent cut- tings samples remains, most notably the exact point of contact between the crater-fill breccia and the crater wall and floor (or down-dropped blocks) in several of the wells. Despite the inherent problems in the interpre- tation of drill cuttings, it is clear that the crater- fill breccia is lithologically variable horizontally as well as vertically. Examination of fragment Figure 7. Exposure of abnormally fractured and deformed Shakopee dolomite, located just outside the SE rim of the Decorah structure (see Fig. 2). View is to north; outcrop face trends approximately east-west. Mapped crater boundary is about 50 m to west (left). This outcrop, the only known surface exposure of visibly deformed pre-impact target rocks close to the structure, shows intense fracturing and deformation that are inconsistent with the typically undeformed character of the same bedrock further away from the structure. At least three sets of closely spaced subparallel fractures are visible, displaying shallow dips to the left, right, and forward (toward the observer). The V-shaped structure, immediately to the left of the geologist’s right hand, may be a recumbent fold whose axial plane dips to the left. The central portion of this structure is a monomict breccia. The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 9 types in several sets of well cuttings revealed a radial cross-basin variation in the breccia unit: carbonate lithologies (dolomite) dominate near the basin rim and siliciclastic (sandstone) frag- ments toward the center (Fig. 8). However, the present data are not adequate to reveal whether this variation reflects processes in the primary impact (e.g., deeper excavation in the center, removing more deeply buried Cambrian sand- stones) or subsequent depositional effects (het- erogeneities in the source areas or sedimentary sorting mechanisms). Crater-Fill Units: The Winneshiek Shale The Winneshiek Shale was originally de- scribed as “a greenish brown to dark-gray finely laminated sandy shale with a significant organic carbon and pyrite content” (Liu et al., 2006, p. 969). In outcrop and core samples, it is a se- ries of alternating sub-mm- to mm-thick silty to sandy shale laminae suggestive of prolonged de- position in a quiet-water environment. The Winneshiek Shale is fully penetrated by 20 drill holes and ranges in thickness from 17 to 27 m. The most reliable shale thickness data come from the two cores CS1 and H2 (62035 and 80238) and a water well (53572), which yielded a natural gamma and cuttings log (Fig. 4). Shale thickness in the cores, which are located 0.6 and 0.3 km from the basin edge and 2.2 and 2.5 km from the basin center (Figs. 3 and 18B), ranges from 17 to 18 m. Shale thick- ness in Well 53572, which is 0.35 km from the basin center, is 26  m as measured from con- sideration of both drill cuttings and a natural gamma log (Fig. 4). All subsurface and airborne electromagnetic data (Kass et al., 2013a, 2013b) suggest that the Winneshiek Shale is the upper- most basin-fill unit across the entire structure. SAMPLE COLLECTION, PREPARATION, AND STUDY METHODS In typical impact structures, the diagnostic indicators of impact-produced shock metamor- phism tend to be located in specific regions within the structure, particularly in units of crater-fill breccias that are deposited in the cra- ter immediately after excavation of the origi- nal bowl-shaped cavity (see e.g., Dence, 1968; Grieve, 1987; French, 1998, Chs. 3, 5). Such units, and the shock-metamorphosed fragments they contain, have been critical in providing convincing evidence for the impact origin of suspected structures, often through samples ob- tained by drilling (Dence, 1968; Dence et al., 1968; Grieve, 1987). At the Decorah structure, access to these crater-fill units is severely lim- ited by the complete lack of crater-fill breccia exposures, and samples of these critical litholo- gies (i.e., the polymict breccias beneath the Winneshiek Shale) could be obtained only from drill holes that encountered the breccia (Figs. 3, 4, and 5). In our study, this limitation is offset by the fact that the identification of diagnostic shock effects (especially PFs and PDFs in quartz) can be suc- cessfully carried out on small samples and even on individual mineral grains (e.g., Stöffler and Langenhorst, 1994; Grieve et al., 1996; Mon- tanari and Koeberl, 2000). The samples studied were obtained from two drill holes on the south- east periphery of the structure that penetrated the polymict breccia underlying the Winneshiek Shale: a rotary drill hole (52450) and a shorter core-drill hole (H2). (See Fig. 18B for locations.) Two samples, one from each hole, were se- lected for detailed study. (In the descriptions below, we follow earlier writers [Stöffler and Langenhorst, 1994; French et al., 2004; French and Koeberl, 2010] in using the nongenetic term planar microstructures [PMs] for all pla- nar or quasi-planar deformation features in quartz, while planar fractures [PFs] and planar deformation features [PDFs] are more strictly defined and regarded as unique and diagnostic shock-wave products that identify meteorite im- pact events.) 1. Sample H2-1-2, a 2.5 by 4.0 cm specimen from a 5-cm-diameter drill core, comes from a depth of 33.36 m in the H2 core (Figs.  5 and 6) and was examined in a standard petrographic thin section (Fig. 9). The sample depth is ~15 m below the top of the breccia. 2. The other sample, from Well 52450, was obtained from a rotary drill hole cuttings sam- ple composed dominantly of dolomite, single quartz grains, minor feldspar, and other miner- als (Fig. 10). Sample depth was an interval of cuttings from 80.8 to 82.3 m, ~8 m below the top of the breccia unit. The quartz grains were generally loose, matrix free, and rounded to bro- ken and angular in shape; the original content of deformed grains showing planar microdeforma- tions (PFs and/or PDFs) was visually estimated to be ≤l% by number. This sample was washed in tap water and oven dried at 70 °C; individual grains in the range of 0.5–1.5 mm in size were then handpicked under a binocular microscope in reflected light at magnifications of less than 40×. Grains with highly developed sets of pla- nar microstructures (PFs and PDFs) appeared white, opaque, and fractured in reflected light (Fig. 10) as a result of internal reflections. These white, opaque grains were handpicked to pro- duce a concentrate in which the percentage of quartz grains with PFs and/or PDFs was in- creased from ≤1% to >75%. The concentrated grains were incorporated into an epoxy plug 2.0 cm in diameter, cemented to a thin section, and ground and polished to a standard thickness of ~0.03 mm for petrographic studies. Petrographic thin sections from both sam- ples were examined on a standard polarizing Figure 8. Graph showing lithologic variation of rock types in drill cuttings from the Decorah impact structure crater-fill breccias and related units as a function of radial distance within the cra- ter. Data from 16 rotary churn-drill holes and 2 cores (e.g., Figs. 3 and 18) located at various radial distances within the structure (X-axis) are presented as the percentage of carbonate lithologies in fragments (Y-axis) as a function of distance (in m) outward from the center of the structure. The data show a pronounced change in the lithologies obtained across the crater: carbonate lithologies appear more abundant (70–100%) in the breccia units near the crater rim, while siliciclastic lithologies dominate closer to the cra- ter center (60–90+%). French et al. 10 Geological Society of America Bulletin, v. 130, no. XX/XX Figure  9. Microscopic views of a thin sec- tion of the breccia underlying the Win- neshiek Shale in the basin of the Decorah impact structure. (Core Hole H2, Sample H2-1-2, depth 33.2–33.3  m; plane-polar- ized light.) (A) Wide-field view of the unit, which consists of diverse, poorly sorted rock and mineral clasts, typically a few tenths of a mm to a few cm in size, enclosed in a matrix of finer clasts and carbonate mate- rial. Gray clasts are carbonate rock frag- ments; smaller white clasts are individual quartz grains, usually single. The diversity of rock types in this unit includes (typical examples indicated by numbered arrows): (1) microcrystalline carbonate rock, with typical grain sizes of 50–150 µm, containing oval structures resembling oolites or possi- bly microfossils (white circular areas in the clast are holes in the thin section); (2) micro- crystalline carbonate rock, composed of granular carbonate (dolomite?) crystals typically 100–300 µm in size; (3) very fine microcrystalline carbonate rock composed of crystals typically 10–50 µm in size, ac- companied by fine dark opaque material (organics?); (4) (in white-outlined box) multiple individual quartz grains contain- ing shock-produced planar microstructures (PMs); two grains (#4, white arrows) have well-developed PMs and display a pale to dark yellowish color producing a so-called “toasted” appearance. (B) Enlarged view of area enclosed in box (inset) of (A), show- ing three large individual quartz grains (white arrows; center and upper center) in a matrix of small carbonate rock fragments, single quartz grains, and finer material. The three indicated quartz grains show multiple sets of PMs interpreted as shock-produced planar fractures (PFs). In contrast to the lower, larger grain (lowest arrow), which appears transparent (white) in thin section, the upper two grains (upper arrows) display a pale to dark yellowish color giving them a “toasted” appearance. The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 11 micro scope (flat-stage) and then transferred to a 4-axis Leitz Universal stage (U-stage) for mea- surement of the polar angles (⊥˄c) between the quartz c-axis and the poles to the various PF and PDF planes, using standard methods (e.g., Robertson et al., 1968; Engelhardt and Bertsch, 1969; Stöffler and Langenhorst, 1994; Grieve et al., 1996; Montanari and Koeberl, 2000, p. 295–300; French et al., 2004; Ferrière et al., 2009). Both samples provided sufficient quartz grains to produce a statistically robust number (> ~100) of polar angle measurements and their angular distributions (Robertson et al., 1968; Ferrière et al., 2009). Only observable PMs were recorded; no correction was made for possible unobservable PMs lying in the “zone of inaccessibility” outside the range of possible U-stage rotations (see, e.g., Engelhardt and Bertsch, 1969; Stöffler and Langenhorst, 1994; Ferrière et al., 2009). Polar angle measurements were initially plot- ted by hand to produce a standard histogram of the number of measurements versus their angu- lar distribution (Robertson et al., 1968; Alexo- poulos et al., 1988; Stöffler and Langenhorst, 1994; Grieve et al., 1996). The individual ori- entation measurements were then replotted on a Wulff (stereographic) stereonet with the quartz c-axis rotated to vertical, producing a so-called “rectified” or “spike” plot. For each measured grain, this plot was then overlain with a stereo- graphic template of quartz orientations in order to assign specific Miller index {hkil} values to the individual planes (for details, see Engelhardt and Bertsch, 1969; Stöffler and Langenhorst, 1994; Grieve et al., 1996; Montanari and Koe- berl, 2000, p. 295–300). All plotting and {hkil} determination mea- surements were done by hand; although several automated methods are now available (e.g., Huber et al., 2011; Losiak et al., 2016), they have not been tested on large populations of measured PFs and PDFs, and we therefore did not apply them in this study. We used the older stereographic template (Engelhardt and Bertsch, 1969) containing 10 quartz {hkil} forms, rather than the recently proposed version containing 15 forms (introduced by Ferrière et al., 2009, Table 1). Using the older template does not re- sult in any serious differences in the orientation patterns obtained (see discussions in Ferrière et al., 2009), and has the advantage of making it possible to compare our petrofabric results di- rectly with those of earlier studies on other im- pact structures (e.g., French et al., 1974, 2004). RESULTS Sample Petrography Sample H2-1-2 This sample is a fine, poorly sorted, possi- bly sedimentary, polymict breccia composed chiefly of a variety of fragments of carbonate and sandstone rocks and individual mineral fragments (Fig. 9A). The lithology is structure- less, and individual fragments show apparently random sorting and orientation. Percentages of different rock and mineral components (visu- ally estimated) vary significantly with location in the thin section. Rock and mineral fragments ≥1 mm in size (and mostly as large as 5–10 mm) are typically angular and blocky and make up ~50% of the sample. Most of these fragments consist of various microcrystalline carbonate lithologies with grain sizes ranging from fine- (50–100 µm) to medium-grained (0.1–0.4 mm). All the carbonate appears to be dolomite, with no calcite present. Some carbonate fragments contain occasional thin layers of small (0.1– 0.5 mm) rounded quartz grains. A few typical carbonate fabrics, such as oolitic to peloidal, are evident in the dolomite clasts; such textures are found in both the Oneota and Shakopee formations. Single quartz grains 0.2–2 mm in size con- stitute ~25% of any given field of view, and a few exotic grains and fragments are present: microcrystalline chert, shale, crystalline quartz- rich and quartz-feldspar metamorphic(?) rocks, and fine-grained quartz-feldspar (volcanic?) rocks with lath-like feldspars forming appar- ently microgranophyric textures. The single mineral grains are overwhelmingly quartz, with very rare feldspar and glauconite. The feldspar grains are fine- to very fine-grained and usually have a rounded grain core with angular feldspar overgrowths, a texture typical for very fine- to fine-grained feldspathic sandstones in the Up- per Cambrian of the Mississippi Valley region (Odom, 1975). Matrix materials (rock and min- eral fragments <1 mm), including single quartz grains, constitute ~50% of the unit, and the in- dividual rock and mineral fragments (≥1 mm), chiefly carbonate rocks, chert and individual quartz grains, occur in a fine microcrystalline carbonate cement (~5%). The breccia consists almost entirely of rock and mineral fragments derived from local sand- stone and carbonate strata that enclose the Dec- orah structure, together with a small percentage of granitic and volcanic(?) lithologies probably derived originally from the underlying Precam- brian crystalline basement. The rarity of these latter fragments (≤1%) in the breccia suggests that they were not derived from direct excava- tion of the basement but were present as isolated pre-impact clasts in the younger sediments (e.g., the Cambrian sandstones) that were excavated by the impact (see Table 1). Individual quartz grains constitute ~25%– 50% of the matrix material and up to 25% of the overall sample, although estimated percent- ages vary significantly (from 10 to 25%) with location in the thin section. The quartz grains are generally single individuals, spheroidal to ellipsoidal in shape, and generally rounded to subrounded (Fig.  9B), although small grains Figure  10. Binocular microscopic views of two quartz grain types (unshocked and shock-deformed) representative of grains found in cuttings sample from Well W52450, depth 80.8–82.3  m. Individual grains were handpicked from the sample under binocu- lar microscope examination. (A) A typical smooth, rounded, and translucent quartz grain characteristic of the majority of quartz found in the coarser-grained Cam- brian and Ordovician sandstones of the study area. Scale bar = 1 mm. (B) A white, opaque quartz grain exhibiting parallel sets of widely spaced fractures. Grains exhib- iting this appearance are distinguishable from the more abundant translucent grains (A) and were concentrated by handpicking for thin sectioning and petrographic exami- nation. Scale bar = 1 mm. French et al. 12 Geological Society of America Bulletin, v. 130, no. XX/XX Figure 11. Shock-produced planar microstructures (“P1 features” or planar fractures ([PFs]), developed in individual quartz grains from crater-fill breccia unit of the Decorah impact structure. (A) Irregular, partly rounded individual quartz grain from the breccia unit un- derlying the Winneshiek Shale, showing three sets of shock-produced PFs: a prominent set parallel to the base (0001) and subordinate sets parallel to the { 2111 } and { 2211 } planes. The lower boundary of the grain shows a steplike character produced by the spalling of quartz fragments along the fracture sets, indicating that the separation planes are indeed open fractures. (Core Hole H2, Sample H2-1-2, depth ~33.2–33.3 m; Grain #4,1; plane-polarized light.) (B) Well-rounded single quartz grain showing three well-developed sets of shock-produced PFs parallel to the base (0001) and to the { 2211 } and { 1210 } planes. Small darker patches along the grain margin (white arrows) are actu- ally small areas where P2 features (planar deformation features [PDFs]) are developed. (From sorted grain mount prepared from drill cuttings, Well “Leon Steinlage” W-52450, depth ~80.8–82.3 m; Grain #11,1; plane-polarized light.) (C) Well-rounded single quartz grain showing three sets of shock-produced PFs parallel to distinct { 1110 } planes in the crystal. (A fourth set of apparent PFs is also observed but could not be indexed [“N/I” = not indexed]). Small, slightly darker, patches along the grain margin (arrows) are actually small areas where P2 features (PDFs) are locally developed. (From sorted grain mount prepared from drill cuttings, Hole “Leon Steinlage” W-52450, depth ~80.8–82.3 m; Grain #20,1; plane-polarized light.) (D) Large well-rounded quartz grain, with a partly angular rim, from the crater- fill breccia unit underlying the Winneshiek Shale. The grain shows three sets of well-developed shock-produced PFs parallel to the { 1010 } and { 1110 } planes; a fourth set of possible PFs (“[N/I]”) could not be indexed. (Core Hole H2, Sample H2-1-2, depth 33.2–33.3 m; Grain #227,1; plane-polarized light.) The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 13 (<0.1  mm) may be angular, possibly because they are fragments derived from larger grains. These quartz grains could be sourced from any quartzose sandstone units in the sedimentary section, e.g., the Ordovician basal Shakopee, or the Cambrian Jordan formations (see Table 1). If the impact process excavated sedimentary rock units deeper than the Lone Rock, the coarser quartz grains could also have been derived from the Wonewoc or Mt. Simon formations. Sample W-52450 The handpicked quartz grains in this sample are typically 0.5–1.5 mm in size, spheroidal to ellipsoidal in shape, and well-rounded to sub- rounded. Out of 96 total grains, 93 were quartz and 3 were carbonate. Out of the 93 quartz grains, 73 (78%) contained measurable sets of planar microstructures (both PFs and PDFs). The remaining 20 quartz grains were not mea- sureable for several reasons: extreme marginal cracking during preparation (16), polycrystallin- ity and small individual grain sizes (3), and ap- parent absence of any planar microstructures (1). Deformation of Quartz Grains In the thin section of sample H2-1-2, the ma- jority of quartz grains show no unusual deforma- tion and display only features characteristic of normal metamorphism (for descriptions, see, e.g., Spry, 1969; Vernon, 2004; French and Koeberl, 2010, and references therein). Most of the quartz grains are undeformed or only slightly deformed; extinction under crossed polarizers is generally sharp (<5°), occasionally slightly undulose (5– 10°), and rarely strongly undulose (>10°). In rare cases, deformation bands or segmented extinc- tion in adjacent areas are evident, but no multiple small domains (“mosaic extinction”) were ob- served. Typical metamorphic deformation lamel- lae (Bӧhm lamellae or MDLs) were observed in only a few grains. The sample of rotary cuttings (sample W-52450) is similar; most of the individ- ual quartz grains show similar features and lack any unusual deformation effects. Significant percentages of quartz grains in both samples show single or multiple parallel sets of planar microstructures (PMs) in various orientations. Such features occur in ≤1% of the grains in sample H2-1-2 (Fig. 9B) and >75% of the individual handpicked grains from sample W-52450. The observed PMs are of several dif- ferent types; some represent the results of nor- mal crystallization and metamorphism; others are due to shock deformation (for discussions, see, e.g., French and Koeberl, 2010, p. 133– 141). For convenience in detection, measure- ment, data reduction, and interpretation, we first classify PMs into two nongenetic categories, P1 and P2, an informal terminology used in study- ing similar shocked sedimentary rocks from the Rock Elm structure, Wisconsin (French et al., 2004). P1 features are larger, darker, more widely spaced, and more continuous across a larger fraction (typically >50%) of an individual grain; they are interpreted as shock-produced open fractures (cleavage), identical to the fea- tures designated as planar fractures (PFs) (Stöffler and Langenhorst, 1994). P2 features designate shorter, narrower, and more closely spaced closed structures that may have a variety of characteristics and origins: healed fractures, subsidiary fractures in feather-fracture features (French et al., 2004; Poelchau and Kenkmann, 2011), and genuine shock-produced planar de- formation features (PDFs) (Stöffler and Lan- genhorst, 1994; Grieve et al., 1996). P1 Features P1 features (or PFs) are the dominant PMs observed in shocked quartz grains from the Decorah structure (Fig. 11). They form multiple sets of diversely oriented planes within single host quartz grains, and they occur in virtually all grains that show any shock-produced PMs. Typically, 1–5 distinct sets of P1 planes are present in individual grains, and larger numbers (≤8) are occasionally present. In both Decorah samples, most grains with P1 features exhibit 2, 3, or 4 distinct sets. Values for sample H2-1-2 are: 2 (25%), 3 (36%), and 4 (11%). Individual P1 planes are generally dark, uniformly planar, between 1 and 2.5 µm thick, and spaced at dis- tances of 10–25 µm. Mutual offsets along inter- secting sets of planes have not been observed. Fracturing and spalling of the host quartz grains has occasionally occurred along the planes of P1 features, forming sharp, step-like grain bound- aries in which the “steps” are continuous with, or parallel to, specific P1 features in the grain (Fig. 12). Within other grains, the development of small rhombic blocks, parallel to existing P1 features, is evident. These observations indicate that the P1 features are themselves discrete open fractures within the quartz grain, along which subgrain fragments have easily separated. The dark color of individual P1 planes may be pro- duced by a filling of exotic material (e.g., clay minerals) within the fractures, although no spe- cific birefringent materials were observed (see French et al., 2004, p. 205). P2 Features P2 features in Decorah quartz grains are gen- erally light-colored and form small, isolated areas typically ≤100 µm in size, commonly located at the corners, or along the edges, of Figure  12. Irregular quartz grain showing two poorly developed sets of shock- produced planar fractures (PFs) parallel to { 1110 }. The upper edge of the grain shows a pronounced steplike boundary produced by spallation of the marginal parts of the quartz grain along the PFs (see also Fig. 11A). (From sorted grain mount pre- pared from drill cuttings, Hole “Leon Steinlage” W-52450, depth ~80.8–82.3 m; Grain #32,1; plane-polarized light.) French et al. 14 Geological Society of America Bulletin, v. 130, no. XX/XX Figure 13. (A) Well-rounded quartz grain showing two sets of shock-produced planar fractures (PFs) parallel to { 1110 } and a third unin- dexed set of less developed PFs ([“N/I”]). Darker marginal patch at the top of grain (white box) is a small area where P2 Features (planar deformation features [PDFs]) are locally developed (see Fig. 13B). (From sorted grain mount prepared from drill cuttings, Hole “Leon Steinlage” W-52450, depth ~80.8–82.3 m; Grain #64,1; plane-polarized light.) (B) Enlarged view of area shown in Figure 13A, showing details of darker patch along grain margin. Dark area is seen to be composed of two intersecting sets of thin, optically distinct, sets of P2 features (PDFs), oriented parallel to the distinctive, shock-produced ω{ 1310 } direction. (From sorted grain mount prepared from drill cut- tings, Hole “Leon Steinlage” W-52450, depth ~80.8–82.3 m; Grain #64,2; plane-polarized light.) (C) Partly rounded to angular quartz grain from the crater-fill breccia unit underlying the Winneshiek Shale, showing good development of two sets of shock-produced P1 Features (PFs) (N-S and NW-SE orientations in photo; actual orientations could not be determined because the grain was not accessible on the U- stage.) In the lower left area of the grain (white outline box and Fig. 13D) P2 features (PDFs) are extensively developed in the areas between the PFs. (Core Hole H2, Sample H2-1-2, depth ~33.2–33.3 m; Grain #155,1; plane-polarized light.) (D) Enlarged view of area of grain (box) shown in Figure 13C. Parallel PFs trend WNW–ESE. PDFs, which trend NNW-SSE and appear as shorter, closely spaced narrow planes in the quartz between the PFs (especially at the lower left center of the picture), show the close spacing and optically distinct appearance from the host quartz that is typical for PDFs. (Actual orientations could not be determined because the grain was not accessible on the U-stage.) (Core Hole H2, Sample H2-1-2, depth ~33.2–33.3 m; Grain #155,2; plane-polarized light.) The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 15 quartz grains, and more rarely within the grains themselves (Fig.  13). Where present, P2 fea- tures generally form multiple sets, typically 2–4 per grain. Individual planes are sharply planar, clear, closely spaced, and locally continuous, and their visibility appears to be enhanced by refractive-index differences (“Becke line ef- fects”) between the planes and the host quartz. The close spacing and optical effects associated with P2 features make measurements difficult, but typical widths appear to be 0.5–1 µm, with spacings of 1.5–2 µm. In the handpicked grains from sample W-52450, P2 features occur in ~60% of the 77 measured grains that also show P1 features. In both samples, P2 features were found alone in only one quartz grain. The characteristics of P2 features in the Decorah samples (multiplicity, narrowness, close spacing, and possible refractive index ef- fects) are virtually identical to those of shock- produced PDFs (e.g., Alexopoulos et al., 1988; Stöffler and Langenhorst, 1994; Grieve et al., 1996; Ferrière et al., 2009), and the orientation patterns obtained from U-stage measurements (described below) support that interpretation. The Decorah P2 features differ from those in deformed quartz from the Rock Elm structure, Wisconsin (French et al., 2004), where P2 fea- tures are typically short, planar to subplanar, and inclusion-decorated, resembling healed frac- tures. The Rock Elm P2 features are commonly associated with P1 features (planar fractures) forming distinctive feather features that have been suggested as diagnostic for low-pressure shock waves in impact structures (French et al., 2004; Poelchau and Kenkmann, 2011). Al- though feather features are common in the Rock Elm rocks (French et al., 2004), they were only rarely observed in quartz grains from our Deco- rah samples (for an example, see Fig. 14), where individual sets of both P1 and P2 features are the most common deformation features. In contrast to the mild deformation observed in the majority of quartz grains from our Deco- rah samples (see above), grains containing P1 and P2 features are significantly deformed themselves. In cross-polarized light, extinction is highly undulose to irregular, often forming a variable mosaic pattern. Such extreme deforma- tion and extinction effects appear restricted to grains that display obvious P1 and P2 features. Rare quartz grains with P1 and P2 features also show a pale-yellow to medium yellow-brown color (“toasting”) in transmitted light in thin section, an effect attributed (Whitehead et al., 2002) to the presence of small fluid inclusions associated with shock-produced PDFs. This “toasting” effect is observed only in a small fraction of grains with P1 and P2 features. The grain darkening produced by the presence of P1 features, combined with the color effects of “toasting” (where present), allow these shock- deformed grains to be quickly distinguished in thin section from the more abundant clear and undeformed grains (Figs. 9A and 9B). Orientation Patterns of Planar Microstructures (P1 and P2 Features) The measurement of planar microstructure (PM) orientations in quartz-bearing samples from suspected impact structures, and the dem- onstration that such orientations are uniquely different from those produced by non-impact processes, have been critical to the recognition of terrestrial impact structures for several decades (see papers in French and Short, 1968; Stöffler and Langenhorst, 1994; Grieve et al., 1996; Fer- rière et al., 2009; French and Koeberl, 2010). To obtain PM orientation data for the Decorah structure, we measured suites of quartz grains containing >100 PMs in both of our Decorah samples (sample H2-1-2: 122 planes/28 grains and sample W-52450: 376 planes/77 grains) (Table 2). Histogram plots of frequency versus polar angle were constructed separately for P1 features, P2 features, and total (P1 + P2) features (Figs. 15A, 15C, and 15E, and 16A, 16C, and 16E). In addition, rectified (“spike”) plots were determined for the same populations (Figs. 15B, 15D, and 15F, and 16B, 16D, and 16F). P1 features (PFs) are well expressed in quartz grains in both samples, with multiple sets ob- Figure 14. (A) Highly rounded elliptical quartz grain showing three sets of planar fractures (PFs) parallel to the (0001), { 1110 }, and { 2211 } planes. A few of the individual fractures show, e.g., in upper right and lower center parts of the grain (white arrows and white outline box), the development of distinctive feather features, in which short parallel planar microstructures (PMs) develop at an angle to the fracture with one end based on the fracture itself. (See enlarged view [inset, white outline box] in Fig. 14B.) (From sorted grain mount prepared from drill cuttings, Hole “Leon Steinlage” W-52450, depth ~80.8–82.3 m; Grain #88,1; plane-polarized light.) (B) Enlarged view of area shown in box in Figure 14A, showing development of short, closely spaced feather features along longer, more widely spaced planar fractures (PFs) parallel to the (0001) and { 1110 } planes. The shorter, more closely spaced feather features themselves are oriented parallel to the { 2011 } and { 2211 } planes. (From sorted grain mount prepared from drill cuttings, Hole “Leon Steinlage” W-52450, depth ~80.8–82.3 m; Grain #88-2, plane-polarized light.) French et al. 16 Geological Society of America Bulletin, v. 130, no. XX/XX served in ≤1% of the grains in a given sample. The P1 orientations show fabrics typical for PFs in shocked sedimentary rocks at numerous im- pact structures (French et al., 1974, 2004; Grieve et al., 1996). This pattern is characterized by strong concentrations of planes parallel to the po- lar angles of 0°, 48°–52°, and 90°, corresponding to the planes c(0001), ξ{1122}, (r,z){1011}, and a{1120} (Figs. 15A, 15B, 16A, and 16B). The three largest peaks of the measured P1 features in W-52450 correspond to the planes (r,z){1011} (39%), c(0001) (15%), and ξ{1122} (12%). The “not indexed” (N/I) values for the different pop- ulations in both samples H2-1-2 and W-52450 (11%–23%) are comparable to those measured from established impact structures (≤10%–20%: Grieve et al., 1996; Ferrière et al., 2009). P2 features in the Decorah samples are in- terpreted as definite PDFs (see above). Their orientations patterns (Figs. 15C, 15D, 16C, and 16D) show significant peaks at polar angles of 0°, 48°, 52°, and 90°, similar to those shown by the P1 features (see above). In addition, the P2 orientations show significant peaks at angles of 23° and 32°, corresponding to the well-known ω{1013} and π{1012} planes that are typi- cal of shock-metamorphosed crystalline rocks (Robertson et al., 1968; Grieve et al., 1996). The P2 orientations in Decorah samples show more and smaller peaks than those for the P1 planes. The three largest peaks for all P2 sam- ples in W-52450 correspond to (r,z){1011} (17%), ω{1013} (13%), and c(0001) (9%). The significant presence of orientations parallel to ω{1013} and π{1012} is consistent with the interpretation of P2 features as PDFs. In the Decorah samples, both P1 and P2 features show strong concentrations at specific polar angles, particularly ~0°, ~50°, and ~90°. Such concentrations are typical for the fabrics produced by shock metamorphism, and the Decorah plots are closely similar to plots from two other established impact structures: the BP site, Libya (French et al., 1974) and Rock Elm, Wisconsin (French et al., 2004) (Fig. 17). The “not indexed” values for P2 features (21%–35%) are higher than those for P1 fea- tures (11%–23%). This situation may reflect a lower degree of measurement precision, which would in turn produce a lower precision of {hkil} assignments. Possible explanations in- clude: (1) the designation as P2 features of multiple features with different origins, e.g., true PDFs, rarer feather-fracture features, and other small unidentified deformations; (2) the difficulty of measuring even true PDFs because of their generally vague, restricted, and patchy character in the quartz grains studied. Despite these difficulties, orientation fabrics for the Decorah samples (Figs. 15, 16, and 17) are closely comparable to those determined for shocked quartz at established impact structures, and the data presented here constitute solid evi- dence for the action of shock waves on these samples and, therefore, for the origin of the Decorah structure by meteorite impact. DISCUSSION Planar Microstructures: Orientation Diagrams, Statistical Evaluation, and Occurrence Use of Traditional Plotting and Data- Reduction Methods Until recently, PM orientation diagrams have been plotted and evaluated by hand, and the use of such plots to identify shock environments and meteorite impact structures has been done largely by inspection and qualitative compari- son, relying on: (1) the uniqueness of shock- produced orientation fabrics; (2) their similari- ties to orientation patterns from experimentally shocked samples or from established impact structures; and (3) the clear differences between the orientations of shock-produced PMs and those produced by non-shock geological pro- cesses (see, e.g., French and Short, 1968; Engel- hardt and Bertsch, 1969; Alexopoulos et al., 1988; Stöffler and Langenhorst, 1994; Grieve et al., 1996; Ferrière et al., 2009; French and Koeberl, 2010). More recently, attempts have been made to develop rapid and accurate com- puterized methods to replace the hand-plotting of orientation histograms and the use of a hand- manipulated stereonet to determine the {hkil} values of individual PM sets (Huber et al., 2011; Losiak et al., 2016). At the same time, a related TABLE 2. PLANAR MICRODEFORMATION FEATURES IN QUARTZ: DECORAH STRUCTURE, IOWA Symbol* {hkil}† values Polar angle (o) P1 P2 P1 + P2 no. (%) no. (%) no. (%) Sample H2-1-2 (Core Sample) c (0001) 0.0 11 15 3 6 14 11 ω {1013} 23.0 4 6 5 10 9 7 pi {1012} 32.4 1 1 3 6 4 3 ξ {1122} 47.7 10 14 6 12 16 13 r/z {1011} 51.7 16 23 8 16 24 20 s {1121} 65.6 3 4 4 8 7 6 ρ {2131} 73.7 4 6 1 2 5 4 x {5161} 82.0 5 7 2 4 7 6 m {1010} 90.0 1 1 5 10 6 5 a {1120} 90.0 0 0 0 0 0 0 Not Indexed 16 23 14 27 30 25 Total 71 100 51 101 122 100 No. of Planes (P) 71 51 122 No. of Grains (G) 28 28 28 P/G ratio 2.5 1.8 4.4 Sample W-52450 (Grain Mount), Total Planar Microstructures c (0001) 0.0 41 15 9 9 50 13 ω {1013} 23.0 7 3 13 13 20 5 pi {1012} 32.4 8 3 6 6 14 4 ξ {1122} 47.7 33 12 8 8 41 11 r/z {1011} 51.7 106 39 18 17 124 33 s {1121} 65.6 7 3 4 4 11 3 ρ {2131} 73.7 13 5 3 3 16 4 x {5161} 82.0 12 4 8 8 20 5 m {1010} 90.0 12 4 8 8 20 5 a {1120} 90.0 0 0 0 0 0 0 Not Indexed 33 12 27 26 60 16 Total 272 100 104 102 376 99 No. of Planes (P) 272 104 376 No. of Grains (G) 77 77 77 P/G ratio 3.5 1.4 4.9 Note: Samples are stored in the collections of the Iowa Geological Survey, IIHR–Hydroscience & Engineering, University of Iowa, Iowa City, Iowa, 52242. *Crystallographic plane. †Miller-Bravais Index. The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 17 development has been the design and use of a new stereonet template containing 15 {hkil} forms (Ferrière et al., 2009) to replace the tra- ditional and established version (Engelhardt and Bertsch, 1969) containing only 10 forms. Computerized data-reduction for PM orien- tation measurements is a highly desirable goal; such methods would produce major savings in time, significant improvements in overall accu- racy and confidence, and the ability to measure and process more and larger batches of PM ori- entation data in shorter periods of time. How- ever, in this paper, we have used the established methods (hand-plotting and the 10-form stere- onet) for the Decorah samples for several rea- sons beyond the obvious ones of convenience and familiarity: (1) the computer methods have not yet been tested and evaluated by comparing large batches of real data on both known and suspect impact structures, and it is possible that significant modifications may still be required; and (2) the old-style orientations can be com- pared directly, both visually or with simple sta- tistical analysis, with similar plots determined for impact structures over several decades (Grieve et al., 1996) (see Fig. 17). Detailed statistical comparisons between the “old” and “new” methods of PM plotting and evaluation have yet to be made, but it is unlikely that our use of the “old” system will compro- mise our conclusions. The five new {hkil} forms in the “new” stereonet (Ferrière et al., 2009) in- volve only a small percentage of the planes mea- sured in a typical sample; for example, only a small fraction of ω{1013} planes in the “old” system would be shifted to the “new” {1014} plane (see Ferrière et al., 2009, p. 934), and this is unlikely to result in major differences in {hkil} assignments between the “old” and “new” stere- onets. Orientation diagrams will appear similar regardless of which stereonet is used, and the identifying characteristics of shock-produced orientations, i.e., extreme concentrations at specific {hkil} planes, will be evident in either stereonet. If more planes can be indexed with the “new” stereonet, the percentage of “not in- dexed” (“N/I”) planes will be reduced (Ferrière et al., 2009, p. 934). Until the “old” and “new” plotting systems are rigorously compared, how- ever, there is no objective basis for determining which one is a more reliable indicator of shock- produced PM orientation fabrics in samples from suspected impact structures. Ambiguities in {hkil} Assignments A significant, although not serious, problem with either system of plotting and data reduc- 80 90 16 0 2 4 6 8 10 12 14 0 10 20 30 40 50 60 70 80 90 N um be r o f P la ne s N um be r o f P la ne s Polar Angle c )( x {5 16 1} m {1 01 0} a {1 12 0} {1 01 3} C (0 00 1) {1 01 2} {1 12 2} r/z {1 01 1} s {1 12 1} p {2 13 1} x {5 16 1} m {1 01 0} a {1 12 0} 16 8 10 12 14 16/71 planes (23%) not indexed [N/I] 14/51 planes (27%) not indexed [N/I] P2 Features P2 Features Grains = 28 Planes = 51 P/G = 1.8 Rectified {hkil} indices 16 0 2 4 6 8 10 12 14 0 10 20 30 40 50 60 70 16 0 2 4 6 8 10 12 14 0 10 20 30 40 50 60 70 80 90 32 0 4 8 12 16 20 24 28 0 10 20 30 40 50 60 70 80 90 N um be r o f P la ne s N um be r o f P la ne s N um be r o f P la ne s N um be r o f P la ne s Polar Angle c )( Polar Angle c )( Polar Angle c )( {1 01 3} C (0 00 1) {1 01 2} {1 12 2} r/z {1 01 1} s {1 12 1} p {2 13 1} {1 01 3} C (0 00 1) {1 01 2} {1 12 2} r/z {1 01 1} s {1 12 1} p {2 13 1} x {5 16 1} m {1 01 0} a {1 12 0} [N/I] Decorah, IA. Core Sample H2-1-2 (Quartz microbreccia) Thin Section, Grain Numbers 201-228 30/122 planes (25%) not indexed [N/I] 0 2 4 6 0 10 20 30 40 50 60 70 80 90 16 0 2 4 6 8 10 12 14 0 10 20 30 40 50 60 70 80 90 Polar Angle c )( Polar Angle c )( [N/I] [N/I] P1 Features (P1 + P2) Features P1 Features (P1+P2) Features Grains = 28 Planes = 71 P/G = 2.5 Grains = 28 (P1+P2) = 122 (P1+P2)/G=4.4 Rectified {hkil} indices Rectified {hkil} indices A B D F C E Figure 15. Graphs of orientations of planar microstructures in quartz grains from crater-fill breccia unit filling the Decorah structure (Sample H2-1-2; depth interval 33.2–33.3 m; see Fig. 9). Histogram plots show frequencies of different angles between quartz c-axis and pole to plane for P1 features (cleavage, fractures: A, B), P2 features (possible planar deformation features [PDFs] and other fea- tures: C, D), and total features (P1 + P2) (E, F). Two types of histograms are shown: (1) unmodified frequency data for polar angles (A, C, E), and (2) “rectified” or “spike” plots (B, D, F) derived from a c-axis-vertical orientation and allowing identification of {hkil} Miller index values for specific planes (for details, see Grieve et al., 1996; Ferrière et al., 2009). Note, in both types of plots, the strong concentrations of planes corresponding to specific orientations, e.g., (0001) (0°), { 1310 } (23°), { 2211 } (48°), and { 1110 } (52°). Such con- centrations are typical and diagnostic for shock-produced deformation features in quartz grains from established impact structures. French et al. 18 Geological Society of America Bulletin, v. 130, no. XX/XX tion is a small number of ambiguities in as- signing exact {hkil} values to certain closely spaced PM sets, because the accuracy of indi- vidual orientation measurements is generally not better than ±  5° (e.g., Grieve et al., 1996; Ferrière et al., 2009). In both Decorah samples, a significant number of grains (e.g., ~10% in sample W-52450), which displayed planes at polar angles of 45–55°, produced different but equally good {hkil} matches with the template, regardless of whether these planes were arbi- trarily assigned to the forms ξ{1122} (48°) or to r/z{1011} (52°). A smaller number of simi- lar ambiguities were noted with other forms. Choices between these possibilities were made arbitrarily, usually by choosing alternate possi- bilities in alternate grains. However, the number of grains at issue is relatively small (only a few grains in each measurement group), and regard- less of how such assignments are made, the dif- ferent choices produce only small changes in the percentages of the particular planes involved. The overall appearance of the orientation pat- tern does not change significantly, and the major shock-produced characteristics are preserved. Rarity of “Feather Features” at Decorah The P2 features at Decorah differ from those at the similar Rock Elm, Wisconsin impact structure (French et al., 2004). At Rock Elm, the features designated as P2 are generally small, healed, quasi-planar fractures, many of which are connected at one end to larger P1 features to form common and distinctive feather fea- tures, which are believed to form at relatively low shock pressures (~7–10  GPa) (French et al., 2004; Poelchau and Kenkmann, 2011). At Decorah, most P2 features appear to be genu- ine PDFs (Fig. 13), which indicate higher shock pressures (>10 GPa; Stöffler and Langenhorst, 1994), while feather features were very rarely observed (e.g., Fig.  14). The difference in ap- parent shock levels in samples from the two lo- calities could be explained in at least three ways: (1) the difference is a random artifact of the small number of Decorah samples examined; (2) the Rock Elm samples were collected from in-place bedrock located below the final crater floor, where relatively lower shock pressures would be expected (Dence, 1968; Robertson, 1975; Robertson and Grieve, 1977); and (3) the Decorah samples consist of rock fragments and individual quartz grains collected from a prob- able unit of crater-fill breccia composed of ma- terial derived from various locations above the original crater floor. Some of the Decorah mate- rial was probably exposed to higher pressures 80 90 16 0 2 4 6 8 10 12 14 0 10 20 30 40 50 60 70 80 90 N um be r o f P la ne s N um be r o f P la ne s Polar Angle c )( x {5 16 1} m {1 01 0} a {1 12 0} {1 01 3} C (0 00 1) {1 01 2} {1 12 2} r/z {1 01 1} s {1 12 1} p {2 13 1} x {5 16 1} m {1 01 0} a {1 12 0} 32 16 20 24 28 33/272 planes (12%) not indexed [N/I] 27/104 planes (26%) not indexed [N/I] P2 Features P2 Features Grains = 77 Planes = 104 P/G = 1.4 Rectified {hkil} indices 80 0 10 20 30 40 50 60 70 0 10 20 30 40 50 60 70 64 0 8 16 24 32 40 48 56 0 10 20 30 40 50 60 70 80 90 160 0 20 40 60 80 100 120 140 0 10 20 30 40 50 60 70 80 90 N um be r o f P la ne s N um be r o f P la ne s N um be r o f P la ne s N um be r o f P la ne s Polar Angle c )( Polar Angle c )( Polar Angle c )( {1 01 3} C (0 00 1) {1 01 2} {1 12 2} r/z {1 01 1} s {1 12 1} p {2 13 1} {1 01 3} C (0 00 1) {1 01 2} {1 12 2} r/z {1 01 1} s {1 12 1} p {2 13 1} x {5 16 1} m {1 01 0} a {1 12 0} [N/I] Decorah, IA. Sample W-52450 grain mount, Grain Numbers 3-93 60/376 planes (16%) not indexed [N/I] 0 4 8 12 0 10 20 30 40 50 60 70 80 90 160 0 20 40 60 80 100 120 140 0 10 20 30 40 50 60 70 80 90 Polar Angle c )( Polar Angle c )( [N/I] [N/I] P1 Features (P1 + P2) Features P1 Features (P1+P2) Features Grains = 77 Planes = 272 P/G = 3.5 Grains = 77 (P1+P2) = 376 (P1+P2)/G=4.9 Rectified {hkil} indices Rectified {hkil} indices A B D F C E Figure 16. Graphs of orientations of planar microstructures in shocked individual quartz grains separated from the microbreccia unit filling the Decorah structure (Sample W-54250; depth interval 80.8–82.3 m.). Histogram plots show frequencies of different angles between quartz c-axis and pole to plane for P1 features (cleavage, fractures: A, B), P2 features (possible planar deformation features [PDFs] and other features: C, D), and total features (P1 + P2) (E, F). Two types of histograms are shown: (1) unmodified frequency data for polar angles (A, C, E), and (2) “rectified” or “spike” plots (B, D, F) derived from a c-axis-vertical orientation and allowing identification of {hkil} Miller index values for specific planes (for details, see Grieve et al., 1996; Ferrière et al., 2009). Note, in both types of plots, the strong concentrations of planes corresponding to specific orientations, e.g., (0001) (0°), { 1310 } (23°), { 2211 } (48°), and { 1110 } (52°). Such concentrations are typical and diagnostic for shock-produced deformation features in quartz grains from established impact structures. The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 19 closer to the impact point, producing numerous grains with definite PDFs and relatively fewer grains with feather features. Geological Structure and Comparisons to Other Impact Structures Apparent Absence of a Central Uplift Our preliminary interpretation of the Decorah structure (Figs.  3 and 18) exhibits several fea- tures characteristic of small, deeply eroded mete- orite impact structures (e.g., French et al., 2004; French and Koeberl, 2010; Kenkmann et al., 2013, 2014, 2017): a generally circular outline, a shallow-basin shape, and a local sequence of anomalous sediments and/or breccias that trun- cate and replace the original regional stratigra- phy. The presence of diagnostic shock-produced PFs and PDFs in quartz grains from the crater- fill breccia establishes a strong and convincing similarity between the Decorah structure and numerous other structures generally regarded as the products of meteorite impact events. However, critical geological information about the structure remains poorly known. Sur- face exposures are rare and limited to slightly deformed rim rocks and to the uppermost part of the post-impact crater-fill sediments (Win- neshiek Shale). Several drill holes provide cut- tings samples of the subsurface units (Figs.  3 and 18), but only one basin-centric hole (53572) apparently penetrates the complete thickness of crater breccia and enters the bedrock beneath (Figs. 3, 4, and 18). The two drill holes (52450 and H2) that provided the critical samples for the discovery and analysis of shock-metamor- phic features in quartz both bottom in the brec- cia layer and provide no information about the crater floor or the sedimentary bedrock units be- low it. The subsurface character of the Decorah structure is therefore uncertain, and no ground- based geophysical surveys, including gravity and seismic methods, have yet been done. The airborne gravity gradient data (Kass, et al., 2013a, 2013b) clearly demarcate an area of low density consistent with the center of the impact structure, and further modeling with this data set may clarify the depth and three-dimensional configuration of the structure. Our preliminary interpretation of the Deco- rah structure (Figs. 3 and 18) shows some sig- nificant differences from impact structures of comparable size formed in similar sedimen- tary targets. The presently preserved Decorah structure apparently lacks a preserved circular, uplifted rim, a condition that may reflect deep erosion of the original structure before depo- sition of the overlying post-crater sediments. A more striking anomaly is the apparent ab- sence of any structural uplift of the subcrater rocks (Wonewoc, Eau Claire, and Mt. Simon formations) in the center of the structure. This tentative interpretation is based only on the evaluation of cuttings from the single well-hole (53572) that apparently penetrated the crater- fill units and entered into the underlying sedi- mentary rocks, which appear to be located at approximately the same level as the equivalent units outside the structure (Fig. 3). The apparent absence of any central uplift in the Decorah structure is surprising. Such uplifts have long been regarded as an integral part of the formation of large (diameter >~2 km) im- pact craters (Dence, 1965; Grieve et al., 1977; Melosh, 1989, Ch. 8), and central uplifts have been identified in nearly all known impact cra- ters in this size range (Grieve and Pilkington, 1996; Kenkmann et al., 2013, 2017), although a few exceptions may exist (Lindström et al., 2005; King et al., 2006; Darlington et al., 2016). Current cratering models (e.g., Grieve, 1991; Grieve and Pesonen, 1992; Grieve and Pilkington, 1996; Melosh and Ivanov, 1999; Kenkmann et al., 2014) suggest that an impact structure of this size (initial diameter ~6  km) in sedimentary rocks would normally develop as a complex structure, consisting of a central uplift of underlying target rocks surrounded by an annular depression (commonly filled with impact breccias and overlying sediments), and an outer rim. In such a complex structure, the maximum stratigraphic uplift of subcrater rocks in the central uplift is ~0.1D, where D is the fi- nal crater diameter. This model implies that a central uplift with a maximum stratigraphic up- lift of ~600 m should have been produced when the Decorah structure formed, but the current limited stratigraphic and drill-core information 80 900 4 8 12 16 20 24 28 0 10 20 30 40 50 60 70 N um be r o f P la ne s Polar Angle c )( Libya, BP Site; Sample U-22.2A (P1 + P2) Features Grains = 26 (P1+P2) = 101 (P1+P2)/G = 3.9 80 900 4 8 12 16 20 24 28 0 10 20 30 40 50 60 70 N um be r o f P la ne s Polar Angle c )( Rock Elm, WI; Samples WRF-98-15 (A,B,C) (P1 + P2) Features Grains = 41 (P1+P2) = 176 (P1+P2)/G = 4.3 64 0 8 16 24 32 40 48 56 0 10 20 30 40 50 60 70 80 90 N um be r o f P la ne s Polar Angle c )( (P1 + P2) Features Grains = 77 (P1+P2) = 376 (P1+P2)/G=4.9 16 0 2 4 6 8 10 12 14 0 10 20 30 40 50 60 70 80 90 N um be r o f P la ne s Polar Angle c )( (P1 + P2) Features Grains = 28 (P1+P2) = 122 (P1+P2)/G=4.4 Decorah, IA. Core Sample H2-1-2 Grain Numbers 201-228 Decorah, IA. Sample W-52450 Grain Numbers 3-93 A B C D Figure  17. Graphs of orientations of planar microstructures (P1 + P2 features) in samples from the Decorah structure, Iowa (W-52450 [Fig. 17A] and H2-1-2 [Fig. 17B]) compared with similar plots for quartz grains in samples from established impact structures, the BP site (Libya) (Fig. 17C) (French et al., 1974, Figure 4, plot 1) and Rock Elm, Wisconsin (Fig. 17D) (French and Cordua, 1999, Fig. 3; French et al., 2004, Fig. 4A). (The BP and Rock Elm plots differ slightly from the original publications because they have been constructed by replotting the original data onto the same templates as the Decorah plots.) All plots show the same development of high concen- trations of planes corresponding to specific orientations, e.g., (0001) (0°), { 1310 } (23°), { 2211 } (48°), and { 1110 } (52°), which are typical and diagnostic for shock-produced deformation features in quartz grains from established impact structures. French et al. 20 Geological Society of America Bulletin, v. 130, no. XX/XX Figure 18. (A) Schematic geological cross section of the Decorah impact structure and pre-impact target stratigraphy, based on drilling re- sults and limited surface exposures (modified from McKay et al., 2011). Vertical exaggeration 10×; datum level set at top of the post-impact St. Peter Sandstone. The section trends parallel to regional strike and has been constructed by projecting stratigraphy determined from mul- tiple wells onto a vertical plane passing in a NW-SE direction through the inferred center of the structure (see Fig. 18B). For simplicity, this reconstruction arbitrarily assumes that the Decorah structure has the general shape of a normal complex crater containing a central uplift of the original crater floor (see Kenkmann et al., 2013, and references therein), although the nature of the subcrater floor and presence of such a central uplift have not been established (see discussion in text: “Apparent Absence of a Central Uplift”). In this cross-section, the Decorah structure is expressed by the presence of an anomalous series of complex and poorly characterized crater-fill units (possible breccias, sand- stones, and other sediments, including the uppermost Winneshiek Shale) that fill the preserved lower part of the original crater basin (area within the closed line C1–A2–C2–C1) (dark solid line and long dashes) and which interrupt and crosscut the earlier stratigraphic units (St. Lawrence–Lone Rock through Shakopee formations). For comparisons between the original and final impact structures, this figure also includes graphical elements of the impact and the resulting crater, assuming that the structure formed as a typical complex crater after the vertical impact of a projectile ~300–400 m in diameter at a velocity of ~15–20 km/sec. At the time of the impact, the original surface (line S1–T1–IP–T2–S2) and the impact point (IP) were located ~300 m above the present surface. (The intervening sediments were eroded after the impact and before deposition of the St. Peter Sandstone.) Excavation and downward displacement of the target rocks by the initial impact formed a deep, paraboloidal transient crater (Dence, 1968; Grieve et al., 1977; Melosh, 1989, Ch. 6) ~4 km in surface diameter and 1.3 km deep, bounded by the line T1–A1–T2 (short dashes). The upper part of the transient crater was excavated to a depth of ~600 m from the origi- nal surface, and an underlying displaced zone extended to a total depth of ~1.3 km from the surface, or deep into the crystalline basement. The point Ao, originally near the bottom of the upper excavated zone, was driven downward to point A1. (Continued on following page.) 30171 32856 31587 25312 53572 32842 52450 58908 1672 25454 60149 31572 H2 core & outcrop well numbersNW SE 00 10030 60 90 120 150 180 210 240 270 300 330 360 390 420 450 480 510 540 570 600 200 300 400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 1600 1700 1800 1900 2000 f tm 25000 0 0 10000 15000 200005000 1000 2000 3000 4000 5000 6000 7000 8000 feet meters C A A A 3 0 1 2 C E 2 C S T STIP 1 1 1 22 Wise Lake - Dunleith St. Peter shale eW li S hn akn e es h i Shakopee Oneota Jordan St. Lawrence - Lone Rock Eau Claire Mt. Simon Precambrian crystalline rocks Vertical Exaggeration = 10x datum = top St.Peter Decorah-Platteville-Glenwood carbonate clast breccia to conglomerate sandstone (breccia?) & shale Qal Wonewoc A A The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 21 (Figs. 3 and 18) does not indicate the presence of even a small central uplift, let alone one of this magnitude. It is unlikely that the absence of a central uplift at Decorah can be explained by post- impact erosion. The central uplifts of more deeply eroded impact structures are generally less prominent and display less stratigraphic uplift than their original maximum, but many complex structures similar to Decorah in size (diameters of 5–7  km) and erosional history preserve stratigraphic uplifts of at least 200– 300 m (e.g., Grieve and Pilkington, 1996; Kenk- mann et al., 2014, 2017). The nearby Rock Elm structure, for example, is similar to Decorah in diameter, target rocks, age, and deep erosional level, and still preserves a visible central strati- graphic uplift of at least 200–300 m (French et al., 2004, p. 204). Current studies of cratering mechanics sug- gest several possibilities, involving the nature of the impact event or the properties of the tar- get, that might explain the absence of a central uplift at Decorah: (a) an unusually low impact angle (<15º from the horizontal; see Pierazzo and Melosh, 2000; Stickle and Schultz, 2012); (b) a marine impact, involving a target of soft, possibly water-filled, sediments covered by a significant thickness of water (see Ormö and Lindström, 2000; Dypvik and Jansa, 2003; Davison and Collins, 2007; Darlington et al., 2016); and/or (c) immediate major erosion of the newly formed crater (and any original cen- tral uplift) by intense post-impact resurge cur- rents (see von Dalwigk and Ormö, 2001; Glims- dal et al., 2007). The existence (or absence) of a central uplift at Decorah is a major issue, not only for the fu- ture study of Decorah itself, but for our current understanding of the exact process by which a large number of similar structures have formed. Resolving this problem will probably require extensive local geophysical investigations in- volving gravity, magnetics, and active seismic methods, ideally supplemented by core drilling through the entire section of crater fill and into the disturbed subcrater rocks. Post-Impact Erosion of the Decorah Structure The currently preserved Decorah structure (see Figs.  3 and 18) shows virtually no relief under its cover of discomformable St. Peter Sandstone. In particular, there is no indication of a preserved uplifted crater rim or of a thick layer of ejecta surrounding the original crater (Melosh, 1989, Ch. 8; Kenkmann et al., 2013). As a result, neither the thickness of post-Sha- kopee, pre-impact sediments, nor the vertical height of the impact point above the present surface (which are approximately the same) can be closely estimated, but the information about the present state of the structure can be used to make estimates about the amount of post-impact erosion of the crater and its surroundings. Current cratering models (Melosh, 1989, Ch. 8; Kenkmann et al., 2013) suggest that removal of the original crater rim and ejecta layer re- quired the erosion of a layer ~300–500 m thick before the eroded structure was covered by the St. Peter Sandstone. By comparison, at the nearby Rock Elm structure there are indications that >300 m of now-eroded sediments younger than the Prairie du Chien group were present at the time of impact and were subsequently eroded (French et al., 2004, p. 214, and refer- ences therein). However, in the Decorah area, such younger pre-impact sediments must have been <300 m thick, or the remaining part of the present bowl-shaped impact structure would not have been preserved. This deep post-impact ero- sion of the Decorah structure, which represents a hiatus of perhaps 15–20 m.y. (Bunker et al., 1988), developed the major unconformity that now separates the youngest preserved pre-im- pact rocks (Shakopee Formation) from the old- est post-impact rocks (St. Peter Sandstone). CONCLUSIONS 1. A circular, largely subsurface, basin-shaped feature ~5.6  km in diameter, which displays anomalous geology, has been identified near Decorah, in northeastern Iowa. The surround- ing regional geology consists of a uniform sec- tion, several hundred m thick, of virtually unde- formed cratonic sediments (chiefly sandstones and carbonates) ranging in age from Upper Cambrian to Upper Ordovician and overlying a basement of Mesoproterozoic crystalline rocks. 2. This circular feature (designated here the “Decorah impact structure”) is expressed at the surface by the presence of an unusual shale unit (the Winneshiek Shale) that is restricted to the circular area and has not been found elsewhere in the region. This unit contains a striking La- gerstätte with a variety of well-preserved fossils (Liu et al., 2006, 2017). 3. Surface exposures of the Decorah struc- ture are few, and its subsurface structure is in- adequately defined, chiefly from examination of water-well drill-hole cuttings, two short cores, and two small surface exposures. The data avail- able outline an apparently basin-shaped fea- ture that extends from the surface to depths of ~200–300 m, truncating units from the Lower Ordovician Shakopee Formation to the Upper Cambrian St. Lawrence–Lone Rock–Wonewoc formations. The basin and its sedimentary fill are overlain disconformably by the Middle Or- dovician St. Peter Sandstone. The structure ap- parently does not penetrate the underlying Me- soproterozoic crystalline basement rocks. 4. The Decorah structure shows several char- acteristics that are consistent with formation by meteorite impact: a generally circular outline, a shallow-basin shape, anomalous cross-cutting relations to geological units outside the struc- ture, and a filling of anomalous sediments not present in the regional stratigraphy. However, it lacks evidence for such typical impact features as an uplifted crater rim and a layer of ejecta surrounding the structure. Figure  18 (continued). In the immediately subsequent modification stage of crater for- mation, the lower part of the transient cavity rebounded upwards, carrying point A1 up to A2 and forming a typical central uplift (Kenk- mann et al., 2013), while the peripheral parts of the transient crater collapsed inward, enlarging the crater diameter from ~4  km to ~6  km. Deposition of the crater-fill sedi- ments into the resulting basin began at this time and continued for a long period after structural movements of the crater itself had ceased. Line C1-C2 marks the upper bound- ary of the crater-fill sediments below the St. Peter Sandstone. (B) Plan-view map showing ground locations of wells whose stratigraphic results have been projected onto a NW-SE– trending vertical plane to produce the cross section shown in Figure 18A. Circle approxi- mates the present preserved exposure of the crater-filling Winneshiek Shale and provides an estimate for the current diameter of the present preserved structure, ~5.6 km. Cross indicates the inferred center of the structure. 91º46´19´´W N 43º18´49´´N IGS well & number inferred center of impact structure H2 core & outcrop 30171 32856 31587 25312 53572 32842 52450 58908 H2 1672 25454 60149 31572 0 1 2 3 4 5 6km B French et al. 22 Geological Society of America Bulletin, v. 130, no. XX/XX 5. Studies of available drill cuttings and drill core samples indicate that the basin of the Deco- rah structure is completely filled by a series of sediments, established breccias, and inferred breccias ~200 m thick that are not observed in the sedimentary sequence outside the structure. Several stratigraphically distinct units can be distinguished on the basis of geophysical stud- ies and drill hole cuttings. One upper unit (thick- ness ~100 m?) is an unusual, poorly sorted, pos- sibly sedimentary, polymict crater-fill breccia composed of fragments of target rock litholo- gies penetrated by the structure. This unit is overlain by the thinner (≤27  m), fine-grained, and finely laminated Winneshiek Shale. Avail- able data suggest that additional lithologies, and a more complex stratigraphy, may also be pres- ent in these basin-fill materials. 6. Two samples of the unusual crater-fill brec- cia, examined by petrographic and petrofabric (Universal Stage) methods, contain significant amounts of single rounded quartz grains, almost certainly derived from underlying Cambrian quartz arenites. A small percentage of these grains (≤1%) display multiple sets of parallel planar microstructures of identical appearance and crystallographic orientation (relative to the quartz c-axis) to shock-produced features observed in rocks from established meteorite impact structures: planar fractures (cleavage) (“P1 features”) and planar deformation features (PDFs) (“P2 features”). Individual grains typi- cally display as many as 1–5 sets of PFs and 2–4 sets of PDFs, both of which are oriented par- allel to specific crystallographic planes in the host quartz, most commonly (0001), {1122}, and {1011} (for PFs) and {1011}, {1013}, and (0001) (for PDFs). 7. Numerous studies by other workers have shown that these planar microstructures are unique and diagnostic shock-metamorphic features, produced by impact-generated shock waves with pressures in the range ≥5–10  GPa (for PFs) and 10–~20  GPa (for PDFs). They provide convincing evidence that the Decorah structure is a meteorite impact structure. No im- pact-related products of higher-pressure shock waves (e.g., shock-produced glasses, impact melts) have yet been identified. The identifi- cation of the Decorah structure as a meteorite impact feature demonstrates again the potential of petrographic and petrofabric studies, even on small or geologically limited samples, to es- tablish the impact origin of structures that are deeply eroded or largely inaccessible. 8. On the basis of its current diameter, we suggest that the original Decorah structure was a complex impact crater with an original diam- eter of ~6 km. However, unlike virtually all im- pact structures of comparable size, the Decorah structure currently shows no evidence for the presence of a structural central uplift, an anom- aly which probably cannot be resolved without more geological and geophysical studies. 9. Based on impact cratering models, we estimate that the original impact point was located in younger, now-eroded sediments ~300–500 m above the present surface. A simi- lar amount of erosion has therefore occurred post-impact, removing any original upraised crater rim or any ejecta layer that surrounded the original structure. This period of deep erosion, which may have lasted as long as 10–20  m.y., ended with the disconformable deposition of the St. Peter Sandstone over the structure and the surrounding region. 10. The age of the Decorah structure has been estimated from a combination of biostratigraphic and radiometric ages as 460–480 Ma, and a re- cent chemostratigraphic study of δ13Corg (Berg- ström et al., 2018) provided a narrower estimate of 464–467 Ma. We therefore suggest, with Bergström et al. (2018), that the Decorah struc- ture may be another member of a growing group of impact structures that apparently represent a “spike” of increased delivery of large and small extraterrestrial bodies to earth during the Middle Ordovician (Schmitz et al., 2001, 2008; Alw- mark et al., 2012) following the breakup of the L-chondrite meteorite parent body in the Aster- oid Belt at 470 Ma (Korochantseva et al., 2007). We suggest that this hypothesis may be tested by obtaining more precise age dates for the Decorah structure (e.g., by the recovery and analysis of shocked zircons [Cavosie et al., 2010]) or by the recovery of diagnostic extraterrestrial chromite grains from the basin-filling units (Alwmark and Schmitz, 2007; Alwmark et al., 2012). 11. Despite the confident establishment of the origin of the Decorah impact structure, major questions about its characteristics, forma tion, and history remain. These include: the nature of the subsurface stratigraphy and structure; the presence or absence of an ex- pected central uplift; the precise age of the impact event; the post-impact environment in which the crater-fill breccia and Winneshiek Shale were deposited; the elapsed time (if any) between the end of deposition of the crater- fill breccia and the start of deposition of the Winne shiek Shale; the relative roles of impact processes and post-impact preservation in cre- ating the large and unusual Winneshiek Lager- stätte now preserved within the structure; pos- sible connections between the Decorah event and other mid-Ordovician impact events, and the subsequent geological and biological his- tory of the structure and its surroundings. 12. The information now available about the Decorah structure provides a solid base from which to plan and execute more sophisticated investigations. The highest priorities for fur- ther studies involve the detailed exploration of the structure, its deformed rocks, and its crater- filling units to determine precisely: the present limits of the structure (e.g., diameter, depth, na- ture of the crater floor); the structural deforma- tion of the deformed target rocks in and around the crater; the lithologies and stratigraphy of the crater-fill materials; and the recognition of other impact-produced lithologies possibly pre- served in the structure. Special attention should be given to the Winneshiek Shale: its stratigra- phy, its depositional environment, its relations to other impact-produced lithologies in the struc- ture, and the timing of its deposition relative to formation of the structure. These explorations will require cooperative multidisciplinary inves- tigations involving the detailed examination of available cuttings and core samples, further core drilling through the crater fill and into the sub- crater rocks, and a range of geophysical studies, especially using gravity, magnetic, and active seismic methods. 13. Further studies of the Decorah structure have the potential to provide specific informa- tion on major questions of meteorite impact mechanics, impact crater formation, the effects of impact on the geological environment, and the geological history of the surrounding re- gion. Although many impact structures of com- parable size have already been identified, the Decorah structure is particularly important as an unusual example of a possible complex im- pact structure apparently formed and enclosed entirely in layered sedimentary target rocks, and further studies of Decorah will help illuminate the specific details of such structures and of how they form. In addition, the presence of a striking fossil Lagerstätte within the crater provides an opportunity to study, on a small scale and over a local region, the possible relations between a meteorite impact event and the preservation of local fauna. Finally, further study of the Deco- rah structure will contribute to understanding its possible relationship to the proposed “spike” of meteorite impact events in the Middle Ordovi- cian and will help illuminate the more general connections between the process of meteorite impact and the geological and biological history of the Earth. ACKNOWLEDGEMENTS BMF thanks the Department of Paleobiology, Smithsonian Institution, for support as a Research Associate during this study. We thank Sarah Pennis- ton-Dorland (Department of Geology, University of Maryland), Tim Gooding (Department of Mineral Sciences, Smithsonian Institution), and Finnegan Marsh (Department of Paleobiology, Smithsonian Institution) for assistance in producing the photo- The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 23 micrographs. RMM and HPL thank Steve and Jane Hildebrand, landowners, for access to outcrop and drill site. Permissions were also provided by the Oneota Country Club, the City of Decorah, Winne- shiek County, and Bruening Rock Products. Jean Young (deceased, 2007), friend, Decorah area resi- dent and geologist, was first to recognize unusual strata beneath the St. Peter Formation near Decorah. Phil Kerr (Iowa Geological Survey) provided graph- ics assistance, Matt Wortel (Department of Earth and Environmental Science, University of Iowa) prepared thin sections, Tony Runkel (Minnesota Geological Survey) provided Minnesota well records, and Ray Anderson discussed impacts. The U.S. Geological Survey Statemap Program supported geologic map- ping in Winneshiek County. We also thank David King and Jeffrey Plescia for careful and detailed re- views which significantly improved the manuscript, and Christian Koeberl for editorial handling. Funding for this research was provided by the Iowa Depart- ment of Natural Resources and by National Science Foundation grants EAR 0921245 (to HPL, RMM and BJW) and EAR 0922054 (to DEGB). REFERENCES CITED Alexopoulos, J.S., Grieve, R.A.F., and Robertson, P.B., 1988, Microscopic lamellar features in quartz: Dis- criminative characteristics of shock-generated varie- ties: Geology, v.  16, p.  796–799, https:// doi .org /10 .1130 /0091 -7613 (1988)016 <0796: MLDFIQ>2 .3 .CO;2. Alwmark, C., and Schmitz, B., 2007, Extraterrestrial chro- mite in the resurge deposits of the early Late Or- dovician Lockne crater, central Sweden: Earth and Planetary Science Letters, v. 253, p. 291–303, https:// doi .org /10 .1016 /j .epsl .2006 .10 .034. Alwmark, C., Schmitz, B., Meier, M.M.M., Baur, H., and Wieler, R., 2012, A global rain of micrometeorites following breakup of the L-chondrite parent body— Evidence from solar wind-implanted Ne in fossil ex- traterrestrial chromite grains from China: Meteoritics & Planetary Science, v. 47, p. 1297–1304, https:// doi .org /10 .1111 /j .1945 -5100 .2012 .01394 .x. Bergström, S.M., Schmitz, B., Liu, H.P., Terfelt, F., and McKay, R.M., 2018, High-resolution δ13Corg che- mostratigraphy links the Decorah impact structure and the Winneshiek Konservat-Lagerstätte to the Darriwil- ian global peak influx of meteorites: Lethaia, v.  51, https:// doi .org /10 .1111 /let .12269. Briggs, D.E.G., Liu, H., McKay, R.M., and Witzke, B.J., 2016, Bivalved arthropods from the Middle Ordovi- cian Winneshiek Lagerstätte, Iowa, USA: Journal of Paleontology, v. 89 (for 2015), p. 991–1006, https:// doi .org /10 .1017 /jpa .2015 .76. Bunker, B.J., Witzke, B.J., Watney, W.L., and Ludvigson, G.A., 1988, Phanerozoic history of the central midcon- tinent, United States, in Sloss, L.L., ed., Sedimentary Cover—North American Craton, U.S.: Boulder, Colo- rado, Geological Society of America, The Geology of North America, v. D-2, p. 243–260, https:// doi .org /10 .1130 /DNAG -GNA -D2 .243. Cavosie, A.J., Quintero, R.R., Radovan, H.A., and Moser, D.E., 2010, A record of ancient cataclysm in modern sand: Shock microstructures in detrital minerals from the Vaal River, Vredefort Dome, South Africa: Geolog- ical Society of America Bulletin, v. 122, p. 1968–1980, https:// doi .org /10 .1130 /B30187 .1. Chetel, L.M., Singer, B.S., and Simo, T., 2004, 40Ar/39Ar geochronology of the Upper Mississippi Valley, Mid- dle and Upper Ordovician Galena Group; sediment accumulation rates and biostratigraphic implications for the history of an epeiric sea: Geological Society of America Abstracts with Programs, v. 36, no. 5, p. 75. Chetel, L.M., Singer, B.S., and Simo, T., 2005, 40Ar/39Ar geochronology of the Upper Mississippi Valley, Mid- dle and Upper Ordovician Galena Group; sediment ac- cumulation rates and biostratigraphic implications for the history of an epeiric sea, in Ludvigson, G.A., and Bunker, B.J., eds., Facets of the Ordovician Geology of the Upper Mississippi Valley Region: Iowa Geological Survey, Guidebook Series, no. 24, p. 55–57. Cohen, K.M., Finney, S.C., Gibbard, P.L., and Fan, J.X., 2013, The International Commission on Stratigraphy International Chronostratigraphic Chart: Episodes, v.  36, p.  199–204 http://www.stratigraphy.org/IC- Schart/ChronostratChart2017-02.pdf (updated version; accessed March 2018). Darlington, V., Blenkinsop, T., Dirks, P., Salisbury, J., and Tomkins, A., 2016, The Lawn Hill annulus: An Ordo- vician meteorite impact into water-saturated dolomite: Meteoritics & Planetary Science, v. 51, p. 2416–2440, https:// doi .org /10 .1111 /maps .12734. Davison, T., and Collins, G.S., 2007, The effect of the oceans on terrestrial crater size-frequency distribution: Insight from numerical modeling: Meteoritics & Plan- etary Science, v. 42, p. 1915–1927, https:// doi .org /10 .1111 /j .1945 -5100 .2007 .tb00550 .x. Dence, M.R., 1965, The extraterrestrial origin of Canadian craters: Annals of the New York Academy of Sciences, v.  123, p.  941–969, https:// doi .org /10 .1111 /j .1749 -6632 .1965 .tb20411 .x. Dence, M.R., 1968, Shock zoning at Canadian craters: Pe- trography and structural implications: in French, B.M., and Short, N.M. eds., Shock metamorphism of natural materials: Baltimore, Maryland, Mono Book Corpora- tion, p. 169–184. Dence, M.R., Innes, M.J.S., and Robertson, P.B., 1968, Re- cent geological and geophysical studies of Canadian craters, in French, B.M., and Short, N.M., eds., Shock metamorphism of natural materials: Baltimore, Mary- land, Mono Book Corporation, p. 339–362. Drenth, B.J., Anderson, R.R., Schulz, K.J., Feinberg, J.M., Chandler, V.W., and Cannon, W.F., 2015, What lies beneath; geophysical mapping of a concealed Pre- cambrian intrusive complex along the Iowa-Minnesota border: Canadian Journal of Earth Sciences, v.  52, no. 5, p. 279–293. Dypvik, H., and Jansa, L.F., 2003, Sedimentary signatures and processes during marine bolide impacts: a review: Sedimentary Geology, v. 161, p. 309–337, https:// doi .org /10 .1016 /S0037 -0738 (03)00135 -0. Earth Impact Database, 2016, University of New Brunswick, Fredericton, New Brunswick, Canada: http://www. passc.net/EarthImpactDatabase/index.html (accessed December 2016). Engelhardt, W.v., and Bertsch, W., 1969, Shock induced planar deformation structures in quartz from the Ries crater, Germany: Contributions to Mineralogy and Petrology, v.  20, p.  203–234, https:// doi .org /10 .1007 / BF00377477. Ferrière, L., Morrow, J.R., Amgaa, T., and Koeberl, C., 2009, Systematic study of universal-stage measurements of planar deformation features in shocked quartz: Impli- cations for statistical significance and representation of results: Meteoritics & Planetary Science, v.  44, p.  925–940, https:// doi .org /10 .1111 /j .1945 -5100 .2009 .tb00778 .x. French, B.M., 1998, Traces of catastrophe: a handbook of shock-metamorphic effects in terrestrial meteorite impact craters: Houston, Texas, Lunar and Planetary Institute, Contribution CB-954, 120 p, (http://www.lpi. usra.edu/publications/books/). French, B.M., 2004, The importance of being cratered: The new role of meteorite impact as a normal geologi- cal process: Meteoritics & Planetary Science, v.  39, p.  169–197, https:// doi .org /10 .1111 /j .1945 -5100 .2004 .tb00335 .x. French, B.M., and Cordua, W.C., 1999, Intense fracturing of quartz at the Rock Elm, Wisconsin “cryptoexplosion” structure: Evidence for meteorite impact [abs.]: Lunar and Planetary Science Conference, 30th, Houston, Texas, Abstracts, CD-ROM, no. 1123. French, B.M., and Koeberl, C., 2010, The convincing iden- tification of meteorite impact structures: What works, what doesn’t, and why: Earth-Science Reviews, v. 98, p.  123–170, https:// doi .org /10 .1016 /j .earscirev .2009 .10.009. French, B.M., and Short, N.M., eds., 1968, Shock meta- morphism of natural materials: Baltimore, Maryland, Mono Book Corp., 644 p. French, B.M., Underwood, J.R., Jr., and Fisk, E.P., 1974, Shock-metamorphic features in two meteorite impact structures, southeastern Libya: Geological Society of America Bulletin, v.  85, p.  1425–1428, https:// doi .org /10 .1130 /0016 -7606 (1974)85 <1425: SFITMI>2 .0 .CO;2. French, B.M., Cordua, W., and Plescia, J.B., 2004, The Rock Elm meteorite impact structure, Wisconsin: Geology and shock-metamorphic effects in quartz: Geologi- cal Society of America Bulletin, v.  116, p.  200–218, https:// doi .org /10 .1130 /B25207 .1. Glimsdal, S., Pedersen, G.K., Langtangen, H.P., Shuvalov, V., and Dypvik, H., 2007, Tsunami generation and propagation from the Mjølnir asteroid impact: Meteor- itics & Planetary Science, v. 42, p. 1473–1493, https:// doi .org /10 .1111 /j .1945 -5100 .2007 .tb00586 .x. Grieve, R.A.F., 1987, Terrestrial impact structures: Annual Re- view of Earth and Planetary Sciences, v. 15, p. 245–270, https:// doi .org /10 .1146 /annurev .ea .15 .050187 .001333. Grieve, R.A.F., 1991, Terrestrial impact: the record in the rocks: Meteoritics, v. 26, p. 175–194, https:// doi .org /10 .1111 /j .1945 -5100 .1991 .tb01038 .x. Grieve, R.A.F., 1997, Extraterrestrial impact events: The record in the rocks and the stratigraphic column: Pa- laeogeography, Palaeoclimatology, Palaeoecology, v.  132, p.  5–23, https:// doi .org /10 .1016 /S0031 -0182 (97)00058 -8. Grieve, R.A.F., 1998, Extraterrestrial impacts on the earth: the evidence and the consequences, in Grady, M.M., Hutchinson, R., McCall, G.J.H., and Rothery, D., eds., Meteorites: Flux with time and impact effects: London, Geological Society Special Publication 140, p. 105– 131, https:// doi .org /10 .1144 /GSL .SP .1998 .140 .01 .10. Grieve, R.A.F., 2001, The terrestrial cratering record, in Peucker- Ehrenbrink, B., and Schmitz, B., eds., Accretion of extraterrestrial matter throughout Earth’s history: New York, Kluwer Academic/Plenum Publishers, p. 379–402, https:// doi .org /10 .1007 /978 -1 -4419 -8694 -8_19. Grieve, R.A.F., and Pesonen, L.J., 1992, The terrestrial im- pact cratering record: Tectonophysics, v. 216, p. 1–30, https:// doi .org /10 .1016 /0040 -1951 (92)90152 -V. Grieve, R.A.F., and Pilkington, M., 1996, The signature of terrestrial impacts: AGSO Journal of Australian Geol- ogy & Geophysics, v. 16, p. 399–420. Grieve, R.A.F., Dence, M.R., and Robertson, P.B., 1977, Cratering process: As interpreted from the occurrence of impact melts, in Roddy, D.J., Pepin, R.O., and Mer- rill, R.B., eds., Impact and explosion cratering: New York, Pergamon Press, p. 791–814. Grieve, R.A.F., Langenhorst, F., and Stöffler, D., 1996, Shock metamorphism of quartz in nature and experi- ment: II. Significance in geosciences: Meteoritics & Planetary Science, v.  31, p.  6–35, https:// doi .org /10 .1111 /j .1945 -5100 .1996 .tb02049 .x. Hawkins, A.D., Liu, H.P., Briggs, D.E.G., Muscente, A.D., McKay, R.M., Witzke, B.J., and Xiao, S., 2018, Ta- phonomy and biological affinity of three-dimensional phosphatized bromalites from the Middle Ordovician Winneshiek Lagerstätte, northeastern Iowa, USA: Palaios, v.  33, p.  1–15, https:// doi .org /10 .2110 /palo .2017 .053. Hergarten, S., and Kenkmann, T., 2015, The number of impact craters on Earth: Any room for further discov- eries?: Earth and Planetary Science Letters, v.  425, p. 187–192, https:// doi .org /10 .1016 /j .epsl .2015 .06 .009. Huber, M.S., Ferrière, L., Losiak, A., and Koeberl, C., 2011, ANIE: A mathematical algorithm for automated index- ing of Planar Deformation Features in shocked quartz: Meteoritics & Planetary Science, v. 46, p. 1418–1424, https:// doi .org /10 .1111 /j .1945 -5100 .2011 .01234 .x. Jourdan, F., and Reimold, U.W., eds., 2012, Impact!: Ele- ments, Special issue, v. 8, no. 1, 60 p. Kass, M.A., Bedrosian, P.A., Drenth, B.J., Bloss, B.R., McKay, R.M., Liu, H., French, B.M., and Witzke, B.J., 2013a, Modeling and inversion results from airborne geophysics over a buried impact structure in Decorah, Iowa, USA: Geological Society of America Abstracts with Programs, v. 45, no. 7, p. 485. Kass, M.A., Bedrosian, P.A., Drenth, B.J., Bloss, B.R., McKay, R.M., Liu, H., French, B.M., and Witzke, B.J., 2013b, Geophysical signatures and modeling results from a buried impact structure in Decorah, Iowa, USA: French et al. 24 Geological Society of America Bulletin, v. 130, no. XX/XX American Geophysical Union Fall Meeting (December 2013), Abstract P34C-04. Kenkmann, T., Collins, G.S., and Wünnemann, K., 2013, The modification stage of crater formation, in Osinski, G.R. and Pierazzo, E., eds., Impact Cratering: Processes and Products: Chichester, UK, Wiley-Blackwell, p. 60–75. Kenkmann, T., Poelchau, M.H., and Wulf, G., 2014, Struc- tural geology of impact craters: Journal of Structural Geology, v.  62, p.  156–182, https:// doi .org /10 .1016 /j .jsg .2014 .01 .015. Kenkmann, T., Sturm, S., Krüger, T., and Salameh, E., 2017, The structural inventory of a small impact crater: Jebel Waqf as Suwwan, Jordan: Meteoritics & Planetary Science, v.  52, p.  1351–1370, https:// doi .org /10 .1111 /maps .12823. King, D.T., Jr., Ormö, J., Petruny, L.W., and Neathery, T.L., 2006, Role of water in the formation of the Late Creta- ceous Wetumpka impact structure, inner Gulf Coastal Plain of Alabama, USA: Meteoritics & Planetary Sci- ence, v.  41, p.  1625–1631, https:// doi .org /10 .1111 /j .1945 -5100 .2006 .tb00440 .x. Koeberl, C., 2014, The geochemistry and cosmochemistry of impacts, in Treatise on geochemistry, 2nd ed.: El- sevier, Reference Module in Earth Systems and Envi- ronmental Sciences, p. 73–118, https:// doi .org /10 .1016 /B978 -0 -08 -095975 -7 .00130 -3. Kolata, D.R., Huff, W.D., and Bergström, S.M., 1996, Or- dovician K-bentonites of eastern North America: Geo- logical Society of America Special Paper 313, 84 p., https:// doi .org /10 .1130 /SPE313. Korochantseva, E.K., Trieloff, M., Lorenz, C.A., Buykin, A.L., Ivanova, M.A., Schwarz, W.H., Hopp, J., and Jessberger, E., 2007, L-chondrite asteroid breakup tied to Ordovician meteorite shower by multiple isochron 40Ar-39Ar dating: Meteoritics & Planetary Science, v. 42, p. 113–130, https:// doi .org /10 .1111 /j .1945 -5100 .2007 .tb00221 .x. Lamsdell, J.C., Briggs, D.E.G., Liu, H.P., Witzke, B.J., and McKay, R.M., 2015a, A new Ordovician arthro- pod from the Winneshiek Lagerstätte of Iowa (USA) reveals the ground plan of eurypterids and chasmatas- pidids: The Science of Nature, v. 102, no. 63, p. 1–8. Lamsdell, J.C., Briggs, D.E.G., Liu, H.P., Witzke, B.J., and McKay, R.M., 2015b, The oldest described eurypterid: a giant Middle Ordovician (Darriwilian) megalograptid from the Winneshiek Lagerstätte of Iowa: BMC Evolu- tionary Biology, v. 15, no. 169, p. 1–31. Lindström, M., Ormö, J., Sturkell, E., and von Dalwigk, I., 2005, The Lockne Crater: Revision and reassessment of structure and impact stratigraphy, in Koeberl, C. and Henkel, H. eds., Impact Tectonics: New York, Springer Publishers, p. 357–388, https:// doi .org /10 .1007 /3 -540 -27548 -7_14. Liu, H.P., McKay, R.M., Young, J.N., Witzke, B.J., McVey, K.J., and Liu, X., 2006, A new Lagerstätte from the Middle Ordovician St. Peter Formation in northeast Iowa, USA: Geology, v. 34, p. 969–972, https:// doi .org /10 .1130 /G22911A .1. Liu, H.P., McKay, R.M., Young, J.N., Witzke, B.J., McVey, K.J., and Liu, X., 2007a, The Winneshiek Lager- stätte: Acta Palaeontologica Sinica, v. 46, supplement, p. 282–285. Liu, H.P., Witzke, B.J., Young, J.N., and McKay, R.M., 2007b, Conodonts from the Winneshiek Lagerstätte, St. Peter Sandstone (Ordovician) of northeast Iowa: Geological Society of America Abstracts with Pro- grams, v. 39, no. 9, p. 63. Liu, H.P., McKay, R.M., Witzke, B.J., and Briggs, D.E.G., 2009, The Winneshiek Lagerstätte, Iowa, USA and its depositional environments: Geological Journal of China Universities, v. 15, p. 285–295 (in Chinese with English summary). Liu, H., Briggs, D.E.G., McKay, R.M., and Witzke, B.J., 2013, The Middle Ordovician Winneshiek Lager- stätte—An unusual setting for exceptional preserva- tion: Geological Society of America Abstracts with Programs, v. 45, no. 7, p. 454. Liu, H.P., Bergström, S.M., Witzke, B.J., Briggs, D.E.G., McKay, R.M., and Ferretti, A., 2017, Exceptionally preserved conodont apparatuses with giant jaw-like elements from the Middle Ordovician Winneshiek Konservat-Lagerstätte, Iowa, USA: Journal of Paleon- tology, v.  91, p.  493–511, https:// doi .org /10 .1017 /jpa .2016 .155. Losiak, A., Golebiowska, I., Ferrière, L., Wojciechowski, J., Huber, M.S., and Koeberl, C., 2016, WIP: A Web- based program for indexing planar features in quartz grains and its usage: Meteoritics & Planetary Sci- ence, v. 51, p. 647–662, https:// doi .org /10 .1111 /maps .12614. Lowman, P.D., 2002, Exploring space, exploring Earth: New understanding of the Earth from space research: New York, Cambridge University Press, 362 p. Ludvigson, G.A., and Bunker, B.J., eds., 2005, Facets of the Ordovician Geology of the Upper Mississippi Valley Region: Iowa Geological Survey, Guidebook Series, no. 24, 129 p. McKay, R.M., 1988, Stratigraphy and lithofacies of the Dresbachian (Upper Cambrian) Eau Claire Formation in the subsurface of eastern Iowa, in Ludvigson, G.A., and Bunker, B.J., eds., New perspectives on the Pa- leozoic history of the Upper Mississippi Valley: Iowa Geological Survey, Guidebook, no. 8, a guidebook for the 18th Annual Field Conference of the Great Lakes Section, Society of Economic Paleontologists and Mineralogists, p. 33–53. McKay, R.M., 1993, Selected aspects of Lower Ordovi- cian and Upper Cambrian geology in Allamakee and northern Clayton counties: Geological Society of Iowa, Guidebook 57, 61 p. McKay, R.M., Liu, H., Witzke, B.J., and French, B.M., 2010, Geologic setting of the Winneshiek Lager- stätte—Decorah, Iowa: Geological Society of America Abstracts with Programs, v. 42, no. 2, p. 89. McKay, R.M., Liu, H., Witzke, B.J., French, B.M., and Briggs, D.E.G., 2011, Preservation of the Middle Or- dovician Winneshiek Shale in a probable impact crater: Geological Society of America Abstracts with Pro- grams, v. 43, no. 5, p. 189. Melosh, H.J., 1989, Impact cratering: A geological process: Oxford, U.K., Oxford University Press, 245 p. Melosh, H.J., and Ivanov, B.A., 1999, Impact crater col- lapse: Annual Review of Earth and Planetary Sciences, v.  27, p.  385–415, https:// doi .org /10 .1146 /annurev .earth .27 .1 .385. Montanari, A., and Koeberl, C., 2000, Impact stratigraphy: the Italian record: Notes in Earth Sciences, v. 93, New York, Springer Publishers, 364 p. Nowak, H., Harvey, T.H.P., Liu, H.P., McKay, R.M., Zippi, P.A., Campbell, D.H., and Servais, T., 2017, Filamen- tous eukaryotic algae with a possible cladophoralean affinity from the Middle Ordovician Winneshiek La- gerstätte in Iowa, USA: Geobios, v.  50, p.  303–309, https:// doi .org /10 .1016 /j .geobios .2017 .06 .005. Nowak, H., Harvey, T.H.P., Liu, H.P., McKay, R.M., and Servais, T., 2018, Exceptionally preserved arthropodan microfossils from the Middle Ordovician Winneshiek Lagerstätte, Iowa, USA: Lethaia, v.  51, p.  267–276, https:// doi .org /10 .1111 /let .12236. Odom, I.E., 1975, Feldspar-grain size relations in Cambrian arenites, upper Mississippi Valley: Journal of Sedimen- tary Petrology, v. 45, p. 636–650. Ormö, J., and Lindström, M., 2000, When a cosmic impact strikes the sea bed: Geological Magazine, v. 137, p. 67– 80, https:// doi .org /10 .1017 /S0016756800003538. Osinski, G.R., and Pierazzo, E., eds., 2013, Impact crater- ing: Processes and products: Chichester, UK, Wiley- Blackwell, 316 p. Pierazzo, E., and Melosh, H.J., 2000, Understanding oblique impacts from experiments, observations, and model- ing: Annual Review of Earth and Planetary Sciences, v.  28, p.  141–167, https:// doi .org /10 .1146 /annurev .earth .28 .1 .141. Poelchau, M.H., and Kenkmann, T., 2011, Feather-features: A low-shock-pressure indicator in quartz: Journal of Geophysical Research, v. 116, B02201, 13 p. https:// doi .org /10 .1029 /2010JB007803. Prior, J.C., 1991, Landforms of Iowa: Iowa City, Iowa, Uni- versity of Iowa Press, 154 p. Robertson, P.B., 1975, Zones of shock metamorphism at the Charlevoix impact structure, Quebec: Geologi- cal Society of America Bulletin, v. 86, p. 1630–1638, https:// doi .org /10 .1130 /0016 -7606 (1975)86 <1630: ZOSMAT>2 .0 .CO;2. Robertson, P.B., and Grieve, R.A.F., 1977, Shock attenua- tion in terrestrial impact structures, in Roddy, D.J., Pe- pin, R.O., and Merrill, R.B., eds., Impact and explosion cratering: Planetary and terrestrial implications: New York, Pergamon, p. 687–702. Robertson, P.B., Dence, M.F., and Vos, M.A., 1968, De- formation in rock-forming minerals from Canadian craters, in French, B.M., and Short, N.M., eds., Shock metamorphism of natural materials: Baltimore, Mary- land, Mono Book Corporation, p. 433–452. Runkel, A.C., 1996, Bedrock geology of Houston County, Minnesota: Minnesota Geological Survey Open-File Report 96-4, 3 pls., scale 1:100,000; text, 13 p. Runkel, A.C., McKay, R.M., and Palmer, A.R., 1998, Ori- gin of a classic cratonic sheet sandstone; stratigraphy across the Sauk II–Sauk III boundary in the Upper Mis- sissippi Valley: Geological Society of America Bulle- tin, v.  110, p.  188–210, https:// doi .org /10 .1130 /0016 -7606 (1998)110 <0188: OOACCS>2 .3 .CO;2. Runkel, A.C., Miller, J.F., McKay, R.M., Palmer, A.R., and Taylor, J.F., 2007, High-resolution sequence stratigraphy of lower Paleozoic sheet sandstones in central North America: The role of special condi- tions of cratonic interiors in development of stratal architecture: Geological Society of America Bul- letin, v.  119, p.  860–881, https:// doi .org /10 .1130 / B26117 .1. Runkel, A.C., Miller, J.F., McKay, R.M., Palmer, A.R., and Taylor, J.F., 2008, The record of time in cratonic interior strata: does exceptionally slow subsidence necessarily result in exceptionally poor stratigraphic completeness? in Pratt, B.R., and Holmden, C., eds., Dynamics of Epeiric Seas: Geological Association of Canada, Special Paper 48, p. 341–362. Schmitz, B., Tassanari, M., and Peucker-Ehrenbrink, B., 2001, A rain of ordinary chondrite meteorites in the Early Ordovician: Earth and Planetary Science Letters, v.  194, p.  1–15, https:// doi .org /10 .1016 /S0012 -821X (01) 00559 -3. Schmitz, B.S., Harper, D.A.T., Peucker-Ehrenbrink, B., Stouge, S., Alwmark, C., Cronholm, A., Bergström, S.M., Tassinari, M., and Xiaofeng, W., 2008, Asteroid breakup linked to the Great Ordovician Biodiversifica- tion Event: Nature Geoscience, v. 1, p. 49–52, https:// doi .org /10 .1038 /ngeo .2007 .37. Schulte, P., et al., 2010, The Chicxulub asteroid impact and mass extinction at the Cretaceous-Paleogene boundary: Science, v. 327, p. 1214–1218, https:// doi .org /10 .1126 /science .1177265. Smith, G.L., and Clark, D.L., 1996, Conodonts of the Lower Ordovician Prairie du Chien Group of Wisconsin and Minnesota: Micropaleontology, v.  42, no.  4, p.  363– 373, https:// doi .org /10 .2307 /1485958. Smith, M.E., Singer, B.S., and Simo, T., 2011, Tiered in- terpolation of radioisotopic and biostratigraphic geochronology resolves the tempo of the Ordovician icehouse to greenhouse transition: Geological Soci- ety of America Abstracts with Programs, v. 43, no. 5, p. 568. Spry, A., 1969, Metamorphic textures: New York, Pergamon Press, 350 p. Stewart, S.A., 2011, Estimates of yet-to-find impact crater population on Earth: Journal of the Geological So- ciety, v.  168, p.  1–14, https:// doi .org /10 .1144 /0016 -76492010 -006. Stickle, A., and Schultz, P.H., 2012, Oblique hypervel- ocity impacts into layered targets: Journal of Geo- physical Research, v.  117, no.  E7, https:// doi .org /10.10292011JE004043. Stöffler, D., and Langenhorst, F., 1994, Shock metamor- phism of quartz in nature and experiment: 1. Basic ob- servation and theory: Meteoritics & Planetary Science, v. 29, p. 155–181, https:// doi .org /10 .1111 /j .1945 -5100 .1994 .tb00670 .x. Tagle, R., and Hecht, L., 2006, Geochemical identification of projectiles in impact rocks: Meteoritics & Planetary Science, v. 41, p. 1721–1735, https:// doi .org /10 .1111 /j .1945 -5100 .2006 .tb00448 .x. Trefil, J.S., and Raup, D.M., 1990, Crater taphonomy and bombardment rates in the Phanerozoic: The Journal of Geology, v. 98, p. 385–398, https:// doi .org /10 .1086 /629411. The Decorah Structure, northeastern Iowa: Geology, formation by meteorite impact Geological Society of America Bulletin, v. 130, no. XX/XX 25 Vernon, R.H., 2004, A practical guide to rock microstruc- ture: New York, Cambridge University Press, 594 p, https:// doi .org /10 .1017 /CBO9780511807206. von Dalwigk, I., and Ormö, J., 2001, Formation of resurge gullies at impacts at sea: The Lockne crater, Sweden: Meteoritics & Planetary Science, v.  36, p.  359–369, https:// doi .org /10 .1111 /j .1945 -5100 .2001 .tb01879 .x. Whitehead, J., Spray, J.G., and Grieve, R.A.F., 2002, Origin of “toasted” quartz in terrestrial impact structures: Ge- ology, v. 30, p. 431–434, https:// doi .org /10 .1130 /0091 -7613 (2002)030 <0431: OOTQIT>2 .0 .CO;2. Witzke, B.J., and Glenister, B.F., 1987, The Ordovician sequence in the Guttenberg area, northeast Iowa, in Biggs, D.L., ed., North Central Section of the Geologi- cal Society of America: Boulder, Colorado, Geological Society of America, Geology of North America, Cen- tennial Field Guide, v. 3, p. 93–96, https:// doi .org /10 .1130 /0 -8137 -5403 -8 .93. Witzke, B.J., and McKay, R.M., 1987, Cambrian and Or- dovician stratigraphy in the Lansing area, northeastern Iowa, in Biggs, D.L., ed., North Central Section of the Geological Society of America: Boulder, Colorado, Geological Society of America, Geology of North America, Centennial Field Guide, v.  3, p. 81–87, https:// doi .org /10 .1130 /0 -8137 -5403 -8 .81. Witzke, B.J., McKay, R.M., Liu, H.P., and Briggs, D.E.G., 2011, The Middle Ordovician Winneshiek Shale of northeast Iowa—Correlation and paleogeographic im- plications: Geological Society of America Abstracts with Programs, v. 43, no. 5, p. 315. Wolter, C.F., McKay, R.M., Liu, H., Bounk, M.J., and Libra, R.D., 2011, Geologic Mapping For Water Quality Proj- ects in the Upper Iowa River Watershed: Iowa Geologi- cal Survey, Technical Information Series, no. 54, 34 p. Science editor: AAron J. cAvoSie ASSociAte editor: chriStiAn Koeberl MAnuScript received 7 SepteMber 2017 reviSed MAnuScript received 1 MAy 2018 MAnuScript Accepted 19 June 2018 Printed in the USA