FIRE ISLAND Ecological Studies of the Sunken Forest, Fire Island National Seashore, New York NPS Scientific Monograph No. 7

The biogeochemical relationships of the maritime forest were quantitatively investigated on the 20 X 30-m plot. The plot was intensively studied to quantify some of the major nutrient compartments and flux rates of the nutrient cycling model used in the study of the Hubbard Brook Forest ecosystem (Bormann and Likens 1967).

Analyses were made of the potassium, sodium, calcium, and magnesium contents of the standing biomass, net primary production, throughfall, stemflow, organic debris, litter fall, ground water, soil minerals, and soil colloids. The nutrient inputs into the system in precipitation and dry fallout were measured outside the plot in the open on the secondary dune crest. Another input component was impacted aerosols washed off the vegetative surfaces in stemflow and throughfall. Direct measurement of nutrient outputs in the ground water were precluded by limitations of time and equipment.

The distribution of available and potentially available nutrients in the compartments of the nutrient cycling model (Fig. 42) influences the rate and extent of ecosystem development (Bormann and Likens 1967). In the Sunken Forest, where the contribution of nutrients from weathering of the rock and soil compartment is assumed to be low, intrasystem nutrient transfers (leaching, decomposition, and uptake) would largely reflect the distributions of nutrients in the organic and available nutrient compartments. In turn, the concentrations of nutrients in the organic and available nutrient compartments are directly influenced by the patterns of meteorologic inputs into the ecosystem.

Fig. 42. Model of nutrient cycling in terrestrial ecosystems (Bormann 1970).

The amounts of potassium, sodium, calcium, and magnesium in the living biomass, organic debris, available nutrient compartment, and soil and rock mineral compartment were analyzed to determine how cations were distributed within the Sunken Forest ecosystem. The weathering, leaching, decomposition, and uptake transfers of these four cations were investigated to determine the magnitude and rate of nutrient circulation within the system. These analyses were undertaken to determine the effects of meteorologic inputs on the intrasystem nutrient relationships.

Organic Compartment

Methods

The potassium, sodium, calcium, and magnesium concentrations of materials analyzed in this study were determined using a Perkin-Elmer Model 403 atomic absorption spectrophotometer, except for precipitation, stemflow, throughfall, and ground-water samples collected between October 1968 and March 1969, which were analyzed using a Perkin-Elmer Model 303 atomic absorption spectrophotometer Spectrophotometric analysis followed the procedures outlined by Perkin-Elmer (1968).

Live tissue samples were taken from plants harvested for production and dimension analysis in July 1969. Preparation of samples for tissue analysis followed the procedures of Likens and Bormann (1970). All material was handled with polyethelyene gloves. Root systems were washed in distilled water rather than detergent since the sandy soil was easily cleansed from root surfaces. No other tissues were washed; therefore, cation concentrations may include some salt-spray aerosols impacted on tissue surfaces at the time of sampling. Tissues from all positions along the stems and roots were analyzed separately, but results are expressed as averages for tissues regardless of position.

The cation concentrations of Ilex opaca, Amelanchier canadensis, and Sassafras albidum tissues are the means of seven to ten samples. Four samples were analyzed for tissues of Prunus serotina, Pyrus arbutifolia, Vaccinium corymbosum, and Viburnum dentatum. The five root systems and five shoots of each herb species were pooled for cation analysis. Calculations of the amounts of cations in the biomass were made by multiplying the tissue weights by the average cation concentrations.

Samples of dead branch wood were taken from sample branches used in the dimension analysis of the living biomass. These samples as well as the >2-mm soil organic matter (litter) samples were analyzed following the same procedures as used for the cation analysis of living tissues.

The amounts of potassium, sodium, calcium, and magnesium in the >2-mm soil organic matter and dead branch material were estimated by multiplying their dry weights and cation concentrations. However, the cations in dead trees were estimated by multiplying the dead tree dry weight by live stem wood cation concentrations. Since the cation concentrations of wood are low and the amount of dead trees in the plot is relatively small, the overestimations due to calculations based on living tissues are assumed to be small.

The nonexchangeable cation concentrations of <2-mm soil organic matter (F and H humus layers) samples were estimated by subtracting the amounts of exchangeable cations (ammonium acetate extraction method) from replicate samples heated in a muffle furnace for 24 hours at 500°C and taken up in 6 N HCl. There were no significant differences (at P=0.05) between the two series of determinations, suggesting the nonexchangeable cation content of <2-mm soil organic matter is very low.

Errors associated with cation analysis of various plant tissues are largely due to real variation between samples expressed by the standard errors (Appendix III). Six replicates of an Ilex opaca wood sample gave the following standard errors: potassium, 2.0%; sodium, 0.6%; calcium, 1.0%; and magnesium, 0.4%. Small errors also result from analytical techniques. The analysis of blanks run with every 12 samples in the cation analysis demonstrated maximum contamination from analytical technique to be less than 0.2% of K+Na+Ca+Mg for the least concentrated tissue (stem wood).

Results and Discussion

There are great differences between species and the cation concentrations of various tissues (Appendix III). The current leaves and twigs of all species have the highest concentrations of the analyzed cations. For perennial tissues, roots have the highest potassium, sodium, and magnesium concentrations, while stem bark has the highest calcium concentration. Stem wood has the lowest cation concentrations of all living tissues, and branch wood plus bark had concentrations less than stem bark but greater than stem wood. Shoots of the herb-layer plants usually had higher cation concentrations than the roots. All tissues, except tree roots, generally have higher concentrations of calcium and potassium than sodium and magnesium.

Differences in the cation concentrations between species are as striking as differences between tissues. Potassium concentrations in Sassafras current twigs and leaves are about twice as high as those in Amelanchier. Roots of Ilex opaca have over twice the magnesium concentrations of Amelanchier or Sassafras roots. The calcium concentrations of Amelanchier tissues are generally 50-100% higher than Ilex or Sassafras.

In contrast to the findings of Scott (1955), the shrub layer on Fire Island has cation concentrations about the same as tree tissues, except for lower sodium concentrations in leaves (Appendix IV). The herb shoots also had lower sodium concentrations than tree leaves (0.11-0.22% vs. 0.26-0.33%), which suggests that the subordinate vegetation may have lower sodium concentrations due to protection by the tree layer from salt-spray impaction.

The amount of cations in the living biomass of the ecosystem is a function of the concentrations and weights of plant tissues. Therefore, the greatest amount of cations is held in the tree layer (85-98%) in general and Ilex opaca in particular (Table 19). The herb and shrub layers each accounted for less than 1% of the total cations in the biomass; however, the above-ground portions of Smilax rotundifolia lianas contained 13% of the total potassium and about 2% of the total sodium, calcium, and magnesium (Tables 19,20).

The distribution patterns of cations contained in the living biomass show greater amounts of potassium and calcium in shoots and greater amounts of sodium and magnesium in the roots (Table 21). The selective absorption of nutrients in the Sunken Forest ecosystem, therefore, may be similar to the model suggested by Collander (1941): cations in the soil solution enter the roots in response to moisture gradients in the plant-soil system but selectivity is exerted by the roots in the endodermis. Root selectivity may be further influenced by ions returning to the root via the phloem and impeding further uptake by mass action. A possible explanation of the relatively large amounts of sodium and magnesium concentrated in root tissues is that these cations, which are abundant in sea water, are present in the maritime forest environment in amounts exceeding those that can be utilized effectively by the shoot. The roots of maritime forest species, therefore, may be selectively excluding the upward transport of much magnesium and sodium while, at the same time, excesses of these two cations are being translocated from the shoot to the roots. Epstein (1966) suggests that plants may have a dual nutrient uptake pattern which permits the retention of sodium in the root while other ions are exported to the shoot. Although unlikely, the excess sodium and magnesium might be due to soil mineral contamination on root surfaces.

Compared to the living biomass, the organic debris accounts for a small amount (3-10%) of the total cations in the organic compartment (Table 19). The organic debris is relatively richer in calcium and magnesium than in sodium and potassium. These differences are probably the result of the retranslocation of mobile ions out of tissues undergoing senescence and the faster removal of sodium and potassium from tissues in leaching and decomposition processes (Stenlid 1958; Attiwill 1968).

The cation distributions within the organic compartment continually change with the transfers between the living biomass and organic debris largely through litter fall. The organic debris component, therefore, probably contains the greatest amount of cations in the autumn after deciduous leaf fall and in early summer after Ilex leaf fall. The >2-mm soil organic matter in the surface 15-cm layer was sampled in mid-July, and the amounts of cations held in the organic debris may be near a maximum at this time.

The organic compartment influences the nature of the available nutrient and rock and soil mineral compartments not only by the circulation of cations by uptake, leaching, and decomposition but also by the production of organic soil colloidal material and by the secretion of organic acids which are important agents in the weathering of soil materials.

Available Nutrient Compartment and Soil Mineral Compartment

Methods

The cation exchange capacities of the <2-mm soil samples and the organic-free ignited soil samples were determined by the ammonium acetate-potassium chloride method, the ammonium acetate leachate from the <2-mm soil samples being retained for exchangeable cation determinations (Wilde et al. 1964). Total digestion of soil samples was accomplished by decomposition with hydrofluoric and perchloric acids (Jackson 1958). Potassium, sodium, calcium, and magnesium were analyzed by atomic absorption spectrophotometry. The pH of the <2 mm-soil samples was determined in the laboratory with a Fisher Accumet model 310 pH meter using a 1:1 soil:distilled water mixture.

Estimates of the cations in the available nutrient compartment were made by multiplying the soil bulk density and the exchangeable cation concentrations. Cation concentrations in soil minerals were determined by subtracting the exchangeable cation concentrations from the digested soil concentrations and correcting for the contribution of organic matter to the soil sample weight. Total amounts of cations in the soil mineral compartment were calculated by multiplying the weight of soil minerals per unit soil volume by the corrected digested soil concentrations (Appendix III).

Results and Discussion: Available Nutrient Compartment

The status of the available nutrient compartment is determined by the soil organic matter. There is a strong positive relationship between the cation exchange capacity and the amount of organic matter in a soil sample (Fig. 41). The surface soil layer, with a higher organic matter concentration and c.e.c., contains 87-96% of the exchangeable cations (Table 22). Within soil layers, a strong positive relationship exists between organic matter and exchangeable cation concentrations largely reflecting the greater exchange capacity of samples with higher organic matter contents (Fig. 41).

The total exchangeable potassium, sodium, calcium, and magnesium represent 2.37 me of cations/100 g of soil, while the c.e.c. is 10.2 me/l00 g soil. Therefore, the soil colloids are only 23% saturated by these four bases. Probably the remainder of the exchange sites are largely saturated with hydrogen ions. The low base saturation, or high hydrogen ion saturation, in the available nutrient compartment is expected in light of the low pH (4.10-4.16), the relatively high soil organic matter content, the humid climate, and the sandy texture of the Sunken Forest soil which permits rapid leaching of bases out of the system (Buckman and Brady 1960; Lutz and Chandler 1946).

The order of abundance of exchangeable cations in the Sunken Forest soil is H⁺ > Ca⁺⁺ > Mg⁺⁺ > Na⁺ > K⁺ (on an equivalence basis). The order of strengths of adsorption for these cations is H⁺ > Ca⁺⁺ > Mg⁺⁺ > K⁺ > Na⁺ (Lutz and Chandler 1946). The greater amounts of sodium (0.23 me/l00 g soil) compared to potassium (0.15 me/l00 g soil) are undoubtedly due to mass action effects of greater amounts of sodium in the maritime environment. The overall composition of the adsorbed cations in the available nutrient compartment resembles that for humid regions in general: H⁺ and Ca⁺⁺ > Mg⁺⁺ > K⁺ and Na⁺ (Buckman and Brady 1960).

Results and Discussion: Rock and Soil Mineral Compartment

Quartz is the predominant mineral in the soil mineral compartment of the Sunken Forest ecosystem, with garnet and magnetite as relatively common accessory minerals. The quartz content of beach sediments comprising the barrier-island chain off the southern shore of Long Island increases westerly from 80-89% at Montauk Point (120 km east of the Sunken Forest) to a nearly uniform 97-98% west of Tiana Beach, 55 km east of the Sunken Forest (U.S. Army Corps of Engineers 1960). The garnetiferous and magnetitic sands as well as other heavy mineral sands in the mid-tide zone show a similar trend, decreasing from 1% at Montauk Point to less than 1% west of Georgia Pond, 84 km east of the Sunken Forest (U.S. Army Corps of Engineers 1960). A microscopic examination of sands from the Sunken Forest area showed the presence of tourmaline as an accessory mineral and the absence of any feldspars (Cowan pers. comm.)

The chemical composition of the dominant sand minerals indicates the soil mineral compartment is a relatively poor source of plant nutrients (Clarke 1924). Quartz (SiO₂) and magnetite (Fe₃O₄), except for impurities they might contain, are not sources of basic cations. Garnet, a ferromagnesium silicate, and tourmaline, a complex barosilicate of aluminum and other bases, may be the source of some calcium, magnesium, and sodium, but the abundance of these minerals is very small compared to quartz.

The soil mineral compartment contains larger amounts of potassium, sodium, calcium, and magnesium (89-94% of the total to 30-cm depth) than any other compartment in the ecosystem (Table 22), due to the far greater weight of soil minerals than organic matter in the system rather than sands having higher elemental concentrations. The exact dimensions of the soil mineral compartment are extremely difficult to measure; theoretically they coincide with the rooting depth of the dominant vegetation. The majority of the roots in the Sunken Forest system appears to be concentrated in the upper 30 cm of soil; however, some roots excavated in the biomass analysis were found at 50-cm depths. The 30-cm depth for the bottom of the soil mineral compartment in the Sunken Forest is somewhat arbitrary, but probably accounts for the majority of minerals that are interacting with the organic and available nutrient compartments.

The amounts of potassium, sodium, calcium, and magnesium in the soil mineral compartment cannot be realistically compared to those in the organic and available nutrient compartments. The cations in the soil minerals, by and large, represent a capital for the ecosystem that is held in a relatively unavailable form. The total cations in the soil minerals become available after long periods of time. In contrast, large amounts of cations in the available nutrient and organic compartments circulate in several year, annual, or even subannual cycles within the ecosystem. The circulation of cations between the soil mineral compartment and available nutrient compartment is influenced largely by the organic compartment and the physical-chemical environment (Bormann and Likens 1967).