BACK TO PROJECT LIST

Biodiversity Impact of Cattail (Typha dominguensis) Dominated Marsh Areas in the Seasonal Wetlands at Palo Verde National Park

Keith Camburn West Mecklenburg High School, Charlotte, North Carolina Mary Frieze Arrowhead School, Pray, Montana Robert Kuhn Centennial High School, Roswell, Georgia Dorothy Ponte Woodbridge Middle School, Woodbridge, New Jersey

Abstract: Cattails (Typha dominguensis) are classified as an invasive species in the freshwater marshes of Palo Verde National Park, Costa Rica. Cattails appear to impact the marsh by creating a floral monoculture in areas where it dominates thus possibly causing a reduction in overall marsh biodiversity. This investigation compares bird and insect abundance (both terrestrial and aquatic) and diversity in the cattail dominated marsh areas to marsh areas lacking cattails. The comparison of these two distinct marsh areas may aid in the assessing the impact of cattail monoculture on marsh biodiversity.

Introduction: The cattail (Typha dominguensis) is an invasive and exotic species in the freshwater marshes of Palo Verde National Park, Guanacaste Province, Costa Rica (Gonzales, 2001). These freshwater marshes, located in the Tempisque River watershed, are seasonally flooded during the wet season. High-water levels within the marsh system reach their peak in September or October, but the marsh partially remains flooded to moist for most of the year (Janzen, 1983). The Palo Verde marshes provide permanent and seasonal refuge for a myriad of resident and migrate bird species. Typha has spread invasively in the marsh since its introduction, thus reducing the amount of open, standing water in cattail monoculture areas. As the root mass of the cattail grows forming a dense layer the need for water increases, thus reducing the amount of open, standing water. As the open, standing water is reduced, and Typha density continues to increase, a monoculture of Typha develops. It is believed that this monoculture and lack of open, standing water affects the roosting and feeding behavior of both resident and migratory waterfowl in the marsh. These waterfowl seek new places with open, standing water, thus travelling to nearby rice fields where they consume the rice crop (Gonzales, 2001). Current forest management practices conclude that by reducing the monoculture of Typha within the marsh, biodiversity and open, standing water will increase, reducing the need for waterfowl to migrate to the surrounding rice fields. However, in some wetland regions of North America, Typha is seen as a catalyst for increasing biodiversity; in some cases Typha is purposely introduced. If a Typha dominant monoculture reduces overall diversity in the Palo Verde marsh, areas of open, standing water should support a higher biodiversity of both insect and bird species. The species richness and diversity of both birds and insects in the marsh may provide good indicators of how Typha induces change in biodiversity.

Methods: The study area was located in the Palo Verde freshwater marsh due south of the office of the Organization of Tropical Studies (Figure 1). This site was chosen due to its large expanses of both Typha meadow and open, standing water devoid of Typha. Sampling stations were chosen by interspersion, pairing ponded and meadow areas along three transects. Two sampling stations were located on each transect. Open, standing-water marsh was defined as being devoid of Typha stands, and Cattail meadows were defined as having Typha stands at a density of five shoots/4m². The sampling site area consisted of a 30m radius around each sampling point defined on each (Figure 1). This radius size was chosen to facilitate bird observation. Insect collection was focused in a smaller area (<2m) around the center point of each station. Sampling occurred over three consecutive mornings for approximately 3 hours each day.

Sampling Methods:

Bird observations were conducted by directly observing the six sampling areas for 10 minutes starting at 6:00 a.m. each morning. The occurrence of each bird species and its total abundance was recorded on three consecutive days. Bird identification was based on Stiles and Skutch (1989). Landed birds were defined as those individuals that actually landed within the study area, whereas fly over birds where observed flying over the study area but without actually landing.

Both aquatic and terrestrial insects were sampled at each station upon completion of the bird observations. To ensure uniformity, samples were collected by the same individuals over the three day period. Aquatic insects were sampled using a benthic dip net for one minute within 2m of the center point of each station. The net was dipped throughout the water column and into the benthic layer of the marsh sampling site using short, even strokes. Terrestrial insects were subsequently collected using a mesh net for one minute within the same sampling area of each station. Long sweeping strokes of the net were used across the surfaces of plants and within the air column approximately 3m above the water surface at each sampling location. In either case, the nets were used in such a manner as to prevent loss of sampled materials. The contents of each net were then carefully placed in a labeled Ziploc bag for identification and enumeration. Insects were identified to Order (Borror, et al., 1989). Unknown insects were defined by morphotype. Although other invertebrates were collected, only insects were included due to uncertainty at identifying other groups. Both larval and adult stages of insects were counted.

Statistical Methods:

Diversity of the ponded marsh (cattail free) and cattail meadow was calculated using the Shannon-Weiner Diversity Index (H’), which is commonly used for a quantitative expression of diversity.

H’= nLogn-? (nLogn+nLogn…)

n

Where n= sample size

However, H’ is a more reliable measure as sampling size increases, so the addition of the calculation of evenness (J’) is often applied.

J’=H’/H’_max, where H’max= Logn

Dominance within the sample was also measured by using 1-J’ as a measure of heterogeneity.

Lastly, the similarity coefficient (CCj) is often used when the data consists of the presence or absence of species, as was seen in our samples (Zar, 1999). In this case, similarity is measured between the two marsh habitats in terms of bird and aquatic insect assemblages. High similarity might indicate that a monoculture presence of Typha does not influence bird and aquatic insect communities in the marsh. The formulae for calculating community similarity that was used is derived from Muller-Dombois and Ellenberg (1977):

Where C = # species in common between communities, S1 and S2 = # of species in communities 1 and 2 respectively.

Diversity was calculated for the AQ (aqueous insects) and AE (aerial insects) for each sample by using the general abundance of 7 orders within each sample. Orders were used instead of species as the highest taxonomic confidence level. For birds, species were used in all calculations as the highest level of taxonomic confidence.

Insect Abundance:  Insects were represented by 7 orders: Odonata, Coleoptera, Plecoptera, Ephemoptera, Hemiptera, Diptera, and Hymenoptera. The hemipterids and dipterids were the most abundant in all samples due to the large abundance of larval midges and giant water bugs (Table 1). Insects sensitive to pollution were used as a rapid bioassessment of the wetlands. Morphospecies present that indicated poor water quality included midges and some mollusks. These morphospecies were primarily found in the ponded water samples. Morphospecies that indicate good to very good water quality include: dragonfly larvae, damselfly larvae, stonefly larvae, mayfly larvae, and riffle bugs. Rapid bioassessment of the wetlands based on the mean indicators present in the samples indicate a good to very good water quality (EPA, 2001).

Insect Diversity:  Diversity of aquatic insects in the ponded habitat was significantly higher that in the cattail meadow (Figure 2). This is most likely due to the high abundance of giant water beetle larvae consistently found in the aquatic cattail meadow samples, proportionate to other species. Dominance (1-J’) values show 86% dominance in the habitat, and a very low (0.14) evenness level. The similarity coefficient (CCj) between the two habitats show a 40% similarity. Aerial diversity for the ponded habitat and the cattail meadow were both very low (0.37 and 0.38, H’max=0.60), but very similar (see Table 3, Figure 3).

Bird Abundance:  During this study a total of 12 bird species, represented by 253 individuals, were observed actually landing within the sample stations or flying over them (Table 2). These species represented a wide range of birds including those which actually nest in the marsh, utilize the marsh as a feeding area, and/or species which fly over the marsh travelling from roosting to feeding areas. The most abundant species observed included Northern Jacana (Jacana spinosa), 54 individuals; Black-bellied Whistling-Duck (Dendrocygna autumnalis), 79 individuals; Groove-billed Ani (Crotophaga sulcirostris), 35 individuals; and Red-winged Blackbird (Agelaius phoeniceus), 57 individuals (see Table 2). These total individuals are cumulative for both sampling days. The eight additional species observed included species which flew into the marsh from surrounding forests (e.g., Hoffmann’s Woodpecker (Melanerpes hoffmannii), species from the surrounding forest which use the marsh for roosting and perhaps some feeding activity (e.g., Common Ground-Dove, (Columbina passerina), marsh species which were observed flying over the study area in route from roosting to feeding areas (e.g., Great Egret, (Casmerodius albus), and those which actually feed over the marsh (e.g., Snail Kite, Rostrhamus sociabilis). Of special note was the occurrence of a fly over Yellow-headed Caracara (Milvago chimachima), a species first observed in Costa Rica in 1973 (Stiles and Skutch, 1989). This occurrence may represent a first sighting of this species in Palo Verde National Park (Dr. Thomas Langen, Clarkston University, personal communication).

Bird Diversity:  The total number of landed and fly over bird species for each of the six sites is presented in Table 2. Bird diversity for landed birds in the ponded marsh habitat as measured by the Shannon-Weiner Diversity Index (H’) was 0.55 (H’max = 1.07) while the cattail meadow exhibited a value of 0.36 (see Table 3 and Figure 5). The evenness (J’) value was 0.51 for the ponded marsh and 0.34 for the cattail meadow (see Table 3). The coefficient of similarity (CCj) showed a 60% similarity between both habitats, where only three of the 12 species enumerated were shared between all sites (Table 3).

Insects-Results indicate that aquatic insects may be more diverse in the ponded marsh (cattail free) than in the cattail meadow habitat. Although the aquatic data is somewhat inconclusive (a consequence of the small sampling size and brief sampling period), the data does show some interesting trends. First, there may be a marked difference between the evenness and diversity of aquatic insect populations. The giant water beetle larvae dominated the cattail meadow samples (Table 1). The abundance of these larvae may help to explain the low overall diversity and evenness results measured. Secondly, the similarity of aquatic species between the two habitats (ponded and cattail meadow) samples may be low due to differences in standing water depth and the availability of light. In addition, more open water naturally occurred in the ponded habitat when compared to the cattail meadow. Lastly, the cattail root systems and the base of the submerged leaves undoubtedly provide a refuge for early larval beetle species that more open, ponded habitat may not. This effectively would cause diversity to decrease overall as a single species or a few species numerically dominant the habitat. This is supported by the high dominance value determined for the aquatic cattail meadow insect data compared to the aquatic insect data from the ponded marsh (0.86 and 0.39, respectively). Aerial insect data showed no significant patterns in terms of overall diversity. Based on the insect data, the floral monoculture of the Typha meadow does exhibit a lower overall diversity of insect orders when compared to the ponded marsh habitat.

Birds-The bird data indicates that the Palo Verde marsh supports a diverse and important avian fauna in northwestern Costa Rica. This is indicated by the occurrence of the four dominant species (Northern Jacana, Black-bellied Whistling-Duck, Groove-billed Ani, and Red-winged Blackbird). The most prevalent and vocal bird in the marsh was the Northern Jacana. This species is associated with ponds and marshes where the species walks over floating vegetation hunting for aquatic insects, small fishes, and snails (Stiles and Skutch, 1989). This species was observed displaying its typical frequent territorial squabbles and copulation was observed on a number of occasions. Also observed was an abandoned nest situated at the edge of Typha stand. The Black-bellied Whistling-Ducks observed during this study were primarily seen resting at the edge of the marsh and/or flying over the marsh. This species, usually feeds on seeds by walking or standing in shallow water (Stiles and Skutch, 1989). Individuals within the Palo Verge marsh appeared to be using this area as a staging area. The occurrence of Groove-billed Ani within the marsh is problematic. This species would normally be associated with cattle, as it would feed on insects being stirred up by these moving animals. The Groove-billed Ani observed within the Typha stands appeared to be travelling in-groups of three to five and did not appear to be displaying feeding behavior within the marsh. The results indicate that the Red-winged Blackbird is preferentially utilizing the Typha stands as a nesting site as old nests as well as young birds were observed in this habitat. This species often forages within aquatic vegetation for insects and spiders (Stiles and Skutch, 1989).

Overall landed bird diversity does indicate that the ponded marsh does harbor a higher overall diversity of birds (Shannon-Weiner Diversity Index values of 0.55 compared to 0.36). However, the two habitats were shown to be 60% similar as measured by a Coefficient of Similarity. In addition a number of species were recorded only as fly overs or were not directly observed within the study areas. Many of these species would utilize these marsh as important feeding and resting areas. These species would include Great Egret, Jabiru (Jabiru mycteria), Wood Stock (Mycteria americana), White Ibis (Eudocimus albus), and Bare-throated Tiger-Heron (Tigrisoma mexicanum). The occurrence or lack of occurrence of these and additional species within the study area is undoubtedly related to the small sample size, time of breeding, and diurnal moments.

The Palo Verde marsh provides a unique, multileveled, habitat for a variety of bird species. Within the marsh it would appear that the monoculture Typha stands physically support a limited number of bird species, with the Red-winged Blackbird being the most conspicuous. The open, standing water areas of the marsh provide critical habitat for the Northern Jacana which utilizes this area for feeding and displaying but would builds its nest in more secluded areas adjacent to the Typha stands. The Black-bellied Whistling-Ducks utilize this marsh for a roosting and staging area. As the Typha continues to expand its dominance in the marsh the Black-bellied Whistling-Ducks will continue to utilize neighboring rice fields as an easy food source thus contributing to yet another man versus wildlife conflict.

In conclusion it would appear that based on aquatic insect and bird data the ponded, open water areas of the Palo Verde marsh support a more diversity fauna than the Typha meadows. This is no doubt related to the lack of habitat complexity associated with a Typha monoculture. This habitat did however, clearly support a number of aquatic insects which may, for a variety of reasons, heavily utilize this habitat. Reasons are no doubt related to the fact that these aquatic insects are using the Typha stands as a nursery area thus limiting their exposure to the multitude of predators in the more open water areas of the marsh. Bird diversity is also less in the Typha meadows. Clearly, the density of Typha stems per unit area and the concomitant "near" lack of open water limits physically the number of bird species, which can utilize this area for feeding. The Red-winged Blackbirds are however, adapted to feeding and nesting within these meadows. The density of the Typha stem would thus protect their nest from overhead predatory birds as well as snakes. These Typha meadows do undoubtedly support a unique bird fauna, which is very difficult to study and observe. Many of the species that would heavily utilize this very protective habitat, such as crakes, rails, and wood-rails, would be very difficult if not impossible to observe within the scope of this study. As the Typha continues to expand its density with the Palo Verde marshes and open, ponded areas become less available, many species will be faced with the need to find more suitable nesting and feeding areas. This appears to be happening at the current time with regard to the Black-bellied Whistling-Ducks which are increasing their foraging activity in the adjacent rice fields. As the open and shallow water zones are "consumed" by the Typha meadows and rice fields continue to proliferate within the Tempisque River valley these ducks will continue to be perceived as agricultural pest.

Acknowledgements:

We wish to acknowledge the continued support and guidance of Dr. Thomas Langen and the rest of the WWF staff, and Mauricio Castillo, GIS Specialist, OTS Palo Verde.