b y
J. II. M A C K I N 1
ABSTRACT
C o rro sio n (la te ra l p la n a tio n ) p e d im e n ts a n d w e ath e rin g —w ash in g (sh e e t w a s h ) p e d im e n ts a re e n d m em b e rs in a n iso m o rp h o n s se ries ; th u s th e d efin itio n o f a p e d im e n t m u s t be d e sc rip tiv e , n o t gen etic. A p e d im e n t is a n e ro sio n su rfa ce , p la n a r o r w ith low local re lie f b u t w ith h ig h o v e ra ll slope, t h a t form s in a rid o r se m i-a rid regions.
F o rm a tio n of p e d im e n ts p rim a rily b y la te r a l p la n a tio n by stre am s w as c h a m p io n e d in th e U n ite d S ta te s b y D ouglas Jo h n s o n . G ra v e l is le ft b e h in d as th e stre a m m ig r a te s la te ra lly . The th ic k n e ss of th e g ra v e l cap on th e c u t b e d ro ck su rface te n d s to a p p ro a c h th e m ax im u m d e p tli o f th e flood sc o u r ch an n el. L a te ra l c u ttin g b y a rid -re g io n stre am s is r a p id com pared to t h a t in h u m id reg io n s because of th e la c k o f b a n k s ta b iliz a tio n b y v e g e ta tio n .
P e d im e n ts form ed d o m in a n tly b y s h e e t w ash occu r on ro c k s th a t w e a th e r to g ru s t h a t is e a s ily re m o v a b le d u rin g th e c lo u d b u rs t ra in fa ll so c h a ra c te ris tic of a rid re g io n s . T h is leaves a s tr ip p e d ro c k su rfa ce o f m in o r re lief b u t w hose o v e rall a p p e a ra n c e is t h a t o f a sm o o th erosion su rfa ce .
T h e s h a rp angle m a in ta in e d betw een p e d im e n t a n d m o u n ta in fro n t is th e r e s u lt of: la te ra l s tr e a m p la n a tio n t h a t tr u n c a te s sp u rs o f th e m o u n ta in fro n t, w ash in g of th e s c a rp , sa p p in g of a r e s is ta n t la y e r a t th e to p , fo rm atio n o f a re s id u a l a rm o r o f b oulders, a n d /o r th e t o r effect.
B o th p e d im e n ts an d p ie d m o n t fans a re c h a ra c te ris tic a lly c o n v ex -u p w a rd in p ro file because b o th a re slopes of tr a n s p o r ta tio n .
O n ly p e d im e n ts fo rm ed re la tiv e to s ta b le o r slow ly lo w e rin g b aselevel a r e d iscu ssed in th is p a p e r. P e d im e n t p ro b lem s n o t d iscu ssed in clu d e: risin g local baselev el; effe c ts o f re tre a t o f m o u n ta in fro n t in u n ifo rm ro c k s; h ig h -lev el P liocene p e d im e n ts of th e R o c k y M o u n tain s.
I. IN TR O D U C TIO N
1 h av e been a stay-at-hom e, an d m y u n d erstan d in g of pedim ents is d istinctly provincial. I h asten to em phasize t h a t if my views differ sh arp ly from those' held b y some of you, it is sim ply t h a t I am not ad eq u ately aw are of concepts and evidence foreign to the w estern U nited S tates an d Mexico. C ertainly much of th e controversy in the U nited S ta te s regarding pedim ents has arisen from th e fa ct th a t people w orking in different places h av e seen different things. If a m an is fam iliar w ith pedim ents of only one origin, and th in k s t h a t all pedi
m ents are th e sam e in origin, he is likely to disagree w ith th o se who have w orked w ith pedim ents elsewhere. M any of our problem s are of th is n atu re ; the disagreem ents betw een different schools of th o u g h t in th e U n ited S tates, over th e years, would have been g re a tly reduced h ad the m en exchanged visits in the field. T h a t is th e reason I am here—to hear views foreign to my th in k in g , and to see y o u r m ountains.
1 T h is v e rsio n of P ro fe sso r M ackin's p a p e r , w ho died on A u g u s t 12, 19(58, is a n e a rly v e rb a l im t e x t o f h is o ral p re s e n ta tio n . T he illu s tra tio n s a re th o se p ro je c te d d u rin g t h e sy m p o siu m .
The a b s t r a c t w as p re p a re d b y D r. W illiam R . M uehlberger.
The first problem is a m a ile r of definition; w h at is a pedim ent? Som e pedi
m ents in our w est are cut by la te ra l stre a m p la n a tio n —th e evidence is conclu
sive. B ut it is e q u a lly certain t h a t o thers are form ed by w eathering and w ashing; th e stream s a c t prim arily as agents of tra n s p o rta tio n , and la te ra l plana
tion is negligible. T hese corrasion pedim ents and the w eathering—washing pedim ents are end m em bers in an isom orphous series; th e relative im por
tan ce of th e tw o processes can v a ry from alm ost zero to alm ost 100 p er cent, depending on conditions. Hence, b ecause pedim ents are polygenetic, they can
not be defined in te rm s of m echanism of origin; th e definition m u st be de
scriptive.
T he pedim ent a b o v e all is an erosion surface, w h e th e r cut in h a rd rock or unconsolidated sed im en ts of any origin, ft is p la n a r or has low local relief, but is ch aracterized b y a high overall slope of the ord er of a few tens or hundreds of m eters per k ilo m e te r. The slope is a slope of tra n s p o rta tio n ; it tends to decrease uniform ly from a “ m o u n ta in m ass” w hich is com m only bu t not necessarily present, to a local baselevel th a t m ay be sta b le or slowly rising or low ering. It can of course be dissected, or buried, or g re a tly modified by proc
esses o th er th a n th o s e th a t form ed it, b u t these changes m erely call for quali
fying adjectives; th e y do not e n te r into the definition.
I t seems to m e re g re tta b le to call an y erosional low land at th e base of a m o u n ta in mass or sc a rp a pedim ent. This m akes p ed im en t synonym ous with
F ig . 1. B ook Cliffs, U ta h ; from U .S . Geol. S u rr ey . W e llin g to n 15' q u a d ra n g le 86
Fig. 2. C u m b erla n d E s c a r p m e n t, K e n tu c k y : fro m U .S . Geol. S u r v e y , B y rd s to w n 15' q u a d ra n g le
piedm ont, a n d we h av e no n eed for synonym s. The p ed im en t is a landform of this not the case th e channel w ould have becom e infinitely w ide and would cease to be a channel. The rock su rface is not a peneplain, strip p e d of its regolith.
Fig. 4. L a te r a l s tre a m p la n a tio n o n a p e d im e n t, H e n ry M o u n ta in s , U ta h ; from G ilb ert, 1877. Fig. 62
Fig. 5. G ra v e l e ap ro c k on a p e d im e n t, H e n ry M o u n ta in s , U ta h ; fro m G ilb e rt, 1877, Fig. 63
89
The area show n in Fig. 7 is in the Colorado P ied m o n t n o rth w est of D enver.
power of a stream to corracle its bed and banks varies directly with its power to transport bed load.
I t has long been know n—it is im plicit in two of G ilbert’s classic papers (1914, 1917) and was brought ou t clearly by R ubey (1938) — t h a t high grad ien t, low discharge stre a m s are far m ore effective agents of bed load tra n s p o rta tio n an d corrosion th a n low g radient, high discharge stream s.
The fact t h a t th e average velocity of m an y stream s te n d s to increase in a dow nvalley d irection (Leopold—Maddock, 1953), as the calib re of the bed load decreases, is seem ingly anom alous. A ctually it is a rem inder t h a t while av erag e velocity is a depth-slope fu n c tio n , modified by a roughness factor, bed load tran sp o rtin g power (and hence corrosive power) varies, not w ith av e rag e velocity b u t w ith bed v elo city ; this relation provides a ra tio n a l ex p lan a tio n for the well docum ented em p irical generalization sta te d above, i.e., t h a t th e high g ra d ie n t, low average discharge stream s of arid regions are far m ore effective ag en ts of tra n s p o rta tio n and corrasion th an th e low g radient, high average discharge stream s of hum id regions. Especially if we recall t h a t th e high g ra d ie n t stream s of th e a rid regions are ch aracterized b y peak flows w hich m ay represent m ost of th e discharge of one or m any y ea rs, it is not su rp risin g th a t corrasive power seems to v ary directly w ith slope, an d , w ithin c e r
tain lim its, inversely w ith av erag e discharge or d ep th .
ill
This general principle, w hich applies p rim a rily to differences in developm ent of planalion surfaces in different clim ates, is confirmed by th e fact th a t, w hen th e re is through-flow ing d ra in a g e and no m a rk e d difference in rock resistance, strea m -cu t surfaces te n d to be w ider n ear th e m o u n tain fro n t th a n fa rth e r out o n the piedm ont; th a t is, th e b re a d th of I lie surfaces decreases w ith th e stre a m grad ien ts, w hich in tu r n reflect a decrease in calibre of load, chiefly b y a
ttri-7 1 /2 ' q u a d ra n g le
tio n . Two places a lre a d y n o te d serve as exam ples; the p la n a tio n surfaces of th e Book Cliffs region (see Fig. 1) decrease m ark e d ly in w id th in the first 10 kilom eters o u t from th e base of th e cliffs, a n d th e R ocky F la ts pedim ents (see Fig. 7), w hich are v irtu a lly continuous for several kilo m eters along th e Boulder-G olden m o u n ta in fro n t, are replaced eastw ard by hills w ith concavo- convex slopes anti soil, w e ath ered in place, t h a t are the an tith e sis of th e p lan ar, gravel-veneered ped im en ts.
A nother an d en tire ly different facto r is significant iu m any places. The b an k s of channels of p eren n ial strea m s of hum id regions are com m only covered by v egetation, ra n g in g from grass a n d shrubs to tre e s; the b an k s of the channels of ephem eral stream s of arid regions are com m only barren. T he riv er engineer is well aware of th e p ro tec tiv e v alu e of b an k v eg e ta tio n ; use of a m at of living willows is a s ta n d a rd and effective procedure in controlling b a n k erosion. Not 92
only is th e v eg etativ e blanket erosion resistan t, b o th above and below th e n o rm al surface; a fte r an episode of scouring th a t dam ages or destroys th e n etw o rk of roots an d branches, it is self-renewing. The b arren b an k s of arid regions, on the o th e r h a n d , are m ore readily cut and th e y stay cu t, to be eroded back still farth er in th e next a tta c k .
F ig . 8. T h ree P e a k s in tru s io n , U ta h , view ed fro m th e so u th
F ig. 9. P e d im e n t su rfa c e , in te rio r o f th e T h ree P e a k s in tru s io n
F ig . 10. P e d im e n t s u rfa c e , in te rio r o f t h e T h ree P e a k s in tru s io n
I t is surely tru e, as we are accu sto m ed to th in k , th a t stra ig h t and angular slopes th a t ten d to m a in ta in th e ir steepness as th e y re tre a t, ra p id runoff and corresponding ex tre m es in stre a m discharge, steep slopes of tran sp o rta tio n , a n d all of the lan d fo rm s and processes th a t we associate w ith arid ity , are indeed ch aracteristic of th a t clim ate. B ut the relations o utlined here indicate t h a t climate, is an indirect cause: n e a rly all arid-hum id geom orphic contrasts are in fact due to th e absence or presence of a v eg e ta tiv e cover, and this, in l u r n , is significant chiefly because it m akes the difference betw een washing a n d creep. There are no extensive la te ra l planes at th e base of th e B lue Ridge e sc a rp m e n t or o th er m o u n ta in fro n ts of our hum id ea st, b u t l th in k th a t there w ould be, with ch an g e in rainfall, if th e m ountains were strip p ed of th eir covers of v egetation a n d th e residual soil held in place b y th e vegetatio n .
P E D IM E N T S F O R M E D C H IE F L Y OR W H O LLY BY W E A T H E R IN G A N D W A SH IN G
In Fig. 8, the esse n tial features are sufficiently clear in profile along the sk y lin e. The peaks a re q u a rtz m on zo n ite porphyry, re sista n t to erosion be
cause th e y escaped d e u te ric a lte ra tio n along th e w estern border of th e Three 94
Fig. I I . P e d im e n t, e as t side o f T h ree P e a k s in tr u s io n
Peaks in tru sio n in southw estern U tah. P edim ents slope both w est (to th e left) and east, from th e axis of th e intrusion. In th e discussion of Fig. 6 we consider
ed part of th e w estern pedim ent, cut in C retaceous sandstone, veneered b y gravel, and clearly formed prim arily by la te ra l planation.
\Yc will sh o rtly exam ine details of the m orphology of the sh arp ly c o n tra ste d type of ped im en t on the east; this d istan t view in Fig. 8 is useful to em phasize its uniform slope, slight ly concave upw ard, all th e more rem arkable because the west half of the slope, w est of the sm all knob on th e horizon, is cut in p o rp h y ry , while th e ea st half is cut in T ertiary alluvium , derived from th e east and displaced dow nw ard against the p o rp h y ry b y a m ajor B asin-R ange fa u lt.
There is no topographic b reak a t th e surface tra c e of the fa u lt.
The rock in Fig. 9, in th e interior of th e intrusion w ith in th e crescentic ridge, is altered porphyry w hich w eathers ra p id ly to crystals an d small aggre
gates of crystals. 1 had th e good fo rtune to he out in the brief cloudburst which fdled the w eathering pits w ith w a te r, and to observe th e process b y which th e surface is form ed, nam ely, th e flushing aw ay, b y sheet w ash, of gruss loosened b y w eathering since th e la st storm .
Figure 10 is an o th er view of the sam e pedim ent. This surface, sparsely covered b y grass and gruss, is about 75 per cent bare rock; it represents a clim ax of the reduction of local relief hy sh eet washing, on a surface w ith an overall slope of 2 to 3 degrees. E x actly as in the case of th e lateral p lan atio n
9 5
ty p e of p ed im en t th e slope is a g raded slope of tra n s p o rta tio n , b u t corrasion is negligible in m odelling th e surface, as is ev id en t (Fig. 11) in a m inor stream channel c h a rac te ristic of th e east T hree P eaks p edim ent. Note, for fu tu re reference, th a t the allu v iu m is chiefly gruss an d th a t th e re are no cut b an k s.
Most local sheet floods m ove th e gruss o n ly sh o rt distances, of the o rd e r of a few h u n d re d m eters. O nly th e 2- or 5 -y e ar flood flows dow n th e entire slope to its base. This flow, w hile ra re, controls a n d gives u n ity to th e entire slope.
in th a t it is produced chiefly by in te rs tre a m erosion processes, is n o t n o r
mally m an tled by alluvium , and is ra re ly p lan ar in d etail, the w e a th e rin g - w ashing p edim ent is analogous to th e s tre a m -c u t valley floor. There are of course in terg ra d atio n s, b u t th e re are basic differences: (1) th e overall slope of the w eathering—washing p ed im en t is steep b ecau se th e controlling stream g rad ien ts are steep for reasons a lre a d y sta te d , w hile th e overall slope of a peneplain is low because th e stre a m g ra d ie n ts are low ; an d (2) relief features te n d to be angular form s produced b y w ashing in c o n tra s t w ith concavo-convex forms produced b y creep.
The q u estio n is the reason for the difference in the re la tiv e im portance of planation versus w eath erin g and w ashing on different pedim ent surfaces. Tin- question is b ro u g h t in to sh arp focus, an d th e range of hypotheses is reduced, by the c o n tra ste d relations on th e opposite sides of Three P eaks. In any u n it area on th e ped im en ts th a t slope eastw ard a n d w estw ard from th a t range, clim ate, rates of w eathering, incidence of local clo u d b u rsts, and m ost o ther factors are su b sta n tia lly th e sam e. B ut it h appens, for reasons of no concern here, th a t nearly a ll—perhaps 90 per c e n t—of th e d ra in a g e of th e axial peaks flows west.
96
Here, then, is a n o th e r factor to be added to coarseness of clasts an d other elem ents favoring p lan atio n on pedim ents; the co rrasiv e power of t he stream s, and th e m agnitude an d duration of through-flow ing floods, is a function of th e areal e x te n t an d relief of th e “ m ountain m a ss” from which th e y issue.
At one extrem e, w e ath erin g and w ashing re la tiv e to planation b y powerful stream s are negligible, as on th e w est pedim ent of T hree Peaks; a t th e other, in th e absence of such stream s, the effects of w e ath erin g and w a sh in g dom i
n ate, as on the east side of Three P eaks, where th e m o u n tain h e a d w a te rs area is lim ited in extent a n d low in relief. The clim ax is illustrated (F ig . 12) by Cima Dom e in th e M ohave D esert (Sharp. 1957) w here pedim ents coalesce to form an alm ost perfect cone w ith no " m o u n ta in m ass” ; the C im a pedi
m ents are the w eath erin g -w ash in g type.
This concept, th a t, o th er things being equal, th e m echanism of origin of a p ed im en t is d eterm in ed b y th e corrasive pow er (or absence thereof) of the stream s traversing it, implies th a t th e dom inant erosional process on a given pedim ent tends to change w ith changes in the e x te n t and relief of th e n u clear m ountain m ass, clim ate, and an y other c o n d itio n th a t affects corrasive pow er. It follows th a t, while th e present is gen erally a clue to th e past, processes now in o p eratio n on a surface are no t necessarily th e sam e in degree or even in kind as those by which it was form ed. I am well aware, as has been stressed b o th in E u rope and A m erica, th a t landform s in many places are ‘•fossil” , h u t this consideration does n o t v itiate com parison of the east and west T hree Peaks P edim ents, n e ith e r of which is fossil.
T H E M OUNTAIN FK O X T
So fa r we have been concerned wholly w ith th e pedim ent su rface. Now we deal w ith another aspect of the p roblem —the te n d e n c y of the m o u n ta in front to rem ain steep, so th a t it meets the pedim ent at an angle or a co ncave upward slope of small rad iu s.
I t is first of all essential to sep a rate those m o u n ta in fronts w h ich are due sim ply to differential erosion of weak and resistan t ro ck , as shown d ia g ra m m a l- ically in A and B, F ig. 13. Ts in all of the diagram s is a m aster s tre a m , stable or slowly degrading, to which all of th e pedim ents are graded; it indicates
Kig. I.‘>. R e tre a tin g m o u n ta in fro n ts : in A , r e s is ta n t ro c k o v e rlie s w eak s h a le ; in B , re s is ta n t rock d ip s u n d e r w eak s h a le a b u ttin g core o f igneous ro c k : in C, fro n t r e tr e a ts i n ig n eo u s rock
i>7
t h a t problem s associated w ith th e rising baselevel are ruled out of ib is discus
sion; tb e pedim ents are ev ery w h ere cut on rock.
In A the m o u n ta in mass is c a p p e d by sa n d sto n e or other re s is ta n t rock re stin g on weak sh ale; re tre a t of th e scarp, and th e m aintenance of its steep
ness, is by sapping; it would occur u n d e r any clim atic conditions, a n d w hether th e w eak rock a t th e base is tr u n c a te d by a p la n a tio n surface, as in th e Book Cliffs (see Fig. 1), o r by a late m a tu r e surface of low relief, as seen in th e Cum
b e rla n d escarp m en t (see Fig. 2). T his scarp is n o t a p art of th e pedim ent problem .
F ig . 14. H o g b ack s a n d p e d im e n t, H e n r y M o u n tain s, U ta h ; fro m G ilb ert, 1877, F ig . 27
In 13 the resistan t rock of tbe m o u n ta in s and the associated hog-back forming s t r a ta , if present, d ip beneath th e w eak rock; th e scarp tends to rem ain in
In 13 the resistan t rock of tbe m o u n ta in s and the associated hog-back forming s t r a ta , if present, d ip beneath th e w eak rock; th e scarp tends to rem ain in