Spatial and temporal abiotic changes along a canopy to intercanopy gradient in central texas juniperus ashei woodlands.
The Edwards Plateau of central Texas comprises approximately 10 million hectares (Gould 1975; Diamond et al. 1995). It is bordered on the north by the High Plains and Rolling Plains, on the west by the Trans-Pecos Region, and on the southern and eastern boundaries by the Balconies Escarpment. In many parts of the Edwards Plateau, especially in the southern portion, Juniperus ashei is a dominant woodland species (Van Auken et al. 1981; Van Auken 1988). Juniperus ashei co-occurs with Quercus fusiformis (= Q. virginiana, Hatch et al. 1990), Q. texana and Diospyros texana in these woodlands (Van Auken et al. 1981).
Juniperus ashei is an evergreen aromatic shrub or small tree (< 9 m) with one or several trunks (Correll & Johnson 1979); it is fire sensitive (Foster 1917; Johnson & Alexander 1974; Fuhlendorf et al. 1996) and likely drought tolerant (Fonteyn et al. 1985; Wayne & Van Auken 2002). Densities of J. ashei in these woodlands in the southern part of the Edwards Plateau are [approximately equal to] 1500 trees/ha (Van Auken et al. 1981; Van Auken 1988) with an estimated canopy cover of 40 to 90% (Van Auken et al. 1981; Smeins & Merrill 1988). Carex planostachys (cedar sedge) occurs under the Juniperus canopy and is an herbaceous species with high cover and wide distribution in these woodlands (Wayne 2000).
These central Texas Juniperus woodlands are fairly open in some places and are associated with glades or small grasslands (Quarterman 1950; Baskin & Baskin 1978, 2000; Quarterman et al. 1993; Terletzky & Van Auken 1996). These open areas are more correctly referred to as intercanopy patches (Breshears et al. 1997a; 1997b; Martens et al. 1997; Reid et al. 1999; Van Auken 2000a; Ware 2002). Additionally, these intercanopy patches may have a high or low cover of herbaceous plants which appears to be related to soil depth (Terletzky & Van Auken 1996; Van Auken 2000a).
Juniperus ashei was present historically in the southern Edwards Plateau region (Foster 1917; Diamond et al. 1995), in areas that offered protection from grassland fires such as steep rocky slopes or outcrops. However, J. ashei, like many other woody species, has increased its density in grasslands over the past 100 to 150 years (Bray 1904; Foster 1917; Diamond 1997; Scholes & Archer 1997; Brown & Archer 1999). Causes of this encroachment are likely due to continuous, heavy grazing by domestic herbivores leading to reduced light fluffy fuel and decreased fire frequency (McPherson et al. 1988; Riskind & Diamond 1988; Diamond et al. 1995; Fuhlendorf et al. 1996; Van Auken 2000b). Anthropogenic factors such as elevated levels of C[O.sub.2] and climatic change are often cited (see Polley et al. 1996) as possible causes of woody plant encroachment, but are not necessary to explain these community changes (Archer et al. 1995; Van Auken 2000b). It is unknown if J. ashei is continuing to encroach into the remaining intercanopy patches, but predictive models indicate that grasslands are maintained with frequent fires (Fuhlendorf et al. 1996).
The physiology and demography of J. ashei in central Texas woodlands and intercanopy patches is poorly understood. Mature J. ashei trees exhibit low stomatal conductance and carbon assimilation during summer drought (Owens & Schreiber 1992; Owens 1996) and high water stress (Fonteyn et al. 1985; Wayne & Van Auken 2002). These trends are reversed in the fall through spring when temperatures are lower and the soil water content is higher. Density of J. ashei seedlings in these woodlands appear to be influenced by spatial and temporal gradients of abiotic factors (Wayne & Van Auken 2002).
In addition, seedling emergence is highest in early winter through early spring; with most emergences occurring beneath the woodland canopy (Jackson & Van Auken 1997), a smaller number of seedling emergences occur at the canopy edge and few in the intercanopy patch. Most J. ashei seedling mortality coincides with summer drought, with the highest mortality in the intercanopy patch, followed by the canopy edge and lowest mortality below the canopy (Jackson & Van Auken 1997; Van Auken et al. 2004). Seedling growth rates on the other hand are highest at the canopy edge and reduced under the canopy. Juniperus ashei seedling water stress is highest during summer drought (< -7.0 MPa), but recovers quickly with small rainfall events (Wayne & Van Auken 2002). Juniperus ashei seedlings at the canopy edge exhibit greater water stress than canopy seedlings during summer drought, but no data is available for seedlings in the intercanopy patches. Carex planostachys, a co-occurring sub-canopy herbaceous species appears to have a water stress response similar to that of J. ashei seedlings (Wayne 2000).
Although several studies have described plant communities in various parts of the Edwards Plateau Region (Van Auken et al. 1981; Van Auken 1988; Terletzky & Van Auken 1996; Van Auken 2000a) none have reported the cause of differences in J. ashei seedling survival or growth, but have suggested various abiotic factors. Van Auken (2000a) reported the presence of a soil depth gradient. Wayne & Van Auken (2002) indicated a xylem water potential gradient in J. ashei woodlands. It is hypothesized that gradients of other abiotic factors occur. These gradients may be responsible for the variation in species density and cover in these Juniperus woodlands. The purpose of this study is to quantify the magnitude and direction of the abiotic gradients from beneath the J. ashei canopy into the intercanopy.
MATERIALS AND METHODS
This study was conducted April through December 1997 on a 1760 [m.sup.2] site in Eisenhower Park, a San Antonio, Texas city park, in northern Bexar County (98[degrees]34'26" W and 29[degrees]37'19" N), located on the southern Edwards Plateau. The park is 128 ha and maintained as a natural area without domestic grazing (> 50 yrs, Eric Lautzenheiser pers. comm.). The site is near the Balconies fault zone and approximately 5 km east of the University of Texas at San Antonio campus. A site was selected representative of a J. ashei woodland with an associated intercanopy patch that appeared to be infrequently accessed by humans. Soil is a clayey-skeletal, smectitic, thermic lithic calciustoll (United States Department of Agriculture 2000) in the Tarrant association--rolling--with a slope of 4.5[degrees] to 13.5[degrees]. Three horizons occur that consist of shallow, clayey, weakly calcareous soil, developed over hard limestone with scattered stones and gravel. The surface horizon ranges from 0 cm to 25 cm in thickness. The subsurface is approximately 20 cm thick, heavily fractured limestone over limestone bedrock (Taylor et al. 1962). Regional climate is classified as subtropical--subhumid with a mean annual temperature of 20[degrees]C (Arbingast et al. 1976). Monthly mean temperature ranges from 9.6[degrees]C in January to 29.4[degrees]C in July (National Oceanic and Atmospheric Administration 1999). Annual precipitation in the study area is 78.7 cm, with two peaks occurring in May and September with monthly means of 10.7 cm and 8.7 cm, respectively. During the study, precipitation was above normal for 1997 at 85.6 cm (National Oceanic and Atmospheric Administration 1999), with a low of 0.0 cm in July, negligible in August, and a high of 18.5 cm in June.
The area vegetation is juniper/oak woodland representative of similar woodlands found throughout this region (Van Auken et al. 1981). The predominant woody vegetation is J. ashei and Quercus virginiana (live oak). Other woody species reported from the area are Q. texana (Spanish oak), Celtis laevagata (hackberry), Diospyros texana (Texas persimmon), Berberis trifoliata (agarita) and Rhus virens (evergreen sumac) (Van Auken et al. 1980; 1981; Terletzky & Van Auken 1996). Carex planostachys (Correll & Johnston 1979) was the dominant herbaceous species below the woodland canopy. The major herbaceous species in the intercanopy patches were Aristida longiseta (red three-awn), Bouteloua curtipendula (side-oats gramma), other [C.sub.3] and [C.sub.4] grasses and a variety of herbaceous annuals (Fowler & Dunlap 1986; Van Auken 2000a).
Measurements of surface and subsurface soil moisture, soil temperature, soil organic content and field capacity were made at each of five positions along six parallel northeasterly transects (41[degrees] azimuth). Frequency and time of measurements are indicated for each factor. The surface horizon of the soil was the upper 2 cm of soil and the subsurface horizon was the lowest 2 cm of soil adjacent to the bedrock. Each transect was 15 m in length and at least 3 m from an adjacent transect. A plumb line dropped from the outermost branch of mature J. ashei trees (2 m above the ground, located directly above each transect) was used to locate the canopy edge (drip line). Surveyor tapes were used to establish the following sampling positions: 10 m inside the canopy (canopy), 5 m inside the canopy (mid-canopy), 0 m inside the canopy (canopy edge), 2.5 m outside the canopy (mid-intercanopy) and 5 m outside the canopy (intercanopy). There were 6 transects by 5 sampling positions for the surface horizon and for the subsurface horizon. Significant differences in soil moisture and soil temperature were detected between the surface and subsurface horizons (ANOVA, SAS Institute 1989). Because the overall mean values between the surface and subsurface were small (< 2[degrees]C for soil temperature and < 5% for soil moisture) surface measurements will be the main focus of this paper.
Soil moisture was determined using the gravimetric procedure and reported as the percent water in the sample on a dry-mass basis (Pearcy 1989; United States Department of Agriculture 1996; Jackson et al. 2000). Soil samples were collected along each transect (n = 6), at each position (n = 5) for the surface and bedrock horizons (n = 2) in April, May, July, August, September, October, and twice in December (n = 8 for a total of 480 samples). Stones and organic litter were removed from the soil surface; soil samples were collected and sealed in plastic bags for transport to the lab. Approximately 40 g of soil was placed in a pre-weighed aluminum planchet, weighed and oven dried at 100[degrees]C to a constant mass.
Soil temperature was measured within two hours after solar noon on the same dates as soil moisture (with the exception of May and the latter December measurement (n = 6 months for a total of 360 samples) using 15 cm long, probe type, analog soil thermometers (Broadbent 1965; Larcher 1995). Surface temperature was measured by inserting the probe 1 to 2 cm into the soil and recording the temperature after five minutes of equilibration. Subsurface temperature was measured by excavating soil to the bedrock and inserting the probe into the lowest 2 cm of exposed soil.
Surface light levels (photosynthetically active photon flux density, [lambda] = 400 to 700 nm,) were measured at solar noon on cloudless days in July, August, October and December (n = 4 months for a total of 120 samples) with a LI-COR[R] (LI-COR Inc., Lincoln, Nebraska) LI-190 SA integrating quantum sensor. Light levels were recorded with a LI-COR[R] LI-1000 data logger in instantaneous mode with 60 s averaging at 5 s intervals. No measurements were made April through June 1997 because of overcast conditions. The quantum sensor was placed level on bare ground at each position and no attempt was made to move or disrupt any woody or herbaceous vegetation over the sensor.
The soil depth profile was measured at the conclusion of this study to minimize potential disturbances to the plants and soil of the study area (Broadbent 1965). Surface litter was removed and measurements were made along each transect at 0.5 m intervals (n = 186) using a 60 by 1 cm rebar driven vertically into the ground until it would not penetrate any deeper. The distance from the top of the rod to the ground was measured and subtracted from 60 cm to obtain the soil depth. Periodically, the rebar was re-measured to ensure the length did not change.
Percent soil organic content was determined for the surface and bedrock horizons (n = 2 for a total of 60 samples) using the loss-on-ignition procedure (Broadbent 1965; United States Department of Agriculture 1996). Excess soil collected from the December 1997 soil moisture sampling was used for the determination of the soil organic content. The soil was air-dried and sieved (#10 mesh), tested for the presence of carbonates (United States Department of Agriculture 1996), oven dried at 90[degrees]C and incinerated in a Fischer Muffle Furnace (Model 58) at 600[degrees]C for 3 hours. The test for presence of carbonates was negative.
Determination of percent field capacity (Broadbent 1965) for the surface and bedrock horizon was made using sieved (#10 mesh), air-dried soil, however only four transects were utilized (n = 2 for a total of 40 samples). The soil was placed level into a perforated aluminum planchet lined with # 1 filter paper, thoroughly wetted for 12 h and drained for 20 minutes. The soil was then oven dried to a constant mass at 100[degrees]C.
The experimental design was factorial for surface light, soil water and soil temperature (position by date). Data were transformed as needed prior to statistical analysis and analyzed with ANOVA (SAS Institute 1989). When significant main effects were detected, data were subset to examine temporal and spatial differences using ANOVA and the Scheffe multiple comparison test ([alpha] = 0.05, SAS Institute 1989). Mean surface values were pooled temporal data (all dates) for each transect position to show the overall spatial differences in surface values. Although ANOVA may indicate that a significant difference occurred the Scheffe multiple comparison test may indicate otherwise because of its conservative nature in computing the minimum significant difference (three examples occurred, SAS Institute 1989; Sokal & Rohlf 1995).
Soil depth was erratic and did not vary significantly from the canopy to the intercanopy patch (F = 0.69, P = 0.8858, Fig. 1). Mean soil depth ([+ or -] SE) ranged from 9.9 [+ or -] 2.3 cm under the full canopy to 7.1 [+ or -] 2.1 cm at the canopy edge and 10.6 [+ or -] 3.1 cm in the intercanopy patch. Soil depth ranged from zero to 40 cm and the overall mean depth was 9.2 [+ or -] 2.5 cm.
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
Overall mean soil organic content varied significantly by position (F = 8.59, P = 0.0001) and ranged from 32.0 [+ or -] 6.9% under the full canopy (Fig. 2) to 16.8 [+ or -] 2.6% at the canopy edge and 12.5 [+ or -] 0.8% in the intercanopy patch. The Scheffe multiple comparison test indicated there was a significant difference in mean soil organic content between the canopy position and both patch positions, but no significant difference between the mid-canopy and the canopy edge positions.
[FIGURE 3 OMITTED]
Overall field capacity varied significantly by position (F = 31.90, P = 0.0001) and ranged from 108.5 [+ or -] 2.8% under the Juniperus woodland canopy (Fig. 2) to 81.3 [+ or -] 2.9% at the canopy edge and 82.9 [+ or -] 1.6% in the intercanopy. The Scheffe multiple comparison test indicated that there was not a significant difference between the canopy and mid-canopy positions but they differed from all other positions. There was no significant difference between means for the canopy edge and the intercanopy positions.
The overall trend in surface light levels, soil temperature and soil moisture are best observed by pooling all surface temporal data for each position (Fig. 3). Mean surface light levels varied significantly by date and position, but the interaction term was not significant (Table 1a). Spatially, surface light levels (Fig. 3a) were lowest below the canopy and mid-canopy positions, 346 [+ or -] 99 [micro]mol * [m.sup.-2] * [s.sup.-1] and 219 [+ or -] 77 [micro]mol * [m.sup.-2] * [s.sup.-1] respectively, were intermediate at the canopy edge and highest in the intercanopy (1183 [+ or -] 149 [micro]mol * [m.sup.-2] * [s.sup.-1]). Mean soil temperature varied significantly by date, horizon, and position, with two significant two-way interactions (Table 1b). The significant interactions were date by horizon and date by position, but the three-way interaction was not significant. Spatially, mean yearly surface temperatures (Fig. 3b) were lowest at the canopy edge (27.6 [+ or -] 1.4[degrees]C), intermediate below the canopy (29.5 [+ or -] 1.8[degrees]C) and highest in the intercanopy (32.6 [+ or -] 2.1[degrees]C). Mean soil moisture varied significantly by date and position, with 3 significant two-way interactions (Table 1c). The three-way interaction was not significant. The general spatial trend for surface soil moisture (Fig. 3b) was highest values below the canopy (43.4 [+ or -] 3.0%), intermediate values at the canopy edge (33.6 [+ or -] 2.2%) and lowest values in the intercanopy (30.3 [+ or -] 2.1%).
Surface light below the canopy did not vary significantly (F = 1.98, P > 0.05) and ranged from 675 [+ or -] 309 [micro]mol * [m.sup.-2] * [s.sup.-1] in July (Fig. 4) to 39 [+ or -] 7 [micro]mol * [m.sup.-2] * [s.sup.-1] in December. At the canopy edge, surface light varied significantly (F = 3.37, P < 0.05) and ranged from 666 [+ or -] 307 [micro]mol * [m.sup.-2] * [s.sup.-1] in July to 78 [+ or -] 17 [micro]mol * [m.sup.-2] * [s.sup.-1] in December; however, the Scheffe multiple comparison test did not detect any significant differences between dates. In the intercanopy, surface light varied significantly (F = 6.88, P < 0.05) ranging from 1614 [+ or -] 302 [micro]mol * [m.sup.-2] * [s.sup.-1] in July to 479 [+ or -] 225 [micro]mol * [m.sup.-2] * [s.sup.-1] in December. The August mean of 1531 [+ or -] 243 [micro]mol * [m.sup.-2] * [s.sup.-1] was significantly different from the October and December means (Scheffe multiple comparison test), but not the July mean.
[FIGURE 4 OMITTED]
Temporal differences in mean surface temperature below the canopy varied significantly (F = 41.37, P = 0.0001) and ranged from 25.6 [+ or -] 1.9[degrees]C in May (Fig. 5a) to a high of 46.5 [+ or -] 3.3[degrees]C in August and a low of 16.0 [+ or -] 0.3[degrees]C in December. Mean surface temperature at the canopy edge varied significantly (F = 53.83, P = 0.0001) and ranged from 25.6 [+ or -] 0.7[degrees]C in May, increased to a high of 39.8 [+ or -] 2.2[degrees]C in July and a low of 16.3 [+ or -] 1.0[degrees]C in December. In the intercanopy, mean surface temperature varied significantly (F = 32.66, P = 0.0001) from 31.0 [+ or -] 0.7[degrees]C in May to a high of 48.8 [+ or -] 1.0[degrees]C in July and a low of 18.1 [+ or -] 0.8[degrees]C in December. Surface soil temperatures followed air temperatures (with a lag) and were high in July and August, and low in December. The highest surface soil temperature was 48.8 [+ or -] 1.0[degrees]C in July in the intercanopy and the lowest was in December at 16.0 [+ or -] 0.3[degrees]C under the canopy. A significant decline from the high soil temperatures seen in July and August for all positions occurred in early September ([approximately equal to] 12[degrees]C), coinciding with a 0.8 cm precipitation on the day preceding temperature measurements. After September soil temperature continued a significant decline to the low values observed in December for all positions except the intercanopy.
[FIGURE 5 OMITTED]
Temporal differences in mean surface soil moisture varied significantly below the canopy (F = 16.94, P = 0.0001) and ranged from 68.4 [+ or -] 7.0% in May (Fig. 5b) to a low of 18.9 [+ or -] 2.1% in July. Following the September precipitation, soil moisture increased to 52.2 [+ or -] 2.1%, followed by a second, but significant decline, and subsequent significant increase to 55.6 [+ or -] 6.8% after a late December precipitation event. The canopy edge and intercanopy locations also varied significantly (F = 42.5, P = 0.0001 and F = 84.2, P = 0.0001) and with the same significant decreases and increases seen below the canopy location. The canopy edge was at 47.7 [+ or -] 3.8% in May, decreased to a low of 12.5 [+ or -] 1.7% in August, increased to 39.5 [+ or -] 1.5% in September and was at 51.0 [+ or -] 2.4% in December. In the intercanopy, mean soil moisture was 43.2 [+ or -] 3.3% in May, declined to 6.8 [+ or -] 0.4% in July, increased to 38.6 [+ or -] 2.3% in September and was at 43.2 [+ or -] 2.0% in December. The overall temporal trend was high surface soil moisture in April-May and low surface soil moisture in June-August.
Soil depth in this study did not indicate a gradient from canopy to intercanopy locations. The very erratic soil depth observations from the Juniperus woodland canopy into the intercanopy patch were likely due to numerous surficial bedrock fractures (Davenport et al. 1996). At the northeastern extent of J. ashei's range, calcareous derived soils are prevalent with rock outcrops common as well as fractures and pockets of deep soil (Quarterman et al. 1993; Ware 2002). These findings in J. ashei woodlands are not unlike those of Pinus edulus/Juniperus monosperma communities of New Mexico where soil depth fluctuated from 33 to 125 cm over distances of 10 m and without any significant differences between canopy and intercanopy locations (Davenport et al. 1996). Other J. monosperma communities such as those in Arizona (Johnsen 1962) and J. pinchotii in north Texas (McPherson et al. 1988) also occur over fractured bedrock. A similar trend of shallow soils over fractured bedrock has been reported for other locations in the Edwards Plateau (Foster 1917; Taylor et al. 1962; Owens & Schreiber 1992). However, gradients of soil depth have been reported in open patch communities in central Texas (Van Auken 2000a) and deeper soils have been confirmed in woodlands compared to intercanopy patches in this same area (Terletzky & Van Auken 1996; Ware 2002).
Specific spatial abiotic gradients were found during this study for soil organic content, field capacity, surface light levels, soil temperature and soil water content. The general trend was a decrease in soil organic content, field capacity, and soil water content from beneath the Juniperus canopy into the intercanopy patch. Surface light and soil temperature followed a reverse trend with high surface light levels and high soil temperatures in the intercanopy patch and lower values beneath the woodland canopy. Temporal differences in surface light, soil temperature and soil moisture were not presented for the mid-canopy and mid-intercanopy positions. However it was noted when examining individual dates the mid-canopy differed little from the canopy, and the mid-intercanopy differed little from the intercanopy (see Wayne 2000).
While surface litter, derived from the overstory, was not measured during this study it does have an influence on soil moisture content as it is incorporated into the soil (Knapp et al. 1993; Breshears et al. 1997b). It was noted that surface litter at the study site was [approximately equal to] 3-5 cm thick below the canopy, thin at the canopy edge, and absent in the intercanopy. In addition, the trend in soil organic content appears to coincide with areas of litter deposition and greater litter depth. High amounts of organic matter have a direct relationship with the soil water holding capacity and soil field capacity (Belsky & Canham 1994; Larcher 1995; Jackson et al. 2000). An additional characteristic of surface litter is that it insulates the soil from atmospheric temperature (Knapp et al. 1993; Breshears et al. 1998). It was demonstrated that soil organic content was low or absent in the intercanopy and increased from the canopy edge into the full canopy position. Similar trends in soil organic content and litter have been noted in African savannas with high levels found proximal to overstory trees (Belsky et al. 1989; 1993). In addition, the same has been found in J. pinchotti communities on the northern Edwards Plateau (Dye 1993; Dye et al. 1995) and west Texas (McPherson et al. 1991), pinon/juniper communities in New Mexico (Davenport et al. 1996) and other savanna communities (Belsky & Canham 1994).
Surface light levels were reduced beneath the Juniperus woodland likely due to light interception by the overstory canopy. This light reduction has been reported in other J. ashei communities on the Edwards Plateau (Yager & Smeins 1999), in oak savannas on the Edwards Plateau (Anderson et al. 2001), in J. monosperma communities in New Mexico (Breshears et al. 1997b; 1998; Martens et al. 2000) and in J. virginiana communities in the eastern North America (Joy & Young 2002). In pinon/juniper communities, differences in surface light levels are related mainly to canopy/intercanopy patch variation (i.e., overstory/no overstory) (Breshears et al. 1997b). Differences in light levels are not only spatial trends but temporal trends as well; and spatial effects are modified temporally. Light levels in pinon/juniper communities varied less temporally beneath the canopy than in the intercanopy patch, but the observed temporal differences were greatest during summer and least during winter. In J. ashei communities (this study), the spatial/temporal trends in light levels are similar to those reported in the Juniperus communities in New Mexico. Temporally light was highest during summer and reduced in winter. Light levels were higher in the intercanopy patch, intermediate at the canopy edge and lower in the canopy positions, which is consistent with pinon/juniper communities in western North America.
Soil temperatures from the canopy to the intercanopy patch followed a trend similar to the surface light gradient, lower soil temperatures below the canopy and highest temperatures in the intercanopy patch. This is consistent with J. monosperma communities in New Mexico (Breshears et al. 1997a) and J. virginiaia communities in eastern North America (Joy & Young 2002). Reduced canopy soil temperature is probably related to the interception of light by the canopy reducing heating of the soil by solar radiation (Helgerson 1990; Belsky et al. 1993; Breshears et al. 1997b). In addition, surface litter probably provides insulation of the soil from atmospheric temperature (Knapp et al. 1993; Breshears et al. 1998). Conversely, the higher soil temperatures in the intercanopy patch are influenced by the lack of overstory shading and absence of surface litter (Breshears et al. 1998). Soil moisture was also higher below the Juniperus canopy and may also play a role in the reduced canopy soil temperatures. High soil moisture also appears to ameliorate high soil temperatures across the entire gradient as noted following small precipitation events (Berndtsson et al. 1996; Wayne & Van Auken 2002).
A specific temporal trend of variable soil temperature was also detected. Peak soil temperatures across the study site were reached in late August; these high temperatures were subsequently modified, [approximately equal to] 20[degrees]C, by a small precipitation event (0.8 cm) in early September followed by a continued seasonal decline, [approximately equal to] 10[degrees]C, from fall through winter. In addition, during fall and winter there was little difference in mean soil temperature along the gradient (see Wayne 2000). Pinon/juniper woodlands in New Mexico followed a similar temporal trend where soil temperatures were elevated in the intercanopy patch (relative to the canopy) during the summer and decline fall through winter (Breshears et al. 1998). Differences were attributed to seasonal air temperatures and the changing angle of the sun.
Trends in soil moisture along the canopy to intercanopy patch gradient were reversed from that described for surface soil temperatures, soil moisture was highest below the canopy and reduced in the intercanopy patch. The exception to this trend was noted after precipitation events when differences between positions were not apparent. Possible causes for differences in soil moisture have been mentioned previously; including the canopy intercepting light resulting in reduced soil temperatures and also the high litter content below the canopy further ameliorating evaporative loss (Yager & Smeins 1999; Anderson et al. 2001; Joy & Young 2002).
Some pinon/juniper woodlands (Breshears et al. 1997a; 1997b; 1998) and oak savannas (Anderson et al. 2001) have lower soil moisture below the canopy and canopy edge then the adjacent patch, but it is unclear whether this was due to canopy interception of rainfall and/or evapotranspiration. With regard to pinon/juniper woodlands the soil moisture trend varies with time such that either patch type, canopy or intercanopy, can have increased soil moisture at some point during the year (Breshears et al. 1997b). Thus, these central Texas Juniperus woodlands were dissimilar from those in New Mexico that had mostly higher soil moisture in the intercanopy. High soil organic content and litter cover below the canopy may account for greater water storage capacity (measured as field capacity, Fig. 2). Runoff during rainfall from small intercanopy areas into canopy areas (Wilcox 1994; Ware 2002) may also increase soil moisture below the canopy and redistribute sediment (and litter) from the intercanopy into the canopy (Reid et al. 1999). Temporally, soil moisture was found to be decreased from spring into summer after cessation of rainfall (from [approximately equal to] 53% to 13% soil moisture), but recharge occurred rapidly (from [approximately equal to] 13% to 44% soil moisture) after small precipitation events (Wayne 2000; Wayne & Van Auken 2002).
Throughout most of the year abiotic conditions at the canopy edge are intermediate (see Wayne 2000; Wayne & Van Auken 2002) to the canopy and patch positions. Differences in aboveground canopy cover appear to explain a considerable amount of the heterogeneity detected in abiotic factors along the gradients in these Juniperus woodlands (Breshears et al. 1997b). Soil depth was not significantly different in this study and does not seem to play a role in the abiotic gradients. Higher J. ashei seedling emergence and survival (Jackson & Van Auken 1997; Van Auken et al. 2004), and high predawn xylem water potential below the canopy (Wayne & Van Auken 2002) seems related to the reduced stress attributable to slightly lower soil temperature and higher soil moisture. Thus, the canopy likely facilitates J. ashei in the early stages of its growth and development (Callaway et al. 1996; Joy & Young 2002). However, the canopy may also hinder J. ashei seedling growth due to light interception and reduced surface light levels (McKinley & Van Auken 2004), more so below the full canopy position then at the canopy edge.
Reduced availability of water and increased soil temperature appears to hinder seedling emergence and survival, while at the same time the increased light likely promotes seedling growth (Van Auken et al. 2004). This anomalous statement appears to explain differences in survival and growth of J. ashei seedlings in these different positions along the gradient. The intercanopy position exhibited the greatest soil temperature and lowest soil moisture, which seems to explain the low emergence and survival of J. ashei seedlings at this position along the gradient. Small precipitation events during late summer also appears to be important in reducing water stress of J. ashei, and other drought tolerant herbaceous species (see Wayne 2000) in these Juniperus communities (Fonteyn et al. 1985; Wayne & Van Auken 2002).
Table 1. F-tables and significance levels from three separate analyses of variance, examining (a) light levels, (b) soil temperature and (c) % soil moisture. Variables examined include the overall model, date (D), transect position (P), soil horizon (H) and the various two and three-way interactions. Transect positions are canopy, mid-canopy, canopy edge, mid-intercanopy patch and intercanopy patch. * = P < 0.05, ** = P < 0.01, *** = P < 0.001, **** = P [less than or equal to] 0.0001 and NS = not significantly different. (a) Light levels. (b) Soil temperature. Source df F Source df F Model 19 6.67**** Model 59 36.90**** Date (D) 3 13.92**** Date (D) 5 365.00**** Position (P) 4 16.37**** Horizon (H) 1 78.55**** D*P 12 1.62NS Position (P) 4 19.80**** D*H 5 2.69* D*P 20 8.38**** H*P 4 0.88NS H*D*P 20 0.49NS (c) Soil moisture. Source df F Model 79 23.77**** Date (D) 7 218.81**** Horizon (H) 1 1.32NS Position (P) 4 39.28**** D*H 7 9.99**** D*P 28 5.66*** H*P 4 2.86**** H*D*P 28 0.53NS
The authors wish to thank E. Lautzenheiser and others with the City of San Antonio Parks and Recreation Department for their cooperation, and for permission to carry out this study in Eisenhower Park. The support of W. and L. Collenback, through a generous scholarship to the senior author is most appreciated. In addition, grants provided by the University's College of Science and Engineering, and the Division of Life Science to the senior author helped make this work possible. Lastly, we wish to thank the Center for Water Research for their support in publishing this work.
Anderson, L. J., M. S. Brumbaugh & R. B. Jackson. 2001. Water and tree-understory interactions: a natural experiment in a savanna with oak wilt. Ecology, 82(1):33-49.
Arbingast, S. A., L. G. Kennamer, J. R. Buchanan, W. L. Hezlep, L. T. Ellis, T. G. Jordan, C. T. Granger & C. P. Zlatkovich. 1976. Atlas of Texas. 5th Ed. Bureau of Business Research, University of Texas, Austin, 179 pp.
Archer, S., D. S. Schimel & E. A. Holland. 1995. Mechanisms of shrubland expansion: land use, climate or C[O.sub.2]. Climatic Change, 29(1):91-99.
Baskin, J. M. & C. C. Baskin. 1978. Plant ecology of cedar glades in the Big Barrens region of Kentucky. Rhodora, 80:545-557.
Baskin, J. M. & C. C. Baskin. 2000. Vegetation of limestone and dolomite glades in the Ozarks and Midwest regions of the United States. Annl. Mo. Bot. Garden, 87:286-294.
Berndtsson, R., K. Nodomi, H. Yasuda, T. Perrson, H. Chen & K. Jinno. 1996. Soil water patterns in an arid desert dune sand. J. Hydro., 185(1-4):221-240.
Belsky, A. J. & C. D. Canham. 1994. Forest gaps and isolated savanna trees--An application of patch dynamics in two ecosystems. Bioscience, 44(2):77-84.
Belsky, A. J., R. G. Amundson, J. M. Duxbury, S. J. Riha, A. R. Ali & S. M. Mwonga. 1989. The effects of trees on their physical, chemical, and biological environments in a semi-arid savanna in Kenya. J. Appl. Ecol., 26(3):1005-1024.
Belsky, A. J., S. M. Mwonga, R. G. Amundson, J. M. Duxbury & A. R. Ali. 1993. Comparative effects of isolated trees on their undercanopy environments in high- and low-rainfall savannas. J. Appl. Ecol., 30(1):143-155.
Bray, W. L. 1904. The timber of the Edwards Plateau of Texas: It's relation to climate, water supply, and soil. United States Department of Agriculture, Bureau of Forestry Bulletin No. 47.
Breshears, D. D., O. B. Myers, S. R. Johnson, C. W. Meyers & S. N. Martens. 1997a. Differential use of spatially heterogeneous soil moisture two semiarid woody species: Pinus edulus and Juniperus monosperma. J. Ecol., 85(3):289-299.
Breshears, D. D., P. M. Rich, F. J. Barnes & K. Campbell. 1997b. Overstory-imposed heterogeneity in solar radiation and soil moisture in a semiarid woodland. Ecol. Appl., 7(4):1201-1215.
Breshears, D. D., J. W. Nyhan, C. E. Heil & B. P. Wilcox. 1998. Effects of woody plants on microclimate in a semi-arid woodland: Soil temperature and evaporation in canopy and intercanopy patches. Int. J. Plant. Sci., 159(6):1010-1017.
Broadbent, F. E. 1965. Organic matter. Pp. 1397-1400, in Methods of soil analysis, part 2, chemical and microbiological properties, (C. A. Black, ed.). American Society of Agronomy, Madison, Wisconsin, 1572 pp.
Brown, J. R. & S. Archer. 1999. Shrub invasion of grassland: recruitment is continuous and not regulated by herbaceous biomass or density. Ecology, 80(7):2385-2396.
Callaway, R. M., E. H. DeLucia, D. M. Moore, R. Nowak & W. H. Schlesinger. 1996. Competition and facilitation: contrasting effects of Artemisia tridentate on desert vs. montane pines. Ecology, 77(7):2130-2141.
Correll, D. S. & M. C. Johnston. 1979. Manual of the vascular plants of Texas. Texas Research Foundation, Renner, Texas, 1881 pp.
Davenport, D. W., B. P. Wilcox & D. D. Breshears. 1996. Soil morphology of canopy and intercanopy sites in a pinon-juniper woodland. Soil Sci. Soc. Am. J., 60(6):1881-1887.
Diamond, D. D. 1997. An old-growth definition for western Juniper woodlands: Texas Ashe Juniper dominated or codominated communities. Gen. Tech. Rep. SRS-15. Asheville, NC: U. S. Dept. of Agriculture, Forest Service, Southern Research Station, 10 pp.
Diamond, D. D., G. A. Rowell & O. P. Keddy-Hector. 1995. Conservation of Ashe Juniper (Juniperus ashei Buchholtz) woodlands of the central Texas hill country. Nat. Areas J., 15(2):189-197.
Dye, K. L. 1993. Effect of mature Redberry juniper on associated herbaceous vegetation. MS. Thesis. Texas A & M University, College Station, Texas, 69 pp.
Dye, K. L., D. N. Ueckert & S. G. Whisenant. 1995. Redberry juniper-herbaceous understory interactions. J. Range Manage., 48(2):100-107.
Fonteyn, P. J., T. M. McClean & R. E. Akridge. 1985. Xylem pressure potentials of three dominant trees of the Edwards Plateau of Texas. Southwest. Nat., 30(1):141-146.
Foster, J. H. 1917. The spread of timbered areas in central Texas. J. Forestry 15:442-445.
Fowler, N. L. & D. W. Dunlap. 1986. Grassland vegetation of the eastern Edwards plateau. Am. Mid. Nat., 115(1):146-155.
Fuhlendorf, S. D., F. E. Smeins & W. E. Grant. 1996. Simulation of a fire-sensitive ecological threshold: a case study of Ashe juniper on the Edwards Plateau of Texas. Ecol. Model. 90(3):245-255.
Gould, F. W. 1975. Texas plants--A checklist and ecological summary. Texas Agricultural Experimental Station, College Station, Texas, 121 pp.
Hatch, S. L., K. N. Gandhi & L. E. Brown. 1990. Checklist of the vascular plants of Texas. Texas A & M University System, Texas Agricultural Experiment Station Publication MP-1655. College Station, Texas, 158 pp.
Helgerson, O. T. 1990. Heat damage in tree seedlings and its prevention. New Forests, 3:333-358.
Jackson, J. T. & O. W. Van Auken. 1997. Seedling survival, growth and mortality of Juniperus ashei (Cupressaceae) in the Edwards Plateau region of central Texas. Tex. J. Sci., 49(4):267-278.
Jackson, R.B., L.J. Anderson & W.T. Pockman. 2000. Measuring water availability and uptake in ecosystem studies. Pp. 199-214, in Methods in Ecosystem Science, (O.E. Sala, R.B. Jackson, H.A. Mooney, R. Howarth, eds.). Springer-Verlag, New York, 421 pp.
Johnsen, T. N. 1962. One-seed juniper invasion of northern Arizona grasslands. Ecol. Monogr., 32:187-207.
Johnsen, T. N. & R. A. Alexander. 1974. Juniperus L. Pp. 460-469, in Seeds of woody plants in the United States, (C. S. Schopmeyer, Tech. Coordinator). United States Department of Agriculture, Forest Service, Agricultural Handbook 450, 883 pp.
Joy, D. A. & D. R. Young. 2002. Promotion of mid-successional seedling recruitment and establishment by Juniperus virginiana in a coastal environment. Plant Ecol., 160(2):125-135.
Knapp, A. K., J. T. Fahnestock, S. P. Hamburg, L. B. Statland, T. R. Seastedt & D. S. Schimel. 1993. Landscape patterns in soil-plant water relations and primary production in tallgrass prairie. Ecology, 74(2):549-560.
Larcher, W. 1995. Physiological plant ecology: Ecophysiology and stress physiology of functional groups. 3rd Ed. Springer-Verlag, New York, 506 pp.
Martens, S. N., D. D. Breshears, C. W. Meyer & F. J. Barnes. 1997. Scales of above-ground and below-ground competition in a semi-arid woodland detected from spatial pattern. J. Veg. Sci., 8(5):655-664.
Martens, S. N., D. D. Breshears & C. W. Meyer. 2000. Spatial distributions of understory light along the grassland/forest continuum: effects of cover, height, and spatial pattern of tree canopies. Ecol. Model., 126(1):79-93.
McKinley, D. & O. W. Van Auken. 2004. Growth and survival of Juniperus ashei (Cupressacae) seedlings in the presence of Juniperus ashei litter. Tex. J. Sci., 56(1):3-14.
McPherson, G. R., H. W. Wright & D. B. Wester. 1988. Patterns of shrub invasion in semi-arid Texas grasslands. Am. Mid. Nat., 120(2):391-397.
McPherson, G.R., G.A. Rasmussen, D.B. Wester & R.A. Masters. 1991. Vegetation and soil zonation patterns around Juniperus pinchotii plants. Great Basin Nat., 51(4):316-324.
National Oceanic and Atmospheric Administration. 1999. National Climatic Data Center, Asheville, North Carolina, USA. Available: http://www.ncdc.noaa.gov/oa/ncdc.html.
Owens, M. K. 1996. The role of leaf and canopy-level gas exchange in the replacement of Quercus virginiana (Fagaceae) by Juniperus ashei (Cupressaceae) in semiarid savannas. Am. J. Bot., 83(5):617-623.
Owens, M. K. & M. C. Schreiber. 1992. Seasonal gas exchange characteristics of two evergreen trees in a semiarid environment. Photosynthetica, 26(3):389-398.
Pearcy, R. W. 1989. Plant physiological ecology: field methods and instrumentation. Pp. 84-96. Chapman and Hall, New York, 457 pp.
Polley, H. W., H. B. Johnson, H. S. Mayeux & C. R. Tischler. 1996. Are some of the recent changes in grassland communities a response to rising C[O.sub.2] concentrations? Pp. 177-195, in Carbon dioxide, populations and communities, (C. Korner & F. A. Bazzaz, eds.). Academic Press, San Diego, California, 465 pp.
Quarterman, E. 1950. Major plant communities of Tennessee cedar glades. Ecology, 31:234-254.
Quarterman, E., M. B. Burbanck & D. J. Shure. 1993. Rock outcrop communities: limestone, sandstone and granite. Pp. 35-86 in Biodiversity of the southeastern United States: upland terrestrial communities, (W. H. Martin & S. G. Boyce, eds.). John Wiley and sons, New York, 528 pp.
Reid, K. D., B. P. Wilcox, D. D. Breshears & L. MacDonald. 1999. Runoff and erosion in a pinon-juniper woodland: influence of vegetation patches. Soil Sci. Soc. Am. J., 63(6): 1869-1879.
Riskind, D. H. & D. D. Diamond. 1988. An introduction to environments and vegetation. Pp. 1-16, in Edwards Plateau vegetation: plant ecological studies in central Texas, (B. B. Amos and F. R. Gehlbach, eds.). Edwards Baylor University, Waco, Texas, 144 pp.
SAS Institute Inc. 1989. SAS/STAT User's Guide, Version 6, 4 Ed., Volume 1 and 2, Cary, North Carolina, 1848 pp.
Scholes R. J. & S. R. Archer. 1997. Tree-grass interactions in savannas. Annu. Rev. Ecol. Sys., 28(6):517-544.
Smeins, F. E. & L. B. Merrill. 1988. Long-term change in semi-arid grassland. Pp. 101-114, in Edwards Plateau vegetation, (B. B. Amos & F. R. Gehlback, eds.). Baylor University Press, Waco, Texas, USA, 144 pp.
Sokal, R. & F. J. Rohlf. 1995. Biometry: the principles and practice of statistics in biological research. 3rd Ed. W. H. Freeman and Company, New York, 880 pp.
Taylor, F. B., R. B. Hailey & D. L. Richmond. 1962. Soil survey of Bexar County, Texas. United States Department of Agriculture. Soil Conservation Survey. Washington D. C.
Terletzky, P. A. & O. W. Van Auken. 1996. Comparison of cedar glades and associated woodlands of the southern Edwards Plateau. Tex. J. Sci., 48(1):55-67.
United States Department of Agriculture. 1996. Soil survey laboratory methods manual. Pp. 169-170, in Soil Survey Investigations, Report No. 42, Version 3.0, 716 pp. Available at: http://soils.usda.gov/technical/lmm/
United States Department of Agriculture. 2000. Natural Resources Conservation Service, Soil Survey Division. Official Series Descriptions. Available: http://ortho.ftw.nrcs.usda.gov/osd/osd.html [2000, November 20].
Van Auken, O. W. 1988. Woody vegetation of the southeastern escarpment and plateau. Pp. 43-55, in Edward's Plateau vegetation, (B. B. Amos & F. R. Gehlback, eds.). Baylor University Press, Waco, Texas, 144 pp.
Van Auken, O. W. 2000a. Characteristics of intercanopy bare patches in Juniperus woodlands of the southern Edwards Plateau, Texas. Southwest. Nat., 45(2):95-110.
Van Auken, O. W. 2000b. Shrub invasion of North American semiarid grasslands. Annu. Rev. Ecol. Sys., 31:197-215.
Van Auken, O. W., A. L. Ford & J. L. Allen. 1981. An ecological comparison of upland deciduous and evergreen forests of central Texas. Am. J. Bot., 68:1249-1256.
Van Auken, O. W., A. L. Ford, A. Stein, & A. G. Stein. 1980. Woody vegetation of upland plant communities in the southern Edwards Plateau. Tex. J. Sci., 32(1):23-35.
Van Auken, O. W., J. T. Jackson & P. N. Jurena. 2004. Survival and growth of Juniperus seedlings in Juniperus woodlands. Plant Ecol., in press.
Ware, S. 2002. Rock outcrop plant communities (glades) in the Ozarks: a synthesis. Southwest. Nat., 47(4):585-597.
Wayne, E. R. 2000. Water relations of Juniperus ashei seedlings and changes in biotic and abiotic factors along an environmental gradient on the Edwards Plateau from under a Juniperus woodland canopy into an intercanopy patch. Unpublished M.S. thesis, University of Texas at San Antonio, San Antonio, Texas, 150 pp.
Wayne, R. & O. W. Van Auken. 2002. Spatial and temporal patterns of Juniperus ashei seedling xylem water potential. Southwest. Nat., 47(2):153-161.
Wilcox, B. P. 1994. Runoff and erosion in intercanopy zones of pinon-juniper woodlands. J. Range Manage., 47(4):285-295.
Yager, L. Y. & F. E. Smeins. 1999. Ashe juniper (Juniperus ashei: Cupressaceae) canopy and litter effects on understory vegetation in a juniper-oak savanna. Southwest. Nat., 44(1):6-16.
Rob Wayne and O. W. Van Auken
Center for Water Research
University of Texas at San Antonio
San Antonio, Texas 78249
RW at: email@example.com