|
<br />58
<br />
<br />F, ZAMORA-ARROYO ET AL
<br />
<br />Transects varied in length, from 62 m near Morelos Dam (kIn 0) to 1465 m at the
<br />southernmost transect, Luis Gonzales (kIn 95), due to widening of the flood plain as it
<br />approached the intertidal zone (Fig. 4). The first transect, Flores Valenzuela (kIn 5) was
<br />dominated by 3-5 m tall P. fremontii and S. gooddingii trees (32'5% cover) and bare soil
<br />(28'6% cover). A fringe ofP. australis (25% cover) grew along the active river channel,
<br />which was not incised in this reach. The Cinco de Mayo transect (kIn 15) was wider and
<br />had a more varied flora than Flores Valenzuela. In its low zone, it was dominated by bare
<br />soil (36'1 % cover) and a mix of small S. gooddingii trees, and B. salicifolia, P. sericea and
<br />T. ramosissima shrubs. The high terrace (Zone ll) of the Cinco de Mayo transect was
<br />mainly bare soil (84%) with a few tall (to 7 m) S. goodingii trees and small shrubs. The
<br />next five sites, extending south to the rail road bridge (kIn 75), were dominated by bare
<br />soil and T. ramosissima, although strands of native trees were found in the back zone
<br />growing along the levees at the Benito Juarez and North-of-Railroad transects. The river
<br />channel was incised along this reach and a low zone was absent, The last transect, Jesus
<br />Gonzales, was in the wide part of the flood plain, where the river had split into several
<br />meandering channels with a well developed low zone. This part of the flood plain
<br />supported large numbers of willows, up to 12 m in height, although T. ramosissima was
<br />the dominant plant.
<br />Plant density data (Fig. 4) showed that juvenile and seedlings of all species were
<br />scattered throughout the transects. Although T. ramosissima and P. sericea seedlings were
<br />the most numerous plants in the plots, P. fremontii seedlings were also abundant in
<br />isolated patches at several transects. Baccharis salicijolia seedlings were a main part of the
<br />understory at some transects.
<br />
<br />Distribution of trees by size and age class
<br />
<br />Detailed tree census data at three transects showed that S. gooddingii (65% of trees
<br />censussed) was more abundant than P. fremontii (35%) in the floodplain (Fig. 5). Based
<br />on the correlation between basal diameter and number of annual rings, three age classes
<br />of trees were apparent: older trees (up to 12 m) with 12-20 annual rings, probably
<br />started during the floods of the early 1980s; younger trees (6-10 m) with 5-7 annual
<br />rings, probably started after the 1993 flood; and juvenile trees (4 m or less) with 2-4
<br />rings, probably started after the 1997 flood. The 1993 age class was the most numerous
<br />for both P. (remontii and S. gooddingii, but P.jremontii (mean age = 7,3 years) tended to
<br />be older than S. gooddingii (mean age = 4,7 years). Although mean ages were differ-
<br />ent, mean heights were similar (8'4 m and 8'1 m, respectively).
<br />
<br />Estimates of plalll distributions based on transects and aerial surveys
<br />
<br />Table 1 gives plant distributions computed by weighted average over the transects and
<br />compares estimates of aerial coverage determined by transect and aerial photographic
<br />methods. Transect results for each species are divided into understory, midstory and
<br />overstory classes based on plant height; these classes correspond approximately to the
<br />groundcover, shrub and tree classes which could be distinguished on the aerial photo-
<br />graphs. Tamarix ramosissima was by far the most abundant plant in the delta, accounting
<br />for 40% of ground cover, followed by S. gooddingii (10'9%) and P. sericea (10'3%).
<br />Transects and aerial photos gave similar estimates of bare soil (35-37%), midstory shrubs
<br />(46-53%) and overstory trees (4'5-7%) but differed in estimates of understory cover,
<br />which was higher in aerial photos than in transect results, Thickly-growing plants such as
<br />P. auszralis and P. sericea often achieved > 2 m height and were placed in the midstory
<br />class in transects, but individual plants of these species could not be distinguished in aerial
<br />photographs so they were classed as understory by the aerial survey method.
<br />
<br />
<br />REGENERATION OF TREES IN RESPONSE TO FLOOD RELEASES
<br />
<br />SO
<br />(a)
<br />40
<br />g 30
<br />0
<br />u 20
<br />10
<br />0
<br />0
<br />
<br />10 20
<br />Height (em)
<br />
<br />
<br />15
<br />
<br />(c)
<br />y = 0.287 x
<br />r' = 0'94
<br />
<br />.
<br />
<br />l:l. 10
<br />c:
<br />'C
<br />'il
<br />'"
<br />c:
<br />~ 5
<br />
<br />
<br />00
<br />
<br />10 20 30
<br />Basal diameter (em)
<br />
<br />59
<br />
<br />60
<br />
<br />30
<br />
<br />
<br />90
<br />
<br />c. "
<br />c:;:.;
<br />C-I.,')
<br />t'\,)
<br />PO'"
<br /><.:;;)
<br />
<br />_ 40
<br />
<br />~
<br />
<br />20
<br />
<br />40
<br />
<br />
<br />30
<br />
<br />40
<br />
<br />=
<br />"'40
<br />(3
<br />
<br />Figure S. (a) Height and (b) basal diameter for 264 cottonwood (-) and willow (---.) trees
<br />surveyed in the Colorado River delta, Mexico. Annual tree rings were counted in a subsample of
<br />tree c~res ~o dete~e age v. bas~l. diameter (c, . = willow, . = cottonwood). A single
<br />regressIon line passing through the ongln was fit to the data to estimate age of trees based on basal
<br />diameter. (d) The trees fell into age classes which appeared to correspond to periods of water
<br />release, marked with arrows: I is the 1997-1999 releases; 2 is the 1993 release; and 3 is the
<br />1983-1986 release.
<br />
<br />T~ble 1. Distribution of species (% ground cover) atfWng understory (< 2m),
<br />,midslory (2'}-6 m) and overslOlj' (> 6m) heighl classesforpkmls in the Colorado
<br /> River delta
<br /> Understory Midstory Overstory Total
<br />Tamarix ramosissima 1'6 (0'2) 38,5 (2-9) 0 40'1 (2'2)
<br />Pluchea sericea 1'0 (0'2) 9'3 (2'8) 0 10'3 (2'1)
<br />Salix gooddingii 0 3'9 (0'7) 7,0 (1'3) 10'9 (1.4)
<br />Populus jremontii 0'3 (0'1) 0.2 (0.1) 0'1 (0'1) 0'6 (0'1)
<br />Baccharis salicifolia 1-4 (0'4) 0'5 (0'2) 0 1'9 (0'4)
<br />Phragmius australis 0 0'7 (0'2) 0 0'7 (0'2)
<br />Comparison of methods:
<br /> Understory Midstory Overstory Bare soil or water
<br />Transects 4'3 (0'5) 53'1(3'1) 7-1 (1'0) 35'5 (1'5)
<br />Aerial survey 12-9 (1'8) 45-6 (2-9) 4,5 (0'6) 37,0 (2'4)
<br />
<br />Values are means and standard errors.
<br />Data for individual species are from nine transects along the river. The % of plants in each height class was
<br />compared for the lI'8IIsect method and by interpretation of 63 aerial photos (13); individual species could nOI
<br />be distinguished in the aerial photographic method,
<br />
|