1 The effect of abiotic factors on the distribution of Cerrorchestia hyloraina (Taltridae) above and below the Monteverde Cheese Factory Vanessa Cervantes Department of Biology, Yale University ABSTRACT Taltrid amphipods are terrestrial crustaceans know n to favor moist habitats and feed off of decaying matter. They have been found infestating caves, beaches, and garden leaf litter. Within the past three years, the Cerrorchestia hyloraina (Crustacea: Amphipoda: Taltridae) population local to the Monteverd e area has exploded due to changes in unknown biotic and/or abiotic factors. Three abiotic factors (distance from river, soil moisture, and temperature) were tested at two different sites: one above and one below the Monteverde Cheese Factory (MCF). Taltri d density was significantly higher below (251.800 Â± 170.811 taltrids/trap) the MCF than above (1.619 Â± 1.910 taltrids/trap, Mann Whitney U = 420.0, P < 0.0001) and soil moisture was the best predictor of taltrid density (Simple regression, R 2 = 0.158, P = 0.0102). Soil temperature was also a significant indicator of taltrid density (Simple regression, R 2 = 0.113, P = 0.0316), and distance from river showed no significant correlation with taltrid density, supporting that C. hyloraina is not water dependent. While abiotic factors are strong determinants of where C. hyloraina live, it may also be changes in the local environment resulting from unknown biotic interactions that may explain the recent highly localized proliferation of this species. RESUMEN Es s abido que los anfÃpodos de la familia Taltridae son crustÃ¡ceos terrestres que prefieren habitats hÃºmedos y comen materia en decomposiciÃ³n. Han sido encontrados infestando cuevas, playas, y mantillo de huertas. En los Ãºltimos tres aÃ±os la poblaciÃ³n de Cer rorchestia hyloraina (Crustacea: Amphipoda: Taltridae) encontrada localmente en el area de Monteverde ha sufrido una explosion debido a cambios en factores biÃ³ticos y abiÃ³ticos desconocidos. Tres factores abiÃ³ticos (distancia del rÃo, humedad del suelo, y temperatura) fueron examinados en dos sitios diferentes: uno encima y otro debajo de la fÃ¡brica de quesos de Monteverde (MCF). La densidad de taltrÃdidos fue mÃ¡s alta significativamente debajo (Promedio = 251.800 Â± 170.811 taltrÃdidos/trampa) de la fÃ¡b rica de quesos que arriba (Promedio = 1.619 Â± 1.910 taltrÃdidos/trampa, Mann Whitney U = 420.0, P < 0.0001) y la humedad del suelo fue la major forma de predecir la densidad de taltrÃdidos (Regresion simple, R 2 = 0.158, P = 0.0102). Temperatura fue una fo rma de predecir la densidad de taltrÃdidos tambien (Resgresion simple, R 2 = 0.113, P = 0.0316), y distancia del rÃo no monstrÃ³ una correlaciÃ³n con la densidad de taltrÃdidos, que sostiene que C. hyloraina no depende de agua. Mientras los factores abiÃ³ticos fueron fuertes determinantes de dÃ³nde C. hyloraina vive, puede ser tambiÃ©n cambios en el ambiente local que resultan de interacciones biÃ³ticas desconocidas que pueden explicar la reciente proliferaciÃ³n altamente localizada de esta especie. INTRODUCTIO N
2 Decaying leaf matter is vastly abundant in the rainforests, and for some organisms, it is a seemingly endless food supply. Taltrid amphipods (Crustacea: Amphipoda: Taltridae), a type of macroinvertebrate who feeds off dead leaf litter, actually thrive o n such an unfavorable food source (Duncan 1994). The local Monteverde species is Cerrorchestia hyloraina (Lindeman 1991). The Taltridae are a type of terrestrial crustacean that live in moist habitats, unable to survive in drier areas (Duncan 1994). Cerror chestia hyloraina lack a waxy coat over their thin skin and absorb moisture from the air; thus, environments that are too dry or too wet can be dangerous for these crustaceans (Fasulo 2001). Because the surrounding environment is crucial to the survival o f these organisms, taltrids possess an escape mechanism that allows them to rapidly get away from unfavorable environmental conditions. This mechanism is synonymous to that of suborder, Gammaridea). They are able to leap abruptly by snapping back their tail like urosomes and attached uropod appendages (Brusca and Brusca 1990), which are usually kept tucked underneath their ventral side. Because this behavior is immediate, and i t is difficult for amphipods to quickly turn their bodies, the directions in which the taltrids jump appear to be random. It is likely for this reason that when soil and leaf litter inhabited by C. hyloraina flood due to heavy precipitation, the C. hylorai na utilize this jumping mechanism in order to escape their saturated environment, making their way probably by chance underneath doorways and into pipe systems. For the past three years, residents downstream from the Monteverde Cheese Factory (MCF) have b een finding large numbers of small shrimp like organisms, pers. comm.). Because land hoppers are active at night (Lindeman 1991) these organisms are often found on mornings after heavy precipitation, desiccated near doorways, bath drains, or any other openings leading to the outside. Residents near the Monteverde Cheese Factory (MCF) have reported that of the past thirty years, C. hyloraina have only been appearing within th e last three, and in extremely high numbers (Trostle pers. comm.). Such high densities are only observed during the wet season, which may mean the soil surface air moisture at this time is optimal for supporting this species. The surge in the C. hyloraina population in the Monteverde area is a recent phenomenon, and their rapid proliferation is likely due to either an improved living environment (e.g. moister wet seasons, loss of predators and/or competitors, introduction of beneficial symbionts, etc.) tha t permits an increased carrying capacity for an established population, or the migration of a new population into an optimal environment. In either case, new optimal environmental factors apparently localized in the Monteverde area are allowing this specie s to flourish. Earlier studies have shown that downstream from the MCF, from 1980 to 1990, the number of the amphipod species was actually steadily decreasing, and in 1990, no amphipod species were found downstream from the MCF (Gill 2000). It was believe d neighboring Rio Guacimal, leading to the flourish of lac + bacteria (probably Lactophilus spp.) due to the 3 4% lactose present in the waste (Gill 2000). The high oxygen consumption of these bacteria through respiration caused the water downstream from the
3 MCF to contain decreased oxygen levels (9.1 mg/L upstream, 7.8 mg/L downstream) and increased carbon dioxide levels (8.0 mg/L upstream, 19.6 mg/L downstream) (Gill 2000) . Lindeman (1991) also showed that no C. hyloraina were found below 1400m. Since then, a hog farm was established around 1993 1994 to manage the excess whey and stop dumping waste directly into the river (Bob Law pers. comm.). Unlike other crustaceans, an imals such as C. hyloraina belonging to the superorder Peracarida carry out direct development, in which peracarids produce embryos that brood in the marsupium (a pouch formed from ventral plates) of their mothers (Brusca and Brusca 1990). They are release d as mancas, or subjuveniles, that are not completely developed, but still look like small adults, possibly differing in only pigment and some proportions (Brusca and Brusca 1990). Peracarids thus lack the free living aquatic larval, or nauplius, stage cha racteristic of other crustaceans (Brusca and Brusca 1990). However, it is not known if the Taltridae spend any stages of their development in water, but if they do, the quality of a nearby water source, in this case the dissolved oxygen content of the Rio Guacimal, could greatly affect their development and abundance. Characteristically, amphipods are known to have symbiotic relationships for their feeding strategies and any effect of the water quality on the abundance of these symbionts could also have a s ignificant affect on C. hyloraina (Brusca and Brusca 1990). Such environmental factors stimulating the recent taltrid population growth can either be biotic or abiotic. While biotic factors such as species interactions are more dynamic than changes in clim ate patterns, and thus are a likely significant factor in the recent proliferation of this species, it is the abiotic factors such as climate and weather that may determine the distribution of C. hyloraina . The purpose of this study was to determine what abiotic factors determine C. hyloraina density. The three abiotic factors tested were distance away from river, soil temperature, and soil moisture. Lindeman (1991) found that C. hyloraina occur at higher densities in moister areas and it was expected that soil moisture would be the major factor determining C. hyloraina density for this study as well. Density was expected to be negatively correlated with distance from river if this species is directly or indirectly dependent on this habitat. MATERIALS AND METHODS This study was conducted at two sites along the Rio Guacimal on the properties of Lindi Maxon (1460 m elevation, above the MCF), and John and Sue Trostle (1380 m elevation, below the MCF). Traps were set up in a grid formation in three columns alo ng the river 10 m apart at 5 m distance increments extending away from the river in forested areas (Figure 1). The traps were set out on July 22, 2004 at approximately noon for the site below the MCF, and on July 27, 2004, at noon for the site above the MC F, and then collected at noon two days after placement. Measurements of dissolved oxygen, soil temperature, and soil moisture for each site were taken on the days of trap collection for each site. The dissolved oxygen of the Rio Guacimal was measured using an Oakton Waterproof Dissolved Oxygen Meter (DO 300 Series; WD 35615 75) at three locations, A, B, and C, ten meters apart along the river at both study sites (Figure 1). Temperature and soil moisture were measured at each trap within a 5 cm distance away from the trap rim at an approximate soil 4 cm depth, using a Digi Sense Digital Thermometer (Type K
4 Thermocouple; 8528 40) for temperature and an LIC Soil Moisture Meter (using a scale of 1 10) for soil moisture. Cerrorchestia hyloraina (Figure 2) samples were collected using pitfall traps set up by placing plastic 16 oz. cups into the soil with the mouth open at soil surface in order to catch organisms crawling along the soil surface (Figure 3). It appeared that the C. hyloraina were most abundant underne ath the leaf litter at the soil surface and within the loose topsoil. Therefore, soil was compacted up to the rim of the cup, and a 3 4 cm high and about 15 cm wide layer of loose soil was also added around the rim of the cup, with a 5 7 cm layer of leaf l itter on top of the loose soil (Figure 3). A half a cup of water was placed into the cup to prevent any C. hyloraina from hopping out of the cup. The number of C. hyloraina between 0 5 mm and the number of C. hyloraina greater than 5 mm were recorded for e ach pitfall trap sampled. Simple linear regressions were used to determine if correlations existed between the numbers of C. hyloraina in each trap and distance away from the river, soil moisture, and soil temperature. Mann Whitney tests were used to dete rmine the significance between average taltrid densities and soil moisture above and below the MCF. RESULTS The twenty trap samples from below the MCF contained a total of 5036 taltrids, and the twenty one trap samples from above the MCF contained a tot al of 34 taltrids. Average taltrid density was significantly higher below (Mean = 251.8 Â± 170.8 taltrids/trap, N = 20) the MCF than above (Mean = 1.6 Â± 1.9 taltrids/trap, N = 21, Mann Whitney U = 420.0, P < 0.0001, Figure 4). Soil moisture was also signifi cantly higher below the MCF (Mean = 8.990 Â± 1.644 units) than above (Mean = 7.805 Â± 1.363 units, Mann Whitney U = 320.500, P = 0.0040). Overall, soil moisture was the most significant factor determining taltrid density, positively correlating with taltrid number (Simple regression, R 2 = 0.158, P = 0.0102, Figure 5c). However, soil moisture was not significantly related to C. hyloraina density when looking at each site separately (Figure 5a and b). When examining both sites, soil temperature was significant ly negatively correlated with taltrid number (Simple regression, R 2 = 0.113, P = 0.0316, Figure 6c). Soil temperature was also found to significantly negatively correlated with soil temperature (Simple regression, R 2 = 0.107, P = 0.0368). Soil temperature was also significantly negatively correlated with taltrid number when looking at below the MCF alone (Simple regression, R 2 = 0.272, P = 0.0183, Figure 6b). Above the MCF, however, soil temperature was negatively correlated with taltrid number as well, but this relationship was not significant (Figure 6a). Between both sites, a positive correlation existed between number of taltrids and distance from the river, but this correlation was not significant (Figure 7c). Below the MCF, distance from river was als o positively correlated with number of taltrids, and this correlation was significant (Simple regression, R 2 = 0.222, P = 0.0362, Figure 7b). However, the opposite effect occurred above the MCF, where distance from the river was significantly negatively co rrelated with taltrid number (Simple regression, R 2 = 0.189, P = 0.0491, Figure 7a). Dissolved oxygen content was very similar between below (Mean = 9.920 Â± 0.059 mg/L) and above (Mean = 10.017 Â± 0.168 mg/L, N = 3 both below and above). Also, the
5 ratio of big to small taltrids (number taltrids > 5 mm: number taltrids 0 5 mm) was found to overall be 0.6402, the ratio below the MCF was 0.6356, and the ratio above the MCF was 1.8333. DISCUSSION As moist air is essential to the survival of these taltrids, it was expected that soil moisture would be a strong indicator of taltrid density, and this hypothesis was supported. Overall, soil moisture was the most significant factor determining taltrid density. These findings help support field observations made at th e site below the MCF. Cerrorchestia hyloraina was observed to mostly inhabit the loose topsoil and leaf litter, appearing to be concentrated at the soil surface. The amount of leaf litter in an area seemed to be a determinant of high taltrid density, proba bly because the denser leaf litter prevented more moisture in the ground from escaping into the atmosphere. Further studies should be conducted to determine if a correlation between taltrid density and amount of leaf litter exists. Taltrid density was fo und to significantly correlate negatively with soil temperature. A significant negative correlation also exists between soil moisture and soil temperature because higher temperatures lead to more evaporation. Because soil temperature and soil moisture sign ificantly correlate on their own, a correlation with one sensitive to the changes in moisture than temperature. Therefore, soil moisture is likely the stronger deter minant of taltrid number over soil temperature. It was believed that the dissolved oxygen content of the river may have some positive effect on taltrid density, but the dissolved oxygen did not greatly vary above and below the MCF. Distance from river als o had no overall significant effect on taltrid density. Below the river, however, there was a significant positive correlation between distance and total number of amphipods. A positive correlation does not support the theory that some part of their develo pment is water dependent. Above the river a significant positive correlation existed between taltrid density and distance from river, and it is likely due to factors other than water quality that is driving this trend. Because the taltrid population has b een fluctuating with the seasons, it is possible that their increased abundance during the wet season is indicative that this is also their mating and reproductive season. The ratio of large to small taltrids was 0.06402, meaning there were much more young er taltrids than older ones, indicating that the population of taltrids is breeding. The dramatic increase in taltrid population below the MCF compared to above the MCF indicates that their recent proliferation may be highly localized. While soil moisture and temperature have been shown to be determinants of taltrid density, it is not know what biotic factors may also be affecting their population growth. The more that is learned about C. hyloraina , the more residents can use this knowledge to prevent the appearance of these taltrids into their homes. ACKNOWLEDGEMENTS
6 I would like to thank Dr. Carlos Guindon for all his help on this project, from trying to catch and separate these little buggers to trying to unscramble the crazy statistic chaos that is John and Sue Trostle for allowing me to work on their property so many times as I tried out multiple different trapping techniques. Thank you also to Leroy and Lindi Maxon for letting me use their property as well. Thanks a lso Margaret and Melissa, my San Luis peeps! You certainly made the walk up La Trocha that much easier. Thanks to the awesome Ollie and Maria for the proofreading, the advising, the slideshow, and all the laughs along the way. And a big shout out to all m LITERATURE CITED Brusca, R. C, and G. J. Brusca. 1990. Invertebrates. Sinauer Associates, Inc., Sunderland, Massachusetts. Duncan, K.W. 1994. Terrestrial T alitridae (Crustacea: Amphipoda). Fauna of New Zealand. 31. Featured Creatures . University of Florida, Florida. Publication number: EENY 220. Gill, D. 2000. The Impac t of the Monteverde LecherÃa on the Quebrada Cuacimal. In Monteverde: Ecology and conservation of a tropical cloud forest. Oxford University Press, New York. pp. 446 447. Lindeman, D. 1991. Natural history of the terrestrial amphipod Cerrorchestia hylorai na Lindeman (Crustacea: Amphipoda: Taltridae) in a Costa Rican cloud forest. Journal of Natural History . 25: 623 638. RÃo Guacimal A B C
7 FIGURE 1. Schematic for location of traps. Dissolved oxygen content of river was measu red at points A , B , and C , ten meters apart along the RÃo Guacimal. Three columns of seven pitfall traps extended perpendicularly from the river in five meter increments. Soil moisture and soil temperature were recorded for each trap. FI GURE 2. General morphology of Amphipoda exterior (subphylum Crustacea, superorder Peracarida). In suborder Gammaridea (beach hoppers, land hoppers, etc.), terrestrial gammarids walk on land with abdomen tucked underneath ventral side. 15 cm loose soil leaf litter 5 7 cm 3 4 cm Expected strata of C. hyloraina habitat
8 FIGURE 3. Schematic for pitfall trap. Traps laid so that mouth opened at base of expected habitat strata for C. hyloraina , maximizing the number of specimens able to be caught. When C. hyloraina fall into water, they are FIGURE 4. Mean average taltrid density above and below the Monteverde Cheese Factory (MCF). Average taltrid density was significantly higher below (251.800 Â± 170.811 taltrids/trap) the MCF than above (1.619 Â± 1.910 taltrids/trap, Mann Whitney U = 42 0.0, P < 0.0001). a
9 b c FIGURE 5. Simple linear regressions of total number taltrids/trap versus soil moisture for (a) above the MCF (Simple regression, R 2 = 0.009,P = 0.6825), (b) below the MCF (Simple regression, R 2 = 0.065, P = 0.2780), an d (c) both sites combined (Simple regression, R 2 = 0.158, P = 0.0102). a
10 b c FIGURE 6. Simple linear regressions of total number taltrids/trap versus soil temperature for (a) above the MCF (Simple regression, R 2 = 0.109, P = 0.1434), (b) below th e MCF (Simple regression, R 2 = 0.272, P = 0.0183), and (c) both sites combined (Simple regression, R 2 = 0.113, P = 0.0316). a
11 b c FIGURE 7. Simple linear regressions of total number taltrids/trap versus distance from river for (a) above the MC F (Simple regression, R 2 = 0.189, P = 0.0491), (b) below the MCF (Simple regression, R 2 = 0.222, P = 0.0362), and (c) both sites combined (Simple regression, R 2 = 0.054, P = 0.1434).
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00390
El efecto de los factores abiticos sobre la distribucin de Cerrorchestia hyloraina (Taltridae) arriba y abajo de la fbrica de quesos Monteverde
The effect of abiotic factors on the distribution of Cerrorchestia hyloraina (Taltridae) above and below the Monteverde Cheese Factory
Taltrid amphipods are terrestrial crustaceans known to favor moist habitats and feed off of decaying matter. They have been found infesting caves, beaches, and garden leaf litter. Within the past three years, the
Cerrorchestia hyloraina (Crustacea: Amphipoda: Taltridae) population local to the Monteverde area has exploded due to changes in unknown biotic and/or abiotic factors. Three abiotic factors (distance from river, soil moisture, and temperature) were tested at two different sites: one above and one below the Monteverde Cheese Factory (MCF). Taltrid density was significantly higher below (251.800 170.811 taltrids/trap) the MCF than above (1.619 1.910 taltrids/trap, Mann-Whitney U = 420.0, P < 0.0001) and
soil moisture was the best predictor of taltrid density (Simple regression, R2= 0.158, P = 0.0102). Soil temperature was also a significant indicator of taltrid density (Simple regression, R2= 0.113, P = 0.0316),
and distance from river showed no significant correlation with taltrid density, supporting that C. hyloraina is not water dependent. While abiotic factors are strong determinants of where C. hyloraina live, it may also be changes in the local environment resulting from unknown biotic interactions that may explain the recent highly localized proliferation of this species.
Es sabido que los anfpodos de la familia Taltridae son crustceos terrestres que prefieren los hbitats hmedos y comen materia en descomposicin. Han sido encontrados infestando cuevas, playas, y mantillo de huertas. En los ltimos tres aos la poblacin de Cerrorchestia hyloraina (Crustacea: Amphipoda: Taltridae) encontrada localmente en el rea de Monteverde ha sufrido una explosin debido a los cambios en los factores biticos y abiticos desconocidos. Tres factores abiticos (distancia del ro, humedad del suelo, y temperatura) fueron examinados en dos sitios diferentes: uno encima y otro debajo de la fbrica de quesos de Monteverde (MCF). La densidad de taltrdidos fue ms alta significativamente debajo (Promedio = 251.800 170.811 taltrdidos/trampa) de la fbrica de quesos que arriba (Promedio = 1.619 1.910 taltrdidos/trampa, Mann-Whitney U = 420.0, P < 0.0001) y la humedad del suelo fue la mejor forma de predecir la densidad de taltrdidos (Regresion simple, R2 = 0.158, P = 0.0102). Temperatura fue una forma de predecir la densidad de taltrdidos tambin (Resgresion simple, R2 = 0.113, P = 0.0316), y distancia del ro no mostr una correlacin con la densidad de taltrdidos, que sostiene que C. hyloraina no depende de agua. Mientras los factores abiticos fueron fuertes determinantes de dnde C. hyloraina vive, puede ser tambin cambios en el ambiente local que resultan de interacciones biticas desconocidas que pueden explicar la reciente proliferacin altamente localizada de esta especie.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone--Monteverde
Seleccin de hbitat
Costa Rica--Puntarenas--Zona de Monteverde--Monteverde
Tropical Ecology Summer 2004
Ecologa Tropical Verano 2004
t Monteverde Institute : Tropical Ecology