Bulletin ofthe Torrey Botanical Club 116(4), 1989, pp. 364-370 Biomass and resource allocation of Typha angustifolia L. (Typhaceae): The effect of within and between year variations in salinity1 D. F. Whigham, T. E. Jordan and J. Miklas2 Smithsonian Environmental Research Center, Box 28, Edgewater, MD 21037 ABSTRACT WHIGHAM, D. F., T. E. JORDAN AND J. MIKLAS (Smithsonian Environmental Research Center, Box 28, Edgewater, MD 21037). Biomass and resource allocation of Typha angustifolia L. (Typhaceae): The effect of within and between year variations in salinity. Bull. Torrey Bot. Club 116: 364-370. 1989.-In a brackish wetland dominated by Typha angustifolia, we found similar aboveground and belowground production at three sites which had very different salinity regimes. There were, however, significant differences between the three areas in shoot density, height, and biomass. Density was greatest in the site with highest salinity while height and biomass were greatest at the site with lowest salinity. There were also differences in the distri- bution of belowground biomass with roots and rhizomes restricted to shallower depths at the site with highest salinity. Long-term study plots in one area have shown that interannual changes in salinity can result in an almost 75% difference in net aboveground biomass. Key words: brackish wetland, Typha angustgolia, biomass allocation, salinity, aboveground biomass, belowground biomass, long-term study. Salinity affects both the physiology (Don- ovan and Gallagher 1985) and distribution (Dawe and White 1986; Deschenes and Ser- odes 1985; Ewing 1986; Lieffers 1984; Roozen and Westhoff 198 5; Shay and Shay 1983; Snow and Vince 1984) of wetland macrophytes and variations from the nor- mal patterns of salinity often result in vege- tation changes (Zedler 1983; Zedler and Beare 1986; Zedler et al. 1986). Vegetation changes also occur as a result of long-term changes in salinity (Brinson et al. 1985; Lief- fers 1984; van Noordwijk-Puijk et al. 1979). In contrast, many types of coastal brack- ish wetlands are exposed to seasonal and annual variations in salinity without any The research reported in this study was funded by the Smithsonian Environmental Sciences Program, Smithsonian Work-Learn Program, and grants DEB- 791 1563, CEE-82-19615,andBSR-83-16948 from the National Science Foundation. * We thank the many people who have helped with various parts of the research. Alan Remde and Eliza- beth Farnsworth performed much of the field work in 1979 and 1984, respectively. Jay O'Neill, Carin Chit- terling, Hank McKellar, Brian Palmer, George Ras- berry, Liza Remenapp, and Jake Weiner helped with various sampling activities. John Bernard and Paul Keddy provided useful comments on the manuscript. The University of Utrecht (The Netherlands) provided space for Whigham during the writing of the paper. Received for publication October 3 1, 1988 and in revised form May 15, 1989. noticeable changes in vegetation. Brackish wetlands in the Rhode River subestuary of Chesapeake Bay, for example, are normally exposed to low salinity water in the winter and spring months and higher salinity in the summer and fall when freshwater runoff from the watershed is low. Despite large interannual variations in salinity, the species composition of brackish wetlands in the Rhode River, dominated by Typha angus- tifolia, has not noticeably changed since our studies began in 1979 (Jordan et al. 1983; Jordan and Correll 1985; Jordan and Whigham 1988). Salinity in the part of the subestuary where T. angustifolia dominates was generally lower than the long-term (1971-1988) annual pattern in the 1970's and higher in the 1980's (Fig. 1). There have also been differences among years, espe- cially during the growing season months. In 1976 and 1980, weekly salinity values fol- lowed the long-term annual pattern. In 1985 and 1986 weekly salinity values increased earlier than usual while in others (1 972 and 1978) it remained low for longer periods of time. What effects, if any, do the differences in salinity have on the wetland? This ques- tion is of interest because, while these wet- lands are known to be very productive (Whigham et al. 1978), there is little infor- mation on how variation in salinity affects production from year to year and place to 19891 WHIGHAM ET AL.: TYPHA ANGUSTZFOLIA 365 Y E A R Fig. I. Patterns of salinity in the low marsh portion of the Rhode River that is dominated by Typha angustifolia. The solid line is the same for each year and represents the long-term annual weekly pattern for the period 1971 to 1988. Crosses represent weekly values of composite samples for the same time period. The samples were taken from the tidal creek that flows through the wetland that was sampled during this study. place within brackish wetlands. In this pa- per, we present data to demonstrate that variations in salinity can have a large influ- ence on the dynamics of Typha angustifolia L. in a brachsh wetland. We present two data sets: 1) a comparison of density, pro- duction, reproduction, and biomass allo- cation patterns for Typha for 1 year in areas of the wetland where salinities were differ- ent, 2) comparisons of 6 years of density, aboveground production, and reproduction data from permanent plots in one area with- in the wetland. Materials and Methods. SITE DESCRIP- TION. The Rhode River, a small subestuary of the Chesapeake Bay, is located a few ki- lometers south of Annapolis, Maryland. The study site was chosen because long-term water quality data were available for the portion of the subestuary where brackish wetlands are common and a distinct salinity gradient exists in the same area. The part of the subestuary considered in this paper is a tidal creek receiving runoff from a wa- tershed of approximately 2286 ha (Jordan et al. 1986). The wetland dominated by Ty- pha angustifolia covers approximately 1 5 ha. At the upstream border, it merges with forested freshwater wetlands (Whigham et spicata (L.) Greene and several other species characteristic of brackish high marshes in Chesapeake Bay (McCormick and Somes 1982). In addition to Typha, Spartina cy- nosuroides (L.) Roth. and Scirpus olneyi are also common. We studied three areas in the wetland: an upstream area 200 m from a freshwater for- ested wetland, a downstream area 100 m from the ecotone with the brackish high marsh, and a middle area half way between. SALINITY GRADIENT. The salinity of in- ,terstitial water was measured with a refrac- tometer in the three areas in 1979. We sam- pled interstitial water which flowed into holes in the substrate that were created by removing material for determination of be- lowground biomass (procedures for bio- mass sampling described in the next sec- tion). Salinity in tidal water at high tide was measured weekly in the three areas between 1980 and 1988 with a Beckman RS5-3 sa- linometer. For comparison of the three areas in the 1980-1988 period, we present salinity data for weeks between Julian Day 1 10 (April 20) and Julian Day 190 (July 9). That time interval coincides with the period of active growth of T. angustifolia (D. F. Whigham unpublished phenological data). a/. 1986) and at the downstream end it COMPARISONS ALONG THE SALINITY GRA- blends with irregularly flooded brackish DIENT. In 1979, we sampled vegetation in wetlands dominated by Scirpus olneyi Gray., each area. An initial survey of the density Spartina patens (Aiton) Muhl, Distichlis of Typha was conducted by counting the BULLETIN OF THE TORREY BOTANICAL CLUB DOWNSTREAM M I D D L E 0 UPSTREAM Fig. 2. Mean salinity between Julian days 110 and 190 for the tidal stream in the upstream, middle, and downstrcarn arcas for 1983-1988. number of shoots in 50 randomly located quadrats (0.25 m2) in each area. Between April and late July, we made four biomass harvests in each area but only report data from the July harvest since our objective is to compare peak aboveground standing crop. All aboveground and belowground biomass was harvested from four quadrats (0.25 m2) in each area. Aboveground bio- mass was harvested first and the heights and diameters of all vegetative and reproductive shoots measured. The plants were then re- turned to the laboratory, dried at 6PC, and weighed. Because we could not separate be- lowground biomass for vegetative and re- productive shoots, aboveground biomass data was combined for the two types of shoots. Belowground biomass was sampled to a depth of 30 cm by removing the sub- strate as a block from each quadrat where the shoots had been harvested. The block was divided into 0-1 0-, 10-20-, and 20-30- cm depth intervals. The material from each depth interval was washed and all living roots and rhizomes (those which were turgid and white) were removed, dried at 60?C, and weighed. All substrate sampling was done at low tide. COMPARISONS AMONG YEARS. Shoot height, shoot density, and the number of reproductive shoots were determined in two sets of long-term study plots in the middle area between 1983 and 1988. The two sets of plots were in different parts of the middle area where separate studies were being con- ducted (Jordan and Whigham 1988; Jordan et al. in press a). Biomass was estimated using the procedures described in detail in Jordan and Whigham (1988). Shoot density was determined in 2-m x 2-m plots which were subdivided into 1-m x I-m subplots. The heights of the 3 tallest plants were mea- sured in each of the subplots. A regression equation was then used to estimate biomass from the density and height data. Results. COMPARISONS ALONG THE SALINITY GRADIENT. Interstitial salinity in 1979 was significantly (P 5 0.00 1) higher (4.5 i 0.3%) at the downstream area. low- Fig. 3. Density (#/m2), shoot height (cm), and shoot diameter (mm) of Typha angustifolia in the upstream, middle, and downstream areas in 1979. Values are means f 1 standard error. Means that are not signif- icantly different share the same superscript. WHIGHAM ET AL.: TYPHA ANGUSTIFOLZA 367 Fig. 4. Aboveground and belowground biomass ex- pressed as g/m2 for the upstream, middle, and down- stream areas in 1979. All values are means + 1 stan- dard error and means which are not significantly different from each other share the same superscript. est (1.0 i 0.17~) at the upstream area, and intermediate (2.0 k 0 .67~) in the middle area in 1979. The same site relationships persisted for salinity in the tidal stream be- tween 1980 and 1988 (Fig. 2). In 1979, shoot density was greater (P < 0.001) in the downstream area but shoot diameter (P i 0.0009) and height (P i 0.0035) were greater in the upstream area (Fig. 3). The contrasting pattern in shoot density compared to shoot diameter and height resulted in similar amounts of above- ground and belowground biomass on an area basis (Fig. 4). Only one flowering shoot (5.8%) was har- vested in the upstream area, no flowering shoots were in the plots sampled in the mid- dle area, and eight (33.0%) flowering shoots were in the plots in the downstream area. There were no significant differences in the ratio of belowground to aboveground biomass but there were significant differ- ences in the distribution of the belowground biomass. The greatest percentage of the rhi- zome and roots biomass occurred within the upper 10 cm at the downstream area (Fig. 5). There was an almost equal division of belowground biomass between the 0-10 and 10-20-cm depth in the upstream area and the middle area was intermediate. COMPARISONS AMONG YEARS. AS shown in Fig. 2, there have been clear differences between the three areas in the salinity of the tidal stream during the growing season and the differences have been most pronounced during years of high salinity. During the 6 years in which we have measured Typha, salinity between April and July has ranged from 0 to almost 9%. There have been large interannual differences at all sites in some years (between 1984 and 1985) and large . 2 :I 4 a 0 3 Fig. 5. Distribution of roots and rhizomes and be- low&ound : aboveground ratios of Typha angustifolia in the upstream, middle, and downstream areas. Depth intervals are as indicated and the values are means of four samples for each area. Means that are not signif- icantly different share the same superscript. 368 BULLETIN OF T H E TORREY BOTANICAL CLUB [VOL. 116 differences only at the upstream area be- tween 1986 and 1987. There were also interannual differences in growth and reproductive characteristics of Typha in the middle area during those same years (Fig. 6). Shoot density, height, and biomass in permanent plots measured be- tween 1983 and 1985 were similar in the two low salinity years (1983-1984) but all decreased significantly (P < 0.05) in 1985 when salinity was high. The number of re- productive shoots in those plots also de- creased in 1985. Differences among years for the plots measured between 1985 and 1988 were all significant (P i 0.05) except for 1985 and 1988 when the salinities were very similar. Density, height, and biomass were all high in 1987 when salinity was low. In 198 5, we sampled both sets of permanent plots and biomass differed significantly (P i 0.05) between them. Discussion. Salinity clearly influences the distribution of macrophytes in many types of wetlands (Adams 1963; Doumlele et al. 1985; Cahoon and Stevenson 1986; Ewing 1986; Roozen and Westhoff 1985; Semen- iak 1983) and long-term monotonic changes in salinity result in dramatic changes in the vegetation (van Noordwijk-Puijk et al. 1979; Zedler et al. 1986). It is also clear that most species of brackish and saltwater wetlands grow better in freshwater (Ewing 1986; Phleger 197 1; Zedler 1983). In contrast, there is very little information on the effects of seasonal and annual variations in salinity on specific vegetation types, although it is known that production of Typha spp. is af- fected negatively by increasing salinity (McMillan 19 5 9). In hypersaline wetlands in southern California, Zedler (1 983) found that a short-term reduction in soil salinity resulted in a 40?/o increase in the production of Spartina foliosa and a 160% increase for Salicornia virginica. Discharges of fresh- water into the hypersaline wetlands resulted in the elimination of halophytes and estab- lishment of Typha. Following the return of hypersaline conditions, production of Ty- pha decreased but the clones have not been eliminated (Zedler and Beare 1986). Ewing (1 986) studied vegetation characteristics along a salinity gradient in the Puget Sound in Washington. The productivity of Scirpus validus, Eleocharis palustris, and Carex lyngbyei was variable but was greatest in the fresher areas. Two other species (Scirpus americanus and S. maritimus) showed very little variation along the salinity gradient, as did Typha in our wetland. This suggests that some species are able to compensate for gradients in salinity. In our study, there were clear differences between areas in shoot density (both asexual and sexual shoots) and size but the differences were opposite in di- rection and the result was that there were no differences in aboveground or be- lowground biomass (Fig. 4). The reasons why shoot density was significantly higher in the two downstream areas are unclear but other studies in the same wetland have also shown that shoot density changes signifi- cantly as the oxygen concentration in over- wintering Typha rhizomes is varied (Jordan and Whigham 1988). The amounts of aboveground and be- lowground biomass measured in this study are within the wide range of values reported for other tidal wetlands (Whigham et al. 1978). The ratios of belowground to above- ground biomass were also within the range reported previously for brackish wetlands and for Typha (Whigham and Simpson 1978). This is the only study that we know of in which the vertical distribution of be- lowground biomass has been shown to vary along a salinity gradient even though there were no differences in total biomass (Fig. 4). Since Typha angustifolia is primarily a freshwater species (McMillan 1959), we ex- pected that roots and rhizomes would have been more evenly distributed vertically in freshwater areas and restricted to the less saline part of the substrate in brackish areas. In our study, however, we found that roots and rhizomes in the brackish area were lo- cated closer to the surface where salinities were higher during the summer. We can only speculate that the upper part of the substrate in the downstream area is less saline at other times of the year, especially in the winter when freshwater inputs are higher. The most striking result of sampling bio- mass in permanent plots has been the an- nual variations in net aboveground biomass and its relationship to salinity during the growing season (Fig. 6). There have been few studies of the affects that annual vari- ations in salinity have on functional pro- cesses in brackish wetlands. During periods WHIGHAM ET AL.: TYPHA ANGUSTIFOLIA Total- IS86 '.. . \ I , I , , , , , , , , 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 S A L I N I T Y ( P P T ) 280- "O- 240- 220- -200- 0 - 180- I- 5 160- - 140- I- 0 120- 0 I m 100- 80 - 6 0 - 4 0 - zo- Fig. 6. Total shoot density and density of reproductive shoots, shoot height, and aboveground biomass for Typha angustifolia in two sets of plots in the middle area. Values for the plots measured between 1983 and 1985 (squares) are the average of 2 samples and the ranges are given. Values for plots measured between 1985 and 1988 (triangles) are means f 1 standard error of 5 samples. All of the means are significantly different from each other exceDt 1985 and 1988. The years associated with the data points are shown and they are in the same 8\ sequence in each part of the figure. ' of brackish water intrusion into normally freshwater forested wetlands in North Car- olina, leafitter production and radial growth of trees decreased (Brinson et al. 1985). Brinson et al. (1 985) also measured higher concentrations of nitrogen in leaf litterfall during the year of brackish water intrusion and suggested that this was due to earlier senescence of leaves prior to the time when nutrients would normally be translocated from the leaves. We have found similar re- sults (Fig. 6). In the long-term study plots, aboveground production decreased with in- creasing salinity. We also found that shoots had significantly higher concentrations of total Kjeldahl nitrogen during years of high salinity (Jordan et al. in press b). With the exception of the studies of Zedler in hyper- saline wetlands in California (Zedler 1983; Zedler and Beare 1986; Zedler et al. 1986), our study is the only one that we know of which shows the significant impact that an- nual variations in salinity can have on bio- mass production in brackish wetlands. The 75% range in aboveground biomass pro- duction was as great as the differences in production among many different types of brackish wetlands (Whigham et al. 1978). Is there any long-term effect of a series of years of either high or low freshwater input? Zedler's studies (Zedler 1983) suggest that wetlands which experience wide variations in salinity are resilient and will return to predisturbance conditions within a few years. Brinson et al. (1985) found that the forested system that he studied returned to predisturbance levels within 1 year. Our data (Fig. 6) also show significant responses from 1 year to the next but there is also a sug- gestion that consecutive years of higher sa- linity may have a cumulative effect on pro- duction. The salinity increase in the middle area between 1985 and 1986 was small (Fig. 2) yet the decreases in shoot height and bio- mass and increase in shoot density were large (Fig. 6). In 1985, we noticed that the shoots of Typha had almost completely senesced by early August. After that time, there was no net carbon gain but there would have been a large respiratory carbon demand by belowground rhizomes and roots because the substrate temperatures were still high (Jordan and Whigham 1988). The utiliza- tion of stored belowground organic matter following senescence in 1985 may have had a negative affect on growth early in the 1986 370 BULLETIN OF T H E TORREY BOTANICAL CLUB [VOL. 1 16 growing season. Growth during the remain- der of the growing season would have also been low due to continued high salinity con- dition. High production rates in 1987 clear- ly demonstrate, however, that the effects of 2 years of high salinity can be reversed quickly and that these systems are very re- silient. Literature Cited ADAMS, D. A. 1963. Factors influencing vascular plant zonation in North Carolina salt marshes. Ecology 44: 445-456. BRINSON, M. M., H. D. BRADSHAW AND M. N. JONES. 1985. Transitions in forested wetlands along gra- dients of salinity and hydroperiod. Jour. Elisha Mitchell Sci. Soc. 101: 76-94. CAHOON, D. R. AND J. C. STEVENSON. 1986. Produc- tion, predation, and decomposition in a low-salin- ity Hibiscus marsh. Ecology 67: 1341-1 350. DAWE, N. K. AND E. R. WHITE. 1986. Some aspects of the vegetation ecology of the Nanoose-Bone11 estuary, Vancouver Island, British Columbia, Can- ada. Canad. J. Bot. 64: 27-34. DESCHENES, J. AND J. B. SERODES. 1985. The influence of salinity on Scirpus americanus tidal marshes in the St. Lawrence River estuary Quebec, Canada. Canad. J. Bot. 63: 920-927. DONOVAN, L. A. AND J. L. GALLAGHER. 1985. Mor- phological responses of a marsh grass Sporobolus virginicus to saline and anaerobic stresses. Wetlands 5: 1-14. DOUMLELE, D. G., B. K. FOWLER AND G. M. SIL- BERHORN. 1985. Vegetative community structure of a tidal freshwater swamp in Virginia. Wetlands 4: 129-145. EWING, K. 1986. Plant growth and productivity along complex gradients in a Pacific Northwest brackish intertidal marsh. Estuaries 9: 49-62. JORDAN, T. E. AND D. L. CORRELL. 1985. Nutrient chemistry and hydrology of interstitial water in brackish tidal marshes of Chesapeake Bay. Est. Coastal Shelf Sci. 2 1 : 45-55. --, W. T. PETERJOHN AND D. E. WELLER. 1986. Nutrient flux in a landscape: The Rhode River watershed and receiving waters, pp. 32-56. In D. L. Correll [ed.], Watershed research perspec- tives. Smithsonian Institution Press, Washington, DC. -- AND D. F. WHIGHAM. 1983. Nutrient flux in the Rhode River: Tidal exchange of nutrients by brackish marshes. Est. Coastal Shelf Sci. 17: 65 1- 667. - AND D. F. WHIGHAM. 1988. The importance of standing dead shoots ofthe narrow leaved cattail, Typha angustifolia L. Aquat. Bot. 29: 319-328. -- AND D. L. CORRELL. In press a. The role of litter in nutrient cycling in a braclush marsh. Ecology. -- AND- . In press b. Effects of nutrient and litter manipulations on aboveground production, nutrient content, and herbivore infes- tation of the narrow leaved cattail, Typha angus- tifolia L. Aquatic Botany. LIEFFERS, V. J. 1984. Emergent plant communities of oxbow lakes in northeastern Alberta Canada. Salinity, water level fluctuations, and succession. Canad. J. Bot. 62: 310-316. MCCORMICK, J. AND H. A. SOMES. 1982. The coastal wetlands of Maryland. Maryland Department of Natural Resources, Annapolis, MD. 24 1 p. MCMILLAN, C. 1959. Salt tolerance within a Typha population. Amer. J. Bot. 46: 521-526. PHLEGER, C. F. 1971. Effect of salinity on growth of a salt marsh grass. Ecology 52: 908-9 1 1. ROOZEN, A. J. M. AND V. WESTHOFF. 1985. A study on long-term salt marsh succession using perma- nent plots. Vegetatio 6 1: 23-32. SEMENIAK, V. 1983. Mangrove distribution in north- western Australia in relationship to regional and local freshwater seepage. Vegetatio 53: 11-31. SHAY, J. M. AND C. T. SHAY. 1983. Prairie marshes in western Canada with specific reference to the ecology of five emergent macrophytes. Canad. J. Bot. 64: 443-454. SNOW, A. A. AND S. W. VINCE. 1984. Plant zonation in an Alaskan salt marsh. 2. An experimental study of the role of edaphic conditions. J. Ecol. 72: 669- 684. VAN NOORDWIJK-PUUK, ., W. G. BEEFTINK AND P. HOGEWET. 1979. Vegetation development on salt marsh plots after disappearance of the tidal factor. Vegetatio 39: 1-13. WHIGHAM, D. F., C. CHITTERLING, B. PALMER AND J. O'NEILL. 1986. Modification of runoff from up- land watersheds-The influence of a diverse ripar- ian ecosystem, pp. 305-332. In D. L. Correll [ed.], Watershed research perspectives. Smithsonian In- stitution Press, Washington, DC. -, J. MCCORMICK, R. E. GOOD AND R. L. SIMPSON. 1978. Biomass and primary production in fresh- water tidal wetlands of the Middle Atlantic coast, pp. 3-20. In R. E. Good, D. F. Whigham and R. L. Simpson [eds.], Freshwater wetlands: Ecological processes and management potential. Academic Press, New York, NY. - AND R. L. SIMPSON. 1978. The relationship between aboveground and belowground biomass of freshwater tidal wetland macrophytes. Aquat. Bot. 5: 355-364. ZEDLER, J. B. 1983. Freshwater impacts in normally hypersaline marshes. Estuaries 6: 346-355. - AND P. A. BEARE. 1986. Temporal variability of salt marsh vegetation: The role of low-salinity gaps and environmental stress, pp. 295-306. In D. Wolfe [ed.], Estuarine variability. Academic Press, New York, NY. -, J. COVIN, C. NORDBY, P. WILLIAMS AND J. BOUND. 1986. Catastrophic events reveal the dy- namic nature of salt-marsh vegetation in southern California. Estuaries 9: 75-80.