Laserfiche WebLink
ache ve using state of the ar[~e ld oroce dure s. <br />I[ offs re [he eoporcuni ry for pro:essf onal Snpu[ <br />an a Conrinuicq halts and to pl:,n !_o ro~reme n[9 on <br />[he reclaimed lands :ac r,er than simply crying co <br />1ml rate the prod ne conci;!ans. <br />P re mf ee Re grne ra tf o^ Race of 8h n:bs <br />Mother concept chat has bee^ considered fa: <br />estahlLs `.1 ag a shrub ,: ~:a :;v scanca rd a 'w led on <br />the es[ica[ed re wens ration rare of shrub s[e¢s. <br />By deflni ng shrub dz rs![y aces rdi r.g to size <br />class, [he ra [e of re?z ne ration can be [heore[1- <br />eally es[ica;ed. A shrub dz nsf ty standard would <br />[hen be set Coln olden[ with [he esri mated rate of <br />regene ra tf on. This approach 1s unproven and very <br />the ore[1 ca 1. Further, 1[ !s al mad a[ duplf ca;i ag <br />[he premine condition without considering [he <br />po[ea[ial [o i¢p rove [he condition. <br />Correlation Areas <br />The Correlation area approach Ss a modiflca- <br />tfon to Che cocparison area con cep[ [hoc ¢ay be <br />usable vhe re 1t is desirable to imitate the pre- <br />mine condi[1 ons. We used rr_ans from [he six <br />comparison areas and the two plan[ comnunt ties on <br />premine areas discussed prz viously (Table 1) [o <br />date coins [he Pca rs on Product `foment Coeffiden[ <br />of Correlation (r). the high co rrc la[f on bervicen <br />[he comparison and premine areas led ro an <br />analyses by linear regression modeling (Sable 8). <br />111th this approach, an ope rotor mould predict [he <br />vegetation success standards for any gl wen year <br />from data obtained in [he same year from the <br />'t omparlson areas. If the values nea su red on the <br />reclaimed areas meet or exceed the lover predic- <br />tion limits of the model ac a given confidence <br />level, vegecaclon success is indl Ca [ed. This <br />method is no[ currently Sn use by any of [he <br />regu la COry agencf es ve reviewed. A li ml[a[ion <br />is Char the accuracy of [he nodal may decrease £f <br />values obtained from [he tonpa ri son areas are <br />ou[aide the range used [o build the model. A <br />ma )or advancaee, eve r, is [hat ff a re ll ab le <br />ce lacionshlo can be described, a str.g le tomcari- <br />son area can coo reivably be used co se[ revegera- <br />clon success s[acda rds for wre Chan one piano <br />cocunicy, or :an qe ales. TaS le 9 shove the <br />result cf re qre sling [he preduc[Son •+a foes from <br />[he Chrez pra.-~ce p!enc coam unicles against the <br />values frm¢ each of the cot oa r!son areas. Lf [he <br />re lotions Sips con:lnue ro held true So future <br />years, Chen a stnq le come la rion area such as <br />you nratn shrub ':o. or Sa sa-gra ss No. 3 could be <br />used to predict [he re ve ge cation sue Less stan- <br />dards for a'_1 three plan[ cormunl [Se6 following <br />mining and re ve geca[f oa. <br />Although the data analyzed ind feats a strong <br />1l near relationship, a '[~ test of the correla- <br />tion coefflcle nts did not shoe they are slgnlfl- <br />can[ at [he 95S confidence level. this Ss <br />because only three years of da [a ve re used to <br />build the node 1. I[ would, therefore, be prema- <br />ture [o Conclude that [he correlation area ton- <br />cep[ is valid. <br />SUfff/.RY AND CONCLUSIONS <br />Revegeta[i on should be evaluated on the <br />basis of its utility far the poscmi nl ng land use. <br />This of 11 no[ always allow testing on the lu sis <br />of how closely S[ rsembles [he premint ng condi- <br />tions. <br />Almost all [he ry r:ent approeche9 for defining <br />rave geta [1 on success s[anda rds are directed <br />[o~•a rds Smi [a tang ,..z premining condicf ons. This <br />does not ne cessarily result in a standard chat <br />bas[ serves [hc pos_minf ng land use ob]et[Ivcs. <br />When standards must .e baled on falta[Sng pre of ne <br />ccndi[ions, i[ disroura gas lnnovaCive land plan- <br />ning and prevents im_roveme nts that could enhance <br />the utility of the ~nd. For example, operators <br />may be required to rr.rurn dense stands of shrubs <br />in [he ea me areas wr.e re land ma nags r5 have been <br />Table 8.-Linear models and cor rela tlon coefficien cs `cr mean <br />annual production from three premine plant can. .:r.tc!es (y) <br />and mean lannual production from corresponding camparis on <br />areas (x). <br />Plant community Linear model Corre lecion <br /> (y - a + bx) coeff lcicn[ <br /> r <br />Sage-grass y - -6.82 + (.21x 0.96 <br />Mountain shrub A y - -8.90 + I.38x 0.97 <br />Nounc nfn shrub 8 y - -8.16 + 0.99x 0.96 <br />11980, 1981 and 1982 produ c[1 on data Croa a coal surface a1~~r .~. no r[hves[ <br />Colorado <br />- /O . <br />T' <br />