|
<br />
<br /> TABLE 2
<br /> M~a" fempUull<FU /oe) <It tit" fill" sluti,ms (1951_/997)
<br /> TABLE J e, "'
<br /> Long""",' ^' "'
<br />Siudel/JIls: The Front Range Climate TrtlFUeCl (J9J2-1997}
<br /> J""~;vy M~. " ., " -22 -\1.8
<br /> Alti'ud~ J...~....y Mia -ILl -6.8 ," -115 -16.1
<br />S"';ODIWn< '"' laLi...de Lon&;.lIIk a....in~ JUlyM.. )1.4 26.' 24,7 '" 12.2
<br /> JUlyM;a 1).0 '" '" ., .,
<br />L008mo11' ,~ 4O"10'N 105"04'W "- ADDualnoDll<'(M.., 25-9 224 22,' 21.4 22.0
<br />..., (Pono:I.....a) 21\l~ 4U'O'll"N 105"22'01"W 1!n1-75 A"/IIIa! rang~ (Mml 24.2 19.2 19,) 16.0 '"
<br />BI(Su8.rloaf) 2591 4U'01'11'N IllS~'43.W 1!n2_14 """......Im..IM"" 18.0 ,., 12,' " "
<br />Cl (Como) ~. 40'0"2'I)'t/ IllS"3,'39"W m'nor..... M...............IMia " ,,' -11.1 -4.) -1.0
<br />01 (N;"'''') 3749 40'0:)')4"1'1 IllS"J7'orw ,"6.1\171 ~...D;""",'R..ng~ 17,) 12.7 12,6 11.7 "
<br />
<br />>01')0''0'1
<br />
<br />10103(1''0'1
<br />
<br />,ugge"mg scnsitivilY oCsucb llSlaiysistotllcaltilUdinalnngc
<br />cxlUl1incd.
<br />
<br />Chmati( Summary and Synopti( Cimatology
<br />
<br />The Colorado Fronl Range bas a continental climate, lying
<br />al 40"N wd approximately 1400 Itm nom the nearest ocean (the
<br />Pacific).ContincnlalilYoCOlDbinedwitbhighclevation,leads to
<br />large ,easonill and diumaltempenrure J'lIDges and a relatively
<br />dry climatc. At the IUgheS! elevations temperalUrel remain low
<br />all year. although sununercircu]ation is weaker and tbe chilling
<br />effect of the Cree-air is reduced. 1bc mean diurnal temperature
<br />range decreases with elevation, NiwOl Ridge beina: CAposcd 10
<br />the upper tropospheric llow at aU timCII and effective]yCAperi-
<br />encina: more maritime-like coodition.s (I"able 2). The annual tem-
<br />pcralWC raDge is relatively invarianl witb c1evation, &I]cast from
<br />AI 10 01. wngmont has a slipt.ly bigbc:r annual range, as il
<br />experiences tempcrafW'C invcniO!l.ll in arctic air masses in winler.
<br />Precipitation is broughl over the Coolineotal Divide from the
<br />we" during winter and sprinj: by upper wesledy 60w bUI mo.stly
<br />falls al higher elevations. Insprina: aDd Call "upalopc" cyclonic
<br />storms OCCW" with easterly winds somelimea bringina: heavy pce_
<br />cipitatioo (often snowfall) 10 the lowerelevatioos, while 01 can
<br />remain clear (Barry, 1992: 312). Such uorms are re'ponsible Cor
<br />the StroDg spring maximum in precipit&tioo at the lower sites
<br />(Barry. 1973), Summer rainCall is u$(I(:iatcd with convective
<br />Slorms which can be violenl with SIrona: winds and hail. The
<br />spatia] distribulion of such rainCall is unevetl and any increase
<br />in mean annual pree'p"auon with elcvlllion is slighl.
<br />
<br />136 I ARCTIC, ANTARCTIC, AND Au>tNE RE.sEARCH
<br />
<br />
<br />TABLE J
<br />
<br />puc~ntag~ of dDily lIl<uilllD and minimD (Jf ~Deh SID"on n,iIDb/e
<br />for analysis
<br />
<br /> M.. Min
<br />LOOK""'''' ~, ~,'
<br />" n,~ '"
<br />" w, W"
<br />e, w, W"
<br />0' w, w,
<br />
<br />FIGURE 1. wC<Jlim, mDp of
<br />Ih~ Colorudo Front Rail!". Th~
<br />fawrMowntarnReu<JrchStUfion
<br />dimal~ SID/Ions are indiwud by
<br />thelrewslOlIlary identification.
<br />AI = Ponduasa /1/95 Ill), Bl
<br />= Swgar/oaf (2591 m), Cl =
<br />Como (3048 m), VI '" Niwa/
<br />Ridge (3749 m). LOll8mollf
<br />(1509111) is wudas Ihebaseline
<br />h;gh p/a;/1S station.
<br />
<br />,
<br />.
<br />
<br />Longman! data (the bueline site) were obtained from the
<br />Colofildo Climate Center at Colorado State University in Foo
<br />Collins, at the web site hllp://CCC.atmos.colostate.cdul. The sta-
<br />tion is in a relatively rural area -15 kIn east oC the foothills and
<br />isrcprescntative'oCtheGreatP]ainsenvitonmenlofeasternCol-
<br />orado (Doesltcn, pers. COmln.. 1998). The Bou]der climate reo
<br />cord. nearer to the otber slotions and an obvious choice, was not
<br />used due to urblllliZilllon and numerous change, in slation 10-
<br />callan.
<br />
<br />--".
<br />
<br />/
<br />
<br />106'1"W
<br />
<br />Data and Melhods
<br />
<br />Dally maximum and minimum temperatures have been re_
<br />corded more or less continuously since October 1952. 11tis paper
<br />concentratCSon the temperatW'erecoros for 1952-1997al the
<br />foW" sitcs AI to 01 and wngmont. At the four MRS Slles lem-
<br />peratures have, for the mOSI part, been recorded using ahygroth-
<br />ennosnpb equipped with a Bourdon tube (Las]eben, 1983). TIus
<br />inslnlmentsitsinside'awell-ventilatedStevensonscreen,painted
<br />while to rellecl radiation. The thermograph charts are replaced
<br />weekly o.r monthly and daily maximum and nunimum temper-
<br />arures arctead by eye. Much dala have alrc:adybcenenteredinlo
<br />the Long-Term Ecological Research Program (LTER) dalabase
<br />(lngcaoll etal" 1997). AtAI and BI the hygrothermographwas
<br />replaced by an elecU'Onic datapod in 1987. Foounately both da-
<br />tapod llSld hygrothennograph were operated simullaneously unlll
<br />1990, so il is possible 10 co","t the datapod readings so thai
<br />they arc compatible with tllc hygrothermograph. Dclal]S of data
<br />handlina: and the production of homogenous records areoullined
<br />in Appendilt I.
<br />Data were screened forlUlusual values. Firsl, values below
<br />andabovelhresholdtempcralWCSwereflagged,andsccond,li.tr-
<br />thcr dubiOllsvalues Were identified through imersitecomparlSOn.
<br />Such values Were either corrected throuSh further investia:alioD
<br />of the oriSinal cham or,ifthis was not possib]e, excluded from
<br />further analysis. Table 3 shows the pen:entage oCpotential dally
<br />m..uima and minima at each slation remaining after screening.
<br />Lowest percentages are shown at Al and BI.
<br />
<br />Results
<br />ABSOLUTE TEMPERATURES
<br />
<br />Daily maximum and minimum lemperarures were p]oned
<br />against time Cor each si.e (Fia:', 2a, 2b). Cw-ves have been
<br />smoothed using "lowess:' alocal]y weia:hted regression based
<br />method (C1eve]and, 1979) whiCh is resistant \0 outliers and
<br />largely locally controlled. Values far away have less influence
<br />on the smoothed value oblained. l1Ic bandwidth (the perioduscd
<br />in derivins a smoothed value for each point) in this case is 1600
<br />d (-5 yr). A longer bandwidth would result in greater smooth-
<br />ing. ww values at th~ extremes oC each record are inJiuenced
<br />by lack ofdala al end points. Missing data, as at Al andBI,
<br />means thai there are gaps in the 'moothed curves within the
<br />recocd. lnsuch cases a straighl line is uscd to bridge the gap.
<br />Si&nificanttemporaJ changes in maxima and minima are
<br />Ii$led in Table 4, calculated using linear regression. AI Longmont
<br />lillle change is evident. although there is a weak. trend towards
<br />wanner maxima (significance ]evel of 4.7%) contrary 10 the find-
<br />ings oC Karl CI aI. (1984). who have shown nia:hnimc minima 10
<br />be increasilll more than daytime maxima OVer mOSI oC the U.S.
<br />Changes at the Cour MRS siles are not always oC the same sign.
<br />AI shows a strong cooling trend by day, BI a strong wanning
<br />trend both by nighland day,CI a trend towards increasing rn.ax-
<br />ima and decreasing minima (an increased diurnal range) and DI
<br />an overall cooling. Relationship, are moslly significam at 0.1%.
<br />WilflIling is thus COncenltllled al middle]evcls (aboul 2500-3000
<br />m) oC the Fronl Range and at Dl lbe lrend is reversed. in agree-
<br />menl with Brown et al. (1992) who found 01 to behave differ-
<br />emly frum Cl ami Ill<: Slali<olns uf Wolf Creek (3244 m) and
<br />ClillUtJl, (3459 m). all al lower elevauons at or below Ireeline in
<br />ioulhernamJcemraIColora<.lo.respective]y.
<br />
<br />SURFACE-BASED UPSE RATES
<br />
<br />Lapse rales were calcu]ated on a daily basis between Lona:.
<br />monl (baseline slation) and each oC the four MRS stations. A
<br />lower slope role (AI vs BI), represcntalive of the upper/lower
<br />mOnlllSlC Coresl transition, and w uppcrs]ope rate (el vs 01).
<br />representalive oCthe ,ubalpinc/alpine transition, were also cal-
<br />CUlaled. As temperaluresdccreasc with elevation the lapse rate
<br />
<br />is usulllly negative. Thus reCernl to an "increase" or "decreasc"
<br />in lapse tates can be confusing. In this ana]y,is... steCl)>l'r lotpsc"
<br />rate is takcn to mean a more negative one with a rapid decrease
<br />oClemperalUlC wuh cleVluion.and a shallower lapse rate as]ess
<br />strongly nca:ative or even positive (a lempcralUre inversion).
<br />Rates get sballoweror steeper overtime depending on the
<br />stations. The overall tempera lUre contrast between Longmonl
<br />andDl is increasing bolh by day and nigbt,but thl,consiSlsof
<br />a weakening temperature gra<.liem on the lower slope (A I versus
<br />B I. or Longmom vs B II and a steepening gradient on the upper
<br />slope (CI versus 01). The Sl(,epening in lapse rate extends to
<br />lower elevations at night (Fig. 3). Warming is concemr~ted r<<:ar
<br />Bl at nighl, an<.l al mi<.ldle levels (BI and CI) during the day.
<br />S]oj>Cs ofbcst-tit lines of lapse ratevstime are also given m
<br />Table ~. The grealcst change is expericnced On the lower slope
<br />(A I versus B I) where the dayume lapse rate weakens by
<br />O.121"C kIn~' yr.' or 5.51oC km"' oVer the whole tecord, The
<br />magnitude oCthe daytime strengthening on the upper~lope (Cl
<br />vs 01) is lin]e more than half of lhis. Nighltime changes are
<br />smallcr.
<br />The mean annual upper slope ]apse rale (ClfDI) shows lwo
<br />periodsoCconttastingbehavior(Fig, 4). From 195210 1974 both
<br />daytime and nigblllme lapse raleS are fairly conSlanl. averag,ng
<br />values of -IO"C km-' and :>-5"C km-'. respectively. After
<br />1974 there is wincreasc in lapse rale variability as well asa
<br />steepening. 1l1c early 1980.. was 0 period of especially steep
<br />lapse rates and the relative cooling al 01 (also in Fig, 2) has
<br />been identifie<.lm previous work (Lasleben. 1996), also occur-
<br />ring at other very high elevation siles in the intermountain west
<br />(wslebcn, 1997). After this anomalous period the lapse rale of
<br />minimum lemperature has weakened to values similar to those
<br />prevailing before the period. b-ut the]apse rate oC maxima has
<br />remainedslrong.
<br />This slreDgthenedlapse raleat the highest elevations is uo.
<br />usual. Many studies in the European A]ps and elsewhere ,how
<br />enhanccd warming at lligh elevauons and thus weakerlapscrales
<br />either in the free.air or ground based (Beniston and Rebelez,
<br />1996; Diu and Graham. 1996: Diu and Brad]ey. 1997). The
<br />findings in Colorado fit in with the ideas of Barry (1990). further
<br />CApoundcd by 'Nilliams et al. (1996). which suggesl that a mOI~I-
<br />er atmosphere associated with global warming could enhance
<br />convection and hia:h--elevation precipitation (includina: snowfaU).
<br />This would depress afternoon maxima preferentially at higher
<br />SileSandstrCngthendaytimC"hlp~rates.ThcincreasedaltilUdinal
<br />snowfall gradielll would enhance nighllime lapse rates. ]n this
<br />hypothesis, it is necessary lhat enhanced precipilalion and eon-
<br />vcction areconcentraled at high elevations. as otherwise. a1loth-
<br />er things being equal, a lDoisler atmosphcre would be as~ociatcd
<br />with weaker]apse rates. al least in lhe free.air. ..
<br />Taking lhe surface upper slope lapse rale (Cl vs Dl). the
<br />daytime strengthening is significant at 1% from September to
<br />May inclu~ive. In contrast. nighuime strenglhening is Oldy slg-
<br />nificanl in 2 mO (October and November) and then only at 5%.
<br />The daytime chanJle corresponds with mOOlhs wllh regular snow
<br />cover. The absence oC a summer re]alion.hip suggests lhat en.
<br />hanced high-elevation convection is unlikely to be a major in-
<br />fluence. However. Increased hlgh-devallon snowfall associaled
<br />with changes in circulation could be a pos,ible cause (Barry.
<br />1990). Alternatively, since mllow is weak in sununer bUI not in
<br />other seasons, the lapse rate change could be aSSOCialed with
<br />chana:es in the vertical lemperarure suucture of the mid-latitudc
<br />westerlies, which dominale the higher elevations for rnost of the
<br />year.
<br />AI lower elevations within the montane forest1.ones. wealr;:-
<br />
<br />N. PEPIN I 137
<br />
|