Spatial variability of micro-climatic conditions within a mid-latitude deciduous forest

Pages 13
Views 5

Please download to get full document.

View again

of 13
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Description
CLIMATE RESEARCH Vol. 15: , Published July Clim Res Spatial variability of micro-climatic conditions within a mid-latitude deciduous forest C. S. B. Grimmond*, S. M. Robeson, J. T. Schoof Atmospheric
Transcript
CLIMATE RESEARCH Vol. 15: , Published July Clim Res Spatial variability of micro-climatic conditions within a mid-latitude deciduous forest C. S. B. Grimmond*, S. M. Robeson, J. T. Schoof Atmospheric Science Program, Department of Geography, Indiana University, Bloomington, Indiana 75, USA C 2-C 6 April CE: ml TS: bf PP: cp ABSTRACT: Micro-climatic differences within forests exert important controls, notably on understory composition, wildlife habitat, and many biogeochemical processes. This study documents the spatial variability and temporal evolution of micro-climatic conditions (air temperature, relative humidity, solar radiation and wind speed) within a mid-latitude deciduous forest, over 3 growing seasons (1995 to 1997). For all sites, in all years, conditions change markedly at the start of the growing season (Days to 15), simultaneous with onset of leaf-out and the development of full canopy cover. Below the fully developed canopy, radiation and wind speed are significantly reduced, both in magnitude and duration, while relative humidity increases. Within the forested ravines, spatial differences in daily maximum air temperature range from.5 to.1 C (average of 2 C) and in minimum air temperatures from to.5 C (average of 1 C). Within the forest, solar radiation varies by 1 W m 2 (when maximum values are W m 2 ), minimum relative humidity varies by 1 to 15%, and wind speed by.5 m s 1. Local weather conditions have a strong influence on the spatial and temporal variability of all the micro-climatic variables considered. In general, differences within the forest are greater on clear, sunny days. Once the canopy closes, the effects of topography and associated aspect/geometry dominate over micro-scale differences due to canopy structure. Less radiation, lower air temperatures, higher relative humidity, and higher wind speeds all are documented at the bottom of the ravines. The differences in micro-climatic conditions measured within the forest are of the same order as those measured in previous studies contrasting open and forest sites. Such spatial variability should be considered in studies of ecological and biogeochemical processes in secondary growth deciduous forests. KEY WORDS: Micro-climate Below-canopy Mid-latitude deciduous forest 1. INTRODUCTION * It is well-documented that forest canopies create distinct understory micro-climates (see examples and explanations in Geiger 1965, Lee 197, Oke 197, Mc- Caughey et al. 1997). Below the forest canopy, solar radiation is reduced, day-time air temperature, at least on sunny days, is decreased, humidity is enhanced, and wind speed is lowered and often effectively decoupled from regional flow (see, for example, the studies of Chen et al and Morecroft et al. 199). Such effects vary with distance from the forest edge, canopy structure, topography (which influences aspect and drainage), and soil type. These micro-climatic differences are important in terms of the structure and function of the entire forest, and are primary influences on understory composition, wildlife habitat, and many biogeochemical processes. Increasing attention is being directed to such effects and also to the potential for greenhouse-induced climate change to alter belowcanopy conditions and thus to affect the ecology and biogeochemistry of forests (Rodenhouse 1992, Root & Schneider 1993, Winnett 199). Although many studies of these effects, and predictions of their consequences, have been conducted, most rely on just a few sites and provide limited documentation of spatial variability of conditions within the forest. Commonly, paired comparisons are made, where an open control site is contrasted with a site in a forest. Notable exceptions do exist, for example, the work of Chen & Franklin (1997), who undertook Inter-Research 13 Clim Res 15: , measurements at multiple sites within a coniferous forest. However, data on the spatial variability of belowcanopy conditions for a range of forest types, which must be considered when evaluating potential climatic impacts, remain limited. The objective of this study is to document the spatial variability of climatic conditions within a mid-latitude secondary growth deciduous forest and their temporal evolution over the growing season. In addition, the effects of specific weather conditions are considered. Of particular interest is air temperature, although information on relative humidity, radiative fluxes (incoming solar and net all-wave), and wind speed is also included. Growing season conditions are considered, both because of their inherent importance and because below-canopy and open-site conditions typically are very similar when trees are not in leaf (see, for example, Morecroft et al. 199 and references cited therein). A large number of measurement sites were used during 1 yr (1997), while additional data from a more restricted number of sites for multiple years (1995 to 1997) are included to provide insight into interannual variability. The results have broad implications for research on ecological and biogeochemical processes within secondary growth deciduous forests. 2. FIELD-SITE DESCRIPTION Climatic data were collected during the growing seasons of 1995 to 1997 in the Pleasant Run Unit of Hoosier National Forest (HNF), south central Indiana (Fig. 1). HNF and surrounding state forests are part of an extensive area of deciduous forests in eastern North America (Braun 1972, Loveland et al. 1995). In addition, HNF represents one of the most continuously forested areas in the Midwestern US (Robinson et al. 1995). The study area (centered on 39 6 N, 6 21 W) consists of 2 contiguous watersheds (ravines) composed of 97 ha of secondary growth deciduous forest. Within a 3 km radius of the center of the field sites, there is over 9% forest cover. The nearest agricultural activity is over 5 km to the north. As a result, the ravines currently have a low level of direct human disturbance. Tree species in the area are diverse. Dominant canopy species are up to 1 yr old and include sugar maple Acer saccharum, American beech Fagus grandifolia, tulip poplar Liriodendron tulipifera, white oak Quercus alba, chestnut oak Quercus cataneifola, and shagbark hickory Carya ovata. The understory shrubs and trees generally consist of flowering dogwood Cornus florida, maple-leaf viburnum Viburnum acerifolium, American beech, sassafras Sassafras albidum, pawpaw Asimina tribola, spicebush Linder benzoin, sugar maple, white ash Fraxinus americana, and slippery elm Ulmus rubra. The topography consists of relatively flat ridge-tops with deeply incised ravines; elevations range from 1 to 25 m (59 to ft) above mean sea level. HNF lies just south of the limit of the Wisconsinan glaciation. 3. DATA AND METHODS 3.1. Field observations. Two sets of data are reported here. First, air temperature was measured at 21 sites in 2 ravines for the growing season of 1997 (Figs. 1 & 2). One ravine is oriented east-west (E-W ravine); the other is oriented south-north (S-N ravine) (Fig. 1). Both ravines have very similar morphometry (Fig. 2). Second, air temperature, humidity, incoming solar and net all-wave radiation, wind speed, soil heat flux and soil moisture were measured at 3 fixed sites within the E-W ravine (Fig. 1) for the growing seasons 1995 to 1997 inclusive. One site was located at the bottom of ravine, one on the east side of the ravine, and the other on the west side (Fig. 1). The first set of data, air temperature for the summer of 1997, provides detailed information on the spatial variability of conditions within the forest. The second set provides a fuller suite of meteorological observations and insight into year-to-year variability in the below-canopy conditions. Shielded HOBO sensors (manufactured by Onset, Pocasset, MA) were used for the study of spatial variability in Manufacturer s specifications on the HOBO sensors indicate a resolution of ~.2 C and an accuracy of ±.7 C. The sensors were placed along transects across the 2 ravines (E-W and S-N), on average approximately 3 m apart (Fig. 1). The sensors were mounted in the lower-most branches of trees at heights of approximately 2.5 m. Data were sampled at.2 Hz, and averaged over 3 min. For the more extensive micro-meteorological stations, deployed at the 3 sites during 1995 to 1997, temperature and relative humidity were measured with Vaisala/Campbell Scientific HMP35C sensors. These sensors have an accuracy of ±.2 C (resolution .1 C) for temperature and an accuracy of ±2% for to 9% and ±3% for 9 to 1% (resolution .1%) relative humidity. Solar radiation was measured with LiCor LI-S sensors (resolution.1 W m 2 ), and wind speed with RM Young 31 wind sentry (threshold.2 m s 1 ; resolution.15 m s 1 ). The sensors were mounted on tripods approximately 1.5 m above the ground. Ambient conditions were sampled at.2 Hz and averages recorded over 3 min. Data were recorded using Campbell Scientific (CSI) 21X dataloggers. All times referred to are Eastern Standard Time. Grimmond et al.: Micro-climatic variability within a deciduous forest 139 %h %h %h %h m 1 Fig. 1. Location of field sites and topography within Hoosier National Forest (HNF). Extent of HNF and Morgan Monroe State Forest (MMSF) shown in inset. Full stations: fixed micrometeorological equipment (measuring temperature, relative humidity, solar radiation and wind speed). Data from stations in the E-W ravine are presented in this paper. Hobo 1997: location of the temperature sensors (numbering sequence indicated) used in the intensive study of spatial variability m m Full stations % h %h%h %h %h %h 1 11 %h %h m%hh hm hh %h m%h %h 1 h Hobo 1997 Contours (1 ft interval) N All sensors were inter-compared at the beginning and end of each field season. All data have been corrected for the inter-instrument differences. After correction, the temperature sensors during the calibration periods produced mean absolute differences that were typically much less than.2 C. Thus, temperature differences in the forest are interpreted in terms of exposure and not due to instrumental error. However, it is important to note that the absolute data from the HOBO sensors and that from the fixed micro-meteorological stations are not directly comparable because of differences in the responses of the sensors and their mounting heights in the forest. Sky view factors (SVF) were determined as a measure of the foliation and architecture of the forest canopy at each site. Hemispheric photographs were taken at approximately weekly intervals using a vertically mounted camera (Olympus OM-1) equipped with a Sigma F fisheye lens. The camera was mounted on a tripod at a height of 1 m, leveled horizontally using a bubble level, and always oriented so that north corresponded to the top of the photograph. Color film ( ASA) was used. After processing, the photographic prints were scanned in black and white to create a square image (5 5 pixels) with the edge of the scanned image defined by the limits of the full hemispheric field of view. Because of the variability in Elevation (m) E-W Transect S-N Transect Distance Along Transect (m) Fig. 2. Topographic profiles of the 2 ravines (E-W and S-N orientations) within HNF (located on Fig. 1). Vertical exaggeration = 2. 1 Clim Res 15: , photographic processing and in the conditions under which the photographs were taken (for example, on occasions the foliage was very brightly lit), each image was visually inspected to ensure that the threshold detection for white (sky) and black (tree, branch, leaf) was consistent with visual interpretation. If necessary, the images were digitally retouched. The image was then analyzed using a program that calculated polar coordinates using 7 concentric rings. The black and white pixels for each ring were then summed and weighted using the method of Johnson & Watson (19) to determine the SVF of the site. A number of sensitivity analyses and independent evaluations were conducted, varying the procedures for scanning and subsequent analysis, for a wide range of photographs to check the consistency of results Cooperative climate data. To place the 1995 to 1997 field-study period in historical context, data from the Columbus, Indiana, cooperative climate station (NOAA CO-OP ID: 177) for 1995 to 1997 are compared with 1961 to 199 normals (see Robeson et al. 199 for further discussion on the selection of this station). The Columbus station is located 36 km to the east of the forest site. Overall, the historic (1961 to 199) variability of (monthly mean) maximum air temperature for Columbus is much lower during the months of June to August than during May. The latter shows a much larger interquartile range (Fig. 3a). Variability in minimum air temperature has a similar pattern (Fig. 3b), although the summer months have slightly more variation in minimum air temperature than in maximum air temperature. During the 1995 to 1997 time period, 1997 appears to be the year that differs most from typical conditions. For maximum air temperature, May, June, and August of 1997 are all well below the lower quartile of 1961 to 199 values. Minimum air temperatures during 1997 did not exhibit the same pattern as maximum air temperature. However, precipitation was fairly high during May, June, and August of 1997, indicating that 1997 was atypically overcast, with lower daytime air temperatures, typical nighttime air temperatures, and abundant precipitation. Other years have inconsistent patterns, although all years have relatively high minimum air temperatures in June. August 1995 has notably high minimum and maximum temperatures. Local weather conditions strongly influence microclimate and its variability at forested sites (Oke 197). A detailed description of synoptic conditions for the growing season of 1997, the period of most intensive study, was compiled. This information is used to interpret changes in the day-to-day evolution of belowcanopy differences in conditions. Specific details are reported below, as appropriate. T max ( C) T min ( C) (a) (b) May June July August. RESULTS.1 Canopy cover May June July August Fig. 3. Box plots of monthly averages of (a) maximum and (b) minimum air temperature for May to September of 1961 to 199 at Columbus, Indiana. The minimum, lower quartile, median, upper quartile, and maximum value of the monthly data over the 1961 to 199 period are indicated. Monthly averages for 1995 to 1997 are plotted for comparison The temporal evolution of leaf emergence and canopy closure, as measured by SVF, is fairly consistent for all years (Fig. ). Canopy cover changes rapidly between Days and 15 (late April through May): Sky View Factor Fig.. Sky view factors (fraction of hemisphere that is open sky) at the fixed climate station sites in HNF during the growing seasons of 1995 to Seasonal evolution of Vegetation Area Index (VAI) (m 2 m 2 ) for the nearby MMSF site in 199 is superimposed (data from Schmid et al. ) VAI MMSF Grimmond et al.: Micro-climatic variability within a deciduous forest 11 SVFs fall from greater than 6% (few leaves/fairly open) to less than % (full canopy cover). From late May through the remainder of the summer, values drop only slightly, down to a minimum of approximately.1. In all years, micro-meteorological and SVF measurements ended before leaf-off. Measurements of canopy structure, specifically Vegetation (leaf and bole) Area Index (VAI) at a nearby site, in the Morgan Monroe State Forest (MMSF) (Schmid et al. ), show a very similar (but inverse) temporal pattern to the SVFs documented at HNF. In 199, VAI at MMSF reached maximum values by Day 12 (May 22), after which values were more or less constant through to late August/September, when the foliage began to fall. Absolute VAI values at MMSF range from ~1.3 for leaf-off (i.e., the bole area index) to ~.5 with full canopy cover. The structure and composition of the forest are very similar at both the HNF and MMSF sites; thus, VAIs are likely to be similar at HNF..2. Below-canopy spatial variability of air temperature (1997).2.1. Trends through the growing season Daily maximum and minimum air temperatures recorded at each location within the 2 (E-W and S-N) ravines through the growing season of 1997 are plotted in Fig. 5. Maximum differences in the maximum and minimum temperatures on each transect (using all possible pairs of stations) also are shown. For the S-N transect, differences in maximum daily temperatures between sites range from a maximum of.1 C down to.7 C (average difference of 2.2 C). Differences between sites are smaller for the E-Woriented ravine (Fig. 5). There, maximum differences do not exceed 3.2 C and the average difference was only 1. C. With respect to minimum temperatures, the largest differences again are for the S-N ravine, ranging up to.5 C (although the average is only 1.1 C). For the E-W ravine, minimum temperature differences are smaller and always less than 3.3 C (average difference of 1. C). As expected, both minimum and maximum temperatures increase over the course of the growing season (temperatures are shown here through the end of August, Day 22). Clearly there is considerable variability on a day-to-day basis, which can be attributed to synoptic disturbances. The diurnal air temperature range (daily maximum daily minimum) at each site increases from approximately 1 C (as low as 5 C) before Day 1, to C for the remainder of the growing season. Differences in maximum air temperatures are greater on warmer days, and depressed on cooler, cloudy days (Fig. 5a). Once maximum air temperature exceeds 23 C, spatial differences in temperature always exceed 1 C, climbing to ~3 C at maximum temperatures of 35 C. Generally, spatial differences in minimum temperatures decrease as minimum temperatures increase. However, there is considerable scatter in the relationship, particularly at lower temperatures (Fig. 5b). With 1 notable exception (Day 19), variability between sites is lower early in the growing season (before Day ), under an incomplete canopy. Maximum temperatures are more spatially variable than minimum temperatures Space-time patterns Daily maximum and minimum temperatures at specific locations give only partial information about differences between sites. For many biological systems, it is not just the maximum or minimum temperature that exerts a control, rather it is the number of hours above or below a specific threshold that is important: the concept of thermal time (Sturman & Tapper 1996). In order to illustrate the evolution and persistence of differences between sites on a day-to-day basis, subsets of data for the 2 ravines for different periods in the growing season were selected. These periods (Days (May 25 3), Days 15 3 (June 7 ), Days (July 13 17), and Days (August 1 23), all in 1997) were chosen to represent (1) different times in the growing season and thus conditions of canopy closure, (2) days with contrasting synoptic conditions, and (3) periods with minimal amounts of missing data. The plots (Fig. 6) show the hourly data for all the sites within each of the 2 ravines (E-W and S-N transects). On the vertical axis is space (distance across the ravine), while the horizontal scale shows time (data for 3 min averages are plotted). To provide a regular spatial grid, and to fill missing data, interpolation between sites is done linearly. These plots show the time and location of the maximum and minimum temperatures and the duration above and below certain values in different areas within the ravines. Before the canopy closes, Days 15 19, temperatures in both ravines are very similar and show consistent patterns through time (Fig. 6a). Synoptic conditions over this 5 d period changed from a low pressure centered over the region on Day 15, to the passage of a cold front on Day 16, lingering precipitation on Days 17 and 1, and the development of high pressure on Day 15. For all days, temperatures in both r
Advertisements
Related Documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x