Cosma 1 Edge e ffect on m oth r ichness a bundance, and p otential pollination a ctivity in a Cos t a Rican cloud f orest Chris topher Cosma Department of Ecology, Evolution, and Marine Biology University of California, Santa Barbara EAP Tropical Biology and Conservation Program, Fall 2016 17 December 2016 ABSTRACT Habitat loss and fragmentation are major threat s to biodiversity, especially in tropical Edge effects are important considera tion s when investigating the impact of forest fragmentatio n on biodiversity as they can lead to large, and often detrimental shifts in population, community, and ecosystem stability. T he taxa specific effects of forest edges are complex, highly variable and poorly understood In this study, I investigate d t he edge effect on moth richness abundance and potential pollination activity in a fragmented tropical cloud forest in Monteverde, Costa Rica by sampling moths with light traps at two different locatio ns: in the forest interior, and at the forest edge. My results suggest that there is a negati ve edge effect on moth richness abundance, and potential pollin ation activity probably caused by wind. By shedding light on the possible edge effects on this extremely diverse and ecologically important t axon, this study provides valuable preliminary insight s to ecolog ists and conservationists working to slow the loss of biodiversity in tropical forests Efecto de borde en la riqueza de polillas, su a bundancia y la potencial actividad de polinizacin en un bosque nuboso de Monteverde RESUMEN La prdida de hbitat y la fragmentacin son las principales amenazas para la biodiversidad, especialmente en los bosques tropicales que albergan a la mayora de las especies del mundo. El efecto de borde debe ser tomado como una importante consideracin al investigar el impacto de la fragmentacin de los bosques en la biodiversidad, ya que pueden dar lugar a grandes cambios a menudo perjudiciales en las poblacion es, las comunidades y la estabilidad de los ecosistemas. Los efectos especficos de los bordes de los bosques son complejos, altamente variables y poco comprendidos. En este estudio, investigu el efecto de borde en la ri queza de polillas, su abundancia y potenciales actividades de polinizacin en un bosque tropical fragmentado en Monteverde, Costa Rica, muestreando las polillas con trampas de luz en dos lugares diferentes: en el interior del bosque y en el borde del bosque. Mis resultados sugieren que hay un efecto negativo de borde en la riqueza de polillas, la abundancia, y la actividad potencial de polinizacin probablemente causada por el viento. Al hacer hincapi en los posibles efectos de borde en este taxn extremadamente diverso y ecolgicamente imp ortante, este estudio proporciona valiosas conclusiones preliminares a los ecologistas y conservacionistas que trabajan para frenar la prdida de biodiversi dad en los bosques tropicales.
Forest edge effect on moths Cosma 2 Among the many current anthropogenic threats to global biodiversity, habitat loss and fragmentation rank as some of the largest. Despite widespread awareness and conservation efforts, deforestation and forest fragmentation in most tropical countries still continues at an alarmi ngly high rate, and is leading to significant species losses (Didham et al 1998). Since the tropics house this represents one of most pressing problems in the conservation of global biodiversity (Fie dler et al. 2007). in t he sense the term in used today, involves both pure habitat loss and the breaking apart of habitat into fragments. Pure h abitat loss has consistent negative effects on biodiversity In contrast habitat fragmentation per se, wh ich involves factors such as patch size and patch isolation, va ries more in its effects on biodiversity, with some of the documented effects being negative, and some positive (Fahrig 2003). Habitat fragments have historically been described using island bi ogeograph y theory. However, habitat fragments are fundamentally different from true islands in that they are not surrounded by completely inhospitable habitat, but instead a matrix of habitat and land use types that may support potential predators and comp etitors for the species living within the fragment (Andren and Angelstam 1988). Along with abiotic factors, this may lead to complex edge effects. A key consequence of habitat fragmentation is the incre ased abundance of edges (Siu et al. 2016). Edges, as defined by Tscharntke e t al. (2002) two habitat types come together documented for abiotic factors ( such as varyin g levels of light, humidity, and wind velocity), as well as effects on veg etation. However, relatively little is known about how animals respond to trop ical forest margins (Fiedler et al 2007). Edge effects are highly variable, and include many interact ing biotic and abiotic factors. This leads to complex and often species spe cific effects. Studies suggest that s ome animals respond positively to forest edges, while many others respond negatively ( Laurance et al. 2002 ). Forest fragmentation may also disrupt biological processes that are important in maintaining biodiver sity and ecosystem functioning (Didham et al. 1996). Of particular significance are those processes involved with reproduction, such a s plant pollinator interactions (Murcia 1996). Negative effects of habitat fragmentation on pollination have been well doc umented in agricultural plots (Chacoff and Aizen 2005). However, little is known about the effects of forest fragmentation on plant pollination in tropical forests, where most pollination involves interactions with animal pollinators. Only two studies by Aizen and Fei nsinger (1994a,b), have been conducted on the topic. These suggest a negative edge effect on pollination in an Argentina dry forest. By influencing the movement and dispersal of insect pollinators, forest fragmentation may have far reaching im pacts for gene flow in plants, as well as in plant anima l community dynamics (Didham et al. 1996). Cleared spaces in forests may act as physical barriers to insect pollinators through exposure to abiotic elements such as increased wind velocity and rain, a s well as biotic elements such as increased predation. Powell and Powell (1987) have shown that even small cleared areas (less than 100 meters) inhibit po llinators from crossing to nearby forested areas The effects of habitat loss, fragmentation, and edge effect on insect pollinators represents an important and understudied area of research. While there are studies on the responses of i nsects to tropical forest fragmentation, the patterns are not well understood, and there is no clear consensus as to why different species react
Forest edge effect on moths Cosma 3 diff erently (Didham et al. 1998). However, e vidence suggests that edge effects may be one of the main factors affecting insect populations in fragment ed forest landscapes (Didham et al. 1996). For example, Jokimaki et al (1998) showed that arthropods are higher at forest edges than forest interiors in the pine dominated forests of Northern Finland. Lepidoptera are one of the most diverse insect orders and play important roles as pollinators, herbivores, and prey. Over 90% of k nown Lepidoptera are nocturnal Lepidoptera, or moths (Ricketts et al. 2001) According to Bawa (1990), moth pollination ranks perhaps only second in importance behind bee pollination in terms of number and diversity of plant species pollinated. Moths have been shown to be essential in mitigating the potential devastating effects of habitat fragmentation on plants by promoting gene flow through pollination (Finger et al. 2014). S tudies have also suggested that moth taxa may be useful indicators of biodiversi ty and habitat disturbance ( Kitching et al. 2000 and Summerville et al. 2004 ). Furthermore, they can be readily sampled in large number s with light traps (Ricketts et al. 2001). For these reasons, moths are particularly useful for studying the effects of habitat fragmentation Habitat fragmentation appears to affect different moth taxa differently (Schmidt and Roland 2006) with some families showing strong negative effects, and others showing strong positive effects (Kitching et al. 2000). Slade et al. (2013) showed that general responses to forest fragmentation differs significantly between moth species, and largely depends on life history traits and landscape characteristics. Similarly, edge effect response s appear to differ be tween moth families (Fied ler et al. 2007). Despite variability in responses, several general trends in moth responses to habitat fragmentation have been identified. For example, a study by Ricketts et al. (2001) found that moths in a disturbed habitat matrix had significantly high er richness and abundance when they were closer to forest fragments. There are several plausible explanations for such observations, including use of forest fragments and habitat edges for resources, shelter from abiotic factors such as wind and rain, and shelter from predators. Considering their importance as pollinators, and their potential as indicators of biodiversity and habitat disturbance, assessing the different responses of moths to habitat loss, fragmentati on, and edge effect is crucial especially in the tropics where they represent a large percentage of the diversity. In this study, I investigate the edge eff ect on moths in a f ragmented tropical cloud forest in Monteverde, Costa Rica. Specifically, I test if there is an edge effect on their richness abundan ce, and potential pollination activity by attracting them with li ght traps and collecting pollen samples at two different locations : one at the edge of a forest bordering an open space, and the other in the forest interior. I also test specifically for the edge effect of wind by sampling the two loc ations in windy conditions, and again in calm conditions. I hypothesize that moths at forest edges bordering open spaces have decre ased access to forest resources, such as the flowers that they feed on for nectar. I also hypothesize that the absence of tree cover at forest edges increases the exposure of moths to wind, inhibiting normal activity such as flight and foraging. I therefore predict that I will see both less richness and less abundance of moths at the forest ed ge than in the forest interior, especially in windy conditions. I also predict that I will s ee a lower percentage of moths carrying pollen and less variety of pollen, at the forest edge comp ared to the forest interior, especially in windy conditions.
Forest edge effect on moths Cosma 4 MATERIALS AND METHODS Study Site The study site is located in Monteverde, Puntarenas Costa Rica. It is approximately 100 meters down the road from the Monteverde Biological Station. The site consists of an open space on the side of a gravel road bordered by protected tr opical cloud forest a type of forest ecosystem characterized by frequ ent low lying clouds The open space is a pproximately 1,000 m 2 relatively flat and clear containing only gravel, dirt, and short grass. The site is approximately 1 500 meters above sea level I se lected two moth attracting locations : one directly on the edge of the forest on the Northwest side of the open space, and the other 30 meters into the forest on the Northwest side of the open space (Fig 1) Figure 1 : GPS image of study site (Google Earth) Open space outlined Moth attra cting locations marked with dots. Monteverde Biological Station visible in top right corner. Data Collection I used a suspende d 2 x 2.5 meter white sheet illuminated with one 20 Watt florescent white lamp and one 20 Watt black lamp to attract moths. Due to possible rain and mist, I suspended a tarp directly above the sheet and light set up on each night I turned the lights on at 5:30 pm ( approximate time of sunset ) on each night of data collection and left them on until 7:30 pm During this two hour time period, I photographed each moth that had a body length greater than one centimeter (tip of head to tip of abdomen), and that landed on either side of the sheet and stayed long enough for me to photograph it on the sheet I then picked up each moth, and pressed a piece of clear tape against the head, proboscis, antennae legs, and underside of thorax areas likely to r eceive pollen during foraging for flower nectar I then transferred each piece of tape to a microscope slide After collecting samples f rom ea ch moth, I transferred them to a plastic bag to avoid recapture where they remained until being released at the end of the data collection for that night I coll ected data for a total of four nights beginning on 2 6 November 2016, and ending on 2 December 2016. At each site, I collected data on one windless night, and one night of strong wind. I alternated between the sites each conse cutive night of data collection A ll data collection
Forest edge effect on moths Cosma 5 nights took place between the third and first quarters of the l unar cycle. Some data collection nights had brief periods of light rain In the laboratory, I searched for pollen grains on each of the slides u sing a compound microscope. I photographed each distinguishable type of pollen on each slide through the microscope lens Afterwards I identified both moths and pollen by morphospecies using the pictures All photographs throughout the experiment were taken with an iPhone 6s Statistics I used c hi s quare tests of independence to test for significant differences between the four night s of data collection in each of three factors: moth abundance, moth morphospecies richness and percentage of moths carrying pollen. Since I sampled two different locations (forest interior and forest edge), and each in two different wind conditions (calm and windy), calculated differences could be due to the effect of the location or the effect of the wind. Therefore, I also used chi square one sample tests for goodness of fit to te st for the effects of wind and location independently on each of the three factors. RESULTS Throughout the study, I identified a total of 88 moth morphospecies. 56 morphospecies were found in the forest interior, and 49 on the forest edge. Strikingly the forest interior and forest edge shared only nine out of the total 88 morphospecies. I also identified 17 morphospecies of pollen: 16 found in the forest interior, and seven found at the forest edge. Out of these, six were shared between both location s (Fig 2 ). Figure 2: Moth and pollen morphospecies comparison between each location forest interior Total moth morphospecies between the two locations was 88. Total pollen morphospecies between the two locations was 17. Moth abundance, measured by the total number of individuals photo graphed, differed significantly between days and locations (X 2 = 10 .179 df = 1, p = 0 .001 ). There were significantly fewer moths at the forest edge in windy conditions compared to calm conditions ( X 2 = 5.453 df =1, p = 0.020 ). In contrast, the forest interior did not differ significantly in moth
Forest edge effect on moths Cosma 6 abundanc e between calm and windy days ( X 2 = 0.012 df = 1 p = 0.912 ) On windy days, there were significantly fewer moths at the forest edge compared to the forest interior ( X 2 = 8.966, df = 1 p = 0.003 ). In contrast, the forest interior and forest edge edge did not differ significantly in moth abundance on calm days ( X 2 = 0.333 df = 1, p = 0.564 ; Fig 3). Figure 3: hour period of moth collection on an individual night. The moth morphospecies richness also differe d significantly between days and locations ( X 2 = 8.615 df = 1 p = 0.003 ), and followed the same trends as abundance There were significantly fewer morphospecies on the forest edge in windy conditions compare d to calm conditions ( X 2 = 5.0, df = 1 p = 0.025 ). In contrast, the forest interior did not differ significantly in number of morphospecies between windy and calm conditions ( X 2 = 1.0, df = 1 p = 0.317 ). Th ere were also significantly fewer morphospecies on the forest edge versus the forest interio r in windy conditions ( X 2 = 3.930 df = 1, p = 0.047 ), whereas th ere was no significant difference between t he locations in calm conditions ( X 2 = 0.545, df = 1 p = 0.460 ; Fig 4). 40 35 41 18 0 5 10 15 20 25 30 35 40 45 Interior Edge Number of Individual Moths Calm Windy
Forest edge effect on moths Cosma 7 Figure 4: hour period of moth collection on an individual night. The percentage of moths carrying pol len also differed significantly between days and locations ( X 2 = 23.120, df = 1, p < 0.00001 ) again following the same trends as richness and abundance At the forest edge, a significantly lower per centage of moths carried pollen in windy conditions compared to calm conditions ( X 2 = 9.696 df = 1 p = 0.002 ). In contrast, within the forest interior, the percentage of moths carrying pollen did not differ between wind conditions ( X 2 = 3.268, df = 1 p = 0.071 ). Whereas the percentage of moths carrying pollen did not differ significantly between locations in calm conditions ( X 2 = 0.098, df = 1 p = 0.891 ), a significantly lower percentage of moths carried pollen on the forest edge compared to the forest interior in windy conditions ( X 2 = 23.728, df =1 p < 0.00001 ; Fig 5) Figure 5: Percentage of moths carrying at least one pollen morphospecies at each location, bar represents one two hour perio d of moth collection on an individual night. 36 30 28 15 0 5 10 15 20 25 30 35 40 Interior Edge Number of Moth Morphospecies Calm Windy 32,5 31,4 48,8 11,1 0 10 20 30 40 50 60 Interior Edge Percentage of Moths with Pollen Calm Windy
Forest edge effect on moths Cosma 8 In the forest interior, I found pollen on a total of 33 moths. There were nine t otal instances where more than one morphospecies of pollen was f ound on an individual moth. In eight of the cases, it was two different pollen morphospecies, and in one rare example, four pollen morph o species occurred on a single moth. In contrast, at the forest edge I found pollen on only 13 moths, and none of these carried more than one morphospecies of pollen (Fig 6 ) Figure 6: Frequency of one or more pollen morphospecies per moth at each location, forest grouped. DISCUSSION The results of this study suggest that moth richness abundance, and pollen carryin g are all significantly lower at the forest edge than the forest interior in windy conditions but not in calm conditions, in part s upporting my initial hypotheses. This provides preliminary evidence for a n egative effect on moth richness abundance, and potential pollination activity at forest edges caused specifically by wind and its associated abiotic changes (e.g. temperature shifts). These results are intuitive, as forest edges are inherently more exposed to the wind and its effects due to the decreased tree cover, which in intact forests acts as a wind break and can provide important benefits to both plants and animals. Wind is perhaps especially pro blematic for aerial insects, particularly those that ar e light and have delicate wings, e.g moths. Cleared spaces in forests may act as physical barriers to moths through exposure to higher wind velocities. Moths may be physically pushed into the forest, or choose to shelter there in windy conditions, which may explain why I found s ignificantly fewer moths at the fores t edge during windy conditions but no significant difference in the forest interior These findings are especially relevant in many tropical countries which are affected by strong seasonal winds, such as the trade winds that affect Costa Rica (including during the data collection period for this study) In a heavily fragmented forest landscape, many cleared areas may become inaccessibl e to certain moths for large portions of the year, which may impact patterns of movement and dispersal. Large scale studies are necessary to assess the population wide impact of cleared spaces and wind on moths. 24 8 0 1 13 0 0 0 0 5 10 15 20 25 30 1 2 3 4 Frequency Number of Different Pollen Morphospecies Per Moth Interior Edge
Forest edge effect on moths Cosma 9 Forests, even small fragments, may provide moths with necessary shelter from wind. Previous studies have shown that moth species use forest corridors and agricultural windbreaks for movement, av oiding the surrounding exposed habitat matrix (Varkonyi et a l 2003 and Coulthard et al. 2016). In this study, not only did the forest interior show no significant difference in moth richness or abundance between the windy day and the calm day (F igs 3 and 4) but also had a higher (though nonsignificant) percentage of m oths carrying poll en (48.8% in windy conditions versus 3 2.5% in calm conditions; F ig 5) This may suggest that moths, even ones that may frequent forest edges in calm conditions, harbor and forage inside the protection of intact forest fragments in windy conditions In addition, it prov ides evidence that even only 30 meters into an intact forest (the distance used in this study) may be enough to provide adequate shelter from wind and access to resources to support moth populations. There have been no previous studies on the edge effect on moth pollination in tropical forests. My results provide the first preliminary evidence for a possible negative edge effect due to wind on potential pollination activity in moths It is important to note that finding pollen o n a moth does not necessarily indicate pollination activity as pollination only occurs when pollen is transferred between conspecific plants and deposited in precise locations. In this respect, my is more directly a measu re of forging activity, as incidence of pollen does indicate that the moths are feeding on flower nectar However, since moths often un knowingly pollinate while foraging on flowers for nectar I use the incidence of pollen on moths as an approximation of t he ir potential pollination activity Acknowled ging the significant decrease in moth abundance at the forest edge during windy conditions, a significantly lower percentage of the moths that were collected here carried pollen compared to the forest edge during calm conditions, or the forest interior even during windy conditions ( Fig 5 ). In addition, there were no instances of multiple pollen morphospecies on moths at the forest edge, whereas in the forest interior there were nine instances (Fig 6 ). Although reduced moth abundance and richness around forest edges themselves may negatively impact moth pollination, these results point to an independent negative effect of wind at forest edges on moth potential pollination activity which could be explained by several possibilities Wind may force moths to spend more of their time fighting the gusts, thus decreasing foraging activity. Alternatively, it may force the moths into inactivity/shelter to avoid possible negative consequences of being caught flying in the wind. Regardless, by negatively affecting moth foraging at forest edges, wind in fragmented forests may impose significant consequences both to the moths and the plants that they pollinate, such as reduc tions in moth fitness, and impeded pollen flow in plants. Given the importance of pollination to ecosystem stability these effects may lead to other ecosystem wide consequences. Future studies must identify the magnitude and specific mechanisms of the forest edge effect on moth pollination. Since w eak artificial lights have small attrac tion radii (le ss than 30 meters) (Fiedler et al 2007) it is likely that the sample s collected from 30 meters into the forest interior consisted of moths that were mostly in the forest, while the sample collected at the forest edge may have consisted of moths that were attracted from the forest immediately bordering the open space, the open space itself and the forest edge With this in mind, i t is important to note that out of the 88 total morphospecies identified, only 9 were shared between the two s ampling locations (F ig 2)
Forest edge effect on moths Cosma 10 There are several plausible explanations for this finding. First, since the windy night spent at the forest edg e resulted in significantly fewer morphopecies it is possible that the full diversity of moths that could be found at the forest edge in calmer conditions are not accurately dep icted. Second, it is possible that some of the morphospecies unique to the forest are specifically adapted to living there, and rarely venture to forest edges or open spaces because of unfavorable abiotic or biot ic conditions ( and perhaps even partly due to the negative effects o f wind). Third, the opposite scenario is possible, where some of the morphospecies of moths that were unique to the forest edge may be better adapted to living there This could be explained by the concept of landscape complementation where certain moths may requ ire different habitat types throughout their life cycle. These species may benefit from the facilitated mobility between habitat types that forest edges provide (Fahrig 2003). T he results of this study, corroborating those of Powell and Powell ( 1987), suggest that even small cleared areas in tropical forest s in this case only 1,000 m 2 may lead to negative effect s on insect populations. The cleared patches may barriers to movement through the landscape due to abi otic and biotic interferences In particular, my results suggest that the abiotic factor, wind may negatively affect moth populations at forest edges surrounding open spaces, possibly even impacting their interspecific pollination i nteractions. Future studies with increased sample sizes are necessar y to provide further support to these preliminary findings With possible effects on patterns of diversity, abundance, dispersal, and gene flow in both moth and plant populations in fragmented forests, the re sults of this study may be relevant to eco system stability and should be investigated further and considered in conservation efforts. ACKNOWLEDGEMENTS I would like to sincerely thank all of the instructors in the Tropical Biology and Conservation UCEAP program, not only for helping me immensely on this project, but for all that they have done for me and th e rest of the group to make study abroad a n unbelievably fulfill ing and enriching experience. This includes my secondary advisor Andres Camacho, Frank Joyce, Federico Chin chilla, Sofia Arce Flores, Justin Welch, and of course my primary advisor, Emilia Tr i an a Emilia as an advisor, source of knowledge, instructor, and friend, was ever helpful to me with anything that I needed for this project an d made the research process exceedingly more enjoyable. I would also like to thank Johel Chaves for letting me use his s power for my lights. Â¡ Muchas Gracias! LITERATURE CITED Aizen, Marcelo A., and Peter Feinsinger. "Forest Fragmentation, Pollination, and Plant Reproduction in a Chaco Dry Forest, Argentina." Ecology 75.2 (1994a): 330 51. Web. Aizen, Marcelo A., and Peter Feinsinger. "Habitat Fragmentation, Native Insect Pollinators, and Feral Honey Bees in Argentine 'Chaco Serrano'" Ecological Applications 4.2 (1994b): 378 92. Web.
Forest edge effect on moths Cosma 11 Andren, Henrik, and Per Angelstam. "Elevated Predation Rates as an Edge Effect in Habi tat Islands: Experimental Evidence." Ecology 69.2 (1988): 544. Web. Bawa, K. "Plant Pollinator Interactions In Tropical Rain Forests." Annual Review of Ecology and Systematics 21.1 (1990): 399 422. Web. Chacoff, Natacha P., and Marcelo A. Aizen. "Edge Eff ects on Flower visiting Insects in Grapefruit Plantations Bordering Premontane Subtropical Forest." Journal of Applied Ecology 43.1 (2005): 18 27. Web. Coulthard, Emma, Duncan Mccollin, and James Littlemore. "The Use of Hedgerows as Flight Paths by Moths i n Intensive Farmland Landscapes." Journal of Insect Conservation 20.2 (2016): 345 50. Web. Didham, Raphael K., Jaboury Ghazoul, Nigel E. Stork, and Andrew J. Davis. "Insects in Fragmented Forests: A Functional Approach." Trends in Ecology & Evolution 11.6 (1996): 255 60. Web. Didham, Raphael K., Peter M. Hammond, John H. Lawton, Paul Eggleton, and Nigel E. Stork. "Beetle Species Responses to Tropical Forest Fragmentation." Ecological Monographs 68.3 (1998): 295. Web. Fahrig, Lenore. "Effects of Habitat Fra gmentation on Biodiversity." Annual Review of Ecology, Evolution, and Systematics 34.1 (2003): 487 515. Web. Fiedler, Konrad, Nadine Hilt, Gunnar Brehm, and Christian H. Schulze. "Moths at Tropical Forest Margins: How Mega diverse Insect Assemblages Respon d to Forest Disturbance and Recovery." Stability of Tropical Rainforest Margins Environmental Science and Engineering (2007): 37 60. Web. Finger, Aline, Christopher N. Kaiser Bunbury, Chris J. Kettle, Terence Valentin, and Jaboury Ghazoul. "Genetic Connect ivity of the Moth Pollinated Tree Glionnetia Sericea in a Highly Fragmented Habitat." PLoS ONE 9.10 (2014): n. pag. Web. Jokimaki, Jukka, Esa Huhta, Juhani Itamies, and Pekka Rahko. "Distribution of Arthropods in Relation to Forest Patch Size, Edge, and St and Characteristics." Canadian Journal of Forest Research 28.7 (1998): 1068 072. Web. Kitching, R.l., A.g. Orr, L. Thalib, H. Mitchell, M.s. Hopkins, and A.w. Graham. "Moth Assemblages as Indicators of Environmental Quality in Remnants of Upland Australian Rain Forest." Journal of Applied Ecology 37.2 (2000): 284 97. Web. Laurance, William F., Thomas E. Lovejoy, Heraldo L. Vasconcelos, Emilio M. Bruna, Raphael K. Didham, Philip C. Stouffer, Claude Gascon, Richard O. Bierregaard, Susan G. Laurance, and Eric a Sampaio. "Ecosystem Decay of Amazonian Forest Fragments: A 22 Year Investigation." Conservation Biology 16.3 (2002): 605 18. Web. Murcia, Carolina. "Chapter 2: Forest Fragmentation and the Pollination of Neotropical Plants." Forest Patches in Tropical La ndscapes By John Schelhas and Russell Greenberg. Washington, D.C.: Island, 1996. N. pag. Print. Powell, A. Harriett, and George V. N. Powell. "Population Dynamics of Male Euglossine Bees in Amazonian Forest Fragments." Biotropica 19.2 (1987): 176. Web. Ri cketts, Taylor H., Gretchen C. Daily, Paul R. Ehrlich, and John P. Fay. "Countryside Biogeography of Moths in a Fragmented Landscape: Biodiversity in Native and Agricultural Habitats." Conservation Biology 15.2 (2001): 378 88. Web. Schmidt, B. C., and J. Roland. "Moth Diversity in a Fragmented Habitat: Importance of Functional Groups and Landscape Scale in the Boreal Forest." Annals of the Entomological Society of America 99.6 (2006): 1110 120. Web.
Forest edge effect on moths Cosma 12 Siu, Jenna C., Daria Koscinski, and Nusha Keyghobadi. "Sw allowtail Butterflies Show Positive Edge Responses Predicted by Resource Use." Landscape Ecology 31.9 (2016): 2115 131. Web. Slade, Eleanor M., Thomas Merckx, Terhi Riutta, Daniel P. Bebber, David Redhead, Philip Riordan, and David W. Macdonald. "Life hist ory Traits and Landscape Characteristics Predict Macro moth Responses to Forest Fragmentation." Ecology 94.7 (2013): 1519 530. Web. Summerville, Keith S., Lisa M. Ritter, and Thomas O. Crist. "Forest Moth Taxa as Indicators of Lepidopteran Richness and Hab itat Disturbance: A Preliminary Assessment." Biological Conservation 116.1 (2004): 9 18. Web. Tscharntke, Teja, Ingolf Steffan Dewenter, Andreas Kruess, and Carsten Thies. "Characteristics of Insect Populations on Habitat Fragments: A Mini Review." Ecologic al Research 17.2 (2002): 229 39. Web. Varkonyi, Gergely, Mikko Kuussaari, and Harri Lappalainen. "Use of Forest Corridors by Boreal Xestia Moths." Oecologia 137.3 (2003): 466 74. Web.