Joints, Tensile Strength and Preferred Fracture Orientation in Sandstones

Document generated on 09/25/2021 1:34 a.m.

Joints, tensile strength and preferred fracture orientation in sandstones. New Brunswick and Prince Edward Island, Canada E.Z. Lajtai and P. Stringer

Volume 17, Number 2, Summer 1981 Article abstract The orientations of tensile fractures in Upper Devonian, Carboniferous and URI: https://id.erudit.org/iderudit/ageo17_2art01 Lower Permian sandstones of southern New Brunswick and Prince Edward Island show correlation at the submicroscopic scale of microcrack alignments, See table of contents the macroscopic scale of joints, and the megascopic scale of airphoto lineaments. Planes of minimum tensile strength and planes of preferred fracture, induced in visibly unfractured sandstone test-samples by line-loading Publisher(s) and point-loading tests, are identified as microcrack alignments. The joints and airphoto lineaments have microcrack alignments parallel to them, but not all Maritime Sediments Editorial Board microcrack alignments are represented at the macroscopic and megascopic scales. ISSN Deformed Carboniferous sandstones east of Saint John have two prominent 0843-5561 (print) orthogonal Joint systems with microcrack alignments parallel to the four joint 1718-7885 (digital) sets. The airphoto lineaments are parallel to only three joint sets. Southeast and northeast striking joints in one orthogonal system are tentatively interpreted as planes of extension and release respectively, related to the direction of Explore this journal compression during the Variscan-Appalachian orogeny. The other orthogonal system, with joint sets striking close to 010° and 100°, is interpreted as post-Triassic in age, possibly related to the presently acting crustal stresses in Cite this article eastern North America. Lajtai, E. & Stringer, P. (1981). Joints, tensile strength and preferred fracture orientation in sandstones. New Brunswick and Prince Edward Island, Canada. Atlantic Geology, 17(2), 70–87.

All rights reserved © Maritime Sediments, 1981 This document is protected by copyright law. Use of the services of Érudit (including reproduction) is subject to its terms and conditions, which can be viewed online. https://apropos.erudit.org/en/users/policy-on-use/

This article is disseminated and preserved by Érudit. Érudit is a non-profit inter-university consortium of the Université de Montréal, Université Laval, and the Université du Québec à Montréal. Its mission is to promote and disseminate research. https://www.erudit.org/en/ Maritime Sediments and Atlantic Geology VOL. 17 AUGUST 1981 NUMBER 2

Joints, tensile strength and preferred fracture orientation in sandstones, New Brunswick and Prince Edward Island, Canada E. Z. Lajtai, department of Geological Engineering University of Manitoba, Winnipeg, Manitoba, Canada RST 2N2 and P. Stringer, Department of Geology, University of New Brunswick P.O. Box 4400, Fredericton, New Brunswick, Canada ESB 5A3

The orientations of tensile fractures in Upper Devonian, Carboniferous and Lower Permian sandstones of southern New Brunswick and Prince Edward Island show correlation at the submicroscopic scale of microcrack alignments, the macroscopic scale of joints, and the megascopic scale of airphoto lineaments. Planes of minimum tensile strength and planes of preferred fracture, induced in visibly unfractured sandstone test-samples by line-loading and point-loading tests, are identified as microcrack alignments. The joints and airphoto lineaments have microcrack alignments parallel to them, but not all microcrack alignments are represented at the macroscopic and megascopic scales. Deformed Carboniferous sandstones east of Saint John have two prominent orthogonal joint systems with microcrack alignments parallel to the four joint sets. The airphoto lineaments are parallel to only three joint sets. Southeast and northeast striking joints in one orthogonal system are tentatively interpreted as planes of extension and release respectively, related to the direction of compression during the Variscan-Appalachian orogeny. The other orthogonal system, with joint sets striking close to 010° and 100°, is interpreted as post-Triassic in age, possibly related to the presently acting crustal stresses in eastern North America.

II yacorrespondance dans les gres du Devonien superieur, du Carbonifere et du Permien inferieur du sud du Nouveau-Brunswick et de l'ile du Prince-Edouard entre 1'orientation des fractures de tension a diverses echelles: microscopique — alignement des microfissures; macroscopique — diaclases; et megascopique — lineaments observes a partir de photographies aeriennes. On reconnait comme alignements de microfissures les plans de force de traction minimale et les plans de fracture preferentielle occasionnes par des essais de charges lineaires et ponct-uelles sur des echan- tillons types de gres prealablement exempts de fractures apparentes. Bien que parallSlement aux diaclases et aux lineaments on retrouve des alignements de microfissures, ce ne sont pas tous les alignements de microfissures qui s'expriment aux echelles macroscopique et megascopique. A l'est de Saint-Jean, les gres deformes du Carbonifere possedent deux systemes distincts de diaclases orthogonales accompagnes d'alignements de microfissures paralleles aux quatre ensembles de diaclases. Les lineaments observes sur photographie aerienne ne sont paralleles qu'a trois ensembles de diaclases. On interprete provisoirement les diaclases de direction sud-est et nord- est dans l'un des systemes orthogonaux comme, respectivement, des plans d'extension et de relache- ment relies a la direction de compression durant l'orogenese varisque-appalachienne. Des orientations se rapprochant de 010° et 100° caracterisent 1'autre systeme orthogonal qui serait d'age post-triasique et peut etre relie aux forces de tension contemporaines agissant sur l'ecorce terrestre dans l'est de l'Amerique du Nord. [Traduit par le journal] INTRODUCTION between joints and microfabric Although the rock occurring be- has been described in sandstone by tween joints is often assumed to Reik and Currie (19 74), and in be unaffected by fracture, there quartz diorite by Swolfs et al. have been a number of studies (1974). Microcrack alignments may demonstrating that fractures, at a exist either parallel or perpen- much smaller scale than joints, do dicular to the maximum principal exist in the seemingly unfractured stress direction(s) of the geo- rock. For example, correlation logical past (Lajtai and Alison 1979). Preliminary comparison of MARITIME SEDIMENTS AND ATLANTIC GEOLOGY microcrack alignments and jointing 17, 70-87 (1981) 71 LAJTAI & STRINGER.

(Lajtai and Alison 1979) nas indi- of five predominantly terrestrial cated that joints have microcracks formations in southern New Bruns- running parallel to them, but not wick and Prince Edward Island necessarily all microcrack align- (Fig. 1), ranging from Upper De- ments become joints. vonian to Lower Permian in age. Microcrack alignments in oriented, The main purpose of the present visibly unfractured sandstone test- investigation has been to estab- samples have been compared with lish a correlation between tensile the strikes of subvertical (75- fractures occurring at the sub- 90° dipping) joints in the flat- microscopic (microcrack align- lying Pictou Formation and the ments) , macroscopic (joints), and gently dipping Perry Formation megascopic (airphoto lineaments) (mainly undeformed rocks of Wil- scales. Joint orientations have liams 1978, see Fig. 1), and in the been measured in the sandstones folded, thrust-faulted and locally

LEGEND

UPPER DEVONIAN CARBONIFEROUS and LOWER PERMIAN O mainly undeformed terrestrial deposits 100 km

j | deformed predomitpredominantly terrestrial deposits 47°-

Fig. 1 - Location map of sandstone formations investigated. Locality 1 - Perry Formation; 2 - Balls Lake Formation; 3 - McCoy Head Formation (Emerson Creek); 4 - McCoy Head Formation (Gardner Creek) ; 5 - Boss Point Formation; 6 - Pictou Formation. Areas of mainly undeformed and deformed Upper Devonian, Carboniferous and Lower Permian rocks are based on Williams (1978). MARITIME SEDIMENTS AND ATLANTIC GEOLOGY 72 cleaved Pennsylvanian Boss Point sponds to the plane of minimum ten and McCoy Head Formations (deformed sile strength, and the second de rocks of Williams 1978). Airphoto termines the tensile strength for lineaments have previously been any plane (McWilliams 1966, Lajtai established (Naing 1976 , Naing and and Alison 1979, Lajtai 1980); Lajtai 1977) east of Saint John in minima on the tensile strength dis southern New Brunswick. tribution curve identify the azi muths of the microcrack alignments. Two methods for identifying pref erentially oriented microcracks Fracture tests were made on sand in rocks are the point-loading and stone samples from flat-lying or the line-loading (Brazilian) tests. gently dipping (0-15° dip) beds in The first produces a preferred four of the formations investi fracture orientation which corre- gated. Discs of sandstone, pre pared by slicing core drilled per pendicular to bedding in the sam ples, were loaded along the axis of the disc in the point-loading tests and along diametral planes in the line-loading tests. The discs were 22.9 mm in diameter and 7.9 mm thick. Thin sections of the sandstone test-samples show no preferred orientation of con stituent grains or grain boun daries , and the fractures produced during the tests are assumed to represent microcrack alignments.

The fracture planes produced in both tests were perpendicular to bedding. In the point-loading tests, the results were plotted as frequency histograms (Figs. 3,4,6 and 8) with the number of fractures counted for each 10° interval of azimuth. Smoothed distribution curves are shown on the same fig ures. In the line-loading tests, the tensile strength was computed at 10° azimuth intervals and dis played to show variations of ten sile strength with azimuth in re lation to the strike of joints and Fig. 2 - Histograms of joint strike fre preferred fractures. quency showing joint sets A1, A2 and B1, B2 in two orthogonal joint systems in Correlation of results between the Balls Lake Formation, 10 km south the two tests is unequivocal only east of Saint John, New Brunswick (upper if the distribution of tensile histogram), and corresponding orthogo strength as a function of orien nal systems in the McCoy Head Formation tation is unimodal, in which case in the Emerson Creek locality, 20 km the peak of the preferred fracture east of Saint John (lower histogram). orientation curve or histogram Broken lines are smoothed distribution corresponds to the trough on the curves. tensile strength distribution 73 LAJTAI & STRINGER curve. In most rocks, however, there is more than one microcrack N = 45 alignment and therefore the ten sile strength distribution curve may have more than one minimum. This condition, however, may not be reflected by the preferred I ------1------1------r ~ 1 fracture distribution unless the tensile strength belonging to each microcrack alignment is of the same order of magnitude. There fore secondary or tertiary micro crack alignments can only be de tected with any reliability by the line-loading technique (Lajtai and Alison 1979, Lajtai 1980).

Joint Terminology In descriptive classification, joints have been divided into sys tematic and nonsystematic joints (Hodgson 1961, Nickelsen and Hough 1967). Systematic joints are planar or broadly curving, occur in sets which cut across other joints and are commonly perpendi Fig. 3 - Joints, preferred fracture orien cular to bedding. Nonsystematic tation and tensile strength in the flat- joints are curviplanar and ter lying Pictou Formation near Cavendish, minate against other joints and Prince Edward Island. Solid circles are bedding planes. In the present individual tensile strength measurements, investigation, systematic and non solid squares are averages per ten degree systematic joints have been dis azimuth interval. In the histogram, the tinguished only in the Pictou and smoothed distribution curve for joint Perry Formations. strikes is a dotted line, and that for preferred fracture azimuths is a broken Presentation of Bata line. Preferred fracture orientation correlates approximately with the strike Joint orientation data for the of the 145° (nonsystematic) joints. The sandstone formations investigated tensile strength minimum at 075° has no are presented in histogram form joint equivalent. (Figs. 2,3,4,6 and 8) showing the number of joints striking within quencies for the two azimuth in each 10° azimuth interval. A tervals adjoining the b interval. smoothed distribution curve is The same smoothing formula has also displayed on each histogram, drawn been applied to the orientation according to the formula data for preferred fracture. Smoothing of the orientation data b ' = [a + 2b + c]/4 emphasizes only the major trends and consequently may eliminate the where b ' is the smoothed frequency minor but nonetheless geologically for each azimuth interval, b is significant trends. For this rea the unsmoothed (actual) frequency son we present both the smoothed for the same interval, and a and c and the unsmoothed (actual) orien are the unsmoothed (actual) fre- tation data. MARITIME SEDIMENTS AND ATLANTIC GEOLOGY 74

JOINTS, TENSILE STRENGTH AND of McCutcheon, fig. D-2 , in Ruiten- PREFERRED FRACTURE ORIENTATION berg et at. 1979) in the Mispec Bay area, 10 km southeast of Saint John, are deformed, with gently Balts Lake Formation to moderately inclined Si and S2 Mississippian(?) sandstones,silt- cleavages, and asymmetrical F2 stones and conglomerates (Balls folds, trending mainly northeast Lake Formation, Alcock 1940, part (Ruitenberg et at. 1975, 1979). of the "Mispek Group" of Hayes and The Mispec Group ranges up to Howell 1937 and the "Mispeck Group" Westphalian B or C (Rast et at. 1978, McCutcheon in Ruitenberg

5. 5 ' et at. 1979), indicating a late Westphalian or younger age for 5.0 - I n=43 the deformation, ascribed to the 4 5 - Variscan-Appalachian orogeny (Rast and Grant 1973). § 4.0 - Joints in the Balls Lake Forma 3. 5 - I > N V / tion along the coast and inland • * near Mispec Bay (Fig. 1, locality 3.0 - v I 2) show two orthogonal systems, 2. 5 - designated A and B, with joint sets striking 015° and 105° desig I I 1------1 I I I I I nated A1, A2, and joint sets striking 060° and 140° designated B1, B2 (Fig. 2). Bedding is mostly gently inclined, locally subhori zontal. The joints are vertical or steep (75° to subvertical dips).

McCoy Bead Formation3 Emerson Creek locality Joints in gently dipping Penn sylvanian sandstone beds (McCoy Head Formation, Hayes and Howell 1937) on the coast southeast of Emerson Creek, 20 km east of Saint John (Fig. 1, locality 3), show two orthogonal systems with joints striking 015° and 105° (A1, A2) and 075° and 155° {B1, B2) , similar to those in the Balls Lake Forma tion (Fig. 2). Pictou Formation Joints striking 025° and 145° are prominent in flat-lying sand Fig. 4 - Tensile strength (above), pre stone beds near Cavendish on Prince ferred fracture orientation and joints Edward Island (Fig. 1, locality 6) (below) in the Perry Formation on the St. which form the youngest beds of Andrews Peninsula and Minister Island, the Pictou Formation, dated as New Brunswick. Orthogonal joint sets Lower Permian (Barss and Hacquebard strike 065° and 155°. Symbols as in 1967). The joints are not ortho Figure 3. gonal; the dihedral angle of about 75 LAJTAI & STRINGER.

60° is compatible with a shear Plumose structures are present origin. However, the two sets of sporadically on the 065° joint joints are unlikely to be conjugate surfaces (Fig. 5). shear fractures; the 025° joints The 155° joints are nonsystematic are systematic and the 145° joints joints. They are curviplanar, are nonsystematic, and neither set varying up to 25° in strike and shows surface features compatible dip from the mean 155"/vertical with a shear origin. The weak attitude. The 155° joints mostly joint orientation peaks at 055° deflect around pebbles, and plumose and 115° (Fig. 3) are orthogonal structures are lacking. The 155° to the dominant 025° and 145° joints commonly terminate against trends, suggesting that two ortho- the 065° joints (Fig. 5). gonal joint systems may be present in the Pictou Formation. A diabase dyke which crosses the St. Andrews peninsula and The Pictou Formation (Fig. 3) the north end of Minister Island has a single preferred fracture (Rhoades 1963) trends 065° parallel orientation around 130°, cor- to the systematic joints. Columnar relating approximately with the jointing is present in the 11 m 145° (nonsystematic) joints. The thick dyke; the columns plunge tensile strength distribution south-southeast at 5-15°, indi- curve has three -troughs. The cating that the dyke contacts have minimum corresponds to the 130° a steep north-northwest dip similar preferred fracture orientation, to that of the 065° joints. The the intermediate one at 0 75° has frequency of the 065° joints in neither joint nor preferred frac- the Perry Formation increases over ture orientation equivalents, 500 m approaching the dyke and while the trough at 010° may cor- the closest (5 to 30 cm) spacing relate with the 025° (systematic) occurs in sandstone on either side joints. of the dyke. The 155° joints in the sandstone continue through Perry Formation the dyke and cut the columns. The Upper Devonian sandstones dip age of the dyke based on the com- gently southeastward on the St. position and freshness of the Andrews peninsula and Minister mineral assemblage is Late Triassic Island (Fig. 1, locality 1) in a or Early Jurassic (Pajari, G.E., thick homoclinal sequence of red Jr., oral comm. 1981). conglomerates, sandstones and siltstones, with some interbedded The Perry Formation has a strong basalts (Perry Formation, McKenzie preferred fracture orientation at and Alcock 1960, Rhoades 1963, about 170° which is subparallel Cumming 1966, Schluger 1973). to the north-south tensile strength Joints strike 065° and 155°, con- minimum (Fig. 4). The nearest stituting an orthogonal system joint trend at 155° is probably (Fig. 4). unrelated. It is more likely that the minor preferred fracture trend The 065° joints, which dip (unsmoothed distribution) at 145°, steeply (75 - 90°) north-northwest confirmed by a minor low in ten- perpendicular to bedding, are sys- sile strength, is the microcrack tematic joints. They are planar alignment corresponding to the 155° and cut across the 155° joints. joints. The 080° major trough on The 065° joints also persistently the tensile strength distribution cut through pebbles, cobbles and curve may correlate with the 065° boulders of igneous rocks in peb- joints, but the major trough at bly sandstones and conglomerates. N-S does not clearly correlate in MARITIME SEDIMENTS AND ATLANTIC GEOLOGY (iii)

Fig. 5 - Joints in the Perry Formation, Minister Island, New Brunswick. Plumose markings on systematic joint surface striking 065°, north end of island (upper photograph). Orthogonal joints exposed on a gently dipping sandstone bed surface, west side of island; 155° joints (parallel to hammer handle) terminate against 065° joints (lower photograph). 77 LAJTAI & STRINGER orientation with the 155° joints; the 080° and N-S tensile strength minima may instead reflect the A orthogonal joint system of the Balls Lake and McCoy Head sand stones east of Saint John.

Boss Point Formation Pennsylvanian sandstones and shales 100 km east-northeast of Saint John (Boss Point Formation, Flaherty and Norman 1941) are de formed by northeast trending major folds. Changes in the dip of bed ding continuously exposed for 3 km along the coast eastward from Alma (Fig. 1, locality 5) define a major asymmetrical syncline verging toward the northwest; the dip of bedding changes from gentle southeastward in the north west limb to northwestward in the Fig. 6 - Histograms showing joints and southeast limb where the dip pro preferred fracture orientation in the gressively steepens until the beds Boss Point Formation, 100 km east-north- are overturned, thereafter dip east of Saint John, New Brunswick. Joints ping steeply southeastward. The strike 040° and 120° on the smoothed dis southeast limb is truncated by a tribution curve, and there is a minor high angle reverse fault on which joint strike at 165°. The strong pre older rocks have been displaced up ferred fracture alignment is approxi to the northwest. The overturned mately parallel to the 040° joints. Sym beds and associated reverse fault bols as in Figure 3. can be traced for at least 10 km inland to the northeast (Flaherty strong preferred fracture align and Norman 1941) . Cleavage^ dis ment at 030°, approximately paral cernible in shale beds at a few lel to the 040° joints (Fig. 6) , places in the coast section dips and has a weaker alignment at 005°. southeast at angles ranging from A weak preferred fracture alignment 20° to 75°. is also present parallel to the 135° joints. Smoothing of the In the Boss Point Formation, the distribution eliminates the north- smoothed joint frequency curve south preferred fracture align (Fig. 6) appears to show a single, ment. Only 20 line-loading tests approximately orthogonal system were done on the Boss Point sand with joints striking 040° and 120°; stone sample which are considered the latter trend is diffuse on inadequate to give a detailed ten the histogram, and may represent sile strength distribution. two sets of joints striking at 110° and 135°. The 135° set would then McCoy Head Formation, be nearly orthogonal to the 040° Gardner Creek locality joints. The smoothed joint fre quency curve also shows a weak Pennsylvanian sandstones and joint trend at 165°. shales of the McCoy Head Formation east of Gardner Creek, 25 km east The Boss Point Formation has a of Saint John (Fig. 1, locality 4) , MARITIME SEDIMENTS AND ATLANTIC GEOLOGY (iii)

Fig. 7 - Structures in the McCoy Head Formation, coast east of Gardner Creek, New Brunswick. Sandstone bed is displaced 20 cm up to the northwest (left) on a low angle reverse fault (upper photograph). Joints in two orthogonal ioint systems exposed on a gently dipping sandstone bed surface strike 017° (parallel to pen), 045°, 100°, and 135° (lower photograph). 79 LAJTAI & STRINGER

appear more deformed than in the B2 r r r 2 \ Emerson Creek locality. Beds have strike of | * strike of been displaced up towards the bedding^ ? thrusts northwest on small scale thrusts /±3.6\ ,<2B\ / iL ,•±2.7 ±2'9'.V± 28 *“ •' (Fig. 7). Slickensides plunge \ :•«j i / southeast at 20° to 35° on the / ±L6 V ’±2.9 ±2.6*n TENSILE STRENGTH ±2.9 thrust surfaces, and cleavage ki parallel to the thrust planes is locally developed in the shales. The thrusts are deformed by north east trending folds which verge to ward the southeast. Bedding is gently inclined in the long limbs of the asymmetrical folds, and is subvertical or overturned south eastward (dipping steeply north west) in the short limbs. Joints measured in the McCoy Head Formation along 800 m of coastal outcrop east of Gardner Creek show two orthogonal systems (Fig. 7) designated A and B, with joint sets A1, A2 striking 010°, 100°, and Bl, B2 striking 045°, 135° (Fig. 8). A visibly unfractured sandstone block from gently inclined bed ding in the McCoy Head Formation received the most comprehensive testing. Instead of spreading the tensile strength tests at 10° in tervals over the 0-180° azimuth range, multiple tests (15 tests each) were made for selected orien tations derived from information on jointing and preferred fracture orientations. Although the stand Fig. 8 - A comparison of trends display ard deviation of the results was ed by tensile strength, preferred frac found to be fairly large, azimuths ture, joints and airphoto lineaments in of tensile strength minima parallel the McCoy Head Formation east of Gardner to each joint trend of the two Creek, New Brunswick. Solid circles are orthogonal systems are apparent (Fig. 8). The azimuths of the means of 15 individual measurements. lowest tensile strength minima Standard deviations are shown for each correlate with the Al and Bl joint mean. The solid arrows are the strikes sets of the two orthogonal systems. of the joints in the two orthogonal joint systems in the outcrop shown in Figure The smoothed preferred fracture 7 (lower photograph). There is a tensile orientation curve (Fig. 8) shows strength minimum parallel to each joint correlation of preferred fracture set, but the strongest minima correlate with the Al, Bl and B2 joint sets, with the joint sets Al, Bl. There is no but not clearly with the A2 joint lineament trend parallel to the Bl joints. set. It should be noted that Broken lines are smoothed distribution point-loading tests, resulting in curves. MARITIME SEDIMENTS AND ATLANTIC GEOLOGY (iii) the determination of a preferred At fracture orientation, can be effec- tive in defining only the plane of McCoy Head Formation minimum tensile strength (Lajtai n = 450 and Alison 19 79). For the three B2 sandstones (Pictou, Perry, and A2 McCoy Head at Gardner Creek) for which both point- and line-loading data are presented, the lowest tensile strength was measured across the plane of strongest preferred fracture orientation. Peaks of lower order on the preferred fracture orientation diagram may also be significant. For the McCoy sandstone, the two major and the one minor peak all correspond to minima in tensile strength. How- ever, not all minima in tensile strength have corresponding preferred fracture peaks, e.g. , A2 of the McCoy Head Formation (Fig. 8), the 080° minimum in the Perry For- mation (Fig.4), and the 075° minimum in the Pictou Formation (Fig. 3) . AIRPHOTO LINEAMENT TRENDS A remarkable feature of the B orthogonal joint system is that the strong B1 joints and parallel microcrack alignment do not produce lineaments identifiable by aerial photograph studies (Fig.9, after Naing and Lajtai 1977). The

Balls Lake Formation has strong AZIMUTH lineaments parallel to the Al, A2 and B2 joints, but no lineament Fig. 9 - Airphoto lineament trends in is identifiable parallel to the four formations. The top three format B1 joints (Fig. 9). The McCoy tions were affected by the Variscan- Head Formation shows a similar Appalachian orogeny. The Al, A2 and B2 lineament distribution curve (Fig. lineament trends correlate approximately 9), although there is also a weak with the strike of corresponding joints lineament parallel to the B1 joints in the Balls Lake and McCoy Head Forma- in the unsmoothed distribution his- tions (Figs. 2 and 8), but a lineament togram (Fig. 8). The Boss Point parallel to the B1 joints is missing. The Formation has lineaments trending Triassic Quaco Formation shows linea- approximately parallel to the 165° ments parallel only to the A orthogonal minor joints (Al lineament trend) joint system. (After Naing and Lajtai and to the 120° joints (B2 trend), 1977). and possibly to the B1 joints (of. Figs. 6 and 9); the A2 lineament aments is about 1.5km, and there- trend is missing. fore it would appear that where lineaments are lacking, the micro- The average length of the line- cracks have grown only to joints 81 LAJTAI & STRINGER. of outcrop scale. The B1 joints The investigation supports the appear to be unique in lacking view that tensile fracture origi- lineaments. Although both joint nates at microscopic or submicro- sets of the orthogonal joint sys- scopic stress concentrations and tem A are represented at the mega- may grow in size to re^ch the scopic scale of lineaments in Pre- macroscopic scale of joints (Lajtai cambrian to Triassic rocks in the 1977) or even the megascopic scale region east of Saint John (Fig.. of lineaments. While earlier formed 10), the 100° (A2) trend is the' fractures propagate, new micro- more prominent. fractures nucleate and grow con-

0 2 4

% Frequency -1st. d. mean Fig. 10 - Airphoto lineament trends in the region east of Saint John, New Brunswick (after Naing 1976). The compilation includes all rocks.

DISCUSSION AND CONCLUSIONS tinuously between them. Conse- The orientations of joints, ten- quently, lineaments and joints sile strength minima, preferred should have a microcrack align- fractures and airphoto lineaments ment parallel or subparallel to show close correlation in the McCoy them. This is indeed the case in Head (Gardner Creek) sandstone the sandstones in the present which received the most compre- study. On the other hand, not all hensive testing, and they show microcrack alignments have joint correlation in the other sandstones or lineament equivalents (e.g., investigated (Table 1); Micro- there is no joint trend corres- crack alignments identified by the ponding to the 075° tensile strength tensile strength minima and pre- minimum in the Pictou Formation ferred fracture orientations are on Prince Edward Island, and north- not always represented by joints east trending lineaments are lack- or lineaments, but joints and line- ing in the Carboniferous formation aments are represented by micro- east of Saint John). crack alignments. Nickelsen and Hough (1967) inter- MARITIME SEDIMENTS AND ATLANTIC GEOLOGY (iii)

preted the approximately orthog- insofar as fracture nucleation onal systematic and nonsystematic and growth would take place from joints in the Carboniferous rocks microscopic stress concentrations of the Appalachian Plateau of and would proceed in an essentially Pennsylvania, designated the fun- compressive macroscopic stress damental joint system3 as develop- field albeit an extensional one ing through a single cycle of co- (i.e. decreasing compression). It axial tectonic loading and unload- is probable that the most effec- ing. The systematic joints, which tive parameter producing a release strike approximately perpendicular joint is the tensile residual to the axial direction of the stress (Friedman 1972, Lajtai 1977, folds, formed early during tectonic Stringer and Lajtai 1979), local- loading while the nonsystematic ized in the cement of a sandstone joints, which strike nearly paral- and there growing in relative in- lel to the axial direction of tensity as the original tectonic folds, are later release - type compression slowly relaxes. fractures. The absence of plumose structures on the nonsystematic or re- The two types of joints in the lease joints of Nickelsen and fundamental joint system repre- Hough (1967, p. 614, p. 626) sug- sent the load-parallel tensile gests that release joints form very fracture of rock mechanics ter- slowly. They noted {ibid., p. 614) minology, termed load-parallel that whereas systematic joints joints herein, and the load per- are common in fresh outcrops, non- pendicular release , or relaxation, systematic joints commonly do not fracture (extension fracture of appear in the high walls of strip Griggs and Handin 1960) , termed mines until after several months release joints herein. The terms or a year of weathering. load-parallel and release specify the macroscopic field of stress existing at the time of formation PERRY FORMATION ORTHOGONAL JOINTS of the joints. The actual stress In addition to joints formed as responsible for both load-parallel a result of tectonic compression and release joints is a tensile and its subsequent relaxation, a stress acting locally around sub- third type of joint results from microscopic Griffith cracks and tensional stress (Nickelsen and flaws (Lajtai 1977). Hough 1967) , associated with crus- The mechanism of release joint tal extension and more specifi- formation is similar to that in- cally with the bending of strata, volved in load-parallel joints i.e. flexures associated with

TABLE 1

PICTOU FORMATION PERRY FORMATION BOSS POINT FORMATION McCOY HEAD FORMATION (Gardner Creek)

Joints (S, systematic; 025° 145° 065° 155° - 040° 120° (165°) 010° 045° 100° 135° N, nonsystematic) (S) (N) (S) (N)

Tensile strength ?010° 075° 130° ?080° 145° 180° 040° 000° 045° 105° 135° minima

(005°) 000° 050° ?115° 135° Preferred fractures - 130° - (145°) 170° 030° 135°

Airphoto lineaments (040°) 120° 160° 000° - 100° 135°

Summary of trends of joints, tensile strength minima, preferred fractures and airphoto lineaments in sandstones of the Pictou, Perry, Boss Point and McCoy Head (Gardner Creek) Formations. Trends in parentheses indicate minor peaks, and question-marks indicate possible correlative trends. Dashes indicate that correlative trends are absent. 83 LAJTAI & STRINGER. vertical movement in the crust ferous sandstones east of Saint (draping across basement faults, John are parallel to the trend of diapiric rise of magma or salt) , folds and thrusts formed by north- or warping of a sedimentary basin west-southeast compression during (Price 1974). Such joints are the Variscan-Appalachian orogeny. termed tensile joints herein. In model tests designed to simu- Again, the adjective (tensile) late tectonic loading during cemen- specifies the macroscopic stress tation and subsequent unloading, field. In load-parallel, release the plane of minimum tensile and tensile joints, the individual strength and strong preferred frac- tensile fracture nucleates from a ture orientation was found to be submicroscopic stress concentra- the plane of release; load-parallel tion at a microcrack. The exten- microcracks may form during the sion of a single microcrack to a loading cycle or possibly during macroscopic joint would depend relaxation (Lajtai and Alison primarily on the large-scale stress 1979). In the Carboniferous sand- field (as opposed to the micro- stones investigated, by analogy, scale of microstress concentra- the northeast striking tensile tion) . fractures would represent the plane of tectonic relaxation; the ORTHOGONAL JOINT SYSTEM B southeast striking tensile fractures would be parallel to the direction of tectonic compression In the Boss Point and McCoy Head (or relaxation). (Gardner Creek) Formations, there is a consistently strong northeast trend of preferred fracture The southeast striking joints orientation (Figs. 6 and 8), and of the orthogonal joint system S in the McCoy Head sandstone the in the Carboniferous sandstones plane of minimum tensile strength east of Saint John are tentatively trends northeast (Fig. 8). The interpreted as load-parallel joints tensile strengths (20 tests) com- and the northeast striking joints puted for the Boss Point sandstone as release joints, related to ranged from 5 to 10 MPa; the low- northwest - southeast compression est value occurred at azimuth 040° , and relaxation respectively dur- again corresponding to the north- ing the Variscan-Appalachian oro- east (Bl) direction. The sand- geny. A difficulty with this in- stone test-sample from the Pictou terpretation is that during fold- Formation on Prince Edward Island ing and thrusting the bulk stresses shows a strong southeasterly pre- would produce vertical extension, ferred fracture orientation (Fig. whereas the subvertical load- 3) unlike any of the other forma- parallel joints imply northeast- tions. In older (Upper Pennsyl- southwest axial extension. How- vanian) Pictou Formation sandstone ever, axial extension, indicated at Fredericton, New Brunswick, 250 by wrench faulting, may result km to the west, however, northeast from continued tectonic compres- trending preferred fractures sion after the folding and thrust- (Lajtai and Alison 1979, fig. 11) ing stage of orogeny (Dewey 1969). may correlate with the Bl joints North-northeast trending, left in the Balls Lake, McCoy Head and lateral wrench faults are present Boss Point Formations. in the deformed Carboniferous rocks west of Saint John (Grant, R.H., The northeast trend of Bl joints, oral comm. 1981) and possibly in tensile strength minimum and pre- the Gardner Creek area (map 21H: ferred fracture in the Carboni- 5E, Ruitenberg et al. 1975). Axial (iii) MARITIME SEDIMENTS AND ATLANTIC GEOLOGY extension may also occur before and lineaments trending approxi- folding and thrusting in orogeny; mately 010° (Al) and 100° (A2) Nickelsen (1979) interpreted vert- are well defined in the Balls Lake ical northwest-southeast trending and McCoy Head rocks, and A1 and joints, which pre-date cleavage A2 microcrack alignments are pres- in folded Pennsylvanian sand- ent in the McCoy Head Formation stones, as extension joints th'at at Gardner Creek (Fig. 8). In the developed during the inception of Boss Point Formation, the lineament layer-parallel shortening under trend at 160°, the weak joint trend the northwest-southeast compres- at 165°, and the unsmoothed pre- sion of the Allegheny otogeny. The ferred fracture distribution at load-parallel joints may, there- 005° (Fig. 6) suggest the exist- fore, have formed under compres- ence of the A1 trend; the unsmoothed sion during an axial extension 110° joints (Fig. 6) suggest an phase of the Variscan-Appalachian A2 trend, but 100° (A2) lineaments orogeny; the release joints re- appear to be absent (Fig. 9). sulted from relaxation of the Strong microcrack alignments in northwest-southeast compression. the 170° (approximately Al) orientation exist in the Perrv Forma- The orthogonal joint system in tion (Fig. 4), and north-south the Perry Formation is similar in preferred fracture has been iden- orientation to the orthogonal sys- tified in the Pictou (Fredericton) tem of northeast and southeast sample (Lajtai and Alison 1979). striking joints (the B system) of In the 100° (A2) direction, no the Balls Lake, McCoy Head and significant microcrack alignments Boss Point Formations. However, exist in either the Perry or Pictou there is no preferred fracture Formation sandstones. parallel to the 065° joints (Fig. 4). Furthermore, the 065° joints The fundamental joint system A in the Perry Formation appear to is interpreted as post-Triassic be associated with intrusion of in age. The bimodal distribution the Late Triassic—Early Jurassic of lineaments in the Triassic dyke, and the 155° joints are in- Quaco Formation (Fig. 9) , adja- terpreted as post-intrusion in cent to the McCoy Head Formation, age; both joint sets apparently clearly suggests correlation with post-date the Variscan-Appalachian the A system; lineaments corres- orogeny. During the Late Triassic, ponding to the B system are absent. the dominant tectonism involved There is evidence that supports a crustal extension perpendicular relatively young age for the A to the Appalachian trend (Rodgers system. In situ strain measure- 1970). The 065° joints in the ments in the Potsdam sandstone, Perry Formation in the St. Andrews northeast of the Adirondacks, sug- area may therefore be tensile gest that the upper crust of eastern joints. The origin of the 155° North America is in a state of joints and their orthogonal re- horizontal compression, the aver- lationship to the strike joints age direction of compression be- is uncertain. ing at 102° (Engelder and Sbar 1976). Focal mechanism solutions of earthquakes suggest that pres- ORTHOGONAL JOINT SYSTEM A ently the maximum principal stress The orthogonal joint system A in most of eastern and central cannot be attributed to any iden- North America is east-northeast tifiable tectonic event. Never- (Sykes and Sbar 19 73), and this theless, it is present in most of stress is tentatively related to the rocks studied. Both joints the presently active driving mec- 85 LAJTAI & STRINGER. hanism of plate tectonics. The occurs along the release joints. strongest trend of lineaments east Although microcrack alignments of Saint John in southern New exist parallel to load-parallel Brunswick is 097±7° (Fig. 10, and joints as well, they appear to be Naing and Lajtai 19 77) . weaker than those associated with release joints. In contrast, A fundamental joint system is lineaments are more numerous in the product of a complete load- the load-parallel joint direction. ing and unloading (release) cycle, and therefore development of a ACKNOWLEDGEMENTS stress release fabric (microcrack alignment, joint and lineament), The authors thank H.V. Donohoe such as the north-south Al trend, Jr. , K.B.S. Burke and P.F. Williams requires the rocks to be in an for critical reviews of the manu- unloaded condition, separated from script. We thank Nick Rast for many stresses deeper in the crust. stimulating discussions throughout Although there is no physical evi- the investigation, and Ismail dence for uncoupling of the sand- Patel for guidance in the field stones investigated from the over the Boss Point sandstones. stressed part of the crust, we The extensive laboratory work would note that at one of the five sites not have been possible without the measured by Engelder and Sbar assistance of Allan Fleming, Jamie (1976), the maximum principal Alison, Gerald Trembath, and Mar- stress direction was at 018°, garet Svab. Darlene Pajari pre- approximately at right angles to pared the figures and Sherri Town- the 102° direction. This suggests send prepared the manuscript. The that at this site the rock has research was supported through indeed been released from the 102° Natural Sciences and Engineering compression. Research Council of Canada Grants A3534 (E.Z.L.) and A7833 (P.S.). The north-south Al trend is marked by strong preferred fracture orientation, low tensile ALCOCK, F. J. 1940. Loch Lomond (west strength, and weak lineament trend, half), Saint John, and Kings Counties, similar to the Bl trend which New Brunswick. Geological Survey of marks the release plane of the Canada, Map 478A. Variscan-Appalachian stress field. BARSS, M. S. and HACQUEBARD, P.A. 1967. This would suggest that the Al Age and the stratigraphy of the Pictou joints are release joints and the Group in the Maritime Provinces as A2 joints are load-parallel joints. revealed by fossil spores. In Geology of the Atlantic Region. Edited by The investigation indicates that E.W.R. Neale and H. Williams. Geo- correlation of tensile fracture logical Association of Canada, Special at all scales (microcracks, joints Paper No. 4, pp. 267-282. and lineaments) can distinguish between load-parallel and release CUMMING, L.M. 1966. Geology of the Passa- joints. When all three scales of maquoddy Bay region, Charlotte County, structure are indentified in a New Brunswick. Geological Survey of sandstone and the tectonic stresses Canada, Paper 65-29. are known from separate structural DEWEY, J.F. 1969. Structure and sequence evidence, there appear to be cer- in paratectonic British Caledonides. tain characteristic features. The In North Atlantic — geology and con- strongest microcrack alignment, as tinental drift. Edited by M. Kay. defined by both tensile strength American Association of Petroleum Geo- and preferred fracture studies, logists, Memoir 12, pp. 309-335. MARITIME SEDIMENTS AND ATLANTIC GEOLOGY (iii)

ENGELDER, J. T. and SBAR, M.L. 1976. Materials, Special Technical Publica- Evidence for uniform strain orientation 402, pp. 175-189. tion in the Potsdam Sandstone, northern NAING, W. 1976. Photogeology of the New York, from in situ measurements. Caledonia area of southern New Bruns- Journal of Geophysical Research, 81, wick. Unpublished Ph.D. Thesis, Uni- pp. 3013-3017. versity of New Brunswick, Fredericton, FLAHERTY, G.F. and NORMAN, G.W.H. 1941. New Brunswick. Albert, Albert County, New Brunswick. NAING, W. and LAJTAI, E.Z. 1977. Air- Geological Survey of Canada, Map 648A, photo lineaments in the Caledonia area with descriptive notes by G. W. H. of southern New Brunswick, Canada. Norman. Maritime Sediments, 13, pp. 107-112. FRIEDMAN, M. 1972. Residual elastic NICKELSEN, R.P. 1979. Sequence of struc- strain in rocks. Tectonophysics, 15, tural stages of the Alleghany orogeny, pp. 297-330. at the Bear Valley strip mine, Shamokin, GRIGGS, D.T. and HANDIN, J. 1960. Obser- Pennsylvania. American Journal of vations on fracture and an hypothesis Science, 279, pp. 225-271. of earthquakes. In Rock Deformation. NICKELSEN, R.P. and HOUGH, vanN.D. 1967. Geological Society of America, Memoir Jointing in the Appalachian Plateau 79, pp. 347-364. of Pennsylvania. Bulletin of the Geo- HAYES, A.O. and HOWELL, B.F. 1937. Geo- logical Society of America, 78, pp. logy of Saint John, New Brunswick. 609-630. Geological Society of America, Special Paper 5, 140p. PRICE, N.J. 1974. The development of stress systems and fracture patterns HODGSON, R.A. 1961. Regional study of in undeformed sediments. Proceedings jointing in Comb Ridge — Navajo Moun- of the Third Congress of the Inter- tain area, Arizona and Utah. Bulletin national Society of Rock Mechanics, of the American Association of Petro- Denver. In Advances in Rock Mechanics, leum Geologists, 45, pp. 1-38. Vol. 1, Part A, pp. 487-496. National LAJTAI, E.Z. 1977. A mechanistic view Academy of Sciences, Washington, D.C. of some aspects of jointing in rocks. RAST, N. and GRANT, R.H. 1973. Trans- Tectonophysics, 38, pp. 327-338. atlantic correlation of the Variscan- 1980. Tensile strength and Appalachian Orogeny. American Journal of Science, 2*73, pp. 572-579. its anisotropy measured by point- loading and line-loading of sandstone. RAST, N., GRANT, R.H. PARKER, J.S.D., Engineering Geology, 15, pp. 163-171. and TENG, H.C. 1978. The Carboniferous LAJTAI, E. Z. and ALISON, J.R. 1979. A deformed rocks west of Saint John, study of residual stress effects in New Brunswick. In Guidebook for Field sandstone. Canadian Journal of Earth trips in Southeastern Maine and South- Sciences, 16, pp. 1547-1557. western New Brunswick. Edited by A. Ludman. 70th Annual Meeting of the McKENZIE, G.S. and ALCOCK, F.J. 1960. New England Intercollegiate Geological St. Stephen, Charlotte County, New Conference, Geological Bulletin No. 6, Brunswick. Geological Survey of Canada, Queens College Press, Flushing, New Map 1096A, with descriptive notes by York, pp. 162-173. F.J. Alcock. REIK, G.A. and CURRIE, J.B. 1974. A study of relations between rock fabric and McWILLIAMS, J.R. 1966. The role of micro- joints in sandstone. Canadian Journal structure in the physical properties of Earth Sciences, 11, pp. 1253-1268. of rock. Testing techniques for rock mechanics. American Society for Testing RHOADES, D.A. 1963. The geology of the 87 LAJTAI & STRINGER. Perry Formation. Unpublished M.Sc. STRINGER, P. and LAJTAI, E.Z. 1979. Thesis, University of Maine, Orono, Cleavage in Triassic rocks of southern Maine. New Brunswick, Canada. Canadian Jour- nal of Earth Sciences, 16, pp. 2165- RODGERS, J. 1970. The tectonics of the 2180. Appalachians. Wiley Interscience, New York, N.Y., 271p. SWOLFS, H.S., HANDIN, J. and PRATT, H.R. 1974. Field measurements of residual RUITENBERG, A.A., GILES, P.S., VENUGOPAL, strain in granitic rock masses. Ad- D.V., BUTTIMER, S.M., McCUTCHEON, S.R., vances in Rock Mechanics, Proceedings and CHANDRA, J. 1975. Geological maps, of Third Congress of the International NTS 21H:4, 5E & W, 6E & W, 10W, HE & Society for Rock Mechanics, 2, pp. W, 12E, 14E, 15E & W; 21G:1E & W, 2E 563-568. United States Committee for and 8E & W, to accompany report, Geo- Rock Mechanics, National Academy of logy and mineral deposits, Caledonia Sciences, Washington, D.C. area. Mineral Resources Branch, New Brunswick Department of Natural Re- SYKES, L.R. and SBAR, M.L. 1973. Intra- sources and Canada Department of Re- plate earthquakes, lithospheric gional Economic Expansion. stresses and the driving mechanism of 1979. Geology and mineral plate tectonics. Nature, London, 245, deposits, Caledonia area. Mineral Re- pp. 298-302. sources Branch, New Brunswick Depart- WILLIAMS, H. (compiler) 1978. Tectonic- ment of Natural Resources and Canada lithofacies map of the Appalachian Department of Regional Economic Ex- Orogen. International Geological Cor- pansion, Memoir 1. relation Program, Project No. 27, the SCHLUGER, P. F. 1973. Stratigraphy and Appalachian-Caledonides Orogen, Cana- sedimentary environments of the De- dian Contribution No. 5. Memorial vonian Perry Formation, New Brunswick, University of Newfoundland, Map Nos. Canada, and Maine, U.S.A. Bulletin of 1 and 2. the Geological Society of America, 84, pp. 2533-2548.

Received: July 13, 1981 Accepted: July 18, 1981 Reviewers: H.V. Donohoe K.B.S. Burke P.F. Williams. W.C. Brisbin (for editors)