factors effecting vegetational zonation on coastal dunes

Factors Effecting Vegetational Zonation on Coastal DunesAuthor(s): Henry J. Oosting and W. D. BillingsReviewed work(s):Source: Ecology, Vol. 23, No. 2 (Apr., 1942), pp. 131-142Published by: Ecological Society of AmericaStable URL: http://www.jstor.org/stable/1931081 .

Accessed: 18/03/2013 05:23

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecology.

http://www.jstor.org

This content downloaded on Mon, 18 Mar 2013 05:23:21 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=esa

http://www.jstor.org/stable/1931081?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


VOL. 23 APRIL, 1942 No. 2

FACTORS EFFECTING VEGETATIONAL ZONATION ON COASTAL DUNES

HENRY J. OOSTING

Duke University, Durhamn, N. C.

AND

W. D. BILLINGS

University of Nevada, Reno

I NTRODUCTION

The coast of North Carolina is dis- continuously paralleled by a series of nar- row, usually elongate islands. These low- lying sandy "banks" may support a sclerophyllous forest in which live oak (Quercus virginiana L.) predominates and which includes numerous character- istic species -of southern affinity. De- scriptions of the general nature of the vegtation are available in several papers: Johnson ('00), Kearney ('00), Coker ('05), Lewis ('18), Wells ('39).

Because of the full exposure to almost constant wind from the ocean, a slight disturbance of the vegetative cover may result in extensive "blow outs" of the unconsolidated soil. Once initiated, the shifting of the sand may not be checked naturally for years. Man's activities have contributed materially to the destruction of cover. Grazing has been particularly serious, for it has indirectly transformed several of the banks into almost barren seas of shifting sand.

Today, grazing has been largely dis- continued and, on government controlled areas particularly, much time and effort have been expended in attempts to stabi- lize the shifting dunes. Miles of brush barriers have been constructed and thou-

sands of acres have been planted with grasses. The success of these attempts cannot yet be evaluated.

One of the most effective native sand- binding grasses is sea oats ( Uniola paniculata L.) which is found in abundance on these coastal dune areas. The impor- tance of this species and its associates as natural stabilizers prompted an investiga- tion of the environmental factors govern- ing their distribution. A portion of Bogue Bank near old Fort Macon at the entrance to Beaufort harbor, North Caro- lina, was chosen for intensive study. The area has been relatively undisturbed in recent years and illustrates all degrees of stability and vegetational development. The exposed position of the bank subjects it to the full sweep of the ocean wind and the pounding surf.

The ocean side of the bank is bordered by a sand beach of varying width which terminates at a low (5-15 ft.) foredune. Behind this is an extensive depression sometimes covered with small dunes in all stages of development and stability, sometimes blown out to a depth very near the water table. This area may be only 150 feet wide but is often 600 feet or more in width, and it terminates inland at a relatively high rear dune which may be 40 feet high. The rear dune forms a

131



132 HENRY J. OOSTING AND W. D. BILLINGS Ecology, Vol. 23, No. 2

ridge down the center of the bank and obviously affords shelter to the leeward, for here vegetation is much less disturbed and woody species predominate, usually forming dense thickets whose tops rarely extend above the crest of the dune.

METHODS

Two representative localities on the ocean side of the bank were selected and on each a meter-wide transect was marked from high tide level over the crest of the rear dune. All plants occurring in the transects were mapped to scale by square meters.

The distribution of the major species (Uniola paniculata L. and Andropogon littoralis Nash) suggested the locations of stations for the study of environmental factors. Five stations were established on each transect, as indicated in figure 1. All determinations of soil moisture, soil temperature, soil pH, air temperature, relative humidity, evaporation, salt content of soil, and wind-borne salt were made at these points. Standard methods were used in the measurement of all but the last-mentioned, heretofore almost neg- lected, factor (Wells, '39).

A "salt trap" was designed for measur- ing the amount of wind-borne salt transported to the various stations in a unit of time. It consisted of an eight-inch square

~~~~~~RNE72 /

~ Z O N E I Ir V

'~~~~~~~~~~~~~~~

2 TRANSECT 1I/

ZONIE I aE x MA Mr 7a

FIG. 1. Diagrammatic profiles along two transects on the dunes at Ft. Macon. Physio- graphic-vegetational zones are delimited as recognized in the study, and the positions of stations used for environmental investigations are indicated by number. Transect one is 110 meters long. Horizontal scale is one half the vertical.

of cheesecloth in the center of which a square decimeter had been accurately marked with India ink. The cloth was stretched on a light wooden frame by means of thumb tacks. The traps were set up at all stations about a foot above the ground and at right angles to the prevailing inshore wind. At the end of an exposure of 12 or 24 hours, the cheesecloth was carefully removed from the frame, the marked decimeter square cut out and placed in a labeled stoppered vial. During this operation, care was taken not to touch the exposed decimeter with the fingers.

In the laboratory the marked square decimeter was soaked in 200 ml. of distilled water for 10 minutes. Duplicate portions of the solution were titrated with silver nitrate using potassium chromate as an indicator. The titrations gave values from which the milligrams of NaCl could be calculated for each square decimeter of cloth. This is assumed to be a fair indication of the relative amount of salt carried to the different stations and of the salt content of the air which comes in contact with the plants distributed along the transects.

As a check on the field studies, healthy, vigorous plants of Uniola paniculata., An- dropogon littoralis, and Spartina patens were taken up from the dunes where they



April, 1942 VEGETATIONAL ZONATION ON COASTAL DUNES 133

were growing naturally and transplanted in uniform (itune sand to one-gallon containers. All were watered for several days to assure their establishment. Be- ginning on August 7, four plants of each species received one of the following treatments:

1. No treatment. 2. Watered with freshwater daily. 3. Watered with freshwater on alternate

days. 4. Watered with seawater daily. 5. Watered with freshwater daily and the

plant sprayed daily with seawater.

Because of rapid percolation, the water added each evening quickly passed through the soil and the excess drained off through the perforated bottoms of the containers. The soil moisture was apparently at field capacity after each watering.

All plants were placed in the open, exposed to full sunlight, and left for the duration of the experiment. The unwatered group was removed to shelter during rains. The experiment was terminated on September 1. Detailed notes were taken on the appearance of the plants on August 19 and 27 and Sep- tember 1.

THE VEGETATION

The profiles representing the lines of the mapped transects suggest five zones which might be significant in determining the distribution of species. These are: (I) the windward side of the foredune; (II) the lee slope of the foredune; (III) the more or less level expanse behind and protected by the foredune; (IV) the exposed long slope of the rear dune; (V) the protected leeward slope beyond the crest of the rear dune.

When these areas are considered in conjunction with the mapped transects each zone can be rather sharply delimited by the variation in species present and the concentrations of numbers of each. It is further apparent that a variation of zone III must be recognized (IIIa) when, within the area, the soil has been blown

out to a depth approaching the water table.

Because of the impracticality of pub- lishing density and cover values for the species in each square meter of mapped transect, only the average values for each zone are given in table I. Density (D) as represented indicates the average numln- ber of individual plants and clumps of plants in each square meter, and cover (C) is the percentage area shaded by plants or clumps of plants when substan- tial enough to map. This combination of values is adopted because many plants are very small. Some of these are merely young, others, such as Leptilon canadense (L.) Britton, are apparently dwarfed by the environment for they are rarely more than an inch or two tall; elsewhere they may attain 2 or 3 feet under favorable conditions.

Size and habit make Uniola paniculata and Andropogon littoralis the most con- spicuous plants on the area. They may grow intermixed but where one is relatively abundant the other is not important. One of the objectives of the transect map- ping was to determine the relative abundance of the two species in different positions on the dunes. The transect data (table I) show that these two species far exceed all others in number and cover and that there are certain zonal concentrations which are probably general. Uniola exceeds Andropogon on the foredune (zones I and II) and on the crest of the rear dune (zone V). Andropogon is more abundant in the protected depression between the large dunes (zone III). On the windward slope of the rear dune (zone IV) either may predominate. On transect 1, Uniola is more important in this zone, but whenever the slope is fairly stable, as on transect 2, Andropogon predominates, especially on the lower portion bordering zone III.

Other species are not so consistent, as may be judged from their average density and cover within the zones. However, it appears that Oenothera humifusa Nutt. and Heterotheca subaxillaris (Lam.)




TABLE I. Average density (D) and cover (C) by zones, of species mapped on transects from high tide to the crest of the rear dune at Ft. Macon

Transect 1 Transect 2

Zones I II III 1I1a IV V I II III IV V

Uniola paniculata L. D 11.8 7.5 3.3 0.4 4.0 10.6 17.8 7.6 4.7 2.8 7.0 C 2.7 6.9 0.3 0.03 9.1 18.9 10.9 12.3 0.3 13.6 15.4

Andropogon littoralis Nash D 2.1 2.5 5.7 6.6 4.1 0.9 0.5 3.2 4.7 8.0 3.7 C 2.2 7.1 7.3 15.4 5.5 2.3 0.7 10.7 10.4 16.2 9.0

Oenothera humifusa Nutt. D 1.1 1.8 7.2 1.8 6.5 4.5 2.8 2.6 1.5 1.8 1.2 C 0.06 0.9 0.1 0.5 0.2 0.3 0.1

Heterotheca subaxillaris (Lam.) D 4.3 0.5 0.5 0.2 1.1 0.8 7.9 3.1 0.06 0.4 0.4 Britt. and Rose C 0.5 0.1 2.2 1.1

Leptilon canadense (L.) Britton D 0.4 1.2 15.1 0.06 6.2 5.1 1.5 6.6 0.5 10.8 1.4

Euphorbia polygonifolia L. D 0.1 0.3 0.3 0.2 0.2 0.8 1.4 0.5 1.0 C 0.2

Fimbristylis castanea (Michx.) D 1.4 11.6 1.9 Vah1 C 0.2 16.6 0.1

Myrica cerifera L. D 0.5 0.2 C 14.7 1.7

Solidago mexicana L. D 0.2 4.7 0.06

Solanum carolinense L. D 0.06 0.06 0.08

Lepidium virginicum L. D 0.1

Smilax bona-nox L. D 0.6 0.2

Parthenocissus quinquefolia D 0.1 (L.) Planch

Physalis viscosa L. D 0.2 0.2

Sabbatia sp. D 0.05

Juncus megacephalus M. A. D 0.1 Curtis

Chenopodium anthelminticum L. D 0.4 0.3 0.1

Xyris sp. D 0.1

Cenchrus tribuloides L. D 0.5 0.06 0.8 0.1 0.2 1.8 0.4

Cakile edentula (Bigel.) Hook. D 0.08

Croton punctatus Jacq. D 0.6

Britt. and Rose both are present in greatest numbers on the foredune. The oc- currence of Leptilon canadense appears to be controlled by soil moisture, for its greatest concentrations are invariably on the outer margins of depressions which are at or near the water table.

There remain several other species char- acteristic of the dune area but of such er- ratic distribution as to merit no discussion. There is, however, a group of species restricted to blowouts which are deep enough to approach the permanent water table. No such habitat occurs on transect




2, but a well developed and rather charac- teristic community is present on transect 1 (zone IIIa). Here Fihnbristylis castatiea (Michx.) Vahl predominates and low clumps of Myrica cerifera L. make considerable cover. Less numerous but typi- cal is Solidago inexicana L., individuals of which grow on the margins of the blowout, abruptly decreasing in numbers with increasing distance from the effects of the water table.

An additional community, less com- monly present in the dunes area and not represented in the transects, is that dominated by Spartina, patens in almost pure stands. This invariably occurs on low level flats behind breaks in the foredunes which permit flooding with sea water during storms; or possibly with exceptionally high tides. This same Spartina society sometimes occurs on broad low beaches in front of the dunes.

THE ENVIRONMENT

Soil Factors

All dune sand areas in Carteret County are classified by Perkins, et al (1938) as Coastal Beach. This miscellaneous soil type is mapped as extending along the outer shore of the banks for the entire length of the county. The sand at Fort Macon is light brown and variable in tex- ture but fine sand predominates. Shell fragments are usually present in abundance.

It seemed plausible that soil moisture might be a factor in determining differences in vegetation. Field water content, therefore, was determined at three different times during the study. Duplicate samples were taken at depths of 10 cm. and 20 cm. at each of the 10 stations and also in the blowout area (zone IIIa). The averages of the duplicate determinations are presented in table II along with average moisture equivalents and wilting percentages. The wilting percentages were determined in the usual manner using seedling oats (Avena sativa L.).

There seem to be no significant differ-

TABLE II. Field water content in per cent of dry weight of soil, moisture equivalent, and wilting percentage* of soils from the ten transect stations

Tran- July27 August 4 August 9 M. E. Wilt. %

Stan 10 20 10 20 10 20 10 20 10 20 cm. cm. cm. cm. cm. cm. cm. cm. cm. Cm11.

1 3.5 2.9 2.6 3.9 3.3 3.9 1.98 2.07 0.215 0.215 2 3.1 5.4 2.9 4.5 2.6 3.8 2.22 2.15 0.121 0.148 3 3.5 3.5 4.8 5.8 3.6 5.6 1.57 1.94 0.122 0.205 3A 16.2 21.1 8.4 13.9 3.9 8.5 2.10 1.87 0.059 0.149 4 2.9 8.1 4.7 5.3 2.2 3.6 1.88 1.99 0.126 0.051 5 3.8 3.6 4.6 4.4 1.7 8.6 2.05 2.15 0.105 0.096

Transect 2

1 3.1 2.4 3.3 3.6 1.0 3.5 1.65 2.04 0.068 0.153 2 4.8 3.5 3.8 5.1 2.4 2.9 1.76 2.38 0.173 0.225 3 2.4 4.2 4.0 4.6 2.2 3.8 1.64 1.58 0.102 0.113 4 4.1 4.6 4.8 5.1 5.2 6.1 1.56 1.66 0.174 0.154 5 1.21 2.3 2.5 2.7 2.5 3.9 1.68 1.84 0.164 0.124

* The wilting percentages are not entirely consistent but, considering the method, the error is apt to be large with such low values.

ences between the moisture contents at the various stations except for the obviously higher percentages recorded from the moist depression at station 3A on transect 1. Usually there is somewhat more moisture at the 20 cm. depth than at 10 cm. The lowest values are exhibited at the first and last stations on each transect.

Although the water content of the soil was low at the times of sample collection some water was always available at all stations and depths since the wilting percentages are even lower. No correlation is apparent between the vegetational zonation and any soil moisture factor between the surface and a depth of 20 cm. This does not preclude the possible deterrent effect of a high water table in the deep blowout areas on the growth of a deep- rooted species such as Uniola paniculata. In fact, this species grows best on the tops of the dunes at a maximum distance from the water table.

Observation of a visible spray of salt water from the surf suggested the possibility that wind-deposited salt on the surface of the soil might be rain-leached into the soil increasing the osmotic concentration of the soil solution. Absorption of water, transpiration rates, and water re- quirements have all been shown to be re-




duced as the salt content of the soil in- creases (Meyer, 1931, and others). If concentrations of salts in the dune sand are sufficiently different at different stations, this might serve as a selective factor in determining the distribution of dune plants.

Soil samples of one hundred grams each were taken at the transect stations from the upper 10 cm. and the second 10 cm. Each sample was soaked in 200 ml. of distilled water for one hour and then 50 ml. portions of the water were titrated and salt calculated as NaCl. The highest salt content found was 1.81 mg. per 100 gm. of soil, which is only 0.0018 per cent of dry weight. There seems to be no reason to suppose that such minute quantities of salt could chemically or osmotically affect the growth or distribution of plants on the dunes.

Considering the amount of salt spray carried over the dunes, it is obvious that leaching must be very rapid to maintain the low salt concentrations at root depths. A possibility also exists that the dune plants may absorb some of this soil salt. Our limited data indicate a variation from day to day in the salt content of the soil and that the concentration at 20 cm. is rather consistently less than at 10 cm. (2 to 40 per cent). There is likewise variation from station to station which is paralleled on the two transects. The windward side of the foredune has the highest concentrations, the crest of the foredune the next highest. This is as might be anticipated in terms of wind-borne salt spray. The crest of the rear dune, which yields the third highest salt spray catches, is lowest in soil salts (0.94 mg. per 100 gm. of soil).

The small areas behind the foredunes which are dominated by Spartina patens are unquestionably related to periodic in- undation by salt water during storms and high tides. Yet, the soil samples taken from them contain no more salts (0.82 mg./100 gm. of soil) than the average on the dunes.

Soil pH throughout the dunes area

ranges from 7.4 to 7.9. Usually there is a greater amount of shell material in samples with the higher values. No consistent differences between stations or depths are apparent.

Temperatures of the dry dune sand, which is never more than partially shaded, are highly responsive to changes in atmospheric temperatures. As in all soils, the highest and lowest dune soil temperatures do not occur simultaneously with high and low air temperatures but lag considerably. The widest temperature variation is found in the surface inch of soil which at night tends to be cooler than the air and by day usually exceeds the air temperature.

Surface soil temperatures of 125-1270 F. are not uncommon in the early after- noon sometime after the sun has reached its peak but at the same time air temperatures do not exceed 95-1000 F.

Soil temperatures at 4 and 10 inches likewise respond to fluctuations of atmospheric temperature, but the magnitude of variation is much less than at the surface and the lag is greater. Highest temperatures recorded at a depth of four inches are 950 F. at 2:00 P.M. and at the same time 890 F. at ten inches. The general tendency is for subsurface temperatures to increase slowly and steadily throughout the day and to decline pro- portionately during the night regardless of minor variations in atmospheric temperature. However, major atmospheric changes may be apparent in soil temperatures even at ten inch depths. On a day when a light rain fell between 1 :00 and 2:00 P.M. air temperatures abruptly dropped 40-50 F., and surface soil temperatures declined 150 almost immediately and continued to fall another 150 during the next hour. In spite of the rapidly declining surface temperatures the sub- soil temperatures continued their gradual rate of increase during the first hour and declined only slightly the second hour. Although surface soil temperatures may exceed four-inch temperatures on sunny days by ten to forty degrees, temperatures




at ten inches are usually no more than five to eight degrees lower than four-inch temperatures.

No soil temperature variations were noted which can be interpreted as factors in the zonal distribution of a species. These observations indicate temperature conditions similar to those of the English dunes studied intensively by Salisbury ('34).

ATMOSPHERIC FACTORS

Because of the intense light, summer air temperatures on the dunes -are relatively high with much fluctuation possible from day to day and from hour to hour. However, maximum temperatures were not measured above 1000 F. Forty year records at nearby Beaufort show an an- nual mean of 63.70 and July and August means of 79.9 and 79.70 F., respectively.

Hourly determinations of air temperature taken simultaneously at all stations during a day show much variation, with no apparent correlations with zones or topography. Other simultaneous determinations taken at isolated times are likewise not significant.

Standard white-bulb Livingston atmometers with rain-traps were used for determining relative evaporation rates at the ten stations. They were set up with the bulbs about one foot above the soil. The instruments could not be left out permanently. However, twelve runs were made between July 26 and August 12 none of which was shorter than 7 hours, and six were for 24 hours. There were 217 hours of exposure in the 18 days. A summary of the results is given in table III.

Several generalizations may be drawn from these data. Evaporation is consistently somewhat higher at stations on transect 2 than at those on transect 1. At all stations, evaporation rates are three to four times as great by day as by night.

The greatest overall loss on both transects is at station 5 atop the rear dune, and the loss at station 2 on the foredune

TABLE III. Average amounts (ml.) of water lost each hour from atmometers exposed at the five stations on each of two dune transects. Day and night losses are presented as determined as well as the average loss per hour on all runs

Day Night Average

Transect I II I I I I I I

Station 1 2.15 2.74 0.67 0.66 1.29 1.35 2 2.74 3.04 0.76 0.61 1.43 1.53 3 2.22 2.89 0.66 0.63 1.24 1.26 4 2.67 2.81 0.61 0.64 1.33 1.32 5 2.37 3.26 0.78 0.80 1.54 1.55

is almost as great. This suggests a possible correlation with the distribution of Uniola, for the species has its best development at or near these stations. The least' evaporation takes place at station 3 sheltered by the foredune. Losses on the windward side of the foredune and rear dune (stations 1 and 4) are all of about the same magnitude.

By day the greatest evaporation is on the foredune (station 2) followed in decreasing order by stations 5, 4, 3 and least of all on the windward side of the foredune (station 1). The night rates of loss indicate a quite different pattern, for station 5 has the highest values followed by stations 1, 2, 3, and 4. The greater daytime evaporation in the exposed situa- tions may contribute to the presence of Uniola and the absence of Andropogon in such places.

Relative humidity determinations with cog psychrometers were made simultaneously on several occasions at all ten stations. Throughout one day such determinations were made at hourly intervals.

As might be anticipated, the data show wide variations and fluctuations both from hour to hour and from station to station. The lowest relative humidity recorded was 54 per cent, on the rear dune. At the same time 84 per cent was recorded for station 3 on the same transect. The highest value, of 88 per cent, was recorded at station 3 on transect 2. On one occasion, the reading was identical at every station.

A single generalization may be made




from the observations. On these dunes, the relative humidity decreases from the foredune inland to the top of the rear dune. On transact 1, the humidity is invariably lower on the beach than on the foredunle but the reverse is true on transect 2.

The data suggest no strong relationship between the distribution of vegetation and relative humidity on the dunes.

Both evaporation and relative humidity are profoundly influenced by the more or less constant wind from the direction of the sea. Other factors being equal, places in the lee of sheltering dunes will lose less water by evaporation than exposed positions in front of or on the crests of dunes. United States Weather Bureau records for nearby Beaufort show the prevailing winds to be inshore (Southwest) during all but the fall months of September through November. Wind velocities are not available for Beaufort, but at Hat- teras, where conditions are comparable although perhaps somewhat more extreme, the average hourly wind velocity is 13.4 miles. Velocities of 50 miles per hour have been recorded there at various times in every month of the year, and there is one record of 80 miles per hour.

That the effect of this wind on dune vegetational zonation is not confined to its evaporational influence is clearly shown by the results obtained with the salt spray traps described above. The amount of salt spray and the distance it is transported are largely determined by wind. The total NaCl intercepted by 5 traps on a transect for any single run ranged from 29 mg. on a calm day to 121 mg. on a windy day when breakers were rolling. Again, the relative amount of spray striking at a station varies from day to day, and this too depends upon wind velocity. Since our concern is with comparative tolerance of species along the transects, absolute values for salt intercepted mean less than relative amounts. The results are therefore expressed for each station as the per cent of total NaCl intercepted on the transect for each run (table IV).

TABLE IV. The amount of wind-borne salt intercepted at each station in each exposure, expressed as per cent of total salt intercepted on the transect during that exposure

Stations

____ 2 3____ 4 1___ - ___ - mg./transect

1 2 3 4 5

Transect 1 July 7-8 40.9 19.6 21.6 5.5 12.5 121.76 Aug. 3-5 40.5 19.5 6.5 9.2 24.3 62.02

6-8 50.1 27.5 8.7 3.7 9.9 52.49 8-9 45.6 23.9 12.2 6.5 11.8 62.79 9-10 45.6 23.8 11.8 4.5 14.2 64.81

10-11 40.6 30.0 13.4 6.0 9.9 29.0

Aver. 43.6 22.8 13.8 5.9 14.0 392.87

Transect 2 July 7-8 51.4 10.7 11.5 5.6 20.9 81.64 Aug. 3--5 39.9 28.4 5.7 0.9 25.1 67.96

6-8 53.8 26.5 5.2 4.8 9.7 46.61 8-9 49.8 23.4 8.5 7.2 11.0 49.67 9-10 42.1 26.8 9.7 8.6 12.9 52.41

10-11 37.3 32.7 9.4 6.7 13.8 30.85

Aver. 46.3 23.1 8.5 5.3 16.7 329.14

Regardless of weather and the amount of salt transported by the wind, the salt catch values for the windward side of the foredune (station 1) are always 40 to 50 per cent of the total salt intercepted on the transect during an exposure of the traps. Then, ordinarily, the crest of the foredune is exposed to the next highest amount of spray and, surprisingly, the crest of the rear dune receives the next highest amount. Successively, the leeward side of the foredune receives relatively much less and the base of the rear dune the very least (only about 15 per cent of the total).

Apparently much of the salt spray is dropped on the foredune as the wind rises from the water. Then as it passes over the depression behind the foredune, there is no obstruction until the rear dune is reached. Some of the spray is lost between the two dunes, perhaps by gravity. However, it would appear that the real force of the wind is deflected upward by the foredune and its first obstruction thereafter is the crest of the rear dune.

Here, then, is a factor which shows




some correlation with the distribution of the two major dune species. Uniola is invariably found in areas of greatest exposure, whether the soil be stable or not. Andropogon predominates on protected areas and on the seaward slope of the rear dune but not on its crest. Assuming that salt spray is the factor limiting or control- ling the two species, one may suppose that the lower portion of the windward slope of the rear dune consistently receives a minimum of salt spray (as at station 4) and that only the crest of the dune is subjected to the heavy spray content of the wind rising from the foredune.

More localized effects are also apparent. In general Andropogon seems to be on more stable soil than Uniola. When both occupy a young dune Uniola is present on the raised portion, Andropogon lower down. It might therefore be con- cluded that Uniola is the pioneer dune former and that Andropogon is succes- sional to it, that where Andropogon is absent the necessary stability has not been attained for its establishment. However, Andropogon occurs on areas both long stable or stabilizing, without Uniola present and with no evidence that it has played a part. Neither can it be demonstrated that Andropogon follows Uniola in its stabilizing activity.

When the presence or absence of the two species is considered in terms of their exposure to salt spray, even these minor variations are explainable if Uniola is tolerant and Andropogon relatively in- tolerant.

PHYTOMETER STUDIES

In table V a summary is presented of the relative responses of the three grass species during the phytometer treatments. Detailed observations were made at the time of transplanting (August 1), and again 12 days after treatment was begun, 8 days later, and finally when the experiment was terminated (September 1). The symbols used in the table indicate

TABLE V. Responses of plants of three dune species to watering and spraying treatments at intervals during a month. The symbols indicate: (0) unaffected; (+) improved in appearance, either in color or new growth; (-) declining, unhealthy or dying. The first symbol in each group compares the plants to their field condition, the second is compared to the first, and the third to the second

Uniola Andro- Spar- pogon tina

No treatment ........ 0 0 + 0 + -Q o- Water daily ......... 0 0 0 0 + 0- 0 0 Water alternate days. 0 0 + ++ - 0 + Seawater daily ....... - O- - - O + Water daily and sea-

water spray daily . .0 0 - - - _ _

the condition of the plants relative to that of the last preceding observation.

No treatment. After the preliminary watering all three species seemed well established. Unwatered Andropogon was apparently favored somewhat by the first decrease in soil water content but com- pletely lost this advantage in the next pe- riod, for the plants were almost dead at the end of the experiment. Uniola and Spartina were quite unaffected at first but as the soil dried out the Spartina looked very unhealthy while the Uniola was in excellent condition although little growth was apparent.

Examination of the root systems at the end of the experiment showed that all plants of all species had made some root growth in the dry sand but that there were no new roots. The water content of the soil in the Andropogon and Spar- tina containers was reduced below any field water contents measured and aver- aged 0.5 per cent in the lower levels of the containers and 0.8 per cent in the upper half. The upper third of the Uni- ola soil was down to 0.14 per cent but the lower two-thirds was still at 2.6 per cent, comparable to several field determinations. Apparently the water requirement of Uniola is much less than that of the other two species.

Daily watering. All plants of all species improved over field appearance in early stages of treatment and thereafter




changed very little. No marked growth was apparent. The root systems of all species grew considerably andl branichinig was pronounced.

Watering alternate days. All plants of all species made more growth and had a better color with this treatment than with any other. Andropogon plants were possibly somewhat more responsive than the others. Root systems were massive and well distributed through the soil available in the containers of all plants.

Seawater daily. Both Uniola and An- dropogon declined under this treatment. At the termination of the experiment, Uniola plants looked unhealthy and pale, and their leaves remained closely rolled at all times. Three Andropogon plants were dead; the fourth bore a single living leaf which was dying at the tip.

Spartina plants were no more affected than by freshwater. The earlier treatments slowed their growth, but on Sep- tember 1 they were all healthy and growing.

Andropogon roots were dead and had made no growth. Many Uniola roots were dead and no growth was apparent. Spartina roots were in excellent condition and had made as much growth as those treated with freshwater.

Daily seawater spray. Andropogon declined steadily under the spray treatment, for although a few new leaves were still being produced on September 1, the total number of active living leaves was continuously being reduced. Uniola seemed at first unaffected by the spray but finally appeared somewhat less healthy than un- sprayed plants. Spartina was totally unaffected.

The phytometer experiments suggest several generalizations. Andropogon and Uniola are relatively unaffected by low soil moisture over considerable periods of time but, because of a probable lower moisture requirement, Uniola is capable of longer survival under extreme conditions. Under comparable low moisture conditions Spartina will die before the

other two species show much indication of suffering. Andropogon cannot live for long when salt water is added to the soil but Uniola is much less affected. Daily watering is less favorable to all three species than watering on alternate days. Probably better growth is possible under wider intervals of watering. Salt spray is seriously injurious to Andropogon, af- fects Uniola only slightly, if at all, and has no effect on Spartina.

DISCUSSION AND CONCLUSIONS

The transect data serve to corroborate the zonal distribution of the major species on the Ft. Macon dune area and to show the relationship of these zones to topo- graphic features. Such a distribution is usually explainable on the basis of exposure, soil, temperature, or moisture re- lationships. Here, however, the usual factors do not seem to control the situation. The soil is uniform or nearly so, soil moisture does not seem to vary character- isticallv for the zones nor sufficiently to have any strong influence. Temperature of the soil and air varies widely and in- consistently. Relative humidity and evaporation rates are greatest at the crests of the fore and rear dunes. None of these factors seems to be strongly correlated with the distribution of the dominant species.

The phytometer studies show that Uni- ola and Andropogon require very little moisture to survive and, in fact, do better under low soil moisture conditions than when watered daily. Thus, the deep depressions are left to more mesophytic species. If the depressions are periodically flooded with salt water Spartina patens is dominant. If moist with seepage water, the low places are predominantly occupied by Fimbristylis castanea. The phytometer experiments further show that the water requirement of Uniola is less than that of Andropogon. This may be a factor con- tributing to the predominance of Uniola on the crests of the dunes where evaporation is greatest and may partially explain




its success as a pioneer dune former in dry and shifting sand. However, the relative tolerance of the two species to salt spray is such that Andropogon dies under exposures which scarcely affect Uniola. Apparently, the general distribution of the two species on the dunes is largely controlled by the extent of exposure of the habitat to wind-borne spray.

Fundamentally, Uniola is the principal pioneer and dune-former. The sturdy, rhizomatous plants of this species survive upon and build up dunes of aeolian sand even in places where wind and salt spray combine to make the environment rig- orous. By vigorous sprouting, the plants grow upward with the dune, finally cap- ping the summit when the sand is stabilized. On- these dune crests, Uniola is subjected to the full force of the spray- laden wind at a maximum distance from the water table. Uniola can grow and reach its optimum development under the constant influence of these inhibitory factors. On the other hand, Andropogon is almost confined to the sheltered habitats nearer the water table, where the fore or rear dune provides protection from the ocean spray. The cause of the intolerance of Andropogon foliage to air-borne salt water and of the comparative tolerance of Uniola to the same factor is a problem which this paper makes no attempt to solve. There is ample evidence, however, that salt spray does play a decisive role in bringing about zonation in coastal vegetation.

SUM MARY

1. The "banks" paralleling the North Carolina coast are elongate sandy islands once well stabilized by vegetation. Breaks in the cover result in blowouts which spread to form extensive dune areas.

2. The dunes on Bogue Bank near old Ft. Macon are representative of a much wider area and were chosen as a site for the study of factors involved in the distribution of dune stabilizing species.

3. Transects mapped from high tide level to the crests of the highest rear

dunes gave quantitative data on the zonal distribution of the dominant species. Within these zones stations were established for the study by standard methods of soil moisture, soil temperature, soil pH, air temperature, relative humidity, evaporation, and soil salt content. Using a special "salt trap," the amount of wind- borne salt passing over each station was determined for different periods of time and under a variety of conditions.

4. Transect studies show Uniola paniculata and Andropogon littoralis as far exceeding all other plants in number and cover, and a zonation in which Uniola predominates on all the exposed ocean side of the foredune and again on the crest of the rear dune. Andropogon is best developed in the lower, protected blowout area between the two dunes.

5. Of the soil factors investigated none seems to have consistent variations which would serve to explain the zonal distribution of the vegetation. The same generalization applies to atmospheric factors with the exception of wind-borne salt. The data show that plants growing on the windward slope and crest of the foredune and those on the crest of the rear dune are exposed to contact with materially more salt spray than those growing elsewhere. The areas of Uniola predominance match the areas exposed to high atmospheric salt content, and Andropogon predominates only in areas protected from salt spray.

6. Phytometer studies corroborated field observations. Daily spraying with sea water was seriously detrimental to Andro- pogon, affected Uniola only slightly, and Spartina not at all. Watering with sea water killed Andropogon, was somewhat unfavorable to Uniola, and did not affect Spartina.

7. The zonal distribution of the major species is therefore explainable on the basis of tolerance to wind-borne salt.

LITERATURE CITED

Coker, W. C. 1905. Observations on the flora of the Isle of Palms, Charleston, S. C. Torreya 5: 135-145.




Johnson, D. S. 1900. Notes on the flora of the banks and sounds at Beaufort, N. C. Bot. Gaz. 30: 405-410.

Kearney, T. H. 1900. The plant covering of Ocracoke Island. Contr. Nat. Herb. 5: 1-26.

Lewis, I. F. 1918. The vegetation of Shack- leford Bank, Carteret County, N. C. N. C. Geol. Surv. Econ. Pap. 46.

Meyer, B. S. 1931. Effects of mineral salts upon the transpiration and water require-

ment of the cotton plant. Amer. Jour. Bot. 18: 79-93.

Perkins, S. O., et al. 1938. Soil survey of Carteret County, N. C.

Salisbury, E. J. 1934. On the temperatures of sand dunes in relation to the vegetation at Blakeney Point, Norfolk. Trans. Nor- folk and Norwich Nat. Soc. 13: 333-355.

Wells, B. W. 1939. A new forest climax: the salt spray climax of Smith Island, N. C. Bull. Torrey Bot. Club 66: 629-634.



factors effecting vegetational zonation on coastal dunes

Documents