Sea urchins reduce fish grazing Beittel 1 Sea urchins reduce fish grazing on algae Alice Beittel University of California Davis Department of Environmental Science and Policy EAP Tropical Biology and Conservation Program, Sprin g 2018 7 June 2018 ______________________________________________________________________________ ABSTRACT : Overfishing of reef ecosystems poses a threat to key trophic interactions that regulate the food competition between fish and the sea urchin, Diadema mexicanum . An important ecosystem engineer, D. mexicanum can help control for positive coral growth at intermediate levels but at high levels can erode coral reefs and at low levels results in algae dominated reefs. Due to the overfishing of echin oderm predators in BahÂ’a de Cuajiniquil, Costa Rica, D. mexicanum populations are less regulated and have the potential to outc ompete other herbivorous fish. My overarching question is how does Diadema mexicanum grazing impact fish feeding intensity? To answer this question , I first examined if fish are grazing on different slopes to understand overlaps between fish and urchin feeding locations . I observed 45 rock patches of three rock slope categories (horizontal, sloped, and vertical) for the number of urchins present, fish visitations, and fish grazing events. There are significantly more fish visitations to rock of slopes 0 30 degrees but there is not a significant difference of number of grazing events between rocks with 0 30 degree, 30 60 degree, and 0 90 degree slopes. A second way to understand competition dynamics between species is to use exclude one competitor from a food source and measure the re sponse of the other competitor. I installed ten exclosures on flat rock substrates to prevent herbiv ory by sea urchins and fish for a period of 5 continuous days. After I removed m y exclosures, I observed more fish grazing events and higher species diversity . Both herbivorous and carnivorous fish fed from the post exclosures plots. Sea urchins are indire ctly competing with herbivorous and carnivorous fish for rock substrate food resources. Conservation of Diadema mexicanum predators will be key to helping restore coral reef systems and rebalance trophic interactions. _____________________________________ _________________________________________ Los erizos de mar reducen la herbivorÂ’a de algas por peces RESUMEN: La sobrepesca de los ecosistemas de arrecifes representa una amenaza para las interacciones trÂ—ficas claves que regulan la competencia aliment aria entre los peces y el erizo de mar, Diadema mexicanum . D. mexicanum , un importante ingeniero de ecosistemas, puede ayudar a regular el crecimiento de coral cuando se encuentra a una abundancia intermedia, pero en niveles altos puede erosionar los arrec ifes de coral y en niveles bajos resulta en arrecifes dominados por algas. Debido a la sobrepesca de depredadores de equinodermos en BahÂ’a de Cuajiniquil, Costa Rica, las poblaciones de D. mexicanum estÂ‡n menos reguladas y tienen el potencial de superar a otros peces herbÂ’voros. Mi pregunta general es: Ã€cÂ—mo influye el pastoreo de Diadema mexicanum en la intensidad de la alimentaciÂ—n de los peces? Para responder esta pregunta, primero examinÂŽ si los peces estaban alimentÂ‡ndose en diferentes pendientes para comprender los traslapes entre los sitios de alimentaciÂ—n de los peces y los erizos. ObservÂŽ 4 5 parches de roca de tres categorÂ’as de pendiente (horizontal, inclinada y vertical) para cuantificar el nÂœmero de erizos presentes, los visitas de peces y los e ventos de pastoreo de peces. HabÂ’an significativamente mÂ‡s visitas de peces a la roca de pendie nt es de 0 a 30 grados, pero no hubo una diferencia significativa en el nÂœmero de eventos de herbivorÂ’a entre las rocas con pendi entes
Sea urchins reduce fish grazing Beittel 2 de 0 30 grados, de 30 60 grados y de 0 90 grados. manera de entender la dinÂ‡mica de co mpetencia entre especies consiste en excluir a un competidor de una fuente de alimento y medir la respuesta del otro competidor. InstalÂŽ diez exclusores sobre roca s plana s para prevenir la herbivorÂ’a de los erizos de mar y los peces durante un perÂ’odo de 5 dÂ’as continuos. Despu ÂŽs de eliminar mis exclusores , observÂŽ mÂ‡s eventos de herbivorÂ’a por peces y una mayor diversidad de especies. Tanto peces herbÂ’voros como peces carnÂ’voros se alimentaron en las parcelas despuÂŽs de quitar los exclusores . Los erizos de mar estÂ‡n compitiendo indirectament e con los peces herbÂ’voros y carnÂ’voros p or los recursos alimenticios en las roca s . La conservaciÂ—n de los predadores de Diadema mexicanum serÂ‡ clave para ayudar a restaurar los sistemas de arrecifes de coral y reequilibrar las interacciones trÂ—ficas. H uman pressures on coastal reef resources jeopardize ecosystem services critical to the stability of coastal economies. With almost 40% of the world's population living within 100 kilometers of coastlines, our interaction with coastal ecosystems through pol icy management, marine harvesting techniques, and tourism has the potential to help or hinder marine resource prosperity (Millennium Ecosystem Assessment 2005). Nearly a third of the world's marine fish species are found in coral reefs and constitute almost 10% of the fish biomass humans consume (Moberg 1999). Mitigating the decline of reef ecosystems is a difficult task for local communities. Most modern reef management fails to reach loca l or global conservation goals (Jac kson 2001, Pandolfi 2003, Millennium Ecosystem Assessment 2005). The ecological and political conservation story of Cuajiniquil, a small fishing town located on the northern pacific coast of Costa Rica, illustrates a socio ecological challenge familiar to many coastal communities: how poorly regulated fishing can dramatically change reef ecology, ecosystem services, and long term economic stability (Rowe 2011) . Selective fishing for large fish at BahÂ’a de Cuajiniquil, Costa Rica, has likely initiated a cas cading effect where a few species dominate the reef ecosystem and create biologically impoverished landscapes with low biodiversity (Millennium Ecosystem Assessment. 2005). In the last 25 years, large herbivorous and carnivorous fish, such as parrot fish, t r iggerfi sh, have experience d population declines due to an increase in fishing pressure from local fishing (Minor La ra , pers onal communication , 11 May 2018 ). In Cuajiniquil, the lack of large herbivorous and carnivorous species has increased competition between other algae grazing fish and urchin species as sea urchin populations increase . Coupled with less predation, the success of D. mexicanum populations is also due to their ability to occupy a variety of niches and their generalist diet (Alvarado 200 9). With the absence of echinoderm eating fish, Diadema urchins have increased in number along the western Pacific Central American coast in the last 20 years and have thus Ã changed reef ecosystems (Alvarado et al. 2012 ) . At high abundances, D. mexicanum grazing reduces the growth rate of calcifying algae and corals through bioerosion ( O'Leary and McClanahan 2010; Alva rado et al. , 2016) . At intermediate abundances, D. mexicanum plays a key role in limiting algal growth which in turn opens for space for coral establishment and growth (Alvarado et al. 2012) . At low abundances, Diadema urchins give way to algae infested co ral reef systems that create unsuit able habitat for coral species ( O'Leary and McClanahan 2010 ). The most abundant sea urchin species along the northern pacific coast of Costa Rica is Diadema mexicanum . ( Alvarado 2008 ). On 9 May 2018, I sampled two randomly selected 20 meter transects at Bajo Rojo, a rock y outcrop two kilometers from Cuajiniquil's main fishing dock, the muelle . At each meter , I measured the distance between the transect tape and the
Sea urchins reduce fish grazing Beittel 3 nearest D. mexicanum individual. The average dista nce to the nearest D. mexicanum individual was 12.8 centimeters, indicating that D. mexicanum individuals are common at Bajo Rojo . Much is known on the impact of high sea urchin abundances on cora l health ( Mumby et al . 2006; Perry et al . 2014 ) . L ess is known about their impact on the feeding of rock grazing fish. In communities such as Cuajiniquil , a spike in sea urchin populations could result in a decrease in the local economic productivity. It is important to understand the role competition pla ys between herbivorous fish and sea urchins to contextualize the consequences of high fishing intensity on the fish remaining ( Kittinger 2014 ) . My research project examine s relationships between the indirect food resource competition of sea urchins and al gae eating fish. My overarching research question is how does Diadema mexicanum grazing impact fish feeding intensity? To answer this question, I have two sub questions to examine specific variables central to understanding relationships between fish grazi ng, food resource availability and D. m exicanum grazing . (1) How does the slope of the rock substrate change the intensity of fish grazing? I will be using slope intensity to compare potential overlap between areas most accessible to sea urchins and areas with the most fish grazing . Analyzing relationship s between urchin accessibility and fish feeding preference will help shed light into where sea urchins and herbivorous fish compete. To understand relationships between slope, fish feeding , and D. mexicanu m grazing, I will randomly evaluate fish feeding intensity and urchin abundance on a range of rock slopes. (2) How does D. m exicanum presence and absence change the amount of fish grazing on a particular rock area? I will experimentally test the impact urc hin grazing has on algal growth by excludi ng sea urchins and fish from a defined area. NATURAL HISOTRY OF FISHING IN CUAJINIQUIL Research was conducted at Bajo Rojo, a rock y outcrop two kilometers from Cuajiniquil, Costa Rica in BahÂ’a de Cuajiniquil. Currently the marine sector of the Ã§rea de ConservaciÂ—n Guanacaste (ACG; Guanacaste Conservation Area) does not include BahÂ’a de Cuajiniquil, but BahÂ’a de Cuajiniquil is located just north of a UNESCO World Heritage Site, Islas MurciÂŽlago (Rowe 2011). Cuaj iniquil is a small fishing village of around 2,000 people located northwestern Pacific coast of Costa Rica. In the 1980 ' s , the federal government constructed a large dock, the muelle , and later attached a fish processing plant in an effort to help provide e conomic stability for the people of Cuajiniquil. The number of fishing vessels in the BahÂ’a de Cuajiniquil quickly jumped from 12 to 180 in a few years. In 2011 there was 812 registered fishing licenses for the 2,235 residents of fishing towns around BahÂ’a de Cuajiniquil. From the early 1980s to 1990s there was a large shift among locals from agriculture to the fishing industry to take advantage of the bay's untapped marine resources. Within a few years, fishermen from larger ports were coming to the muelle to buy and export local catch . An influx of profit and "middlemen" between Cuaijiquil fishers and the larger pacific market allowed local fishers to upgrade to more efficient fishing equipment and increase catch sizes . While community outreach and education of the benefits marine conservation has been largely suc cessful, poor law enforcement of no take zones (areas where fishing is prohibited) and regulation of key species by INCOPESCA ( Instituto Costar ricense de Pesca y Acuicultura ; Costa Rican Institute of Fishing and Aquaculture) coupled with historical and continued fishing pressures has reduced populations of large sized fish species to locally threatened or critically endangered levels (Rowe 2011) . Fish population i mbalances have resulted in high sea urchin populations that have overgrazed sea weed populations (Minor La ra, pers onal comm unication , 11 May 2018 ) . Bajo Rojo was once an area where fishers would come to harvest seaweed, parrotfish, triggerfish, snapper s and other large fish species . A lack of desirable fish causes fishermen to travel further distances to obtain profitable catch size (Minor La ra, personal
Sea urchins reduce fish grazing Beittel 4 comm unication , 11 May 2018 ) . My study aims to understand the implications of D. mexicanum high abundances on the grazing of remaining fish populations. MATERIALS AND METHODS Slope Research Design: To observe how rock slope affects the intensity of fish grazing intensity, I observed fish feeding intensity on 45 randomly selected rock patches from 10 May to 18 May 2018 . I observed a rock substrate patch of approximately 0.5 m by 0.5 m for five minutes. My observation time totaled to 225 hours. I randomly chose each rock substrate patch by using a stopwatch with pre determined direction guideli nes. I assigned odd numbered milliseconds to west and even numbered milliseconds to east. After randomly stopping the stopwatch, I swam in the direction indicated by the millisecond number for a length of time indicated by the seconds number. I recorded 18 in situ observations and 27 video observations (total sample size was 45) on a GoPro Hero 3. For each observation period, I recorded the number of grazing events per species. I defined one grazing event when I observed a fish touch its mouth to the rock s urface and release. To record fish visitations to the site, I recorded fish species abundance of any fish that came within approximately 10 cm above of the focal rock substrate. I recorded an estimation of the number of urchins present on the rock substrat e patch, categorized the algae cover (low, medium, and high ), and categorized the slope of the rock into three groups horizontal (0 30 degrees), sloped ( 30 60 degrees ) , and vertical ( 60 90 degrees) . Exclosure Research Design: I installed ten exclosures to experimentally test the effect of the removal of sea urchins on fish grazing intensity. With the help of Juan Carlos Castro , a local mariculture employee, we constructed ten 0 .5 m by 0.5 m exclosure using two inch PVC tubes an d connected the tubes using PVC T shaped connectors. We secured plastic netting to the tops of the squares so that the netting completely enclosed one PVC side (Figure 1 ). The squares of the plastic mesh measured 2 cm x 2 cm . Figure 1 . Schematic illustrating the exclosure observation timing, installation, and removal. 0.5 meter by 0.5 meter PVC exclosures were installed on horizontal rocks for a minimum of five days to exclude fish and urchin grazing on the rock substrate. On 13 May 2018 between 8am and 2:30 pm I installed ten exclosures stretching from west to east along the south facing side of Bajo Rojo. Before and after each installment, I took ten minute videos of the exclosure area to record the species richness and abun dance of fish grazing events and fish visitations to the exclosure area. I used the same parameters for determining fish grazing events and visitations as the slope research design.
Sea urchins reduce fish grazing Beittel 5 Six of ten exclosures remained in place for a minimum of 120 continuous h ours (5 days) and I used in my analysis. I discounted the remaining four due to repeated movement by tidal surge. I removed the six successful exclosures 19 May 2018 between 8 am and 10 am and repeated the 10 minute observation period using GoPro video imm ediately after removing the exclosures. All exclosures used in the statistical analysis were located on a near zero degree rock slope. I measured the depth of each exclosure on 15 May 2018 using a transect tape tied to a weight. I rounded to the nearest half meter and calculated the depth at high tide for each exclosure using the high tide and low tide depths and times for 15 May 2018 ( Figure 2 ) . Table 2 . Summary of total time exclosures were secured in place and exclosure depth. Exclosures are 0.5 meters by 0.5 meters and are constructed with 2 inch (5cm) PVC pipes. A plastic netting (holes 2 centimeters in width) was tied to the top of the PVC frame. Ex closures were placed on horizontal rocks and secured in place for a minimum of five continuous days (120 hours). Statistical Analysis I used a o ne way ANOVA to test the relationship between total number of fish visits to an observation site and each of the three slope categories . Using a Tukey Kramer test, I tested the difference among the means to understand the significant difference between each slope category. I performed a one way ANOVA to test the relationship between total number of grazing event s by slope category. For my exclosure sites, I used a paired Wilcoxon Signed Rank Test to assess if there was a difference between the number of grazing events post and pre exclosure placement . RESULTS Slope vs Grazing I observed a higher average number of fish grazing events, fish visitation events, and urchin individuals on horizontally sloped rocks. I observed statistically significant higher fish visitations to horizontal rocks ( 0 30 degree slope) than sloped rocks ( 30 60 degree slope) . T here was not a significant difference between fish visitations to vertical rocks (60 90 degree slope) compared to horizontal and sloped rocks (O ne way ANOVA; F = 4.45; P < 0.05 ; Tukey Kramer Comparison of Means ; Figure 3 ) . I did not observe a statistically significant difference in the number of grazing events across the three slope categories ( One way ANOVA; F = 0.70; P > 0.5). The horizontal slope category had the highest average number of grazing events, visitation events , and aver age number o f urchins.
Sea urchins reduce fish grazing Beittel 6 Figure 3 . Variation of the average number of individuals across three slope categories: horizontal (0 30 degrees), sloped (30 60 degrees), and vertical (60 90 degrees) and three observation sets. First set: There is no significant difference of fish grazing events between horizontal, sloped, and vertical rocks. Second set: Red asterisks correspond to the Tukey Kramer test showing the significant difference between the number of fish visitations to horizontal and sloped categories. Third set: There is no significant difference of the number of sea urchins present between horizontal, sloped, and vertical rocks . Horizontal sloped rocks have a higher average number of grazing, visitations, and urchins present. All data points are recorded from a five minute observation period on randomly selected rock patch es . Exclosu re I observed a statistically significant higher number of fish grazing events after I removed the exclosures than before exclosures placement ( Wilcoxon Signed Rank; W = 0; N = 6; P < 0.05; Figure 4 ). I observed no statistically significant relationship between the depth of the exclosures and the number of grazing events pre and post exclosure; however, there is a stronger relationship between depth and pre exclosure grazing events than post exclosure grazing events ( Figure 5 ) . Mexican Hogfish and Acapul co Gregory individuals had the highest increase in grazing events post exclosure ( Figure 6 ). Acapulco Gregory individuals had the highest increase in visitation events post exclosure (Figure 7).
Sea urchins reduce fish grazing Beittel 7 Figure 5. Depth of each exclosure by total number of fish grazing events pre and post exclosure placement. Exclosure depth and number of fish grazing events are not significantly correlated. Depth measurements quantified at high tide from 15 May 2018. PRE EXCLOSURE POST EXCLOSURE Figure 4. Change in the number of grazing events of f ish individuals between pre and post exclosure placement. There are significantly more fish grazing events after removal of the exclosure.
Sea urchins reduce fish grazing Beittel 8 Figure 6. Species breakdown of the change in the number of fish grazing events pre and post exclosure placement. Acapulco gregory damselfish (herbivorous) grazing events increased after the removal of the exclosures in all but one site. The mexican hogfish (carnivo rous) was not present in pre exclosure observations but is present in post exclosure observations. Figure 7 . Species breakdown of the change in the number of fish visitation events pre and post exclosure placement. Acapulco Gregory damselfish visitatio ns increased in all but one site. There is more species richness and abundance in after the removal of the exclosures.
Sea urchins reduce fish grazing Beittel 9 DISCUSSION Slope Observations With my slope observations , I observed that areas with horizontal slopes ( 0 30 degree s) had the highest average of grazing events, visitations events, and number of sea urchins present. Overlapping preferences for horizontal sloped rocks between fish and sea urchins indicates strong indirect competition for rock substrate based food resources . Increased popu lation levels of D. mexicanum have the potential to eliminate preferred feeding habitat for fish. Former EAP student Brittany Jellison (2009) , reported that higher densities of D. mexicanum aggregations are found on flat rock than creviced or rocky substra tes. Herbivorous, algal consuming fish tend to have oral kinematics that allow feeding on flat rocks to be less energy expensive (Michel 2014; Rupp and Hulsey 2014). I observed statistically significant more fish visitations to horizontal sloped rocks than s loped and vertical rocks and more fish came to the observation areas as visitors than grazers. T here could be interspecific interactions, energy budget limits, varying levels of food availability, or perceived risks that are preventing visiting fish from grazing. If algal consuming fish are faced with dramatically less algae on flat rocks (due to increased urchin grazing and abundance), their oral morphological and kinematic characteristics may not be able to adapt fast enough to a novel food distribution on the reef. For Diadema species, ease of access to food is not the only reason why they may prefer horizontal slopes. First, shallow sloped rocks allow for urchins to move longer distances and at faster paces during the night t o escape predation and rea ch food resources. Second , shallow sloped rocks allow more aggregations and in turn reduce t he success of predation. Third , shallow sloped rocks allow more urchin aggregations and increase the probability of mating success during spawning because sperm and eggs are released closer to each other ( Alvarado 2009). Future research can analyze the oral kinematics of herbivorous fish to understand their physical limitations and the extent of their adaptability to graze on different sloped rocks. Integrating fish morphology, trophic interactions, benthic geography, and population abundances, can provide valuable insight into helping quantify and predict the impact D. mexicanum could have on higher trophic level fish species valuable to local fishin g communities. Exclosure Observations Diadema mexicanum urchin grazing has a strong impact on fish feeding. I observed a statistically significant increase in fish grazing when urchins were excluded from the rock substrate ( Figure 4 ) . In addition to an i ncrease in number of fish grazing on the areas, there was a visible amount of algae colonization on the plastic netting, rock substrate, and PVC tubing (Figure 8 ) . I can deduce that there was not only an increase in algal abundance, but also other rock dwe lling organisms because the top two species who increased in grazing events post exclosure are carnivorous and herbivorous. The Acapulco Gregory damselfish , an herbivore, increased in visitation and grazing events at a majority of the sites. The Mexican Ho gfish , a carnivore, had the second highest increase in grazing post exclosure . Herbivorous and carnivorous fish are competing with sea urchins for food. D. mexicanum grazing impacts a wider range of fish than I initially expected and has the potential to c reate negative trop h ic cascading ef fects on higher trophic levels .
Sea urchins reduce fish grazing Beittel 10 Figure 8 . Close up image of algal growth on PVC pipe exclosure frame and on rock substrate . In addition to the explanation that increased food resources drives increased grazing, I have two alternate explanations for why there is an increase in the number of grazing events. First, the increase in algae and other rock colonizers was great enough to attract fish to the exclosure site after the exclosure was removed. Gil and Hein (201 7) describe how fish feeding behavior operates on a feedback loop. Fish spend more time feeding when there are more species present. It is possible that with a newly exposed food resource , more fish are attracted to the food or the unfamiliarity of the fre shly colonized rock substrate and , in turn , socially att ract more species to the patch. Under this concept, d eclines in grazing fish species due to overfishing could threaten the feeding abilities of other fish who depend on interspecific feeding cues and further reduce reef fish diversity. Future research examining implications of overfishing of key algal grazing species on social foraging feeding may provide insight into which species may be ecologically imperative to apply catch limit regulations . Secon d , the act of removing the exclosures may have attracted more fish to the untouched food resource. Acapulco Gregory individuals had the highest increase in visitations to the exclosure areas post removal. Acapulco Gregory fish are home to the genus Stegastes who are found to be some of the most vocal reef fish ( Kennedy et al 2010). Their vocalizations and the behavioral curiosity of other fish could have created more visitation and grazing events after exclosure removal. These two driving forces coul d contribute to the positive feedback loop explained by Gil and Hein. Future research can control for the removal process's potential disturbance by observing fish grazing at an identical open frame PVC pipe. Diadema mexicanum grazing has an impact on roc k substrate resource availability. Without effective conservation of sea urchin predators, Bajo Rojo and similar reef ecosystems have the potential risk of losing fish species important to the long term sustainability of ecosystem services beneficial to th e fishing industry. More research is needed to explicitly understand how indirect grazing competition on rock substrate resources impacts higher and lower trophic levels. Research of overfished areas such as BahÂ’a de Cuajiniquil can be valuable in creating ecologically informed and relevant conservation policies to help create sustainable costal fishery systems. ACKNOWLEDGEMENTS Thank you Frank for pushing me to think and make decisions independently, for giving me half the puzzle and making me draw and p ut together the second half. I learned so much this way and am so grateful to have had the opportunity to learn from you. Thank you for your patience through my many questions and moments of utter confusion. Thank you Fede for helping me think clearly , unc over a new appreciation for plants, and making me laugh (you are
Sea urchins reduce fish grazing Beittel 11 always spot on with the jokes). Thank you Juan Carlos Castro for dedicating his Saturday morning to help me construct my exclosures, teaching me a whole bunch of new super strong knots with r elinga, and helping me construct the exclosures in record time. Thank you, Anabel, for helping me tie rocks to my exclosures and hand them down to the life jacket escalator. To Minor and family, you all are the wheels to the train, or I guess I should say the motor to the boat. Thank you for spending hours on the boat with us, being patient while I practiced my Spanish, and answering all my many questions about fishing in Cu a jiniquil. HUGE shout out to the Cuajiniquil crew. Couldn't have done it without the constant la ughter. Remember that one time? Shout out to Calla and Michael for helping be videotape fish after fish after fish after fish and enduring many jelly stings. Thank you Jenna for helping me think through my ideas , putting things in new perspecti ves, and always knowing how to make me laugh even through the sunburns and jellies. Shout out to the concrete rock I picked up from the muelle, without you there would be no video observations. You rock. Science rocks.
Sea urchins reduce fish grazing Beittel 12 LITERATURE CITED Alvarado, J.J. 2 008. Seasonal occurrence and aggregation behavior of the sea urchin a stropyga pulvinata (Echinodermata: Echinoidea) in BahÃµÂ«a Culebra, Costa Rica. Pacific Science. 62: 579 592. Alvarado, J.J., et al. 2012. Reconstruction of Diadema mexicanum bioerosion imp act on three Costa Rican Pacific coral r eefs. Biologia Tropical 60: 121 32. Alvarado, J.J., et al., 2016. Bioerosion by the sea urchin Diadema mexicanum along Eastern Tropical Pacific coral reefs. Marine Ecology. 37: 1088 1102. Gill, M.A. and Hein, A.M. 2017. Social interactions among grazing reef fish drive material flux in a coral reef ecosystem . Journal of Animal Ecology. 69: 494 503 Jackson , J.B.C., et al. 2001. Historical overfishing and the recent collapse of coastal ecosys tems. Science 293, 629 Ã 638. Jellison , B. 2009. Movement and aggregation behavior of the sea urchin Diadema mexicanum. University of California Education Abroad Program: Tropical Biology and Conservation. Kennedy E.V., et al. 2010. Spatial patterns in reef generated noise re late to habitats and communities: Evidence from a Panamanian case study. Journal of Experimental Marine Biology and Ecology. 395: 85 92. Kittenger , J.N. 2014. Marine historical ecology in conservation: applying the past to manage for the f uture. Universit y of California Press. 112. Lara, Minor . 11 May 2018. Cuajiniquil. Personal communication . Michel , K.B. 2014. Functional anatomy and kinematic s of the oral jaw system during terrestrial feeding in Periophthalmus barbarus. Journal of Morphology. 275: 1145 60. Millennium Ecosystem Assessment . 2005 . Coastal s ystems. Millennium Ecosystem Assessment. 19: 515 543. Moberg, F. and Folke , C. 1999. Ecological goods and services of coral reef ecosystems. Ecological Economics 29: 215 233. Mumby , J.P. et al., 2006 Re visiting the catastrophic die off of the urchin Diadema antillarum on Caribbean coral reefs: Fresh insights on resilience from a simulation model . Ecological Modelling. 196: 131 148. O'Leary, J.K., and McClanahan T.R. 2010 Tropic cascades result in large s cale coralline algae loss through differential grazer effects. 91: 3584 Ã 3597 . Pandolfi , J. M. , et al. 2003. Global trajectories of the long term decline of coral reef ec osystems. Science 301, 955 Ã 958. Perry , C.T., et al., 2014 Changing dynamics of C aribbe an reef carbonate budgets: emergence of reef bioeroders as critical controls on present and future reef growth potential . Proceedings of the Royal Society B. 281: 1 9. Rowe , C. 2011. Fishing away marine conservation: poverty, resource dependence, and poor management in C uajiniquil, Costa Rica. Department of Environmental Studies of Amherst College . Rupp , F.P. and Hulsey , C.D. 2014. Influence of substrate orientation on feeding kinematics and performance of algae grazing Lake Malawi cichlid fishes . Journal of Experimental Biology. 217: 3057 3066 .