1 The effects of habitat heterogeneity in the parasocial orb weaver Metabus gravidus ( Araneae: Araneidae) Sara Wilbur Department of Biology, University of Puget Sound ABSTRACT Metabus gravidus (Araneae: Araneidae) is a colonial, orb weaving spider, conspicuous along streams and rivers in Central America. These spiders construct individual webs along anchor lines spanning small waterways, placing foraging spiders i n areas of high insect traffic Riparian habitats are characterized by heterogeneity of physical factors, such as rainfall, temperature, and cloud cover. The purpose of this study was to determine the effects of habitat heterogeneity on daily M. gravidus presence M. gravidus in dividuals in webs were counted daily in nineteen colonies along La Quebrada Mquina in Monteverde, Puntarenas province, Costa Rica. Abiotic conditions, such as previous 24 hour rainfall, temperature, cloud cover, rainfall during measurements, and wind inte nsity were recorded. None of the abiotic factors considered in this study had a significant impact on the numbers of M. gravidus present. Although abiotic factors are normally important d eterminants in spider establishment the unique habitat occupied by M. gravidus is likely to render these usually damaging factors insignificant. RESUMEN Metabus gravidus (Araneae: Araneidae) es una araa que es colonial y hace redes crcula res a travs de las quebradas y ros de Amrica Central. Estas araas construyen redes individua le s sobre redes de anclaje a travs de quebradas peque as entonces las araas estan en los reas con muchos insectos para cazarlos. Los hbitat s de ros son heterogeneous en las caracteristicas fsicas como lluvia, tempuratura, y la s nubes. El objet iv o de este proyecto era determinar el efecto de hbitat s heterogeneo s en los n meros diarios de M. gravidus Los i ndividuos de M. gravidus en las redes eran contados cada da en diezinueve colonias en La Quebrada Mquina en Monteverde, Punt arenas, Costa Rica. Condiciones abiticas, como la lluvia vienti cuatr o horas despus, tempuratura, la s nubes, lluvia y viento, eran registrados. Nunca de los condiciones abiticos consideraban en este proyecto tenan un impacto considerable en los numeros de M. gravidus que estaban presente s Aunque normalmente condiciones abiticos son determinantes importantes en el establecimiento de las araas, el hbitat nico donde vive M. gravidus probablemente deja estos factores afuera por que usualmente sus daos son insignificantes. INTRODUCTION Metabus gravidus ( Araneae: Araneidae ) is a colonial orb weaving spider found in riparian habitats throughout Central America (Spencer 2008 ). While colon i alism can aid in juvenile rearing, prey capture, and defense from predators and kleptoparasites, M. gravidus is a parasocial spider, living in loosely organized web colonies of five to 70 spiders lacking any cooperation between individuals (Buskirk 1975b; Foelix 1996). Several advantages for colonial living have been postulated, including group defense against both predators and feeding areas, exploitation of a resource not available to solitary individuals, and for juveniles to learn proper fo raging methods. Fur thermore, a group can modify the environment to make it more favorable for the colony in ways individuals cannot (Buskirk 1975b).
2 Group h abitat alteration by M. gravidus to improve foraging conditions for individual spiders is especially beneficial in the heterogeneous environments these spiders colonize. The physical setting of Monteverde Costa Rica encompasses a wide range of environmental conditions ( Clark et al. 2000). the continental divide, geology, and hydrology produce high abiotic diversity that in turn has a direct effect on great biological diversity Three seasons are recognized in Monteverde: the dry season (February April), the wet season (May October), and the transition season (November January). The tra nsition season is characterized by strong northeasterly trade winds, wind driven precipitation, and mist, which increase with elevation (Clark et al. 2000). These abiotic patterns typical of the transition season are intensified in riparian systems, where strong, highly variable biophysical gradients produce particularly heterogeneous environments ( Na iman et al. 2005). Thus, organisms that colonize riparian habitats must be adept at foraging and surviving in the face of high environmental fluctuation to mai ntain healthy populations. Normally, s pider habitats are strictly defined and are limited by physical conditions such as temperature and precipitation (Foelix 1996). Although there is some plasticity in individual web placement due to seasonal and microhabitat changes, M. gravidus aggregations are re latively permanent in location (Buskirk 1975b). Foraging efficiency due to colonality in M. gravidus may also provide eff ective hunting in environmentally unfavorable conditions, such as heavy rainfall or low temperatures. Previous studies have explored many aspects of M. gravidus behavior, including interactions within colonies (Buskirk 1975a; Gerst 1999), habitat selection (Buskirk 1975b; Gillespie 1987; Melchiors 2005), and prey capture (Kerzicnik 1993; McLaughlin 2008; Spencer 2008; ), but there is an incomplete understanding of the important effects that abiotic conditions have on M. gravidus colony characteristics. I hypothesize d that with an increase in 24 hour rainfall and a decrease in temperature M. gravidus presence will not be affected due to their ability to adapt to the heterogeneity characteristic of riparian habitats. METHODS Study site The study are a was a 300 meter stretch of La Quebrada M quina near the Estacin Biolgica in Monteverde, Puntarenas province, Costa Rica. This stretch of stream occupies a steep valley and has a rocky or gravelly bottom Boulders, fallen logs, and branches line the riverbanks and often cross the entire stream at narrower s ections. Small waterfalls occur at varying locations along the stream and the water is generally fast moving, although some small pools s till the water in several locations High cloud cover, bursts of wind and associated mist, occasional rain, and temperatures from 16 21C characterize typical Monteverde morning weather. Measurements The effects of rainfall and temperature on number of spiders present were measured in nineteen M. gravidus colonies ranging from 1 23 webs per colony. Measurements were taken between 0900 and 11 00 daily for two weeks in November 2009. Millimeters of rainfall were recorded at the start of each sampling, rep resenting 24 hours of rainfall. Weather patterns were also noted, including temperature, current level of precipitation, degree of cloud cover, and general
3 wind/mist strength At each site, number of webs with spiders was recorded. General physical characteristics of colony placement, s uch as riverbank vegetation, presence of boulders or fallen trees, and speed of current, were also noted. Data analysis A one wa y ANOVA was performed to compare daily average dif ferences in number of spiders on webs per colony. Regression analyses were u s ed to measure the effects of rainfall an d temperature on the number of webs with spiders. T tests were performed to analyze differences in spider numbers betwe en sunny and cloudy conditions, rainy and dry conditions and windy and still conditions All re ported errors are standard error (JMP, 2002) RESULTS General habitat and colony observations Colonies were typically anchored across narrower stretches of the stream, while colonies situated in wider sections were usually anchored along riverbank vegetation parallel to the water. Anchor lines were often crisscrossed between fallen logs or branches and were sometimes attached to rocks or boulders in the stream (Figure 1) Colonies were often found immediately downstream from a small waterfall, stretching over or along a still pool or rapids generated by the waterfall. Although daily numbers of webs and spiders varied slightly webs were typically constructed in similar locations within the colony. Adult M. gravidus were almost exclusively found within 20 Effects of abiotic factors on spider presence Daily fluctuations in number of spiders per colony were not different th roughout the study period for sixteen of nineteen colonies measured ( one way ANOVA, F 1,161 = 0.0346, p = 0.8256 ) Colonies ten, thirteen, and nineteen had significantly different daily fluctuations in number of spiders throughout the study period (one way ANOVA; colony ten : F 1,7 = 14.5385, p = 0.0066; colony thirteen: F 1,6 = 6.3253, p = 0.0456; colony nineteen : F 1,6 = 7.0187, p = 0.0381 ). The number of spiders in webs was not affected by temperature (R 2 = 0.000, p = 0.9672 ; Figure 2 ). Temperatures ranged from 17.5 to 20.2C. Most daily counts of spiders in webs ranged from 17 to 42, although there was one day (18C) where no spiders were fo und. Rainfall did not affect the number of spiders encountered ( R 2 = 0.414, p = 0.06 15 ; Figure 3 ) The amount of r ainfall in a 24 hour period ranged from zero to eight millimeters Generally cloud cover, rainfall during measurements, and presence or absence of wind did not affect the number of spiders encountered. Presence or ab sence of clouds had a marginal effect on the average number of spiders present in webs ; u nder sunny conditions, the mean value of spiders present was 2.33 0.54, while the mean value of spiders present under cloudy conditions was slightly lower at 1.27 0.17 (t test, t = 1.870, df =1 88 p = 0.06 30 ; Figure 4) Presence or abs ence of rain during data collection did not affect the number of spiders present (t test, t = 1.040, df = 1 88 p = 0.2998 ; Figure 5 ) Mean spider number under rainy conditions was 1.03 0.37 while mean spider number under dry conditions was 1.46 0.19 Similarly, the
4 presence or absence of wind during measurements did not affect the number of spiders encountered (t test, t = 0.295, df = 1 88 p = 0.7683 ; Figure 6 ). In the presence of wind, mean spider number was 1.33 0.22 while in still conditions mean spider number was 1.43 0.26. DISCUSSION The purpose of this study was to examine the effects of daily precipitation and temperature on numbers of M. gravidus individuals present in webs. In addition daily numbers of spiders per colony were examined to see if group size was maintaine d throughout the experiment. Finally, the effects of abiotic factors such as cloud cover, rainfall during measurements, and wind presence were analyzed in relation to number of spiders present. This wide range of abiotic factors and their effects on spider p opu lation size provides broad insight s into the effects of habitat heterogeneity in M. gravidus Habitat analysis Buskirk (1975b) found that M. gravidus colonies are not randomly distributed across streams or on bank vegetation, but are associat ed with stream characteristics such as narrow stream width, steady current, and ov erhanging ledges or prominences; my data are concurrent with these findings My data concerning relatively per manent M. gravidus aggregations with slight daily variation in sp ider numbers within colonies is also in agreement with previous M. gravidus studies, in addition to the relative location of adults and juveniles within the colonies (Buskirk 1975b; Peterson 2007). While Peterson (2007) also studied M. grav idus aggregations in La Quebrada Mquina Buskirk (1975b) analyzed habitat preference in the Guacimal River, located in another area of Puntarenas county. Data agreement between my study and the study conducted in the Guacimal River indicate that M. gravid us colonies are constructed with preference to similar stream characteristics, irres pective of the different water bodies M. gravidus colonize. Abiotic factors and effects on daily colony size Across nineteen colonies, sixteen did not show significant di fference s in daily colony size, indicating that, in general, daily fluctuations in abiotic conditions did not have a significant effect on colony size throughout the study period. This suggests that spiders are permanent members of a given colony and const ruct individual webs amidst the same anchor lines day to day In terms of groups whose size s were significantly affected throughout the study period, c olonies ten and thirteen were relatively small juvenile aggregations, ranging from zero to three and zero to five spiders, respectively. Fluctuations in these coloni es were likely due to site fidelity differences between juveniles and adults, which show higher specificity for a specific site than do juveniles (Buskirk 1975b). Colony nineteen the third group with significant daily variation in spider number, was a n aggregation of four adult spiders. The anchor lines constructed by these spiders were the longest of any colony with few connections to boulders or vegetation and could have been more affected by rain or wind t han other colonies with shorter, better connected anchor lines Temperature did not have a significant effect on num ber of spiders present This could be attributed to the narrow range of dai ly temperatures throughout the study; the lowest temperature
5 was 17.5C while the highest was 20.2C, not even spanning a full 3C. The small difference between minimum and maximum temperatures is most likely the reason for similar spider numbers per colon y. The absence of spiders at 18 C is possibly due to the heavy rain falling during data collection, likely knocking down webs that had been constructed or completely preventing spiders from foraging. Similarly, t he effect of 24 hour rainfall on daily numbers of spiders was marginally significant; generally, spider numbers decreas ed with a n increase in rainfall Although previous 24 hour rainfall may ha ve been responsible for some web damage, M. gravidus is diurnal spending nights under stones or logs after tearing down and eating webs (Buskirk 1975b). If rain ceased before sp iders began to build their webs, 24 hour rainfall would not be a likely factor contributing to decreased spider numbers. The negative trend s een in my data is more likely due to damage to anchor lines rather than to individual webs. Abiotic factors such a s temperature and 24 hour rainfall are clearly not affecting spider presence and foraging behavior. M. gravidus numbers were not significantl y affected by abiotic conditions such as cloud cover, rainfall and presen ce or absence of wind during measurements. W hile collecting data we ather patterns fluctuated often and quickly; on a typical day, cloudy conditions were the predominant pattern, although the sun would often briefly break through Quick glimpses of sun were likely due to fluctuations in wind intensity, which was also a typ ical phenomenon during data collection. Winds may negatively affect organisms by directly exerting physical disturbances on their prey capture constructions (Liao et al. 2009). However, M. gravidus was not affected by absence or presence of wind. Though th e study period was characterized by heavy night winds, wind intensity was never as strong as from the evening prior and may not have been intense enough to disrupt web structure or discourage spiders from foraging. In terms of rainfall during measurements, similar numbers of spiders present during both rainy and dry conditions may be due to differences in web angle and width between web radii. Juveniles build webs with radii that are closer together than those built by adults (Buskirk 1975b) While rain mig ht affect webs with tight radii, this is not an issue with M. gravidus juveniles as they construct webs perpendicular to the water surface. Adult webs, built parallel to the stream or river over which they are erected, may not be affected by rainfall as th e radii are very widely spaced, lessening the chance that rain drops would damage web structure. Furthermore, because with a steady current (Buskirk 1975b) adu lt M. gravidus may be used to occasional splashes from the stream or river beneath, rendering rainfall an insignificant abiotic factor in terms of potential web damage. Finally, juvenile spiders typically build webs under bank vegetation (Buskirk 1975b), t hese webs may be further protected from rainfall. M. gravidus maximizes foraging efficiency by constructing relatively permanent anchor lines across water bodies upon which individual webs can be constructed. Thus, anchor lines permit the placement of webs in areas of hig h insect traffic, allowing exploitation of this otherwise inaccessible prey resource. Because abiotic factors are important in determining habitat selection and foraging behavior in all spider species (Foelix 1996), it is worthwhile to study the effects of rainfall, temperature, cloud cover, and wind intensity to see if they incur damage on individual webs. If fewer spiders were present during times of environmental stress, this would indicate that they were saving reso urces (energy, silk, etc.) until conditions were more favorable. However, if abiotic factors did not hinder M. gravidus foraging, as was found in this study, this would indicate that typically unfavorable conditions for spider foraging were not damaging en ough to avoid foraging in this species implying that it was more energetically
6 favorable to hunt more or less continuously, even in the face of wind, rain, temperature fluctuations, or degree of cloud cover. To help complete understanding of the effects o f abiotic conditions on M. gravidus foraging characteristics, the effects of rainfall, temperature, wind, etc. could be tested on insect abundances and types. Including prey type data with general abiotic impacts on foraging would add anot her layer of know ledge on these unique predators. ACKNOWLEDGEMENTS Special thanks goes out to Anjali Kumar for her numerous suggestions, unwavering patience, statistical tutorials, and suitable study site hunting. Gratitude must also be extended to Pablo Allen, Jos Carlos Caldern, and Yimen Araya for providing me with essential supplies and advice along the way. Finally, without the support of La Estacin Biolgica Monteverde and the allowance granted to me to us e La Quebrada Mquina, the unique natural history of these spid ers could not have been studied. LITERATURE CITED B USKIRK R.E. 1975a. Aggressive display and orb defence in a colonial spider, Metabus gravidus Animal Behavior 23: 560 567. ___________. 1975 b Colonality, activity patterns and feeding in a tropical orb weaving spider. Ecology 56: 1314 1328. C LARK K.L., R.O. L AWTON AND P.R. B UTLER Monteverde: Ecology and Conservation of a Tropical Cloud Forest Eds. N.M Nadkarni and N.T. Wheelwright. New York: Oxford University Press, 2000. 15 38. F OELIX R. R. Biology of Spiders 2 nd ed. New York: Oxford University Press, 1996. G ERST K.L. 1999. The Effect of Spider Size on Colony Dynamics in the Colonial Orb Weaving Spid er, Metabus gravidus UCEAP Monteverde Tropical Biology Fall 1999. G ILLESPIE R.G. 1987. The Mechanism of Habitat Selection in the Long Jawed Orb Weaving Spider Tetragnatha elongata (Araneae, Tetragnathidae). Journal of Arachnology 15: 81 90 K ERZICNIK L. 1993. The relationship of the water surface and prey capture with the colonial orb weaving spider, Metabus gravidus CIEE Monteverde Summer 1993: 154 165. L IAO C., K. C HI AND I. T SO 2009. The effects of wind on trap structural and material properties o f a sit and wait predator. Behavioral Ecology 20: 1194 1203. M C L AUGHLIN J. 2008. Differential response to vibration stimulus in Metabus gravidus (Araneidae) communities. CIEE Monteverde Spring 2008: 76 83. M ELCHIORS K. 2005. Colonial placement behaviors of Metabus gravidus CIEE Monteverde Summer 2005: 107 114. N AIMAN R.J., J.S. B ECHTOLD D.C. D RAKE J.J. L ATTERELL T.C. EEFE AND E.V. B ALIAN Ecosys tem Function in Heterogeneous Landscapes. Eds. G.M. Lovett, M.G. Turner, C.G. Jones and K.C. Weathers. New York: Springer, 2005. 279 309. P ETERSON L. 2007. Effect of colony size on age structure and behavior of Metabus gravidus (Araneidae) in the Montever de Cloud Forest. CIEE Monteverde Spring 2007: 124 130. S PENCER S. 2008. The Dynamics of Prey Capture of the Colonial, Orb Weaving Spider Metabus gravidus (Araneidae) in Monteverde, Costa Rica. CIEE Monteverde Summer 2008: 86 91.
7 FIGURES F IGURE 1. Schematic of a typical M. gravidus colony Locations of (1) egg case, (2) spiderlings, (3) older juveniles, (4) retreat for colony occupants, (5) denser aggregation upstream, and (6) open edge of colony downstream are shown (adopted from Buskirk 1975b).
8 F IGURE 2 The effect of temperature on n umber of spiders Temperature was taken at the beginning of each day of measurements and spiders were counted if they were waiting in a web. Temperatures ranged from 17.5 to 20 .2 C and daily numbers of spiders enco untered ranged from zero to 42. Number of spiders was not affected by temperature M. gravidus foraging may not have been affected by fluctuations in temperature due to the small range of daily temperature s. 0 5 10 15 20 25 30 35 40 45 17 17.5 18 18.5 19 19.5 20 20.5 Number of spiders Temperature (degrees C)
9 F IGURE 3 The effect of rainfall on number of spiders in webs. Rainfall was recorded as the amount that had fallen in the 24 hour period prior to the time of measurement, while s piders were counted if they were waiting in a web. Rainfall ranged from zero to eight millimeters and daily numbers of spiders ranged from zero to 42. Amount of rainfall had a marginally significant affect on the number of spiders present; in general, fewer spiders were counted with heavier rainfall. M. g ravidus only hunt from dawn to sunset; therefore, if rain ceased before spiders began to build their webs, it is unlikely that previous 24 hour rainfall would impede foraging and spider presence. 0 5 10 15 20 25 30 35 40 45 0 2 4 6 8 10 Number of spiders Rainfall (ml)
10 F IGURE 4 Mean number of M. gravidus present during sunny days (n = 2) and cloudy days (n = 6). There were fewer spiders pres ent on cloudy days than on sunny days although this difference was only marginally significant. Small differences in means between sunny and cloudy conditions were likely due to rapid tur nover in amount of cloud cover; normally, mostly cloudy conditions dominated measurement days, although the sun would break out briefly. Likewise, on mostly sunny days, clouds would often block the sun for brief periods of time. 0 0.5 1 1.5 2 2.5 3 3.5 Sunny Cloudy Average number of spiders
11 F IGURE 5 Mean numbers of M. gravidus present during rainy days (n = 2) and dry days (n = 6). Differences betw een means were not significant. Rain presence was recorded if it was raining during data collection. Adult M. gravidus build horizontal webs with widely ra dii, while juveniles build vertical webs with narrowly spaced radii. Both web types appear to effectively leading to the assumption tha t webs are designed to tol erate splashes from the stream or river over which they are erected. Finally, juvenile M. gravidus build webs under vegetation, further protecting them from rainfall. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Rain No rain Average number of spiders
12 F IGURE 6 Mean numbers of M. gravidus present during windy days (n = 5) and still days (n =3). Differences bet ween means were not significant. Although the presence or absence of wind was based on the major pattern throughout the collection day, frequent, sudden fluctuations in wind intensity is likely to account for the insignificant diffe rences between the means. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 Windy No wind Average number of spiders
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 M39-00322
Los efectos de la heterogeneidad del hbitat de la araa Metabus gravidus (Araneae: Araneidae)
The effects of habitat heterogeneity in the parasocial orb-weaver Metabus gravidus (Araneae Araneidae)
Metabus gravidus (Araneae: Araneidae) is a colonial, orb-weaving spider, conspicuous along streams and rivers in Central America. These spiders construct individual webs along anchor lines spanning small waterways, placing foraging spiders in areas of high insect traffic. Riparian habitats are characterized by heterogeneity of physical
factors, such as rainfall, temperature, and cloud cover. The purpose of this study was to determine the effects of habitat heterogeneity on daily M. gravidus presence. M. gravidus individuals in webs were counted daily in nineteen colonies along La Quebrada Mquina in Monteverde, Puntarenas province, Costa Rica. Abiotic conditions, such as
previous 24-hour rainfall, temperature, cloud cover, rainfall during measurements, and wind intensity, were recorded. None of the abiotic factors considered in this study had a significant impact on the numbers of M. gravidus present. Although abiotic factors are normally important determinants in spider establishment, the unique habitat occupied by M. gravidus is likely to render these usually damaging factors insignificant.
Metabus gravidus (Araneae: Araneidae) es una araa colonial que hace sus redes visibles a lo largo de las quebradas y los ros de Amrica Central. Estas araas construyen las redes individuales sobre las redes de anclaje a lo largo de las quebradas pequeas, entonces las araas estn en las reas con muchos insectos para cazarlos. Los hbitats de los ros son heterogneos en las caractersticas fsicas, como las precipitaciones, la temperatura, y la nubosidad. El objetivo de este proyecto era determinar el efecto de los hbitats heterogneos en los nmeros diarios de M. gravidus. Se contaron diariamente a los individuos de M. gravidus en las redes de diecinueve colonias a lo largo de La Quebrada Mquina en Monteverde, Puntarenas, Costa Rica. Se registraron las condiciones abiticas, tales como las precipitaciones veinticuatro horas antes, la temperatura, la nubosidad, las precipitaciones durante las mediciones y la intensidad del viento. Ninguno de los factores abiticos considerados en este proyecto tena un impacto significativo en los nmeros de M. gravidus que estaban presentes. Aunque normalmente las condiciones abiticas son determinantes importantes en el establecimiento de las araas, el hbitat nico donde vive M. gravidus probablemente deja estos factores afuera por que usualmente sus daos son insignificantes.
Text in English.
Costa Rica--Puntarenas--Monteverde Zone
Seleccin de hbitat
Costa Rica--Puntarenas--Zona de Monteverde
Tropical Ecology Fall 2009
Ecologa Tropical Otoo 2009
t Monteverde Institute : Tropical Ecology