89 Prospects and Limitations of Citizen Science in Invasive Species Management: A Case Study with Burmese Pythons in Everglades National Park Bryan G. Falk1,*, Ray W. Snow2, and Robert N. Reed3 Abstract - Citizen-science programs have the potential to contribute to the management of invasive species, including Python molurus bivittatus (Burmese Python) in Florida. We characterized citizen-science–generated Burmese Python information from Everglades National Park (ENP) to explore how citizen science may be useful in this effort. As an initial step, we compiled and summarized records of Burmese Python observations and removals collected by both professional and citizen scientists in ENP during 2000–2014 and found many patterns of possible significance, including changes in annual observations and in demographic composition after a cold event. These patterns are difficult to confidently interpret because the records lack search-effort information, however, and differences among years may result from differences in search effort. We began collecting search-effort information in 2014 by leveraging an ongoing citizen-science program in ENP. Program participation was generally low, with most authorized participants in 2014 not searching for the snakes at all. We discuss the possible explanations for low participation, especially how the low likelihood of observing pythons weakens incentives to search. The monthly rate of Burmese Python observations for 2014 averaged ~1 observation for every 8 h of searching, but during several months, the rate was 1 python per >40 h of searching. These low observation-rates are a natural outcome of the snakes’ low detectability— few Burmese Pythons are likely to be observed even if many are present. The general inaccessibility of the southern Florida landscape also severely limits the effectiveness of using visual searches to find and remove pythons for the purposes of population control. Instead, and despite the difficulties in incentivizing voluntary participation, the value of citizen-science efforts in the management of the Burmese Python population is in collecting search-effort information. Introduction Laypersons have a long history of voluntarily participating in biological research by generating and sometimes analyzing data (Dickinson et al. 2010, 2012; Silvertown 2009). Success stories of citizen-science initiatives include the annual Christmas bird counts (Butcher et al. 1990, Link et al. 2006), the determination of complex protein structures (Dill et al. 2012, Good et al. 2013), and Charles Darwin’s work (Silvertown 2009). Citizen-science programs are potentially useful in the management of invasive species because of the associated reduction in labor costs and potential for early detection of newly introduced species (Crall et al. 1US Geological Survey, Everglades National Park, Homestead, FL 33034. 2National Park Service, Everglades National Park, Homestead, FL 33034. 3US Geological Survey, Fort Collins Science Center, Fort Collins, CO 80526. *Corresponding author - bfalk@usgs.gov. Manuscript Editor: John Willson Everglades Invasive Species 2016 Southeastern Naturalist 15(Special Issue 8):89–102 Southeastern Naturalist B.G. Falk, R.W. Snow, and R.N. Reed 2016 90 Vol. 15, Special Issue 8 2010, Dickinson et al. 2012, Gallo and Waitt 2011). Here, we explore how citizen science might be useful in the management of an established invasive species, Python molurus bivittatus Kuhl (Burmese Python) in Florida. The Burmese Python population in Florida is a widely recognized example of a problematic invasive animal population (Dorcas et al. 2012, Dove et al. 2011, Mc- Cleery et al. 2015, Snow et al. 2007a). Pythons colonized Everglades National Park (ENP) via the pet trade during or prior to the 1990s (Meshaka et al. 2000, Willson et al. 2011), and the population has since spread beyond ENP boundaries across much of the greater Everglades ecosystem (Hunter et al. 2015). This invasive snake preys upon nearly 40 species of native vertebrates (Dove et al. 2011, Falk and Reed 2015, Snow et al. 2007b) and has been implicated in declines and local extirpations of several of these prey populations (Dorcas et al. 2012, McCleery et al. 2015). Unfortunately, we lack effective management tools (Reed and Rodda 2009), and many questions remain (e.g., What are the trends in population size? How many hours of searching are necessary before observing a python?). Although not collected to answer these questions, citizen-scientist–generated records of Burmese Python observations and removals may contain some answers. Python records from ENP and surrounding areas collected by both professional and citizen scientists are archived by biologists for the National Park Service (NPS) and the US Geological Survey (USGS) in a combined NPS/USGS database that contains ENP python records from the mid-1990s through the present. Though ENP provides annual summaries of these records from both in inside and outside the park on its website (www.nps.gov/ever/learn/nature/burmesepythonresearch. htm), records have only been explored and summarized through 2005 (Snow et al. (2007a). Another source of python records is the Early Detection and Distribution Mapping System (EDDMapS; www.eddmaps.org). Individuals submit observations of invasive species on the EDDMapS website or through a free smartphone-application, and these observations are verified by experts and made available online. The EDDMapS database duplicates some records from the NPS/USGS database, but the extent of redundancy between these databases is unclear. These records may convey information about the Burmese Python population, particularly the effect of a cold event on survivorship. The death of many freeranging pythons in southern Florida during January 2010 was attributed to several days of atypically cold temperatures (Mazzotti et al. 2011), and concurrent studies at higher latitudes with captive animals suggested that pythons may lack behavioral and/or physiological characteristics necessary to withstand cold temperatures (Avery et al. 2010, Dorcas et al. 2011). Trends in the annual number of python records in the NPS/USGS database were consistent with the hypothesis that the 2010 cold snap resulted in population declines, but whether these records are suitable proxies for population size has not been explored. Citizen-science efforts may further enhance the research value of the python records by contributing search-effort data. Citizen scientists have already played a significant role in gathering information about invasive pythons in ENP via the park’s Authorized Agent Program (AAP). The AAP is composed of citizen scientists who Southeastern Naturalist 91 B.G. Falk, R.W. Snow, and R.N. Reed 2016 Vol. 15, Special Issue 8 are authorized as NPS volunteers to collect Burmese Pythons in the park. Participants surrender all captured pythons and provide capture location, date, and time to ENP, and though they have substantially contributed to the number of Burmeses Python records from ENP, volunteers have not been required to submit search-effort data. Our goals were to examine existing datasets and leverage citizen-science efforts to better characterize the Burmese Python population in ENP. We compiled and summarized Burmese Python records from ENP through 2014 using the NPS/USGS and EDDMapS databases, asking whether the records are consistent with a die-back following the 2010 cold event, and exploring other potential explanations for the spatial and temporal patterns in the records. We also collected search-effort data from the AAP in 2014 to better understand intra-annual variation in python-observation rates and suggest ways in which citizen-science programs can contribute to the management and research of the Burmese Python population in Florida. Methods Summary of Burmese Python records from ENP through 2014 Our first goal was to assemble a complete-as-possible database of Burmese Python records for ENP. We merged all ENP python records from the NPS/USGS and the EDDMapS databases (accessed 10 April 2015) into a single database. We examined each record in the resultant database to identify duplicate records, paying particular attention to the date, location, and reporter identity of each record. We flagged and discarded duplicates prior to analysis. We characterized each record in the new database in several ways. Each record was categorized as an observation or a removal. We considered an “observation” as a credible python sighting for which the animal was neither captured nor found dead, and a “removal” as an instance when the animal was either captured or found dead (i.e., “removed” from the ecosystem). We used the collecting-locality information to discriminate ENP records from those originating elsewhere (e.g., Big Cypress National Preserve, US Highway 41, South Florida Water Management District). If within ENP, we further binned records into a more specific area that referred to eastern ENP, northern ENP, main park road, and “other” (Fig. 1). The first 3 area categories included only observations on or immediately adjacent to roads or in areas accessible by vehicle, and the “other” category included observations not associated with these areas. We included information on snout–vent length (SVL) and sex when available. Using this newly compiled database, we summarized Burmese Python records from ENP through 2014. We examined the number of annual python records from each general location that were in the NPS/USGS database to learn what proportion of the records were from ENP. All following analyses focused only on the ENP records. These summary analyses included: (1) the annual number of python removals and observations, (2) the annual number of python removals from each area, and (3) the monthly number of python removals from each area. We were also interested in identifying any age-related patterns in the python records, and summarized the annual proportion of python removals consisting Southeastern Naturalist B.G. Falk, R.W. Snow, and R.N. Reed 2016 92 Vol. 15, Special Issue 8 of animals approximately ≤1 year old. We performed the same summary using animals removed only during July–September of each year—when most young snakes are observed—to assess whether a lack of young pythons in any year was a result of differences in search effort when the young snakes are relatively abundant. We classified animals as ≤1 year old using SVL, and assumed that the shortest SVL measured during May (at least 1 month before Burmese Pythons hatch in Florida; Snow et al. 2007a, 2010) of any year represented the approximate minimum SVL attainable in 1 year. These calculations included only Figure 1. Map of southern Florida. The green-filled circles represent locations of Burmese Python observation and removal records from Everglades National Park during 2000–2014, most of which occurred on roads where pythons were relatively easy to see. Records from outside the park are not included. The dotted lines delineate the 3 primary geographicarea categories (main park road, north, and east), with the remaining areas combined in an “other” category. We used these 4 categories to group the python records. Southeastern Naturalist 93 B.G. Falk, R.W. Snow, and R.N. Reed 2016 Vol. 15, Special Issue 8 animals that were removed and subsequently measured and not animals that were removed without being measured. We explored another potential source of variation in the python records to identify alternative hypotheses that could also explain the observed trends. We could not test for differences in search effort among years because search-effort data were unavailable. Instead, we used chi-square tests to ask if the proportions of records from each area were significantly different from year to year when accounting for sample size (i.e., we tested the hypothesis that the proportion of records from each area—whatever they may be—remained stable over time). A change in the proportion of records from each area over time indicates that temporal changes across areas have occurred in python density, detectability of individuals, and/or search effort. Any of these alternative explanations would render the python records a poor proxy for the population itself. Using citizen science to collect search-effort data We made several changes to the AAP in 2014 to further explore the effectiveness of citizen science as a management tool for Burmese Pythons. We renewed authorization for interested past participants, and recruited new potential agents by word-of-mouth and through inquiries to the ENP website. We required that agents submit effort data for all searches, including the date, start time, duration, and location, and used these data to calculate an average python-observation rate (i.e., the number of hours searched per python observation) for each month. Even though most of these pythons were also removed, we used the term “observation” because we were interested in how often pythons were encountered rather than capture efficiency. We were interested in whether agents were more likely to search when observation rates were high, and so we tested for a correlation between monthly python observation rates and search effort using Pearson’s r. We evaluated agent participation by calculating the number of hours spent searching by each agent in 2014. Finally, we calculated the number of python records contributed by the AAP in ENP for 2014. Results Summary of Burmese Python records from ENP through 2014 We compiled a database of 1412 Burmese Python records from ENP through 2014. The NPS/USGS database contained a total of 2475 records, 1337 (or 54%) of which were from ENP. We identified a total of 886 records in the EDDMapS database as originating in ENP, and 807 (or 92%) were duplicates from the NPS/ USGS database. Four EDDMapS records appeared to be duplicates of other EDDMapS records, and the remaining 75 EDDMapS records were not in the NPS/ USGS database. Of the 1412 unique ENP records (from both databases), 11 were removals without collection information (e.g., python carcasses found in freezers at ENP law-enforcement offices without associated data). A total of 8 records were recorded prior to 2000, the year that the population was first recognized in the literature as established (Meshaka et al. 2000), and 1393 records were Southeastern Naturalist B.G. Falk, R.W. Snow, and R.N. Reed 2016 94 Vol. 15, Special Issue 8 collected during 2000–2014. At least 76% of these records from 2014 were generated by citizen scientists, but we had insufficient information to characterize the citizen-scientist contributions in prior years. The number of python records from ENP is variable over space and time. Most records were removals and not observations, and though the annual total for all records was highest in 2009, more pythons were removed from ENP in 2014 than in any previous year (Fig. 2). Most pythons were removed from the main park road area of the park, but there was a large increase in removals from the eastern Everglades region during 2013–2014 (Fig. 3A). Monthly removal records had a bimodal distribution, with the highest peak in python removals during August, and a second peak during January (Fig. 3B). The removals from each area generally co-varied from month to month, but proportionally more records originated from the North Everglades area during the winter (Fig. 3B). The annual proportion of young pythons markedly increased during 2013– 2014 (Fig. 4). The minimum SVL for any python collected from ENP in May was 105 cm, and so we assumed that snakes with a SVL of ≤100 cm were ≤1 year old. During 2003–2011, an average of 19% (± 6.9% SD) of all pythons measured were ≤100 cm SVL, but no pythons ≤100 cm SVL were measured in 2012. In 2013 and 2014, 78% and 53% of measured pythons were ≤100 cm SVL, respectively. The absence of observations of young pythons in 2012 likely results from a small overall sample size; a total of only 4 pythons were measured during July–September 2012 (when small pythons are likely to be observed), but the number of measured pythons ranged from 9 to 123 during those months in other years (data not shown). Additional pythons were removed, but not measured, and it is unknown whether there was a size bias for which individuals were left unmeasured in some years. Figure 2. Annual numbers of Burmese Python removals and observations from Everglades National Park during 2000–2014. Most records are removals. The highest number of records occurred in 2009, and the most removals occurred in 2014. Southeastern Naturalist 95 B.G. Falk, R.W. Snow, and R.N. Reed 2016 Vol. 15, Special Issue 8 Figure 3. Geographic origin of (A) annual and (B) monthly Burmese Python removals in Everglades National Park. Most annual and monthly removals have been from main park road (MPR). Proportionally more pythons were removed from the eastern area of the park during 2013–14, and proportionally more pythons were removed from the northern area of the park during the winter months. Southeastern Naturalist B.G. Falk, R.W. Snow, and R.N. Reed 2016 96 Vol. 15, Special Issue 8 We were unable to reject differences in detectability or search effort among areas as alternative explanations for the patterns in the Burmese Python records. The proportion of pythons collected from each area was variable over years, χ2 (13, n = 36) = 462.10, P much less than 0.001. Using citizen science to collect search-effort data Citizen scientists in the AAP searched for Burmese Pythons for a total of 1127 h and contributed 148 (74%) of all python records for ENP in 2014. The annual Figure 4. (A) Number and (B) proportion of young Burmese Pythons measured in Everglades National Park during 2003–2014. Pythons measuring ≤1 m were considered ≤1 year of age and “young” (see text), and the proportion of these animals markedly increased beginning in 2013. Southeastern Naturalist 97 B.G. Falk, R.W. Snow, and R.N. Reed 2016 Vol. 15, Special Issue 8 python observation rate was ~8 h per python observation, and this rate varied over the year (Fig. 5). Observation rates were highest in the summer months, reaching 1 python per 3.7 h in August (i.e., a rate of 0.27 pythons/h; Fig. 5). Monthly observation rates were 1 python per 15–40 h in December, March, April, and June, and 1 python per >40 h during January, February, and May (Fig. 5). Participation was variable among agents, and generally low. A total of 36 individuals participated in the AAP during 2014, but the number active at any one time fluctuated over the year as we recruited new participants (a total of 11 in 2014) and others left. One individual voluntarily left the program because of the new reporting requirements, another was excused after exhibiting prohibited behavior, and 5 were no longer interested in participating or moved from the area. The monthly totals for search effort were positively correlated with the monthly rate of python observations (r = 0.81; P = 0.002). One participant contributed 890 h (79%) of all program search-effort and 127 (86%) of all python observations for the program. The second-most–productive participant contributed 100 h (9%) of all search effort and 9 (6%) python observations. Another 9 participants each contributed an average of 15 (1%) search hours and 1.3 (1%) python observations. The remaining 25 program participants in 2014 did not search at all during the year. Discussion There is considerable temporal and spatial variation in the observation and removal records for pythons in ENP, some of which is consistent with a population Figure 5. The average monthly number of Burmese Python observations per hour in Everglades National Park during 2014, as calculated from data submitted by citizen-scientist participants authorized to search in the park. The highest observation rates occurred in August, when young pythons were commonly encountered. Python observation rates were lowest in the winter and early spring. Southeastern Naturalist B.G. Falk, R.W. Snow, and R.N. Reed 2016 98 Vol. 15, Special Issue 8 die-back and re-growth after a cold event; nonetheless, it is difficult to distill meaning from these patterns without additional information. We collected search-effort data for the first time in 2014, and the rate of python observations was generally low, with less than 1 python observed for every 40 h of searching during 3 months of the year. Citizen-scientist participation was generally low (most authorized individuals did not search at all) and strongly correlated with the rate of python observations. Summary of Burmese Python records from ENP through 2014 Citizen-scientist–generated observation and removal records may enhance the detection of newly introduced populations (Simpson et al. 2009), but without search-effort information, these records alone may have limited value for management of invasive species in the areas where they are known to be established. Python removals peaked in 2014 (Fig. 2), which may have been a result of more pythons in ENP (i.e., population-size growth), or simply a reflection of differences in search effort. For example, 1 AAP participant contributed a majority of removals from ENP in 2014, most of which resulted from highly intensive searcheffort in the eastern region of ENP; this activity explains the recent increase in removals from that area (Fig. 3A). Likewise, the recent trend of increasing python removals may be explained by this same AAP participant’s concurrent enormous effort. These few examples illustrate how interpretation of removal data without information on search effort can lead to spurious conclusions. We were able to show that the python records are biased, but the records are nonetheless consistent with the hypothesis that the January 2010 cold event caused a die-back in Florida’s Burmese Python population. Focusing on records only from the main park road, python removals generally increased until reaching a peak in 2009, and after a decline through 2012, the number of records only began to increase in 2013–2014. A large influx of young pythons also occurred beginning in 2013, three years after the cold event. Coincidentally, it takes a female Burmese Python 3 y from hatching to become reproductive (i.e., 1 python generation; Reed and Rodda 2009), suggesting that Florida’s python population was re-growing after a die-back. In any case, some proportion of pythons clearly survived a cold event that caused the death of many others. How these individuals may have survived (e.g., whether it was a physiological or behavioral mechanism) and whether they were able to confer the same advantage to their offspring remains an interesting and important area of future research. The summary monthly data reveal the seasonality of python observations in ENP (Fig. 3B). Total python removal and observation rates are highest in August, when young pythons are most commonly seen. This pattern contrasts with Snow et al.’s (2007a) summary, which indicated that in Florida through 2005, August had the fewest monthly python removals. Though snake-activity patterns may have changed over time, the differences probably arose because the data from that study included areas outside the park where pythons are more likely to be found in the winter. During the winter months, pythons are most easily found along primitive roads (e.g., levees) where adult snakes occasionally bask (observations of young snakes are relatively infrequent during this time). There are few such areas in ENP, Southeastern Naturalist 99 B.G. Falk, R.W. Snow, and R.N. Reed 2016 Vol. 15, Special Issue 8 most notably the L-67ext levee (Fig. 1). The wintertime productivity of the L-67ext (and the relatively poor wintertime searching conditions in other areas) explains the higher proportion of python records from the northern area of ENP during the winter months. Using citizen science to collect search-effort data Low or non-existent search effort by most AAP participants reveals the limitations of citizen science as a control tool for pythons. At a glance, the AAP appears to be successful: 148 pythons were removed and 1127 h were spent searching by program participants in 2014. Nevertheless, nearly all this work was conducted by 2 individuals, and without their contributions, the AAP would have produced only 137 h of search effort and 12 pythons in 2014. The strong positive correlation between search effort and the rate of python observations suggests that participants are more likely to search for pythons during periods of the year when the probability of success is high. Again, however, this correlative pattern is primarily derived from pythons removed by a single participant. The remaining participants searched too infrequently to analyze their collection data separately, and our finding should be viewed as preliminary. Program productivity is almost certainly hindered by the low python observation rates; participants donate their time, accrue associated expenses (e.g., vehicle fuel, field gear), and if the python observation is the reward, it is rarely bestowed. Is it possible to increase productivity of citizen-science programs through better incentivizing? There are currently no formal rewards for participation in the AAP, and most participants seem to be motivated by a desire to observe and capture a python in the wild, a general interest in field herpetology, and/or a desire to gain additional field experience. During many months of the year, the latter 2 motivations may be satiated long before encountering a python. The authorization to search for pythons in ENP allows participants to use snake hooks, spotlights, and other “herping” gear otherwise prohibited in a national park, and so is mildly incentivizing for some people. Nonetheless, neither the ability to legally search for reptiles in ENP nor a more general desire to assist with ecosystem conservation appears to be sufficient reward to inspire participants to maintain enough interest to generate significant contributions to python research or management. Many zero-effort participants in 2014 contributed substantially in prior years, suggesting that whatever caused initial interest may have waned over time. Replacing inactive participants with new, eager participants seems like a logical solution, but efforts were generally unsuccessful in 2014; some new recruits contributed search-effort data, but most contributed none. In the end, the local pool of reptile-focused potential participants with a willingness to both persistently search and record search-effort data is small and unlikely to increase markedly. The success of the AAP in 2014 is owed to 2 dedicated participants who are uniquely motivated, and unfortunately rare. One participant uses the python searches to provide an opportunity for military veterans who assist with the searches to find relief from post-traumatic stress disorder. The other enjoys finding and observing Southeastern Naturalist B.G. Falk, R.W. Snow, and R.N. Reed 2016 100 Vol. 15, Special Issue 8 snakes in the wild, and has been actively searching for pythons for over a decade without losing interest. We acknowledge that other uniquely motivated individuals may exist and be able to contribute to the AAP in the future, but emphasize that if a citizen-science program is going to substantially contribute to the research and management of Burmese Pythons, it cannot rely on a few over-achieving participants. Low observation-rates are a significant challenge to research and management of Burmese Pythons The low observation rate is a significant challenge to the research and management of Burmese Pythons. During January, March, and May in 2014, for example, the equivalent of at least one 40-h workweek was necessary to remove just 1 python (i.e., a fulltime employee tasked with searching for pythons would likely have found ≤13 of them during those 3 months). These low sample sizes make it diffucult to estimate population size, delimit the geographic extent of the population (but see Hunter et al. [2015]), or answer other questions essential to successful management of the invasive python population. The low rate of python observations is almost certainly due to their low detection- probability for human searchers. In a study where human searchers were tasked with finding pythons in a 31 m x 25 m outdoor natural enclosure, they were successful ≤1% of the time (Dorcas and Willson 2013). These rates may be even lower in the wild. Improving observation rates by human searchers may ameliorate the costs and difficulties in python research and management, but observation rates may already be so low that considerable increases in efficiency will have little or no practical effect. The low detectability of pythons and the limited accessibility of the southern Florida landscape are obstacles to the use of visual searches as a management tool for Burmese Pythons. Nearly all pythons removed from ENP during 2000–2014 were found on roads and levees where the animals are easy to see; the python locations in Figure 1 are essentially plots of the roads in ENP. Pythons located just off the road and in the vegetation are very difficult to find (Dorcas and Willson 2013). Even if all pythons could be eradicated from areas adjacent to roads, the vast amount of roadless areas in southern Florida (both inside and outside ENP) might offer sufficient suitable habitat to sustain the population and serve as a source for continual recolonization of the roadside areas. We suggest that the potential value of visual searches and citizen-science programs like the AAP is in providing search-effort data for Burmese Pythons. In 2014, we collected python search-effort data from citizen scientists for the first time and were able to provide preliminary answers to questions about the temporal changes in python observation rates in ENP and the behavior of volunteer searchers. More rigorous data-collection protocols and a huge increase in search effort would provide further insights, such as estimates of population-size trajectories (Link and Sauer 1998, Royle 2004) and occupancy (MacKenzie et al. 2002, 2006), but there is little evidence that such a large, sustained effort can be leveraged from a citizen-science program for Burmese Pythons in Florida. Southeastern Naturalist 101 B.G. Falk, R.W. Snow, and R.N. Reed 2016 Vol. 15, Special Issue 8 Acknowledgments Thomas Rahill (The Swamp Apes), Anthony Flanagan, and other members of the Authorized Agent Program contributed a significant amount of search effort and data, for which we are extremely grateful. Tylan Dean facilitated the changes to the Authorized Agent Program. Members of Kristen Hart’s and Frank Mazzotti’s labs transposed many of the python records into a digital database. J.D. Willson, Brian Smith, and 2 anonymous reviewers provided many helpful comments that improved our manuscript—thank you. Everglades National Park, the Greater Everglades Priority Ecosystem Science Program, and the USGS Invasive Species Science Program provided funding. Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the US Government. Literature Cited Avery, M.L., R.M. Engeman, K.L. Keacher, J.S. Humphrey, W.E. Bruce, T.C. Mathies, and R.E. Mauldin. 2010. Cold weather and the potential range of invasive Burmese Pythons. Biological Invasions 12:3649–3652. Butcher, G.S., M.R. Fuller, L.S. Mcallister, and P.H. Geissler. 1990. An evaluation of the Christmas Bird Count for monitoring population trends of selected species. Wildlife Society Bulletin 18:129–134. Crall, A.W., G.J. Newman, C.S. Jarnevich, T.J. Stohlgren, D.M. Waller, and J. Graham. 2010. Improving and integrating data on invasive species collected by citizen scientists. Biological Invasions 12:3419–3428. Dickinson, J.L., B. Zuckerberg, and D.N. Bonter. 2010. Citizen science as an ecological research tool: Challenges and benefits. Annual Review of Ecology, Evolution, and Systematics 41:149–172. Dickinson, J.L., J. Shirk, D. Bonter, R. Bonney, R.L. Crain, J. Martin, T. Phillips, and K. Purcell. 2012. The current state of citizen science as a tool for ecological research and public engagement. Frontiers in Ecology and the Environment 10:291–297. Dill, K.A., and J.L. Maccallum. 2012. The protein-folding problem, 50 years on. Science 338:1042–1046. Dorcas, M.E., and J.D. Willson. 2013. Hidden giants: Problems associated with studying secretive invasive pythons. Pp. 367–385, In W.I. Lutterschmidt (Ed.). Reptiles in Research: Investigations of Ecology, Physiology, and Behavior from Desert to Sea. Nova Science Publishers, New York, NY. 547 pp. Dorcas, M.E., J.D. Willson, and J.W. Gibbons. 2011. Can invasive Burmese Pythons inhabit temperate regions of the southeastern United States? Biological Invasions 13:793–802. Dorcas, M.E., J.D. Willson, R.N. Reed, R.W. Snow, M.R. Rochford, M.A. Miller, W.E. Meshaka, P.T. Andreadis, F.J. Mazzotti, and C.M. Romagosa. 2012. Severe mammal declines coincide with proliferation of invasive Burmese Pythons in Everglades National Park. Proceedings of the National Academy of Sciences 109:2418–2422. Dove, C.J., R.W. Snow, M.R. Rochford, and F.J. Mazzotti. 2011. Birds consumed by the invasive Burmese Python (Python molurus bivittatus) in Everglades National Park, Florida, USA. The Wilson Journal of Ornithology 123:126–131. Falk, B.G., and R.N. Reed. 2015. Challenges to a molecular approach to prey identification in the Burmese Python, Python molurus bivittatus. PeerJ 3:e1445 Gallo, T., and D. Waitt. 2011. Creating a successful citizen-science model to detect and report invasive species. BioScience 61:459–465. Southeastern Naturalist B.G. Falk, R.W. Snow, and R.N. Reed 2016 102 Vol. 15, Special Issue 8 Good, B., and A. Su. 2013. Crowdsourcing for bioinformatics. Bioinformatics 29:1925–1934. Hunter, M.E., S.J. Oyler-Mccance, R.M. Dorazio, J.A. Fike, B.J. Smith, C.T. Hunter, R.N. Reed, and K.M. Hart. 2015. Environmental DNA (eDNA) sampling improves occurrence and detection estimates of invasive Burmese Pythons. PloS One 10:e0121655. Link, W.A., and J.R. Sauer. 1998. Estimating population change from count data: Application to the North American Breeding Bird Survey. Ecological Applications 8:258–268. Link, W.A., J.R. Sauer, and D.K. Niven. 2006. A hierarchical model for regional analysis of population change using Christmas Bird Count data, with application to the American Black Duck. The Condor 108:13–24. Mackenzie, D.I., J.D. Nichols, G.B. Lachman, S. Droege, J. Andrew Royle, and C.A. Langtimm. 2002. Estimating site-occupancy rates when detection probabilities are less than one. Ecology 83:2248–2255. Mackenzie, D.I., J.D. Nichols, J.A. Royle, K.H. Pollock, L. Bailey, and J.E. Hines. 2006. Occupancy Estimation and Modeling: Inferring Patterns and Dynamics of Species Occurrence. Academic Press, San Diego, CA. 344 pp. Mazzotti, F.J., M.S. Cherkiss, K.M. Hart, R.W. Snow, M.R. Rochford, M.E. Dorcas, and R.N. Reed. 2011. Cold-induced mortality of invasive Burmese Pythons in south Florida. Biological Invasions 13:143–151. McCleery, R.A., A. Sovie, R.N. Reed, M.W. Cunningham, M.E. Hunter, and K.M. Hart. 2015. Marsh Rabbit mortalities tie pythons to the precipitous decline of mammals in the Everglades. Proceedings of the Royal Society of London B: Biological Sciences DOI: 10.1098/rspb.2015.0120. Meshaka, W., W.F. Loftus, and T. Steiner. 2000. The herpetofauna of Everglades National Park. Florida Scientist 63:84–103. Reed, R.N., and G.H. Rodda. 2009. Giant constrictors: Biological and management profiles and an establishment-risk assessment for nine large species of pythons, anacondas, and the Boa Constrictor. US Geological Survey open file report 2009-1202. Washington, DC. Royle, J.A. 2004. N-mixture models for estimating population size from spatially replicated counts. Biometrics 60:108–115. Silvertown, J. 2009. A new dawn for citizen science. Trends in Ecology and Evolution 24:467–471. Simpson, A., C. Jarnevich, J. Madsen, R. Westbrooks, C. Fournier, L. Mehrhoff, M. Browne, J. Graham, and E. Sellers. 2009. Invasive species information-networks: Collaboration at multiple scales for prevention, early detection, and rapid response to invasive alien species. Biodiversity 10:5–13. Snow, R., K. Krysko, K. Enge, L. Oberhofer, A. Warren-Bradley, and L. Wilkins. 2007a. Introduced populations of Boa constrictor (Boidae) and Python molurus bivittatus (Pythonidae) in southern Florida. Pp. 365–386, In R. Henderson and R. Powell (Eds.). Biology of Boas and Pythons. Eagle Mountain Publishing, LC, Eagle Mountain, UT. 438 pp. Snow, R.W., M.L. Brien, M.S. Cherkiss, L. Wilkins, and F.J. Mazzotti. 2007b. Dietary habits of the Burmese Python, Python molurus bivittatus, in Everglades National Park, Florida. Herpetological Bulletin 101:5–7. Snow, R.W., A.J. Wolf, B.W. Greeves, M.S. Cherkiss, R. Hill, and F.J. Mazzotti. 2010. Thermoregulation by a brooding Burmese Python (Python molurus bivittatus) in Florida. Southeastern Naturalist 9:403–405. Willson, J.D., M.E. Dorcas, and R.W. Snow. 2011. Identifying plausible scenarios for the establishment of invasive Burmese Pythons (Python molurus) in southern Florida. Biological Invasions 13:1493–1504.