Effective citizen science implementation reveals ecological and conservation insights for two amphibian species    

Effective citizen science implementation reveals ecological and conservation insights for two amphibian species | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 24 February 2026 V2 Latest version Share on Effective citizen science implementation reveals ecological and conservation insights for two amphibian species Authors : Omer Darel 0009-0006-5603-3575 , Olga Rybak , Asaf Ben Levy , Gabi Kolodny , Tamar Kis-Papo , Nirit Lavie Alon , Rotem Vidan , and oren kolodny 0000-0002-0095-693X [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.173376560.04983997/v2 Published Journal of Zoology Version of record Peer review timeline 352 views 204 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Over 40% of amphibian species are threatened, urbanization being a major threat to their survival. Conservation efforts to sustain viable amphibian populations within urban spaces may play a meaningful part in protecting amphibian species. To explore the factors that influence the viability of urban amphibian population in Mediterranean environments, we used a capture-recapture analysis applied to a large dataset collected in a multi-year citizen-science program, focused on two urban Salamandra infraimmaculata populations, and to a second dataset of two Bufotes variabilis populations in an urban park and in a nearby nature reserve. Individuals of both species have individually-unique patterns of dorsal spots, allowing for noninvasive recapture identification. Using the salamander dataset, we developed a length-based age-estimation method and discovered a prolonged period of increased vulnerability throughout their first years of life, even after reaching sexual maturity, a finding with important implications for management. Additionally, the two case studies suggest that the creation of fish-containing artificial water bodies in Mediterranean habitats can have highly detrimental impacts on amphibian populations. The study uncovered population-specific information, such as unknown breeding sites and population size estimates, of conservation importance, and demonstrates the utility of citizen science in study and conservation of urban ecology. Using citizen science to protect threatened amphibian populations in Mediterranean urban spaces Omer Darel 1* Olga Rybak 2 , Asaf Ben Levy 2 , Gabi Kolodny 2 , Tamar Kis- Papo 3 , Nirit Lavie Alon 2 , Rotem Vidan 2 , Oren Kolodny 1\RL* 1 Department of Ecology, Evolution and Behavior, The A. Silberman Institute of Life Sciences, The Hebrew University of Jerusalem, Jerusalem \RL9190401, Israel Correspondence: [email protected] ; [email protected] 2 The Society for the Protection of Nature in Israel 3 The Israeli Ministry of Education Abstract Over 40% of amphibian species are threatened, urbanization being a major threat to their survival. Conservation efforts to sustain viable amphibian populations within urban spaces may play a meaningful part in protecting amphibian species. To explore the factors that influence the viability of urban amphibian population in Mediterranean environments, we used a capture-recapture analysis applied to a large dataset collected in a multi-year citizen-science program, focused on two urban Salamandra infraimmaculata populations, and to a second dataset of two Bufotes variabilis populations in an urban park and in a nearby nature reserve. Individuals of both species have individually-unique patterns of dorsal spots, allowing for noninvasive recapture identification. Using the salamander dataset, we developed a length-based age-estimation method and discovered a prolonged period of increased vulnerability throughout their first years of life, even after reaching sexual maturity, a finding with important implications for management. Additionally, the two case studies suggest that the creation of fish-containing artificial water bodies in Mediterranean habitats can have highly detrimental impacts on amphibian populations. The study uncovered population-specific information, such as unknown breeding sites and population size estimates, of conservation importance, and demonstrates the utility of citizen science in study and conservation of urban ecology. Introduction 1.1 background and research goals The world is urbanizing at an increasing rate (Desa, 2018; Seto et al., 2013 \RL ), a process that threatens ecosystems globally (Lin & Fuller, 2013). Amphibians, the most threatened land vertebrates (Cushman, 2006; Hamer & McDonnell, 2008; Stuart et al., 2004), exhibit heightened susceptibility to urbanization-driven environmental changes (Clevenot et al., 2018; Cushman, 2006; Hamer & McDonnell, 2008; Hamer & Parris, 2011; Schoch, 2009). Urban environments present challenges such as pollution, fragmentation, and habitat degradation. Amphibians, with their permeable skin, limited mobility, and dual reliance on aquatic and terrestrial habitats, are especially susceptible to these threats. (Hamer & McDonnell, 2008). Recent years have witnessed a surge in ecologists’ interest in urban ecosystems (Colléony & Shwartz, 2019; Lin & Fuller, 2013), driven in part by the growing recognition of urban ecosystems’ importance for mental health of urban dwellers, as well as appreciation of the importance of regular exposure to nature for maintaining positive views and sentiments towards nature, thus enhancing public engagement and support for conservation efforts(Colléony & Shwartz, 2019; Kay et al., 2022; Miller, 2005; Pyle, 2002, 2003). The abundance of studies resulting from this increased interest have enriched our knowledge and understanding of urban ecosystems (Beninde et al., 2015; Shwartz et al., 2013, 2014). Several studies have shown that urban ecosystems may contain a similar number of amphibian species to that found in rural ecosystems (Hamer & McDonnell, 2008; Hamer & Parris, 2011; Parris, 2006). However, their presence does not guarantee that these species sustain viable populations in urban environments(Clevenot et al., 2018; Hamer & McDonnell, 2008). Specifically, studies concerning the suitability of stormwater reservoirs and garden ponds for amphibian reproduction have shown mixed findings (Clevenot et al., 2018; Hamer & McDonnell, 2008). These studies raise concerns over urban water bodies potentially acting as ecological traps—habitats, often artificial or human-altered, unable to support certain species but mistakenly perceived as suitable, thereby luring individuals from adjacent habitats and harming the population (Clevenot et al., 2018; Demeyrier et al., 2016). Understanding how urban environments can sustain viable populations of endangered amphibian species can be a vital for mitigating the threats posed by rapid urbanization (Hamer & McDonnell, 2008; Oertli & Parris, 2019; Vignoli et al., 2009). To achieve this, it is crucial to extensively measure proxies of population viability in urban amphibian populations. Such measures include reproductive success, juvenile recruitment, and adult survival rate (Clevenot et al., 2018; Hamer & McDonnell, 2008; Hamer & Parris, 2011; Parris, 2006). \RL Yet, linking these attributes to the environmental conditions may be challenging; Certain types of urban breeding sites can sustain a viable population in one setting, while functioning as an ecological trap in another, with no single dominant factor determining the outcome (Clevenot et al., 2018). Additionally, most studies of urban amphibians are focused on North America and Europe, limiting the existing knowledge’s applicability to other parts of the world (Oertli & Parris, 2019). The ecology of amphibians in the semiarid Western Mediterranean region significantly differs from the well-studied temperate zones (Blank & Blaustein, 2012): Breeding occurs primarily in ephemeral ponds and small rock pools sustained by rainfall or in small springs, which typically remain dry for most of the year \RL (Degani & Kaplan, 1999; Ferreira & Beja, 2013); Tadpoles in the Mediterranean often exhibit phenotypic plasticity, allowing them to adjust the timing of their metamorphosis to unpredictable hydroperiods (Goldberg et al., 2012); Additionally, the harsh summers in the region pose desiccation threats, leading many amphibians to estivate in burrows or karstic cavities for months (Degani & Warburg, 1978; Elron, 2007; Ferreira & Beja, 2013; Shemesh et al., 2022). Some studies indicate that the limiting factor for many of these populations may be the availability of aquatic breeding sites (Ferreira & Beja, 2013; Vignoli et al., 2009). There are eight known amphibian species in the southern levant (Israel and the Palestinian Territories), only one of which is not locally threatened (Dolev & Perevolotsky, 2004; Meiri et al., 2019). \RL For most of these species, this region represents the southern edge of their distribution(Degani et al., 2019).While edge populations are often the most threatened by climate change (Cahill et al., 2013; Hampe & Petit, 2005), they are also considered invaluable for the survival and evolution of species during climatic shifts (Hampe & Petit, 2005). Nevertheless, edge populations remain under-studied (Cahill et al., 2013). The projected population density and resulting urbanization rates in the state of Israel are exceedingly high (Strategic Plan for Housing (Hebrew), 2017; OECD, 2022; Worldometers.info, 2022). Since the beginning of the 20th century, an estimated 95% of the ephemeral water bodies in the country have been lost to development, and the remaining aquatic habitats are often highly degraded (Levin et al., 2009; Mendelssohn, 1985). The expected effects of climate change on the ephemeral water bodies in the area are also abnormally severe, with an expected decrease in precipitation and intense warming leading to a rapid shortening of their hydroperiods (Givati & Rosenfeld, 2013). Due to its location in a region that is expected to experience some of the most significant climatic shifts in the coming decades, the southern levant may serve as a possible indicator of the changes expected in other parts of the world (Givati & Rosenfeld, 2013). However, data on its amphibian populations is scarce, particularly in urban environments (Degani & Kaplan, 1999; Dolev & Perevolotsky, 2004; Oertli & Parris, 2019; Segev et al., 2010) Understanding the factors determining the suitability of urban habitats for sustaining viable amphibian populations is crucial to harnessing their potential for amphibian conservation. (Clevenot et al., 2018; Hamer & McDonnell, 2008; Hamer & Parris, 2011; Parris, 2006; Semlitsch, 2002). Previous studies showed that roads, the introduction of predatory fish, and invasive amphibian competitors are some of the most consistent hazards.(Clevenot et al., 2018; Hamer & McDonnell, 2008; Hamer & Parris, 2011; Kats & Ferrer, 2003). Most existing literature is focused on the community level,but identifying the specific factors that jeopardize each amphibian species requires understanding of species-specific habitat utilization, movement patterns, and life history traits. (Hamer & McDonnell, 2008; Scheffers & Paszkowski, 2012). The study of these traits usually requires intensive long-term surveys, limiting the ability of many researchers in academia to provide this crucial information (Hamer & McDonnell, 2008; Parris, 2006; Scheffers & Paszkowski, 2012). The use of citizen science in ecology has long been established as an effective tool for research and education (Campbell et al., 1999; Palmer et al., 2021; Rae et al., 2019; Rowley et al., 2019; Sullivan et al., 2014; Westgate et al., 2015). As urbanization increases, more people are denied the opportunity to have regular contact with nature. This process, dubbed the ”extinction of experience” (EOE), has worrying physical and mental health implications, and reduces the publics’ emotional involvement in conservation (Colléony et al., 2020; Colléony & Shwartz, 2019; Miller, 2005; Pyle, 2003). People who lack familiarity with amphibians in their vicinity were shown to hold more negative views towards the amphibians in their environment, and a higher tendency to directly harm those amphibians (Pavol & Fančovičová, 2012; Riós-Orjuela et al., 2020; Sousa et al., 2016). The hands-on experience provided by participating in citizen science projects was found to promote understanding and sympathy towards amphibians, among other wildlife (Sousa et al., 2016). In multiple cases, the participants of citizen science projects reported that their knowledge of the focal species, as well as their willingness to share this knowledge with other members of their community, or participate in conservation efforts of the focal species, have all increased after they participated in these programs (Peter et al., 2019). Aims In this paper we first present a case study of an intensive, multi-year citizen science project, which focused on two urban salamander populations in the city of Haifa. This project was aimed to gather extensive data to guide future conservation efforts. Over six years, local volunteers regularly located and documented dozens of salamanders, using their distinctive dorsal spot patterns for identification. The resulting dataset allowed us to perform a multi-year capture-recapture analysis on both populations, gathering insights about the population size, structure, and stability in the study sites, as well as deepening our understanding of the life history, movement patterns, and population ecology of these urban amphibians. Additionally, we analyzed a capture-recapture survey following a similar protocol on a smaller scale, which we conducted on two toad populations within and near the city of Jerusalem. By combining findings from both surveys, we assessed species-specific and general responses of amphibians to urban environmental factors. This study highlights how citizen science can yield valuable conservation insights while fostering community engagement with local wildlife. Study species The species Salamandra infraimmaculata (Marten, 1885) , the near Eastern fire salamander, is considered to be “Near Threatened” globally (Papenfuss et al., 2008), and endangered in Israel (Dolev & Perevolotsky, 2004). After mating, the females gestate for several months before spawning 10-200 larvae at aquatic breeding sites (Sharon et al., 1997). The juveniles usually spend several months in the aquatic phase of their lives, but the age and size at the time of the metamorphosis into the terrestrial adult phase vary significantly with environmental conditions (Goldberg et al., 2012). Most adult salamanders do not travel further than 10 meters during their foraging activity, though rare dispersal events of over 1 kilometer were documented (Bar-David et al., 2007). Adult salamanders usually reach their nearly-maximal body-size by the 9 th or 10 th year of their lives (Warburg, 2007; Warburg, 2008). Israeli S. infraimmaculata populations breeding in ephemeral and permanent water sources were found to differ significantly in their population size, with the former having typical population sizes of 30 to 130, and the latter sometimes numbering over 500 (Segev et al., 2010). An analysis performed on the distribution of salamanders in the Carmel found that their occupancy of urban and rural breeding sites is similar (Blank & Blaustein, 2012). Bufotes variabilis is a green toad species native to the southern levant (previously classified as Bufotes viridis , Bufo viridis or Pseudepidalea viridis )(Meiri et al., 2019) \RL . Toads of this species usually, but not exclusively, breed in ephemeral water bodies (Elron, 2007). The mating takes place in the water, during which the female may lay more than 8000 eggs (Elron, 2007). The tadpoles complete their metamorphosis within 3 months on average and reach sexual maturity at the age of two to three years. The mean lifespan is six years for males and four years for females(Elron, 2007). This toad species was once the most commonly sighted amphibian species in Israel (Bodenheimer, 1935; Nevo & Schneider, 1976), but its population declined significantly over the last decades (Dolev & Perevolotsky, 2004; Elron, 2007), and it is now considered to be locally endangered (Meiri et al., 2019). \RL Both B. variabilis and S. infraimmaculata have visible dorsal patterns which is unique to each individual and can therefore be used as a substitute for marking individual animals (Burgstaller et al., 2020; Warburg, 1994). Results Preface In this section we present the results of our analysis of two datasets. The first is the product of the Haifa Salamander Monitoring project, a citizen science program that aimed to monitor two S. infraimmaculata populations within the city of Haifa. The survey was first executed in 2016, in the Haifa educational zoo and the adjacent stream, Wadi Lotem. This population was previously monitored to some extent by the staff of the Haifa zoo and increasing concerns regarding its decline prompted a more thorough survey in the form of this citizen science program. This salamander population is thought to have originally been breeding in the small, ephemeral water bodies along the Wadi Lotem stream; another possibility is that it was introduced prior to the 1980’s from elsewhere via escapees from zoo captivity. Several additional potential breeding sites have existed at least since the 1980’s in the form of artificial ponds within the zoo. Some of these ponds contain invasive predatory fish, as well as dense vegetation and algae. It is feared that the exotic vegetation covering the water surface leads to anoxic conditions during certain times of the year, possibly rendering these ponds unsuitable for salamander breeding and even turning them into ecological traps. In the following year, 2017, the survey was extended to an additional site, a stream named Wadi Ahuza, three kilometers to the north of Haifa zoo. This site suffered a wildfire in 2016, and one of the project’s goals was to assess the wildfire’s impact on the resident salamander population. The main component of the project was a terrestrial survey, which included two constant routes in Wadi Lotem/Haifa zoo and one constant route in Wadi Ahuza. All surveys took place within a day of rainfall, and all salamanders that were spotted by the volunteers were measured using ruler by one of the trained team leaders, weighed, and photographed. The surveyors additionally inspected the potential breeding sites for the presence of both adult salamanders and salamander larvae during each survey night. The source of our second dataset is a survey we conducted of two B. variabilis populations around the city of Jerusalem, following the same protocol as the salamander surveys, though without the assistance of volunteers. The focal site was the Gazelle Valley urban nature park within the city of Jerusalem. Like the Haifa zoo/wadi Lotem salamander population, the toads of Gazelle Valley had originally bred in the small, ephemeral water bodies along the two streams passing through the site. In 2015, an urban nature park was established in Gazelle Valley, and several artificial ponds, both ephemeral and permanent, were established. As in Haifa zoo, the presence of invasive predatory fish in these ponds, as well as a large population of the Levant water frog ( Pelophylax bedriagae ) which arrived at the site since the introduction of permanent water bodies, are feared to turn these ponds into potential ecological traps. As a control, we performed a survey following the same protocol at a site in which existed, according to our preliminary survey, a large population of B. variabilis : the Ein Hemmed national park, 8 kilometers northwest of Gazelle Valley. The presence/absence of adults and tadpoles of B. variabilis was documented in six additional sites around Jerusalem. 2.1 S. infraimmaculata 2.1.1 Salamander citizen science project Between February 2016 and February 2022, the salamander monitoring project’s volunteers performed 89 surveys, 28 in Wadi Ahuza and 61 in Wadi Lotem. Surveys were performed during rainfall or within a day of its occurrence. The volunteers captured altogether 212 visually distinct individuals, 162 in Wadi Ahuza and 50 in Wadi Lotem. The mean number of individual salamanders encountered per survey was 14.96 ±10.60 SD (473 capture events in total) in Wadi Ahuza and 3.87 ±3.22 SD (294 capture events in total) in Wadi Lotem, with the difference in frequencies being statistically significant for α = 0.05 (Welch Two Sample t-test: t = -5.41, df = 29.65, p << 0.001). The mean length of salamanders captured in the Lotem population (M = 24.98, SD = 3.44) was significantly larger than the length (M = 22.36, SD = 3.25) of salamanders captured in the Ahuza population (Welch Two Sample t-test: t = -4.4924, df = 63.299, p-value = 3.055e-05). The lengths of males (M=23.24, SD=2.50) and females (M=23.72, SD=2.21) did not differ significantly (Welch Two Sample t-test: t = -1.3645, df = 156.01, p-value = 0.1744). The proportion of sighted females in the Lotem population (0.26, n = 47) and in the Ahuza population (0.36, n Count Data: p-value = 0.2238). In Wadi Ahuza, salamander tadpoles were sighted in both the ephemeral and permanent water bodies along the stream during all survey years. In Wadi Lotem / Haifa zoo, during the sampling season of 2016-2021, tadpoles were only found in one water body, a small artificial pond within the zoo. 2.1.2 Movement and activity patterns To understand the salamanders’ habitat-use and site fidelity, and to identify possible dispersal events, we explored the salamander’s movement patterns by calculating the distances between all capture points for 68 S. infraimmaculata individuals for which the exact capture location (within 1-\RL20 meters) was specified in more than one capture. Of the individuals for which more than two exact locations were documented, 85% were found in the exact same location (within 1-\RL20 meters) on a least two different occasions, 54% were found in the same location three times, and 19% were found in the same location more than five times. 37% of these individuals were found in the same exact location in more than one winter. All but five salamanders were captured while in terrestrial activity. 64.8% of documented capture locations were within 100 meters of a known breeding site. Additionally, we calculated the maximal displacement for each of these individuals using the maximal distance between two of its capture locations (see figure 1). We conducted a two-way ANOVA to test the effect of sex and population of origin on an individual salamander’s maximal documented displacement, with no statistically significant effects found for population (F(1) =0.12, p =0.73) and sex (F(1) =0.24, p =0.62) and no significant interaction found between their effects (F(1) =0.01, p =0.93). A Figure 1 – S. infraimmaculata movement patterns- A: A histogram of maximal documented displacement [m] of individuals with two or \RL more than documented exact locations (n=68). B: Exact capture locations of 12 individuals from the Wadi Ahuza population. Points of the same color mark repeated captures of the same individual. Consecutive capture points of each individual are connected by lines. The individuals displayed were those with the four highest and four lowest maximal documented displacement values in the Wadi Ahuza population. Some points have been slightly shifted to enhance their visibility. 2.1.3 Growth rate and age estimation The age structure of a population can be crucial to our understating of its viability. Previous studies have described the growth of tadpoles or juveniles of this species and adults of other species in the same genus as an exponential function (Najbar et al., 2020; Warburg et al., 1979). However, no reliable method for estimating the age of adult salamanders of the species S. infraimmaculata is currently available. While we did directly examine the differences in length distribution between populations and over time (see figure 2C), not all of the variance in length is necessarily related to the age structure. To address this issue, we derived a rough estimate based on our own data. The growth rate of S. infraimmaculata is nonlinear, and we therefore used the repeated measurements of 84 individual salamanders over multiple years to quantify the link between their length and their estimated growth rate. We used it to develop a formula for estimating the age of S. infraimmaculata individuals using their total length. Similarly to Warburg (2008) we performed a linear regression (see figure 2A) between the salamanders’ growth rate ( [length at year x+1]- [length at year x] ) and length from snout to tail tip ( ([length at year x+1] + [length at year x])/2 ) on all individuals which were measured in two consecutive years. For individuals that were measured in 2 consecutive seasons more than once, we used the earliest measurement, to account for the scarcity of measurements of shorter (and therefore, younger) individuals in the data. Measurements of two individuals indicated that they shrunk by 2.5 cm between sampling seasons. Based on existing knowledge of the species (Warburg, 2008) these were marked as errors that were likely to be caused by tail truncation due to injurie, and therefore excluded from further analysis. The fitted regression model was: \begin{equation} \left(1\right)\ \ growth\ rate=\ -0.345*\left(\text{length}\right)+9.087\nonumber \\ \end{equation} If we mark the length of a salamander at the age t as\(L_{\left(t\right)}\), we can express the linear model as: \begin{equation} \left(2\right)\text{\ \ }\frac{d}{\text{dt}}L_{\left(t\right)}=\ -0.345*L_{\left(t\right)}+9.087\nonumber \\ \end{equation} The overall regression was statistically significant (R 2 =0.44, F (1,82) =66.9, p = 3.081e-12, see Fig. 2A). we used the estimated slope and intercept to solve equation 2 (see Methods) and acquire the growth function, \(L_{\left(t\right)}\), with the growth coefficient (k) set at -0.345 and the asymptotic length (A) set at 26.373, using the length upon completion of metamorphosis from Goldberg et al. (2012) as\(L_{\left(0\right)}\): \begin{equation} \left(3\right)\ L_{\left(t\right)}=-20.703*e^{-0.345t}+26.373\nonumber \\ \end{equation} We used permutation tests to determine if these parameters’ estimations differed significantly between sexes and populations. Between males and females, the differences in both the growth coefficient (10,000 iterations, ∆k =0.028, p = 0.77) and the asymptotic lengths (10,000 iterations, ∆A = 0.97, p = 0.38) were non-significant. Between the Ahuza and Lotem populations the differences in both the growth coefficient (10,000 iterations, ∆k =0.060, p = 0.51) and the asymptotic lengths (10,000 iterations, ∆A = 0.336, p = 0.74) were non-significant as well. The 95% CIs for parameters estimated from the data were estimated using bootstrap and were corrected for bias (10,000 iterations, k :(estimate = -0.345, upper = -0.268, lower = -0.414), A: (estimate = 26.373, upper = 27.433, lower =25.645)). The CI for the starting length was derived by simulating a normal distribution using a mean (=5.67) and SD (=1.04) from existing literature (Goldberg et al., 2012). We then used the reversed function for \(L_{\left(t\right)},\)\(L_{\left(\text{length}\right)}^{-1}\) , as an estimate of a salamander’s age based on its total length. However, the function \(L_{\left(t\right)}\) approaches an asymptote, from a certain point \(L_{\left(\text{length}\right)}^{-1}\) becomes extremely sensitive to small differences in length, producing unreliable results. We set the upper limit of detectable growth, based on the resolution of the length measurements in the survey, as \(\frac{d}{\text{dt}}L_{\left(t\right)}<0.5\frac{\text{cm}}{\text{year}}\) . We, therefore, determined the upper limit of estimable ages, using the derived parameters, to be 7.35 years, with the respective length being 24.92 cm. Salamanders longer than that were given the age of 7.35 years but excluded from further analysis regarding the age composition of the two populations since they are irrelevant for comparing the recruitment of juveniles during the survey years. We performed a two-way ANOVA on the estimated ages with respect to both the population and survey year, both the population’s effect (p<0.01) and the survey year’s effect(p<0.01) were found to be significant, though their interaction was not (p = 0.143). We then performed Tukey’s HSD test, and a significant difference was found within the Ahuza population (see figure 2D), with the mean age in the year 2020 higher than in the year 2017 (difference =1.27, adjusted p-value =0.02). Three other significant differences were found between Ahuza and Lotem subgroups. Figure 2 – S. infraimmaculata growth rate and aging A : Individual salamanders’ mean length over two consecutive years (x-axis) and the difference in mean length between the two years (y-axis). The points show the raw data for all salamanders that were measured in two consecutive years (n = 68). The trend line and area around the curve represent the fitted linear regression (Growth rate~ Length) and its 95% CI. B : the growth function L(t) which was derived from the linear regression. The curve represents the estimated mean length as a function of the salamander’s age and the area around the curve represents a bootstrap 95% CI for that mean. The colored points show the data for all salamanders that were measured more than once (n = 106), with each salamander’s initial age estimated by its length upon its first measurement. The ages of salamanders that were longer than the upper limit we set for age estimation (24.92 cm) were adjusted to show the likely age distribution in the population more accurately, using the mean adult mortality /year estimated in our capture-recapture analysis (0.177) (see text). Points of the same color mark repeated measurements of the same individual. The consecutive points of each individual are connected by lines. C : Scatterplot of the lengths of all measured individuals in the two studied salamander populations in the years 2017-2020. D : Scatterplot of the estimated ages of all measured salamanders that were short enough for age estimation (length < 24.92 cm). The mean age of the individuals in the Wadi Ahuza population in 2017 was significantly lower than the age of the individuals of the same population in 2020 (Tukey’s honest significance test, see text). 2.1.4 Capture-Recapture analysis To estimate demographic parameters such as survival rate \RL and population size, we performed a capture-recapture analysis using the ”robust design – Huggins p and c ” model structure in the program MARK (White & Burnham, 1999). We constructed 28 a priori capture-recapture population models, with different combinations of length, sex, air temperature, relative humidity, sampling season, and sampling occasion as predictor variables. Of these models, the data conclusively supported (total AICc weight > 0.99) models that included an effect of length on both capture probability (ß = 0.104449879, 95% CI = 0.060749107 (lower) - 0.148150723 (upper), LOGIT link function parameter) and survivorship between consecutive years \RL (ß = 0.0565, 95% CI = 0.040207031 (lower) - 0.072509265 (upper), LOGIT link function parameter), and both of these effects were found to be statistically significant (p ß=0 < 0.05) in all of the supported models. The effects of air temperature and relative humidity were both included in all the supported models and were both found to be statistically significant (p ß=0 < 0.05). The average survival rate between consecutive years weighted by each model’s AICc weight was 0.775 (95% CI = 0.699 (lower) - 0.837(upper)) for the Wadi Ahuza population and 0.766 (95% CI = 0.683 (lower) - 0.832(upper)) for the Wadi Lotem population. The estimated population size for each sampling season did not change significantly over time in either population (figure 3A). Though females were underrepresented (19% of observations were females, 70% \RL males, and the rest were of unknown sex or juvenile). all models containing an effect of sex on capture probability failed to produce a ß sex parameter estimate, possibly due to the sample size for females being too small. These models were therefore excluded from the analysis, but this bias (which possibly stems from inherent bias in the capture method caused by the males’ conspicuous behavior during the mating season, for example: Hall, 1977; Marsh & Goicochea, 2003; Marvin, 1996; Salvidio, 2008) may have resulted in an underestimation of the population size by our analysis. If we were to assume that this underestimation is proportional to the sex bias among our captured individuals, and that there are as many females in the population as males, the corrected population sizes would be larger by approximately 50%. The population size was additionally estimated based on the non-parametric approach described in (Chao et al., 1992). See all population size estimates in appendix table 1. Figure 3: Parameters estimated using Capture-Recapture analysis . Estimated sizes of the two studied S. infraimmaculata populations from 2017 to 2020 (A) and of the two studied B. variabilis populations in 2020 (B). \RL The error bars represent 95% CIs. C: the link function between Length and the probability of survival between consecutive years estimated using the Capture-Recapture analysis (LOGIT (-0.0565 * L) = S), the area around the curve represents 95% CI. D: Survival rates of salamanders along their lifetime simulated (10,000 iterations) using the derive age-to-length and length-to-survival rate function parameters and their respective CIs. The area around the curve represents the range between the 5% and 95% quantiles of the simulation results. 2.1.5 The effect of age on the survival rate of S. infraimmaculata Using the estimate and the CI for the effect of length on the survival rate of individual salamanders that we derived from the Capture-Recapture analysis (ß = 0.0565), we plotted the link between length and survival rate (figure \RL3C). We then simulated the survival rate for each age group by randomly sampling values from within the 95% confidence intervals of the estimated age-to-length and length-to-survival rate function parameters (10,000 iterations, see figure 3 D). 2.2 B. variabilis 2.2.1 Toad survey We conducted 21 nocturnal surveys overall in search of B. variabilis, twelve in Gazelle Valley urban nature park and nine in Ein Hemmed national park, between August 2020 and May 2021. In Gazelle Valley and Ein Hemmed, we captured 18 and 79 individuals respectively, and all individuals had distinct enough patterns to allow identification upon recapture except three individuals captured in Ein Hemmed. Over 76% of the individuals captured in Ein Hemmed were males caught in or nearby a water body, displaying characteristic courting behaviors (floating at the water’s surface, emitting mating calls). No individuals captured in Gazelle Valley were inside a water body. Two individuals emitting mating calls inside a water body were observed during the survey but not captured. B. variabilis tadpoles were observed in Ein Hemmed in three surveys, in multiple water bodies inside the park, numbering in the thousands on each occasion. In Gazelle Valley, only one group of 435 B. variabilis tadpoles was observed on one occasion (manually counted from a photograph), but no tadpoles or metamorphs were observed in the same water body in subsequent surveys. We observed 16 metamorphs in Ein Hemmed. The discrepancy between the many thousands of tadpoles that were observed and the relatively small number of metamorphs can be possibly accounted for by the metamorphs’ tendency to spend most of their time in hiding in the weeks following the metamorphosis (Elron, 2007). No metamorphs were observed in Gazelle Valley. We could not detect metamorph recaptures due to the lack of a reliably stable and distinct dorsal pattern at this age. The mean number of individual toads encountered per hour of surveying was 10.56 ±9.71 SD in Ein Hemmed and 1.69 ±2.96 SD in Gazelle Valley, with the difference in frequencies being statistically significant for α = 0.05 (Two-Tailed t-test Two-Sample Assuming Unequal Variances: t = 2.50, df = 8, p = 0.037). 2.2.2 Capture-Recapture analysis From the six models developed for the Ein Hemmed population, the two models supported by the data (total AICc weight > 0.99) included an effect of survey duration on capture probability, which was statically significant in both models (p ß=0 0.05). The average estimate for the Ein Hemmed population size, weighted by the AICc, was 395 (95% CI = 125 (lower) - 666(upper)) (figure 3B). Female \RL captures were rare in the Ein Hemmed survey (14% of observations), and the estimated parameter for the effect of sex on capture probability was too small to affect the population size estimation, leading to a possibility of an underestimation of the population size by our model. If we were to assume that this underestimation is proportional to the sex bias in our captures, namely that there are as many females as males, the corrected population size for Ein Hemmed would be approximately 50% larger (~600 individuals). Such a scenario is reasonable, considering the behavioral differences between males and females during the breeding season: males are found at the breeding site for long periods of time – both along the season and during each specific night – while females usually arrive at the breeding site, lay eggs, and leave. For the Gazelle Valley population, the estimated population size was 40 (95% CI = 23 (lower) – 106 (upper)) (figure 3B). Due to the small sample size (21 total observations, 3 individuals captured more than once) we did not analyze the effect of environmental and individual covariates on capture probability in this population. 3. Discussion In this section, we discuss the conclusions from our analysis, and their possible implications for the conservation of amphibian species in urban environments and outside of them. We further discuss the potential that our results show for exploring phenomena and dynamics that were previously out of reach by using intensive, non-invasive citizen science surveys. Finally, we propose a framework for future research programs, leveraging citizen science’s unique advantages to inform conservation efforts. One useful result of our analysis is population size estimates. While the sizes of both S. infraimmaculata populations were stable over time, the Wadi Lotem population was consistently smaller than the Wadi Ahuza population. The population size in Wadi Lotem is similar to other nearby populations which breed exclusively in ephemeral ponds, despite the presence of permanent water bodies at the site, opposite to the trend described in Segev et al. (2010) . Some attributes of these water bodies, such as the presence of predatory fish, possibly prevents them from being suitable breeding sites. Many of our results agreed with previous studies on the salamanders of the Carmel mountain range, including population sizes, maximal displacements, site fidelity, and the decrease in growth rate over time (Bar-David et al., 2007; Segev et al., 2010; Sinai et al., 2020; Warburg, 2008). This agreement suggests that this citizen science project is similar in reliability and precision to traditional surveys conducted by trained researchers, all while adding the educational and societal advantages offered by citizen science. Warburg (2008) studied the link between length and growth rate of S. infraimmaculata in the Carmel for decades. With a similar methodology, we managed to replicate Warburg’s results regarding the changes in the growth rates of these salamanders over the course of their lives, with most of the salamanders’ growth occurring in their first eight years of life. Warburg determined that the degree ofvariance does not allow for reliable length-based age estimation in this species. Similarly, we avoid drawing conclusions from our length-based age estimation regarding specific individuals, or from the differences in length between the two studied populations. However, we assert that in a specific population, a change in the length distribution over time, is a reliable measure for changes in juvenile recruitment. This type of trend, which we found in the Wadi Ahuza population, could be an informative tool for research and conservation. Used cautiously, the age estimate produced by our formula can be informative on its own for populations of a similar genetic makeup and habitat, such as other salamander population in the Carmel. Females were relatively scarce in our data. The underrepresentation of females was observed in other amphibian studies (Wainwright et al., 2023) and may stem from amphibian males’ conspicuous behavior during the breeding season (Green, 2013). However, the fact that the vast majority of the individuals we observed were male is a possible indication of an additional threat: males of the genus Salamandra are often found conspicuously standing in a specific display posture during the breeding season, often in relatively open spaces (Manenti et al., 2017), and as a result males of this genus spend much more time than females on roads (Vincenz & Reyer, 2005). Our surveys were conducted mainly along broad trails and on dirt roads, and this behavioral difference is thus probably the reason we detected far more males than females. However, this result leads us to the realization that roads near breeding sites may accordingly constitute an ecological trap, affecting males in particular, above and beyond the appreciated risk of roadkill that salamanders face when crossing roads during the breeding season. This possibility is supported by a previous study on the effect of roadkill on S, infraimmaculata in Israel, which showed that males are significantly more likely to be found dead by the roadside (Sinai et al., 2020B). We quantified the underrepresentation of juvenile salamanders in our survey, a result which may be applicable in future studies, improving our approximations of the age structures of S. infraimmaculata populations. The underrepresentation of juveniles could have been caused, among other factors, by the juveniles being smaller and more cryptic than the adults. This is in sharp contrast with the adults’ breeding and courting behaviors which make them easier to detect. This underrepresentation of juveniles may have made it difficult to detect the rate of juvenile recruitment in the statistical analysis via MARK. However, the sizes of both populations were stable during this period, while the estimated mortality was over 10%, suggesting that recruitment is occurring to some extent, but can also indicate an error in our assessment. Our analysis shows that S. infraimmaculata do not reach their peak survivorship until the age of eight years, when most of their growth is completed. Our estimates suggest that only ~8% of individuals that completed metamorphosis would reach that stage of life. Relatively low survivorship during their first years of life might mean that a loss of adult salamanders cannot be compensated for by juvenile recruitment without several consecutive years of successful breeding and terrestrial habitats stable enough to allow a sufficient fraction of the sensitive juvenile salamanders to complete their growth. A possible major implication for future conservation efforts is that adult, fully grown salamanders should be thought of as a vital resource for sustaining the population, that should be prioritized alongside the breeding sites themselves. However, due the underrepresentation of females and juveniles in our study, and the fact that we did not have any individuals with known age to with which to ground our estimates, our conclusions regrading life-history must be considered with caution. Similarly to the salamanders, the two toad populations differed in size; in both cases, the smaller population was found where the aquatic breeding sites were repeatedly disturbed and had a consistent presence of predatory fish. In contrast, the larger population of both species had access to multiple small, stable, and diverse breeding sites. The Ein Hemmed toad population was both the largest amphibian population in this study and the only population that showed clear signs of successful breeding. Not surprisingly, Ein Hemmed is the only site located within a protected area. Gazelle valley had the most diverse amphibian community, yet the most threatened species in it ( B. Variabilis ) had a small population that showed no signs of successful breeding. As we repeatedly emphasized in this work, the breeding site’s characteristics prove to be critical; a major concern regarding both Gazelle Valley and Wadi Lotem is that the fish-containing, non-ephemeral ponds function as ecological traps. If, as we suspect, a substantial proportion of the breeding efforts of those population is concentrated in these low-quality breeding sites, the resulting low recruitment could be highly detrimental to the continued existence of these populations. This damage to recruitment may have also contributed to the relative scarcity of juveniles at the two sites. It might be possible to effectively block the breeding adults from accessing these traps in some cases, but barring that, eradication of the invasive predators from the ponds is likely to be the only solution that would allow the toad and salamander populations to survive in the long run. Amphibians in the levant readily breed in small, ephemeral aquatic habitats. The presence of permanent ponds, which are often host to invasive predators and competitors can be detrimental for the local species. Particularly, the invasive Mosquitofish ( Gambusia affinis ), which was shown to prey on tadpoles of both S. infraimmaculata and B. variabilis in laboratory experiments (Elron, 2007; Segev et al., 2009), might be a leading cause of population collapse. The implication arising from these aspects of our study is that conservationists in this region should consider providing the target populations with multiple small and diverse breeding sites. This could minimize the chances that one large unsuitable pond will doom the entire population. The two species discussed here differ in their conservation status, and each species is sensitive to a different set of threats. While S. infraimmaculata is capable of breeding in ephemeral ponds, these habitats only support relatively small populations when compared with deeper permanent ponds (Sinai et al., 2020). The possible reliance of this species on permanent ponds might mean that providing multiple small ephemeral breeding sites will not be as effective for S, infraimmaculata as it may be for B. variabilis . The two species also differ in their sensitivity to fragmentation: S. infraimmaculata in Israel are heavily impacted by human disturbance, such as roadkill (Sinai et al., 2020B), and low connectivity might have a severe effect on the population size (Sinai et al., 2020). B. variabilis are more resilient and were even shown to be positively associated with urban areas in certain cases (Blank & Blaustein, 2012). This difference was also apparent in genetic analyses of the two species in Israel, which showed that S. infraimmaculata populations are more influenced by fragmentation than B. variabilis populations (Goldberg, 2013). One of the main challenges in identifying threats to amphibian populations is trapping enough individuals multiple times, a prerequisite for the study of movement patterns, population dynamics, and life history traits. The Haifa salamander monitoring project addressed this with a simple format. Here, a small group of trained researchers and a larger team of volunteer trackers formed an effective surveying unit that sampled its research sites consistently, thoroughly, and over a long period of time. A common criticism of surveys based on citizen science is that their resolution is often limited to presence/absence or to rough estimates of abundance data. The approach developed in the Haifa salamander monitoring project overcame these challenges and managed to provide details on the population structure and life history, which are paramount to making informed management decisions. Together with our green toads’ survey, our analysis highlights that species’ abundance is not a sufficient measure of a site’s suitability for sustaining amphibian populations, and that additional measures may be crucial for informed management. Finally, we suggest an outline for future monitoring of urban amphibians, based on these projects and the conclusions stemming from them. In the salamander survey, each individual animal was manually measured. The measurements were conducted by several operators. Though each of the operators was specifically trained for safe handling and precise measurement, this was still a likely source of measurement error. Another consequence of this measurement method was inconsistency, mainly in the lack of a snout-to-vent length data for many captures. Due to this inconsistency, we based our age estimates on the total length, which is less suitable for age estimation in salamanders (Lunghi, 2022). Additionally, physically Handling any wild animal has a potential of causing it distress and even lasting damage, which is of a particular concern in citizen science projects. In the toad survey, both recognition of recaptures and length measurements were achieved using pictures alone. Since salamander SVL can also be measured reliably using photography (Lunghi et al., 2020), we assert that there may be no need for capturing and handling animals for such capture-mark-recapture surveys. A possible promising venue for future citizen science programs may be to outline specific survey routes at multiple sites and allow volunteers to survey the sites that are in their immediate vicinity independently. The volunteers can be instructed regarding how to photograph the amphibians alongside standard scale cards while causing minimal interference in the animals’ activities, allowing them to then survey the adults and tadpoles along specific routes on a regular basis. By continuously recruiting volunteers to regularly survey many sites, it may be possible to gather the crucial data conservationists will require to effectively protect the amphibians of the Mediterranean. Salamandra infraimmaculata Citizen science project The Haifa Salamander Monitoring Project was initiated and led by local volunteers, collaborating with several local organizations dealing with conservation, research, and environmentalism. An initial group of volunteers with research and conservation backgrounds designed the survey protocol and routes and recruited a larger team of volunteers, trained them to track and safely capture the salamanders, and supervised the teams performing the surveys. Each team of volunteers included a team leader from the core group of volunteers who was specifically trained in safely handling the animals, performing the required measurements and determining the animals’ sex. The practical training was conducted in the field by the head of the salamander project (O. Rybak) and only upon its completion a team leader would be allowed to handle salamanders independently. The rest of the volunteers were required to go through an orientation session with the head of the project before taking part in the surveys. The volunteers walked in teams of 3-4 along a constant route in each survey site, always at nighttime, scanned the path using flashlights, and documented every salamander they spotted along their path. They measured its length (snout to tip of tail, and occasionally the snout-to-vent length was measured additionally), weighed it, photographed its unique dorsal spot pattern for later individual identification, and documented the capture location. All of the measurements above were performed by a specific team member with the required training. The volunteers additionally documented the temperature and relative humidity during each survey. The data includes the observations from the main survey (Wadi Lotem: 2016-2022, Wadi Ahuza: 2017-2021), casual observations of salamanders in the city of Haifa during the same years, and observations of salamanders captured from the Wadi Lotem population by members of the Haifa Zoo staff during the years 2007-2009. The volunteers additionally noted possible hazards to the salamanders such as feral cats, discarded trash, signs of water contamination, or other possible threats the were spotted during the survey, as well as the presence of salamander larvae or other amphibian species in the breeding sites. In addition to the engagement of the project participants, the project included additional efforts to inform and engage the community, including posting signs in the breeding sites, free lectures about the salamanders’ biology, social media activity, and the operation of information booths around the city of Haifa. Interviews and questionnaire of the project volunteers were not conducted for this study. The surveys and all handling of animals were conducted under the Israel Nature and Parks Authority (INPA) permits numbers 2018/42061, 2019/42368, and 2020/42648, only researchers and volunteers who were specifically listed in these permits and received the training described above handled the studied animals, and all methods were carried out in accordance with relevant guidelines and regulations. No animals were harmed in this study, and none were held in captivity. All handling in the field (photography, weighting, and measuring of salamander length) followed the guidelines required by the ethical committee of the Hebrew University of Jerusalem. Study site The S. infraimmaculata survey took place within the Carmel Mountain range, which is the southernmost edge the global distribution of S. infraimmaculata . The Carmel was originally covered by Mediterranean woodland, with numerous caverns and springs as suitable habitats, shelters, and breeding sites for salamanders of this species. The urbanization of the region which occurred gradually since the mid-20 th century altered and disturbed many of these habitats, and its effects on the salamander population are not fully understood (Degani et al., 2019). The habitats in the study sites were both altered by their proximity to the city of Haifa, including wildfire, water pollution, water body alterations, discarded waste, and the presence of feral cats and dogs. Concerns regarding the state of the resident salamander populations motivated the Haifa Salamander Monitoring Project. Data analysis We manually recognized recaptures of individuals from the pictures taken during the surveys using their unique dorsal spot patterns. The data from the last salamander sampling season in Wadi Ahuza (winter of 2021-2022) was not yet processed at the time of this analysis and was therefore excluded from it., In comparisons we conducted between the two population or between males and females which were not directly concerned with growth rate, we averaged multiple measurements of the same individuals when more than a single measurement of an individual was available (unless otherwise stated). The data was analyzed in the Rstudio platform (Allaire, 2012) using the packages toolsForAtlas, ggplot2 (Wickham et al., 2016), Dplyr (Mailund, 2019),tidyverse (Wickham & Wickham, 2017) ggpubr (Kassambara & Kassambara, 2020), Boot (Canty, 2002), leaflet (Cheng et al., 2019), and MASS (Ripley et al., 2013). Growth rate and age estimation To determine the salamanders’ ages, we sought to derive a function that describes their growth throughout their lives (will be marked as L (t) , noting the snout to tip-of-tail length as a function of time). In this analysis, we could not use direct regression methods since we do not have any data about any of the salamanders’ ages. However, the L (t) function derived for similar species (Najbar et al., 2020) and for larvea of salamanders of this species (Warburg et al., 1979) were exponential. Therefore, if we assume the L (t) function for the adults of this species is exponential as well, we can assert that a linear link exists between a salamander’s growth rate (\(\frac{d}{\text{dt}}L_{\left(t\right)}\)) and its length (L (t) ), as is generally true for simple exponential functions . The differential equation representing this link is shown below:: \begin{equation} \frac{d}{\text{dt}}L_{\left(t\right)}=\ a*L_{\left(t\right)}+b\nonumber \\ \end{equation} Since we had multiple observations of salamanders from consecutive years, a linear regression between annual growth ( [length at year x+1]- [length at year x] ) and the average length during the year of growth length ( ([length at year x+1] + [length at year x])/2 ) could be performed to derive a and b. The solution for the differential equation is: \begin{equation} L_{\left(t\right)}=Ce^{\text{at}}-\frac{b}{a}\nonumber \\ \end{equation} Where a is the growth constant (k), - \(\ \frac{b}{a}\) is the asymptotic length and C is the difference between the asymptotic length and the initial length. The initial length, meaning the mean length of salamanders when they complete their metamorphosis, was the only parameter which we did not derive directly from our data, and was instead taken from a previous study on this species in northern Israel(Goldberg et al., 2012). From the same set of parameters, a reverse growth function- length to age – could be derived : \begin{equation} L_{\left(\text{length}\right)}^{-1}=\frac{\ln\left(\frac{length+\frac{b}{a}}{C}\ \right)}{a}\ \nonumber \\ \end{equation} and used to convert the salamanders’ measured lengths to age estimations. However, since \(L_{\left(t\right)}\) approaches an asymptote, from a certain point \(L_{\left(\text{length}\right)}^{-1}\) becomes extremely sensitive to small differences in length, producing unreliable results. We determined the upper limits of the reliability of \(L_{\left(\text{length}\right)}^{-1}\ \) as the point at which the yearly growth rate - \(\frac{d}{\text{dt}}L_{\left(t\right)}\)- is 0.5 cm/year since 0.5 cm was the measurement resolution in the survey. Using 67 cases in which the same individual was measured twice within the same month we additionally calculated the mean measurement error, which was on the same scale as the measurement resolution (±0.66 cm, SD = 0.67). Capture-Recapture analysis For the Capture-Recapture analysis, we excluded all observations that did not occur during the standard-effort surveys, and all individuals that did not have a distinct enough pattern for identification upon recapture. For surveys in which air temperature, relative humidity, or survey duration was missing, the average for all other surveys was used. Since our planned model structure for the S. infraimmaculata dataset required the individual covariates (sex, length) will be listed for each sampling season for all individuals, and not all individuals were captured and measured in all seasons, the missing lengths were estimated using the growth function we derived from the data. All datasets were analyzed using the program MARK (White & Burnham, 1999). For the S. infraimmaculata data, we constructed 28 a priori population models, with different combinations of length, sex (individual covariates), air temperature, relative humidity (environmental covariates), sampling season, and sampling occasion (time dependency) as predictor variables the probabilities for individual being captured in a specific occasion, surviving from one sampling season to the next, immigrating between sampling seasons, and emigrating between sampling seasons. The models were implemented using the ”robust design – Huggins p and c ” (under the assumption that captures do not alter the salamander’s behavior in a way that affects future captures: p = c ) model structure available in MARK. To display individuals that were longer than the maximal length allowing age estimation (24.92 cm) on the growth curve in fig. 2B, we used the estimated mean survival rate for fully-grown adults (0.823) to simulate the age distribution of the subpopulation of salamanders longer than 24.92 cm (under the assumption that the recruitment is constant for this subpopulation) and used this distribution to produce a rough estimation of their age, based on the rank order of their lengths. Bufotes variabilis survey The study sites of the B. variabilis survey were the Gazelle Valley Urban Nature Park inside the city of Jerusalem and the Ein Hemmed National park in the adjacent Judean Hills region. Gazelle Valley was a highly disturbed habitat for several years, with large amounts of waste inside and outside of the water bodies. The urban nature park on the site was established in 2015, and vast rehabilitation efforts were devoted to the aquatic habitats, including a water suspension and circulation system which added permanent ponds of differing sizes to the preexisting small ephemeral water bodies. However, it is possible that the ambient pollution present in the urban runoff water which flow into those water bodies, as well as the presence of the invasive predatory fish species Gambusia affinis , does not allow some amphibian species to breed in them, to a degree they may constitute ecological traps. Our survey sought to assess the influence of these changes on the preexisting amphibian population, by determining the population size and composition and comparing them to those of a control population. The chosen control population was the toad population of Ein Hemmed National park, located near a suburban area of the Judean Hills, in which a similar water suspension and circulation system have existed for over a decade, and in which large numbers of B. variabilis were observed in our preliminary survey. MARK output does not include exact p-values for the estimated effects of environmental and individual covariates on survival and capture probability, and instead shows 95% CI for the parameter value. Where discussing the significance of different estimated effects in this analysis, we note cases in which the 95% CI of the parameter does include 0 as p ß=0 < 0.05 since the two are equivalent. All CIs as a part of the full model outputs (see appendix table 2). Field survey We performed a series of nighttime surveys in Gazelle Valley urban nature park and in Ein Hemmed national park in order to assess the population size and composition of B. variabilis in both sites, while documenting their breeding activity, possible hazards, and the presence of other amphibian species. We recorded all amphibian observations and their location but only members of the species Bufotes variabilis and Hyla savignyi were photographed alongside a scale bar, and their snout-to-vent lengths were measured using the scaled photographs. We included in the route the approachable shorelines of all known water bodies in both sites and recorded the presence of tadpoles and eggs, as well as breeding activity by adults of the surveyed species. We performed the surveys along a constant route, during the evening hours, and documented the air temperature, relative humidity, survey start time, and duration for each survey. No more than two people took part in each survey. We additionally performed two daytime surveys in the water bodies of Gazelle Valley Urban Nature Park and recorded the amounts of amphibian tadpoles, eggs, and other marine organisms acquired after 10 net swipes, in three different depth categories . For the B. variabilis data from Ein Hemmed National park, we constructed 6 a priori models with combinations of survey duration, sampling occasion, sex, and length as predictor variables of the probability of an individual being captured on a specific occasion. The models were implemented using the ”Closed capture - Huggins p and c ” (also assuming p = c ) available in MARK. The B. variabilis data from Gazelle Valley Urban Nature Park was analyzed using the ” Closed capture - full-likelihood” model structure available in MARK. Author contributions O.D. carried out the Bufotes variabilis surveys, conducted the analysis, and wrote the manuscript. O.K and O.D. planned the study. O.K. supervised the research and commented on the manuscript. O.R, A.B.L, G.K and T.K.P planned and executed the Salamandra infraimmaculata surveys and commented on the manuscript. N.L.A and R.V accompanied the Salamandra infraimmaculata surveys on behalf of the SPNI and commented on the manuscript. Acknowledgements We first and foremost thank the Haifa Salamander survey volunteers who, with great dedication, carried out the salamander surveys in the coldest and rainiest nights of winter. We thank Liran Sagi, for his valuable guidance and assistance with the MARK analysis, All members of the Kolodny lab for their help and input, and Assaf Shwartz, Avi Bar Massada, Rachel (Ekly) Ben Shlomo, and Amir Balaban for extensive discussions that inspired this study. We thank the Israel Nature and Parks Authority, Society for Protection of Nature in Israel, ”Yarok Ba’Lev” NGO, the Haifa Educational Zoo, the Municipality of Haifa, and Gazelle Valley Urban Nature Park for their essential cooperation and assistance in carrying out the surveys and the management of the data from the Haifa Salamander survey, and to the Haifa community of the SPNI in particular for their help and support of the Haifa Salamander project over the years. We thank Yael Hammerman, Avishai Shoresh, Noemi Cohen Kappach, Chen Chaim, for their help and support in carrying out the toad surveys, analyzing preliminary datasets, and involvement in various phases of the citizen science project. Conflict of Interest (required for all article types) The authors have no conflict of interest to disclose. Data availability statement All summary statistics of the data, intermediate analyses, and full raw data, are available in the supplementary material. Funding This study was supported by the Israeli Science Foundation (ISF; 1826/20), the United States – Israel Binational Science Foundation (BSF), and the Minerva Center for Population Fragmentation. Bibliography Allaire, J. (2012). RStudio: integrated development environment for R. Boston, MA , 770 (394), 165–171. Bar-David, S., Segev, O., Peleg, N., Hill, N., Templeton, A. R., Schultz, C. B., & Blaustein, L. (2007). Long-distance movements by fire saLamanders (Salamandra infraimmaculata) and impLications for habitat fragmentation. Israel Journal of Ecology and Evolution , 53. Beninde, J., Veith, M., & Hochkirch, A. (2015). Biodiversity in cities needs space: a meta‐analysis of factors determining intra‐urban biodiversity variation. Ecology Letters , 18 (6), 581–592. Blank, L., & Blaustein, L. (2012). Using ecological niche modeling to predict the distributions of two endangered amphibian species in aquatic breeding sites. Hydrobiologia , 693 (1), 157–167. Bodenheimer, F. S. (1935). Animal life in Palestine . Burgstaller, S., Gollmann, G., & Landler, L. (2020). The green toad example: a comparison of pattern recognition software . Cahill, A. E., Aiello-Lammens, M. E., Fisher-Reid, M. C., Hua, X., Karanewsky, C. J., Yeong Ryu, H., Sbeglia, G. C., Spagnolo, F., Waldron, J. B., & Warsi, O. (2013). How does climate change cause extinction? Proceedings of the Royal Society B: Biological Sciences , 280 (1750), 20121890. Campbell, Alastair., Australia. Environment Australia. Biodiversity Group., & Natural Heritage Trust (Australia). (1999). Declines and disappearances of Australian frogs . Biodiversity Group, Environment Australia. Canty, A. J. (2002). Resampling methods in R: the boot package. The Newsletter of the R Project Volume , 2 (3), 2–7. Chao, A., Lee, S. M., & Jeng, S. L. (1992). Estimating population size for capture-recapture data when capture probabilities vary by time and individual animal. Biometrics , 201–216. Cheng, J., Karambelkar, B., Xie, Y., Wickham, H., Russell, K., Johnson, K., Schloerke, B., & Agafonkin, V. (2019). Package ‘leaflet.’ Clevenot, L., Carré, C., & Pech, P. (2018). A Review of the factors that determine whether stormwater ponds are ecological traps and/or high-quality breeding sites for amphibians. Frontiers in Ecology and Evolution , 6. Colléony, A., Cohen-Seffer, R., & Shwartz, A. (2020). Unpacking the causes and consequences of the extinction of experience. Biological Conservation , 251 , 108788. Colléony, A., & Shwartz, A. (2019). Beyond assuming co-benefits in nature-based solutions: A human-centered approach to optimize social and ecological outcomes for advancing sustainable urban planning. Sustainability , 11 (18), 4924. Cushman, S. A. (2006). Effects of habitat loss and fragmentation on amphibians: A review and prospectus. Biological Conservation , 128 (2), 231–240. Degani, G., & Kaplan, D. (1999). Distribution of amphibian larvae in Israeli habitats with changeable water availability. Hydrobiologia , 405. Degani, G., Mendelssohn, H., Warburg, M. R., Shkolnik, A., Nevo, E., Goldberg, T., Pearlson, O., & Sharon, R. (2019). The Fire Salamandra (Salamandra infraimmaculata) and the Banded Newt (Triturus vittatus) along the southern border of their distribution . Scientific Research Publishing, Inc. USA. Degani, G., & Warburg, M. R. (1978). Population structure and seasonal activity of the adult Salamandra salamandra (L.)(Amphibia, Urodela, Salamandridae) in Israel. Journal of Herpetology , 437–444. Demeyrier, V., Lambrechts, M. M., Perret, P., & Grégoire, A. (2016). Experimental demonstration of an ecological trap for a wild bird in a human-transformed environment. Animal Behaviour , 118 , 181–190. Desa, U. N. (2018). Revision of world urbanization prospects. UN Department of Economic and Social Affairs , 16 . Dolev, A., & Perevolotsky, A. (2004). The red book: Vertebrates in Israel . Israel nature and parks authority. Elron, E. (2007). Amphibian population decline: The green toad (Bufo viridis) test case, biological and ecological processes [PhD thesis]. Tel Aviv University. Ferreira, M., & Beja, P. (2013). Mediterranean amphibians and the loss of temporary ponds: Are there alternative breeding habitats? Biological Conservation , 165 , 179–186. Givati, A., & Rosenfeld, D. (2013). The Arctic Oscillation, climate change and the effects on precipitation in Israel. Atmospheric Research , 132 , 114–124. Goldberg, T. (2013). Characteristics of the Habitats of Amphibian Species in Northern Israel and the Genetic Variability of Salamandra Infraimmaculata and Pseudepidalea Viridis Larvae [PhD thesis]. University of Haifa. Goldberg, T., Nevo, E., & Degani, G. (2012). Phenotypic plasticity in larval development of six amphibian species in stressful natural environments. Zoological Studies , 51 (3), 345–361. Green, D. M. (2013). Sex ratio and breeding population size in Fowler’s Toad, Anaxyrus (= Bufo) fowleri. Copeia, 2013(4), 647-652.\RL‏ Hall, R. J. (1977). A population analysis of two species of streamside salamanders, genus Desmognathus. Herpetologica , 109–113. Hamer, A. J., & McDonnell, M. J. (2008). Amphibian ecology and conservation in the urbanising world: A review. In Biological Conservation , 141(10), 2432–2449. Hamer, A. J., & Parris, K. M. (2011). Local and landscape determinants of amphibian communities in urban ponds. In Ecological Applications , 21( 2). Hampe, A., & Petit, R. J. (2005). Conserving biodiversity under climate change: the rear edge matters. Ecology Letters , 8 (5), 461–467. Strategic Plan for Housing (Hebrew), (2017). https://www.gov.il/he/departments/policies/2017_dec2457#divFiles Kassambara, A., & Kassambara, M. A. (2020). Package ‘ggpubr.’ R Package Version 0.1 , 6 . Kats, L. B., & Ferrer, R. P. (2003). Alien predators and amphibian declines: review of two decades of science and the transition to conservation. Diversity and Distributions , 9 (2), 99–110. Kay, C. A. M., Rohnke, A. T., Sander, H. A., Stankowich, T., Fidino, M., Murray, M. H., Lewis, J. S., Taves, I., Lehrer, E. W., Zellmer, A. J., Schell, C. J., & Magle, S. B. (2022). Barriers to building wildlife-inclusive cities: Insights from the deliberations of urban ecologists, urban planners and landscape designers. People and Nature , 4 (1), 62–70. Levin, N., Elron, E., & Gasith, A. (2009). Decline of wetland ecosystems in the coastal plain of Israel during the 20th century: implications for wetland conservation and management. Landscape and Urban Planning , 92 (3–4), 220–232. Lin, B. B., & Fuller, R. A. (2013). Sharing or sparing? How should we grow the world’s cities? Journal of Applied Ecology , 50 (5), 1161–1168. Lunghi, E. (2022). Doubling the lifespan of European plethodontid salamanders. Ecology, 103(2), e03581.\RL‏ Lunghi, E., Giachello, S., Manenti, R., Zhao, Y., Corti, C., Ficetola, G. F., & Bradley, J. G. (2020). The post hoc measurement as a safe and reliable method to age and size plethodontid salamanders. Ecology and Evolution, 10(20), 11111-11116.\RL‏ Mailund, T. (2019). Manipulating data frames: dplyr. In R Data Science Quick Reference 109–160. Manenti, R., Conti, A., & Pennati, R. (2017). Fire salamander (Salamandra salamandra) males’ activity during breeding season: Effects of microhabitat features and body size. Acta Herpetologica, 12(1), 29-36.\RL‏ Marsh, D. M., & Goicochea, M. A. (2003). Monitoring terrestrial salamanders: biases caused by intense sampling and choice of cover objects. Journal of Herpetology , 460–466. Marvin, G. A. (1996). Life history and population characteristics of the salamander Plethodon kentucki with a review of Plethodon life histories. American Midland Naturalist , 385–400. Meiri, S., Belmaker, A., Berkowic, D., Kazes, K., Maza, E., Bar-Oz, G., & Dor, R. (2019). A checklist of Israeli land vertebrates. Israel Journal of Ecology and Evolution , 65 (1–2), 43–70. Mendelssohn, H. (1985). Winter rain pools of Israel: neglected habitats in need of protection. Israel Land and Nature , 11 , 74–77. Miller, J. R. (2005). Biodiversity conservation and the extinction of experience. Trends in Ecology & Evolution , 20 (8), 430–434. Najbar, A., Konowalik, A., Halupka, K., Najbar, B., & Ogielska, M. (2020). Body size and life history traits of the fire salamander Salamandra salamandra from Poland. Amphibia-Reptilia , 41 (1), 63–74. Nevo, E., & Schneider, H. (1976). Mating call pattern of Green toads in Isreal and its ecological corelate. Journal of Zoology , 178 (1), 133–145. OECD. (2022). “Family Indicators.” Oertli, B., & Parris, K. M. (2019). Review: Toward management of urban ponds for freshwater biodiversity. Ecosphere , 10 (7). Palmer, A., Reynolds, S. J., Lane, J., Dickey, R., & Greenhough, B. (2021). Getting to grips with wildlife research by citizen scientists: What role for regulation? People and Nature , 3 (1). Papenfuss et al. (2008). Salamandra infraimmaculata. IUCN : Red List of Threatened Species. Parris, K. M. (2006). Urban amphibian assemblages as metacommunities. Journal of Animal Ecology , 75 (3). Pavol, P., & Fančovičová, J. (2012). Tolerance of amphibians in Slovakian people: A comparison of pond owners and non-owners. Anthrozoos , 25 (3), 277–288. Peter, M., Diekötter, T., & Kremer, K. (2019). Participant outcomes of biodiversity citizen science projects: A systematic literature review, Sustainability (Switzerland), 11(10). Pyle, R. M. (2002). Eden in a vacant lot: Special places, species, and kids in the neighborhood of life. Children and Nature: Psychological, Sociocultural, and Evolutionary Investigations , 305–327. Pyle, R. M. (2003). Nature matrix: reconnecting people and nature. Oryx , 37 (2), 206–214. Rabou, A.F.N.A., Yassin, M.M., Al-Agha, M.R., Hamad, D.M., Ali, A.K.S. (2007): The herpetofauna of the Gaza Strip with particular emphasis on the vicinity of Wadi Gaza. In The Islamic University Journal, 15(1), 111-135. Rae, M., Miró, A., Hall, J., O’Brien, K., & O’Brien, D. (2019). Evaluating the validity of a simple citizen science index for assessing the ecological status of urban drainage ponds. Ecological Indicators , 98 , 1–8. Riós-Orjuela, J. C., Falcón-Espitia, N., Arias-Escobar, A., Espejo-Uribe, M. J., & Chamorro-Vargas, C. T. (2020). Knowledge and interactions of the local community with the herpetofauna in the forest reserve of Quininí (Tibacuy-Cundinamarca, Colombia). Journal of Ethnobiology and Ethnomedicine , 16 (1). Ripley, B., Venables, B., Bates, D. M., Hornik, K., Gebhardt, A., Firth, D., & Ripley, M. B. (2013). Package ‘mass.’ Cran r , 538 , 113–120. Rowley, J. J. L., Callaghan, C. T., Cutajar, T., Portway, C., Potter, K., Mahony, S., Trembath, D. F., Flemons, P., & Woods, A. (2019). FrogID: Citizen scientists provide validated biodiversity data on frogs of Australia. In Herpetological Conservation and Biology 14(1). Salman, I., Salsaa, M., & Qumsiyeh, M. (2014). Distribution and cytogenetics of amphibians from the occupied Palestinian territories (West Bank of Jordan). In Jordan Journal of Natural History,1, 116-130 . Salvidio, S. (2008). Temporal variation in adult sex ratio in a population of the terrestrial salamander Speleomantes strinatii. The Herpetological Journal , 18 (1), 66–68. Scheffers, B. R., & Paszkowski, C. A. (2012). The effects of urbanization on North American amphibian species: Identifying new directions for urban conservation. Urban Ecosystems , 15 (1), 133–147. Schoch, R. R. (2009). Evolution of life cycles in early amphibians. Annual Review of Earth and Planetary Sciences , 37 , 135–162. Segev, O., Hill, N., Templeton, A. R., & Blaustein, L. (2010). Population size, structure and phenology of an endangered salamander at temporary and permanent breeding sites. Journal for Nature Conservation , 18 (3), 189–195. Segev, O., Mangel, M., & Blaustein, L. (2009). Deleterious effects by mosquitofish (Gambusia affinis) on the endangered fire salamander (Salamandra infraimmaculata). Animal Conservation , 12 (1), 29–37. Semlitsch, R. D. (2002). Critical elements for biologically based recovery plans of aquatic‐breeding amphibians. Conservation Biology , 16 (3), 619–629. Seto, K. C., Parnell, S., & Elmqvist, T. (2013). A global outlook on urbanization. In Urbanization, biodiversity and ecosystem services: Challenges and opportunities , 1–12. Sharon, R., Degani, G., & Warburg, M. R. (1997). Oogenesis and the ovarian cycle in Salamandra salamandra infraimmaculata Mertens (Amphibia; Urodela; Salamandridae) in fringe areas of the taxon’s distribution. Journal of Morphology , 231 (2), 149–160. Shemesh, H., Dener, E., & Sadeh, A. (2022). Bedrock may dictate the distribution of the fire salamander in the southern border of its global range. Israel Journal of Ecology and Evolution, 69(1-2), 5-9.\RL‏ Shwartz, A., Muratet, A., Simon, L., & Julliard, R. (2013). Local and management variables outweigh landscape effects in enhancing the diversity of different taxa in a big metropolis. Biological Conservation , 157 , 285–292. Shwartz, A., Turbé, A., Julliard, R., Simon, L., & Prévot, A. C. (2014). Outstanding challenges for urban conservation research and action. Global Environmental Change , 28 (1), 39–49. Sinai, I., Segev, O., Koplovich, A., Templeton, A. R., Blaustein, L., & Blank, L. (2020). Relationships among breeding site characteristics and adult population size of the fire salamander, Salamandra infraimmaculata. Hydrobiologia , 847 (14), 2999–3012. Sinai, I., Oron, T., Weil, G., Sachal, R., Koplovich, A., Blaustein, L., … & Blank, L. (2020). Estimating the effects of road-kills on the Fire Salamander population along a river. Journal for Nature Conservation , 58 , 125917.\RL‏ Sousa, E., Quintino, V., Palhas, J., Rodrigues, A. M., & Teixeira, J. (2016). Can environmental education actions change public attitudes? An example using the pond habitat and associated biodiversity. PLoS ONE , 11 (5). Steinfartz, S., Veith, M., &Tautz, D. (2000). Mitochondrial sequence analysis of Salamandra taxa suggests old splits of major lineages and postglacial recolonizations of Central Europe from distinct source populations of Salamandra salamandra. In Molecular Ecology , 9. Stuart, S. N., Chanson, J. S., Cox, N. A., Young, B. E., Rodrigues, A. S. L., Fischman, D. L., & Waller, R. W. (2004). Status and trends of amphibian declines and extinctions worldwide. Science , 306 (5702), 1783–1786. Sullivan, B. L., Aycrigg, J. L., Barry, J. H., Bonney, R. E., Bruns, N., Cooper, C. B., Damoulas, T., Dhondt, A. A., Dietterich, T., Farnsworth, A., Fink, D., Fitzpatrick, J. W., Fredericks, T., Gerbracht, J., Gomes, C., Hochachka, W. M., Iliff, M. J., Lagoze, C., La Sorte, F. A., … Kelling, S. (2014). The eBird enterprise: An integrated approach to development and application of citizen science. Biological Conservation ,169, 31–40. Vignoli, L., Mocaer, I., Luiselli, L., & Bologna, M. A. (2009). Can a large metropolis sustain complex herpetofauna communities? An analysis of the suitability of green space fragments in Rome. Animal Conservation , 12 (5), 456–466.Vincenz, B., & Reyer, H. U. (2005). Road Effects on a Fire Salamander Population (Salamandra salamandra terrestris) [PhD thesis], University of Zürich.\RL‏ Wainwright, T., Trevena, M., Alewijnse, S. R., Campbell, P. D., Jones, M. E., Streicher, J. W., & Cooper, N. (2023). Sex biases and the scarcity of sex metadata in global herpetology collections. Biological Journal of the Linnean Society , 129.\RL‏ Warburg, M. R. (1994). Population ecology, breeding activity, longevity, and reproduction strategies of Salamandra salamandra: an 18 year long study of an isolated population on Mt. Carmel Israel. Mertensiella , 4 , 399–421. Warburg, M. R. (2007). Longevity in Salamandra infraimmaculata from Israel with a partial review of life expectancy in urodeles. Salamandra 43(1), 1- 21. Warburg, M. R. (2008). Growth-rate in Free-roaming Salamanders Salamandra infraimmaculata: A Long-term Study. Current Herpetology , 27 (2), 61–69. Warburg, M. R., Degani, G., & Warburg, I. (1979). Growth and population structure of Salamandra salamandra (L.) larvae in different limnological conditions. Hydrobiologia , 64 (2), 147–155. Westgate, M. J., Scheele, B. C., Ikin, K., Hoefer, A. M., Beaty, R. M., Evans, M., Osborne, W., Hunter, D., Rayner, L., & Driscoll, D. A. (2015). Citizen Science Program Shows Urban Areas Have Lower Occurrence of Frog Species, but Not Accelerated Declines . White, G. C., & Burnham, K. P. (1999). Program MARK: survival estimation from populations of marked animals. Bird Study , 46 (sup1), S120–S139. Wickham, H., Chang, W., & Wickham, M. H. (2016). Package ‘ggplot2.’ Create Elegant Data Visualisations Using the Grammar of Graphics. Version , 2 (1), 1–189. Wickham, H., & Wickham, M. H. (2017). Package tidyverse. Easily Install and Load the ‘Tidyverse . Worldometers.info. (2022, October 6). Countries in the world by population (2022) . Https://Www.Worldometers.Info/World-Population/Population-by-Country/. Information & Authors Information Version history V1 Version 1 09 December 2024 V2 Version 2 24 February 2026 Peer review timeline Published Journal of Zoology Version of Record 28 Oct 2025 Published Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords ecological experiment ecosystem ecology freshwater vertebrate Authors Affiliations Omer Darel 0009-0006-5603-3575 Hebrew University of Jerusalem View all articles by this author Olga Rybak Society for the Protection of Nature in Israel View all articles by this author Asaf Ben Levy Society for the Protection of Nature in Israel View all articles by this author Gabi Kolodny Society for the Protection of Nature in Israel View all articles by this author Tamar Kis-Papo The Israeli Ministry of Education View all articles by this author Nirit Lavie Alon Society for the Protection of Nature in Israel View all articles by this author Rotem Vidan Society for the Protection of Nature in Israel View all articles by this author oren kolodny 0000-0002-0095-693X [email protected] Hebrew University of Jerusalem View all articles by this author Metrics & Citations Metrics Article Usage 352 views 204 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Omer Darel, Olga Rybak, Asaf Ben Levy, et al. Effective citizen science implementation reveals ecological and conservation insights for two amphibian species . Authorea . 24 February 2026. DOI: https://doi.org/10.22541/au.173376560.04983997/v2 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.173376560.04983997/v2","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9fe2132729ce06f3',t:'MTc3OTE4Mjk0MQ=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();