PERMFILE65938 - Laserfiche WebLink

Home Browse Search

Table 1. Premining condition of transectsa CottonwoodRespvnf•1NaterTableDeclines 351 Premining condition (1991) Trartsects Controls 4 Mining affected 2 9 Surface (Nl 12 Density (trees/ha) Mean 1088 [tango 75-2026 Standard devtation 625 S[cm size (cmt/vee) Mean 2972 Range 605-4901 Standard deviation 1349 Depth to groundwater (m) Mean 2.40 Range 1.52-3.61 Standard deviation 0.69 'Meant are avc2gea of all sur6en with aces for each tranacct Saitterthtvaite's approximation for degrees of freedom. We conducted two types of transact comparisons. Prem- ining conditions of the tranxers were compared by examining 1991 differences in average surface values of tlepth m water table, density of cottonwood trees, and s've of cottonwood trews. We made a total of eight prerrtinirlg pairwlse tuts and used a 13onferroniadjrrsted signifimrrre level of 0.00625 for these tests. Response over the interval from the end of [he growing season in 1991 [o the end of the growing season br 1994 was examined by making between-transact comparisons of surface slues for water table level, Popular survivorship, radial stem growth increase of surviving trees, normalized branch increment of survive ing trees, and percent live crown volume of surviving veer. We made a total of ten pairwise comparisons of [Wining response and used a Bonferroni-adjusted signifi- cance level of 0.005 for these tests. A maximum likelihood logistic regression was fit to the binary survival of individual trees across all mansects in 1993 using an iterative reweighted least squares algorithm (Prot Logistic of SAS Ver. 6.09, SAS 1990). The independent variable in this regression was the absolute change in live crown volume index between 1991 and 1992, excluding those trees dying in 1992. One-way analysis of variance was used to examine differences in leaf area and specific leaf mass between aces in three defined water table classes. Results Premining Conditions In June 1991, 698 live and 57 standing dead trees were located and tagged. Stem density and stem siu 7 10 5 412 1i71 85fi 59570 95-2175 516-1240 179 ti55 394 7950 2940 9781 9895-11,982 908-5573 2744-4814 285] 1475 ]49 2.15 2.48 2.59 1.29-9.16 2.06-2.94 2.45-2.66 0.79 0.?6 0.08 (cross-sectional area) of live trees in the mining-affected transacts (2 and 3) were intermediate between control transacts 1 and 4 (Table 1). Average u-ansect densities ranged from 412 to 1088 trees/ha for transacts 4 and 1, respectively, and individual surface densities ranged from 59 aces/ha for a surface on transact 4 to 2175 trees/ha for a surface on ttansect 2, The T tests indicated no significant di$ercnce in initial density between transeet Sand the pooled controls (T= -0.058, dj= 22, P= 0.954) or between uartsect 2 and the pooled controls (T = 0.695, df = 27, P = 0.49). Avenge stem size on individual surfaces ranged from 605 ems/tree for a surface on conrol transact 1 m 11,382 cmY/tree for a surface on control transact 4. Average values for the mining affected tansects 3 and 2 were 3781 and 2940 ctllr/tree, respectively, whereas control transact 1 had slightly smaller stems at 2372 [cots/tree and control transact 4 had somewhat lazger stems at 7350 tmR/tree. Average stem sizes across surfaces were IIot signifitxndy different between transact 3 (T= 0.53, N = 22, P = 0.60) or transact 2 (T= 1.47, dj= 26.8, P= 0.15) atd the pooled convols. Average depths to the water table were similaz for the transacts before mining (Table 1), and the water table surface remained level across the valley bottom at each transact. Both control transacts had surfaces with shal- lower {1,29 and 1.52 m for tansects 4 and 1, respec- tively) and deeper water tables (9.16 and 3.61 m for ttansecrs 4 and 1, respectively) than either of the mining-affected transacts (extremes of 2.06 and 2.94 both on [ransect 2). There were no significant differ- ences in the average 1991 depths to the water table between ttansect 3 (T= 1.34, df= I9.7, P= 0.20)'or