FALSE RING FORMATION IN BALDCYPRESS (TAXODIUM DISTICHUM)
SAPLINGS UNDER TWO FLOODING REGIMES
P. Joy Young ~, J. Patrick MegonigaF, Rebecca R. Sharitz ~ and Frank P. Day 3 ' Savannah River Ecology Laboratory
Aiken, South Carolina 29802 2 Dept. of Botany, Duke University
Durham, North Carolina 27706 J Dept. of Biology, OM Dominion University
Norfolk, Virginia 23529
Abstract: Baldcypress saplings were subjected to two flooding regimes of continuous and periodic inundation for three years to allow comparison of ann.ual ring characteristics. Basal stem discs were examined for the number and nature of intra-annual response features, such as false rings. The formation of latewood was also compared for trees from each flooding regime. Radial growth was significantly greater in saplings from the continuously flooded treatments for years 2 and 3. No significant difference was found between treatments for the total number of response features present; however, periodically flooded saplings developed a sig- nificantly higher number of major false rings. Trees from this regime also consistently produced a greater number of thick-walled, latewood-type cells associated with false rings and annual rings.
Key Words: annual rings, baldcypress, false rings, flooding, growth response.
INTRODUCTION
Tree rings have long been used to determine the ages of trees and to reconstruct climatic histories (Stokes and Smiley 1968, Fritts 1976). Tree species used in climate history studies were often chosen because they were long-lived and had well-defined latewood mar- gins, few false or missing tings, and were climatically sensitive (Stokes and Smiley 1968, Fritts 1976). Al- though baldcypress, Taxodium distichum (L.) Rich., is the longest living species in the southeastern United States with trees greater than 1700 years old (Stahle et ai. 1985a, 1988), dendrochronologists historically be- lieved baldcypress to be unsuited for use in climatic studies. This belief was due to its propensity to form false rings, to have locally absent or merging rings (Fritts 1976, Bowers 1981, Stable 1985a, Duever and Mc- Collom 1987), and also because it was thought to be climatically insensitive due to its wetland habitat. Al- though response features such as false rings or locally absent tings can present problems in dendrochrono- logical analysis, especially when ring counting (I_arson 1956, Beaufait and Nelson 1957, Fritts 1976, Bowers 1981, Duever and McCollom 1987), recent studies in Arkansas, North Carolina, and Florida have shown the feasibility of using baldcypress in ecological studies and
293
in creating long, climatically sensitive chronologies (Stahle et. al 1985a,b, 1988).
Studies concerning the formation of discontinuous or locally absent rings in trees have been conducted (Larson 1956); however, little work has been done to determine the specific causes of false rings. Defoliation, insect infestation, and drought have been cited as el- ements that decrease growth of baldcypress (Larson 1963, Zahner et al. 1964, Fritts 1966, Goyer and Len- hard 1988, Goyer et al. 1990), but a direct cause-effect relationship in the formation of false rings has not been established (Stahle et al. 1985a). If the factors that cause response features such as false tings could be determined, the presence of these features might allow the identification of past intraseasonal events respon- sible for such variations in growth.
We examined the intra-annual characteristics of the growth rings of two- and three-year-old baldcypress saplings subjected to two flooding regimes and com- pared the effect of flood regime on the formation of response features and annual ring formation.
METHODS
Baldcypress seeds were germinated in early 1986.
Twenty-five seedlings were transplanted into each of
294 WETLANDS, Volume 13, No. 4, 1993
I I L . 1 I /
o~m l c.', z~,',', ~,'.
I 4¢m 5 m 'l c m = 0 . 0 7 m m
Figure 1. Cross-section of a continuously flooded, two-year- old sapling stem showing a minor response feature (125 x ).
two mesocosm units in the outdoor rhizotron facility at the Savannah River Ecology Laboratory', Aiken~
South Carolina. Each unit measures 2.8m x 2.8m × 1.5m in depth. These large mesocosms permitted rel- atively free root expansion and allowed for the control of flooding as well as ensuring that the treatments were influenced by the same environmental factors.
One mesocosm was continuously flooded to a level 10cm above the soil surface. The second was period- ically flooded according to a schedule designed to ap- proximate natural cycles (Megonigal and Day 1992):
Jan.-Feb., 10cm above the soil surface; March-May, 20cm below the soil surface; June--Sept., 50cm below the soil surface; Oct.-Dec., 20era below the soil surface.
Saplings from each treatment were harvested in late October according to the following schedule: periodi- cally flooded--9 in 1986, 8 in 1987 and 8 in 1988;
continuously flooded--8 in 1986, 9 in 1987 and 8 in
•
I
4~ ~ ~ ~,.
I I- .
Ocm lcm 2cm 3cm
l I
I I
4cm 5crn
l c m = O . O 7 m m
Figure 2. Cross-section of a periodically flooded, two-year- old sapling stem showing a moderate response feature (a) accompanied by resinous cells (b), (I 25 x).
1988 (Megonigal and Day 1992). Only those trees har- vested in 1987 and 1988 were used tor growth ring analysis. The saplings were cut and a lcm wide disc was removed from each stem at a point approximately 20cm above ground level. Discs were sanded with pro- gressively finer grades of sandpaper, beginning with 320 and ending with 600. Discs were then buffed using lambswool.
Lines indicating the cardinal directions were lightly drawn with pencil on the surfaces of the discs. Ori- entation of the lines was determined from a reference mark that had been made on each stem prior to har- vest. The entire surface of each disc was scanned using a Hensen University Model Incremental Measuring Machine with attached stereo dissecting microscope.
The position, distance from the cambium, and com- pass direction of each response feature was recorded.
4& 5&
l c m = 0 . 0 7 m m
Figure 3. Cross-section of a periodically flooded, two-year- old sapling showing a major false ring (125 x).
These features were defined and grouped into three categories:
1) minor--not a false ring--any variance in the cells not generally visible to the unaided eye such as cell size, and cell wall thickness, and also resinous areas and compression wood (Figure 1)~
2) moderate--double or false rings characterized by more distinct variances in cell size and cell wall thickness but not having a distinct terminal edge (Figure 2a); and
3) major--false rings that at some point along their length had characteristics typical of annual rings-- a gradual decrease in cell size and increase in cell wall thickness with a distinct margin defining the terminal boundary of the feature followed by thin- walled cells with large lumens (Figure 3, 4).
These categories were further divided into contin- uous and discontinuous, depending on whether the
I I
Ocm l c m
l c m = O . O 7 m m
Figure 4. Cross-section of a continuously flooded, two-year- old sapling showing the boundary of an annual ring (125 x).
feature continued for the circumference of the disc or was present for only a segment of the circumference.
In addition, the distinctness of'the annual ring bound- ary was recorded. Boundaries were considered distinct if the latewood cells comprising the terminal rows were similar in appearance and were tbllowed by clearly distinguishable, large lumened earlywood cells.
A t-test was used for comparisons between flooding regimes for the type of features as well as the total number of features. The relation between flooding re- gimes and the number of terminal latewood cells was evaluated by a 2 × 3 Chi Square test.
RESULTS AND DISCUSSION Response Features
No significant differences in the total number of re- sponse features existed between flooding regimes (Ta-
296 WETLANDS, Volume 13, No. 4, 1993 Fable I. Differences in the number of growth features in
continuous flooding versus periodic flooding treatments.
Minor IVloderatc Major Total
1986 ns ns ** ns
1987 ns ns ** ns
1988 ns * ** ns
1986-87 ns
1986-87-88 ns
ns = no significance.
* = significant (p < 0.05).
** = highly significant (p < 0.01 ),
ble 1); however, significant differences (p<0.01) were noted when only the number of major false rings was considered (Table 1). Saplings from the continuously flooded treatment did not develop major false rings, whereas saplings from the periodically flooded treat- ment developed an average of 1.76 major false rings per tree for the three years of growth (Figure 3, Table 2). There was no pattern in the incidence of continuous response features or false rings, as the majority of re- sponse features in all categories were discontinuous.
Continuous response features, however, appeared to be more likely within specific individuals than within treatments or feature categories.
Significant differences (p<0.05) occurred between flooding regimes in the number of thick-walled, late- wood-type cells associated with response features.
Thick-wailed cells comprising false rings in trees from the periodically flooded regime appeared in greater number, displayed a gradual but marked decrease in size, and had thicker cell walls than those from the continuously flooded regime.
Despite different flooding regimes, most response features were accompanied by resinous cells (Figure 2b), and there were some features that were pro-
nounced and common to the majority of trees in the study. The most obvious of these were: 1) a response feature attributed to transplant shock, 2) an area of compression wood in the northeast quadrant of the 1986 annual growth, 3) a series of five response fea- tures, or false rings, in the 1987 annual growth, and 4) a 4-6ram band of earlywood in 1987 terminated by what appeared to be a false ring in gross examination.
Under magnification, however, this feature appeared to be only a minor decrease in the size of the large lumened, thin-walled cells that made up this 4-6ram band. This band ofearlywood is most likely related to the decrease in density following the 1986 fall harvest.
Although the total number of response features is similar for saplings from both treatments, indicating that trees from both treatments did respond in some way to the same environmental factors, the degree of the anatomical response was affected by the flooding regime. The effects of environmental factors respon- sible for the formation of response features appear to have been mitigated by the continuous flooding, Trees from the periodically flooded regime had more pro- nounced and a greater range of anatomical responses to environmental factors than did trees from the con- tinuously flooded regime.
Annual Radial Growth and Latewood Formation Saplings from the periodically flooded treatment had a significantly greater (p<0.05) number of cells with thicker cell walls constituting the latewood than sap- lings from the continuously flooded treatment. The 1986 and 1987 annual rings of saplings from the pe- riodically' flooded regimes were distinct and easily identified, and the 1988 annual tings were complete at the time of harvest (Table 3). In contrast, many sap- lings from the continuously flooded treatment had an-
"lable 2. Mean number o1 response features in baldcypress saplings under continuous and periodic flooding treatments.
1986 1987 1988
CF PF CF PF CF PF
Response Mean Mean Mean Mean Mean Mean
feature (STD) N {STD) N (STD) N (STD) N (STD) N (STD) N
Minor 5.75 17 5.34 16 6.03 17 4.77 16 3.16 8 3.72 8
(1.08) (0.68) t0.94) (0.56) (0.62) (0.74)
Moderate 1.21 17 1.48 16 2.01 17 2.52 16 0, t3 8 0.53 8
( ~. i 3) {0.2g) {0.38) {0.28) {0.13) {0.25)
Major 0.00 17 O 60 16 O.00 17 0.94 16 0.00 8 0.22 8
(0.00) (0.20) (0.00) (0.32) (0.O0) {0.13)
(-'V- ('onlinuous flooding, PF = Periodic flooding.
N = number of trees in sample/year,
icm=O,O'Tmm
Figure 5. Cross-section of a continuously flooded, two-year- old sapling showing the boundary of an annual ring which has latewood cells with a minimal increase in cell wall thick- ness and inconsistent or minimal decrease in cell size ( 125 x ).
nual rings for 1986 and 1987 that had areas where no latewood cells existed, as well as areas where the ter- minal latewood cells decreased only slightly in size and had a minimal increase in cell wall thickness (Figure 5). These types of areas resulted in rings that were indistinct (Table 3). These patterns are consistent with those noted by Bowers (1981 }, who found that a change in the hydrology of an area from permanently flooded to seasonally or periodically flooded, increased the density of latewood and resulted in more abrupt ring boundaries in baldcypress. In addition, terminal late- wood formation of the 1988 annual rings for trees from this treatment was apparently incomplete at the time of harvest (Table 3). The lack of a well-defined late- wood zone and the incomplete 1988 annual ring for saplings from the continuously flooded regime may indicate a longer growth interval, as well as a more
Table 3. Number of baldcypress saplings tiom continuous and periodic flooding treatments showing distinct, indistinct annual rings.
Growth 1986 1987 1988
Icature** CF PF CF PF CF PF
Indistinct [5 2 I2 2 8 1
Distinct 2 14 5 14 0 7
** = feature exists al some poinl along the circumfbence.
CF = Continuous flooding.
PF = Periodic flooding.
gradual cessation of the growth phase. Differences in the length of the growth phase were noted by B. D.
Keeland, U.S. Fish and Wildlife Service (pers. com.), who found that baldcypress in shallow, continuously flooded areas grew for a longer period during the season than trees in other hydrologic regimes.
Individuals from the continuously flooded treatment had substantially greater annual growth than those from the periodically flooded treatment, averaging a differ- ence in annual radial increase of 45% for 1987 and 103% for 1988 and an increase in basal area of 51%
for 1987 and 168% for 1988. This resulted in a differ- ence in cumulative area increase of 107% (Table 4).
These results are consistent with those of Eggler and Moore (1961 ), Dickson et al. (1965), Broadfoot (1967), and Broadfoot and Williston (1973), who reported ra- dial growth increases as great as 50-100% for trees in areas of shallow water impoundment. Stahle et al.
(1992) and Young (unpublished data) found a surge in radial growth in baldcypress trees for the years im- mediately following an impoundment. This is in con- trast, however, to the findings of Duever and Mc- Collom (1987), who found inundation of baldcypress to result in decreased growth.
A positive correlation (p < 0.05) existed between an- nual growth and response feature formation for 1986 and 1987, with the saplings that grew the most having the greatest number of response features. This trend, however, did not continue for the 1988 growth. By the third year of growth, the maturing saplings had de- veloped extensive root systems that were adapted to each flooding treatment (Megonigal and Day 1992) and most likely became less sensitive to minor environ- mental changes. Other studies have shown similar physiological adaptations to flooding (Hosner and Boyce 1962, Hook el al. 1970).
CONCLUSIONS
Even though specific factors that cause the formation of false rings in baldcypress remain unclear, the hy- drology of the site will determine the degree to which a reaction will be expressed in saplings. If this can be
298 WETLANDS, Volume 13, No. 4, 1993 Table 4. Annual growth in baldcypress saplings under continuous flooding and periodic flooding.
1986 1987 1988 Cumulative
CF PF CF PF CF PF CF PF
Ring width* 0.68 0.86 1.44 0.99 1.87 0.92 3.99 2.77
Basal area** 1.44 2.33 12.67 8.38 35.76 13.34 49.87 24.05
(T Continuous flooding.
PF = Periodic flooding.
* Ring width (cm) is the mean of four radii.
** Areas (cm ~) were calculated on the basis of mean ring width.
shown in mature trees, then knowledge of the inherent growth characteristics associated with different hydro- logic regimes can be used. Trees found in continuously flooded areas would be expected to produce few major false rings, possibly due to the moderating effects of the water, and although ring boundaries may not al- ways bc distinct, the possibility of misinterpreting a false ring would be minimized. Individuals from pe- riodically flooded sites would be expected to produce moderate-to-major false rings, indicating a greater sen- sitivity to intraseasonal events. If erossdating of these trees with those from a shallow, continuously flooded area nearby could be successfully accomplished, the presence of false rings or response features could pro- vide a chronology that would contain reliable infor- mation on both annual and past intraseasonal events.
ACKNOWLEDGMENTS
We thank John Bailey, Kent Bryant, and other tech- nicians for assistance in collecting samples and Bob Keeland, Dave Stahle, and colleagues at SREL who reviewed early drafts of this paper. Two anonymous reviewers also provided valuable suggestions. This re- search was supported by contract DE-AC09-76SR00- 819 between the U. S. Department of Energy and the University of Georgia.
LITERATURE CITED
Beaufait, W. R. and T. C. Nelson. 1957. Ring counts in second growth baldc)q3ress. Journal of Forestry 55: 588.
Bowers, L. ,i. 1981. Tree ring characteristics of baldcypress growing in va~'ing flooding regimes in the Barataria Basin, Louisiana.
Ph.D. Dissertation. Louisiana State University, Baton Rouge, LA, USA.
Broadfoot, W. M. 1967. Shallow-water impoundment increases soil moisture and growth of hardwoods. Soil Science Society of America Proceedings 31: 562-564.
Broadfoot, W. M. and H. L. Williston. 1973. Flooding effects on southern forests. Journal of Foreslr3'. Sept. 1973: 584-587.
Dickson, R. E., ,I. F. Hosner, and N, W. Hosley, 1965. The effects
of four water regimes upon the growth of four bottomland tree species. Forest Science 11:299-305.
Duever, M. ,i. and J. M. MeCollom. 1987. Cypress tree-ring anal- ysis in relation to wetlands and hydrology, p. 249-260. In: G,C, Jacoby, Jr. and J.W. Hornbeck (eds,) Proceedings of the Inter- national Symposium on Ecological Aspects of Iree-Ring Anal- ysis. U,S. Department of Energy Publication CONF-8608144.
Eggler, W. A. and W. G. Moore, 1961. The vegetation of Lake Chicot, Louisiana, alter eighteen years impoundment. The South- western Naturalist 6:175-183.
Fritts, H. C. 1966. Growth rings of trees: their correlation with climate. Science 154: 974-979.
Fritts, H.C. 1976. Tree Rings and Climate. Academic Press. New York, NY, USA.
Goyer, R. A. and G. J. Lenhard. 1988. A new insect pest threatens baldcypress. Louisiana Agriculture 31:16-17.
Goyer, R. A., G. J. Lenhard, and J. D. Smith. 1990. Insect her- bivores of a bald-cypress/tupelo ecosystem. Forest Ecology and Management 33/34:517-521.
Hook, D, D,, O, G. Langdon, J. Stubbs, and C. L. Brown. 1970.
Effect of water regimes on the survival, growth, and morphology of tupelo seedlings. Forest Science 16:304--311.
Hosner, J, F, and S. G. Boyce. 1962. Tolerance to water saturated soil of various bottomland hardwoods. Forest Science 8:180-186.
Larson, P.R. 1956. Discontinuousgrowth rings in suppressed slash pine. Tropical Woods 104: 80-99.
karson, P. R. 1963. The indirect effect of drought on tracheid diameter in red pine. Forest Science 9:52~62.
Megonigal, J. P. and F. P. Day. t992. Effects of flooding on root and shoot production of bald cypress in large rhizotrnn cells. Ecol- ogy 73: 1182-1193.
Stahle, D. W., E. R. Cook, andJ. W. C. White. 1985a. Tree-ring dating of baldcypress and the polential for millennia-long chro- nologies in the southeast. American Antiquity 50: 796-802.
Stahle, D. W., M. K, Cleaveland, and J. G. Hehr. 1985b. A 450- year drought reconstruction for Arkansas. United States. Nature 316: 530-532.
Stahle, D. W., M. K. Cleaveland, and J. G. Hehr. 1988. North Carolina climate changes reconstructed from tree rings: A.D. 372 to 1985. Science 240: 1517-1519.
Stable, D. W., R. B. VonArsdale, and M. K. Cleaveland. 1992.
Tectonic signal in baldcypress trees at Reelfoot Lake, Tennessee.
Seismological Research Letters 63: 439447.
Stokes, M. A. and T. L Smiley. 1968. An Introduction to Tree- ring Dating. The University of Chicago Press, Chicago, IL, USA.
Zahner, R., J. E. Lotan, and W. D. Baughman. 1964. Earlywood- lalewood features of red pine grown under simulated drought and irrigation. Forest Science 10:361-370.
Manuscript received 31 March 1993; revision received 27 July i 993;
accepted 30 August 1993.
