Back to Forests of the Central Appalachians | Feedback

Soil Temperature and Forest Type, III

RF Mueller, March 2004

Abstract

    Soil and water temperatures were determined for the third season (2003) at Ramsey's Draft, the Shenandoah Valley Oak - Hickory Ridge and Mesic Slope Forests and other locations in the Central Appalachians. The results are of particular interest due to an unprecedented rainy season in which the mean yearly precipitation was virtually doubled in many areas.

    While the soil and water temperatures at Ramsey's Draft adhere closely to those of the more "average" year of 2001, early growing season soil temperatures of the Oak-Hickory Ridge and Mesic Slope Forests lie substantially below those of 2001. This is especially true of the Mesic Slope Forest, which is dominated by northerly aspects and has soils rich in clay and organic matter, in contrast to the sandy soils at Ramsey's Draft. It is concluded that although the Ramsey's Draft soils by and large have greater access to moisture than those of the Slope soils, as a consequence of their proximity to the Stream / spring related water table, their sandy nature allows more rapid drainage of excess water to occur. Based on these data, as well as those from other Appalachian forests, it is concluded that the quantity of precipitation, particularly in the growing season, may have an important effect in certain forests by lowering soil temperatures.

Introduction

    Our data from 2001 and 2002 (Mueller, 2002, 2003) are consistent with a strong relation between certain forest types and soil temperatures. Also of interest is the response of soil and water temperatures to varying seasonal air temperatures and soil moisture conditions throughout the 2001 and 2002 growing seasons. Particularly marked was the response of the 2002 data to unusually high air temperatures and the severe drought conditions in that season.

    For descriptions of methods, terrain, floras and forest types the reader is referred to the previous work ( Mueller, 2002).

    Conditions in 2003 were unprecedented in the region, particularly with respect to the amount of rain, which was virtually double the yearly average in many parts of the Central Appalachians. Moreover, substantially lower than average air temperatures also accompanied this precipitation. However we shall see that there apparently is no simple relation between air and soil temperatures, with the relation being also dependent on soil character.

    While the data for 2003 are not as extensive as those of the previous years-partly as a consequence of weather conditions-they do span the most critical part of the growing season.

Ramsey's Draft

    The 2003 soil and water temperature data from Ramsey's Draft are shown in Table 1, and the composite data for 2001 (Mueller,2002) and 2003 have been plotted in Figure 1.

    It should be mentioned that the last three surveys (6-10-03, 7-17-03, 8-23-03) are from only the first five soil temperature stations. Notable also is the similarity in the temperature spreads to those of the two previous years.

    Figure 1 shows that despite the universally wet conditions in 2003, both the soil and Deep-source Spring water temperatures are very similar to those of the more "normal" year 2001, and in the case of the Deep-source Spring the data points essentially fall on the same smooth curve, given an uncertainty in precision of + or- 0.5 deg C. Additionally, the last 2003 Spring data point (for 8-23-03) falls late enough in the season to show the difference in character of this curve from that of the drought period of 2002., which plunges to lower values this late in the season as a result of the truncation of the shallow water component of the Spring (Mueller, 2003). While the 2003 soil temperature data points do not all fall on lines joining the 2001 data points, they come close, except for the last ( 8-23-03 } point, which seems a little high, but may still approximate a hypothetical 2001 value, given the absence of data for the latter year here.

    Date Temperature
    (Degrees C)
    Spread
    (Degrees C)
    4-15-03 soil 9.0
    spring 8.0
    stream 9.0
    8.5 - 10.0
    5-20-03 soil 11.5
    spring 11.0
    stream 11.0
    11.5 - 12.5
    6-10-03 soil 14.0
    spring 12.5
    stream 14.5
    13.5 - 15.0
    7-17-03 soil 17.5
    spring 15.5
    stream 17.0
    17.0 - 18.0
    8-23-03 soil 19.5
    spring 17.0
    stream 18.5
    19.0 - 20.0

    Table 1: Soil and water temperatures and temperature spreads for Ramsey's Draft. Soil temperatures are median values; water temperatures are single values. The last three surveys (6-10-03, 7-17-03, 8-23-03) included only the first five temperature stations.

    Figure 1: Soil and Deep-source Spring temperatures for Ramsey's Draft. The 2001 data are indicated by "O" (soil temperatures) and "@" (Spring temperatures). The 2003 data are indicated by "X" (soil temperatures) and "+" (Spring temperatures). Soil temperatures are median values; water temperatures are single values. The 2001 and 2003 data sets for the Deep-source Spring are fitted by a single smooth curve.

Oak-Hickory Ridge Forest

    The 2003 soil temperature data for the Oak-Hickory Ridge Forest are shown in Table 2. These data have been plotted in Figure 2 with the more "normal" 2001 trend shown for comparison. It is clear from these data that the 2003 soil temperatures fall below those of 2001, with the largest difference at the start of the record in April. Because of the paucity of data, little can be said about the period after June, except that the single July point seems to indicate convergence with both the 2001 and 2002 (Mueller, 2003) data in mid summer.

    Date Temperature
    (Degrees C)
    Spread
    (Degrees C)
    4-2-03 9.5 8.5 - 11.0
    4-13-03 8.0 7.0 - 11.0
    4-28-03 11.5 10.0 - 14.5
    5-14-03 14.5 12.5 - 16.0
    6-6-03 14.5 13.5 - 16.0
    6-23-03 16.5 15.5 - 18.5
    7-21-03 20.0 19.0 - 20.5

    Table 2: Median soil temperatures and temperature spreads for 2003 in the Oak-Hickory Ridge Forest

    Figure 2: Median soil temperatures (crosses) for 2003 in the Oak-Hickory Ridge Forest. The 2001 soil temperature trend is shown for comparison.

Mesic Slope Forest

    The 2003 soil temperature data for the Mesic Slope Forest are shown in Table 3 and these have been plotted in Figure 3, with the 2001 trend shown for comparison. It should be noted that four surveys, as indicated in the Table caption, are based on only the first five stations with a northerly aspect, and omit the two ridge crest stations, which have nearly flat aspects and generally register the highest temperatures.

    It is clear by comparison with Figure 2 that the soil temperatures for this forest show a greater difference from the 2001 trend than those of the Oak-Hickory Ridge Forest. Also of interest is the apparent replication of the early May maximum that appears in both 2001 and 2002 and which presumably marks the end of the bare-branch period of rapid warming.

    Date Temperature
    (Degrees C)
    Spread
    (Degrees C)
    4-12-03 6.5 6.0 - 7.5
    4-24-03 9.0 8.5 - 10.0
    5-3-03 13.0 13.0 - 13.5
    6-2-03 12.0 11.5 - 14.0
    6-22-03 15.0 15.0 - 16.5
    7-20-03 18.0 17.5 - 19.0
    9-16-03 16.5 15.5 - 17.0

    Table 3: Median soil temperature and temperature spreads for 2003 in the Mesic Slope Forest. It should be noted that surveys done on 5-3 - 03, 6-22-03, 7-20-03 and 9-16-03 are based on only the first five stations.

    Figure 3: Median soil temperatures (circles) for 2003 in the Mesic Slope Forest. The 2001 soil temperature trend is shown for comparison.

Additional Soil Temperatures from the Mountains

    Additional single-value soil temperatures for four locations in the Virginia and West Virginia mountains are shown in Table 4. These values have been plotted in Figure 4 with the composite 2001-2003 soil temperature trend for Ramsey's Draft. The reader is referred to our descriptive sections on the Virginians for Wilderness Web Site for these locations.

    Chimney Hollow is actually represented by two identical values that were obtained at slightly different locations. Also, since this hollow is located near Ramsey's Draft, and like the latter, is subject to cool subsiding air, the similarity in values is not unexpected. While the two areas differ in elevation, the fact that Ramsey's Draft opens to the southwest and Chimney Hollow to the northwest, probably tends to compensate for the lower elevation of the latter.

    The values of the July-early August data points for the East Fork of the Greenbrier and Fanny Bennett seem a little high given the distinct boreal character of much of their floras. However, the late August values for the East Fork of the Greenbrier and Locust Spring / Laurel Fork (Virginia) are more in line with what is expected. The datum from Locust Spring / Laurel Fork is in fact from Red Spruce Forest, a habitat known for its frigid soils (Fleming and Moorhead, 1996).

    Date LocationTemperature
    (Degrees C)
    6-11-03 Chimney Hollow 14.0
    7-31-03EFK Greenbrier 17.0
    8-1-03 Fanny Bennett 17.0
    8-25-03 EFK Greenbrier 16.0
    8-25-03 Laurel Fork 15.5

    Table 4: Single value soil temperatures from several locations in the Virginia and West Virginia Mountains.

    Figure 4: Single value soil temperatures (circles) from several locations in the Virginia and West Virginia mountains. The composite 2001 and 2003 soil temperature trend from Ramsey's draft is shown for comparison. Note that the "Laurel Fork" referred to here is located in the northwest of Highland County, Virginia.

Conclusions

    Of considerable interest is the difference in response of the soil temperatures at Ramsey's Draft from those of the Shenandoah Valley Oak-Hickory Ridge and Mesic Slope Forests. It seems likely that this difference is due largely to differences in soil character. The soils at Ramsey's Draft are dominantly sandy, occupy interstices between coarse alluvial cobble and are not particularly rich in organic matter. By contrast the Shenandoah Valley forest soils, being derived from limestone, are far less stony and tend to be rich in clay and organic matter, This is particularly true of the Mesic Slope Forest, which, in addition, is dominated by northerly aspects. The Ramsey's Draft soils are dominantly low-lying and not far above the steam and spring- governed water table, which confers a constant supply of moisture not available to upland soils. And this characteristic is compatible with the common occurrence of such moisture-demanding species as Wood Nettle (Laportea canadensis), Blue Cohosh (Caulophyllum thalictroides) and Horse Balm (Collinsonia canadensis), of which only the latter is found in even the moistest parts of the Mesic Slope Forest. However, the sandy and stony character of the Ramsey's Draft soils also enables excess water to drain away rapidly, in contrast with the Valley soils, in which the abundant clay and organic matter acts to retain excess moisture. As a consequence we should not be surprised that abundant precipitation results in a marked temperature lowering in the Valley soils but has little or no effect in those of Ramsey's Draft.

    The similarity of the 2001 and 2003 Deep-source Spring data, that enables their trend to be approximated by the same smooth curve, is more difficult to explain. However it may be that in more "normal" years such as 2001, as well as in years of abundant moisture such as 2003, the Spring flow is considerably influenced by a relatively shallow component which is diminished only in severe drought years such as 2002.

    As indicated previously (Mueller,2003), a high yearly precipitation may have a profound effect on the flora. This is indicated by the occurrence of such boreal species as yellow Birch (Betula alleghaniensis) in mixed mesophyte forests in the Cumberland Mountains (Braun, 1950) and other locations of the Southern Appalachians (Newell et al, 1997 and Newell and Peet, 1998) at moderate elevations. In the case of the virgin old growth forests described by Braun, this effect may have also been enhanced by the tall canopies which provided an additional cooling effect of shade cast through a greater vertical distance. Also, a number of occurrences described by Newell et al and Newell and Peet fall on valley flats where they may have been influenced by subsiding cool air. It would be of further interest to determine if the low 2003 soil temperatures in the Mesic Slope Forest occasion the appearance of additional cool-climate species in that forest.

Acknowledgements

    The writer is pleased to acknowledge the continued support of Patagonia Corporation and our fiscal sponsor Save America's Forests. He is also grateful to Fred Huber and to Ted Green and Jim Morrow of the US Forest Service for providing documents used in this study, and to Robert Hunsucker for all his instructions on the Appalachian flora.

References

    Braun, E. Lucy (1950) Deciduous Forests of Eastern North America. Macmillan Publishing Co., New York, N. Y.

    Fleming, Gary P and William H. Moorhead (1996) Ecological Land Units of the Laurel Fork Area, Highland County, Virginia. Technical Report 96-08, Virginia department of Conservation and Recreation, Div. of Natural Heritage, Richmond, Virginia 114 pp plus appendices.

    Mueller, R. F. (2002, 2003) Soil Temperature and Forest Type and Soil Temperature and Forest Type II. Forests of the Central Appalachians Project. Virginians for Wilderness Web Site.

    Newell, Clair L., Robert K. Peet and Jonathan C. Harrod (1997) Vegetation of the Joyce Kilmer - Slickrock Wilderness, North Carolina. Report to the U S Forest Service, Dept of Agriculture, National Forests of North Carolina.

    Newell, Clair L. and Robert K. Peet (1998) Vegetation of the Linville Gorge Wilderness, North Carolina. Castanea 63(3) 275-322.


The graphs on this page are plotted directly from XML data using a custom Flash component whose source code is available for non-commercial purposes. If you'd like to use it for your own data-plotting, download the .FLA file. Questions about this code should be directed to Gus Mueller.


Back to Forests of the Central Appalachians | Feedback