Evaluating invasive reptile control efforts: nocturnal bulk removal of green iguanas (Iguana iguana Linnaeus, 1758) in Florida, USA

The green iguana (Iguana iguana Linnaeus, 1758) ranges from Mexico to Brazil and portions of the Lesser Antilles. In Florida, iguanas are an invasive species responsible for numerous adverse impacts, prompting a demand for removal. Removal efforts have been inconsistent throughout Florida and the effects of removal on population size have not been evaluated. We conducted diurnal count surveys between 2018 and 2019 to estimate iguana abundance across two populations, treatment and control, along the C-11 Canal in Broward County, FL. We applied a nocturnal removal method for iguanas in the treatment population between the first and second period to assess its effect on the overall population. We removed 808 iguanas from the treatment population. We observed decreasing iguana captures with continued removal efforts, which was corroborated with an impact on daytime abundance estimates from count surveys. However, a seasonal reduction of iguana abundance was observed in the control population, with the number of iguanas observed in the control population not significantly different than the treatment population by the last period. While reductions of iguanas in the control population were not as extreme as those observed for the treatment population, this indicates that the biology of invasive iguanas likely plays a role in seasonal variation in abundance and stresses the need to fill knowledge gaps concerning the ecology of iguanas in Florida. Our results demonstrate why one-time removal efforts, or repeated efforts conducted over a short duration, are insufficient to control population growth, highlighting the need for continuous and persistent efforts.

Invasive species (nonnative species that are breeding, spreading, and causing impacts) pose a significant global threat to biological diversity and ecosystem function by altering trophic interactions, competitive dynamics, and ecological processes such as seed dispersal and nutrient cycling (Simberloff, 2011, 2013; Simberloff et al., 2013). Invasive reptiles and amphibians have become a pervasive threat to ecological integrity and human health and safety in Florida, United States, and pose a threat to restoration efforts for Everglades ecosystems in South Florida (Dorcas et al., 2012; Sementelli et al., 2008; National Research Council, 2005). Of particular concern are the potential negative impacts on native plants and animals as well as the physical damage to critical water control infrastructure and associated maintenance costs of repair (Ferriter et al., 2007; Ketterlin and Hardin, 2010). Florida has more nonnative reptiles and amphibians than anywhere else in the world, with at least 180 introduced species, more than 60 of which are established species (i.e., breeding; Krysko et al., 2016),. There are more nonnative lizards reproducing in the wild than native species in Florida (Engeman et al., 2011); the four largest invasive species of lizards breeding in Florida are from Africa (Nile monitor, Varanus niloticus), South America (Argentine black and white tegu, Salvator merianae), Central America and Mexico (spiny tailed iguana, Ctenosaura spp.), and Mexico, South, and Central America (green iguana, Iguana iguana) (Engeman et al., 2011).

The green iguana (Iguana iguana Linnaeus, 1758) is a large, diurnal lizard which ranges from northern Mexico through southern Brazil and portions of the Lesser Antilles (Ketterlin and Hardin, 2010). They are an established invasive in numerous locations including Florida and Texas, United States and several Caribbean and Pacific islands (Townsend et al., 2003; Garcia-Quijano et al., 2011; Thomas et al., 2013; Rivera-Milán and Haakonsson, 2020). Green iguanas have been documented to have adverse effects where they have become established. These include damaging native, ornamental, and agricultural plants, dispersing non-native plant seeds, fouling patios, pools, and canals with feces, and damaging infrastructure such as runways, seawalls, and levees via burrowing (Sementelli et al., 2008; Ketterlin and Hardin, 2010; Govender et al., 2012; Rivera-Milán and Haakonsson, 2020; Claunch et al., 2025). As a result of these negative impacts, green iguanas were designated by Florida statute as a Prohibited species on 29 April 2021, thus restricting their possession and sale with few exceptions.

Green iguana populations in southern Florida have expanded and increased in size since a historic freeze event in 2010 (Meshaka et al., 2004; Krysko et al., 2016). Resulting high densities of green iguanas on private and public property, plus their adverse impacts, have prompted an increase in demand for removal efforts. Current methods include hand grabbing, live traps, noosing, tongs, netting, and shooting (Kern, 2004; Krysko et al., 2007; Ketterlin and Hardin, 2010; Krauss et al., 2014; Wasilewski et al., 2022). Trapping is most effective during the day, when they are active, and all other methods can be applied during either the day or night. Most reported green iguana control efforts have focused on diurnal methods of removal (Smith et al., 2007; Thomas et al., 2013; Krauss et al., 2014; Rivera-Milán and Haakonsson, 2020), though a recent study focused on nocturnal removals as a means of remaining inconspicuous among other practical considerations (Wasilewski et al., 2022).

In Florida, diurnal removal methods are suitable for small, localized populations of green iguanas, and if applied in a sustained manner these methods may reduce local populations (Smith et al., 2007). However, the need for removal of green iguanas has expanded to large populations (hundreds to thousands) and large areas of public lands such as those managed by the South Florida Water Management District and the U.S. Army Corps of Engineers for flood control, water supply, and natural resource management. Nocturnal removal of green iguanas offers the opportunity for bulk removal over large areas by taking advantage of the tendency of green iguanas to rest in trees at night, especially over water. Sleeping green iguanas are visible and vulnerable to removal using a variety of techniques (Wasilewski et al., 2022).

Catch-per-unit-effort (CPUE), where effort is defined by time, distance, or monetarily, is a common index of abundance used to evaluate effects of removal efforts (Frazer et al., 2012; Moorhouse et al., 2014; Favaro et al., 2020). However, using CPUE implies a relationship with population size and that detectability, wariness, temporal, and environmental factors do not influence the proportion of individuals detected (Eversole et al., 2015; Farris et al., 2021; Nafus et al., 2025). Scant research has been done to evaluate efficacy of invasive reptile removal efforts (Kraus, 2009) though efforts for brown tree snakes (Boiga irregularis) in Guam have been an exception (Rodda et al., 2007; Christy et al., 2010; Nafus et al., 2025). Recent efforts by Rivera-Milán and Haakonsson (2020) and Rivera-Milán et al. (2022) have used monitoring and modeling to evaluate harvesting of green iguanas in the Cayman Islands, finding that inefficient removal (below 60%) can help to rapidly increase an iguana population as it would release it from density dependent factors.

To replicate an actual iguana removal effort, we used an on-going contracted iguana removal effort in South Florida to assess green iguana abundance across two populations (treatment and control) during four periods (T0 = pre-removals -January – February 2018, T1 = immediately post-removal -May 2018, T2 = three months post-removal -August 2018, and T3 = 12 months post-removal -May 2019) and evaluated effectiveness of a onetime (between T0 - February and T1 - May) nocturnal removal effort at the treatment population along the C-11 Canal in Broward County, Florida. We hypothesized that: 1) nocturnal removals would result in bulk iguana removals (≥ 25 iguanas per removal event); 2) iguana counts in the control population would be higher than in the treatment population following removals showing a population impact; and 3) reduced iguana counts in the treatment section would be maintained for a short period of time (at least three months, T2) post-removal as new individuals can repopulate the area due to the lack of continuous effort. However, we also hypothesized that seasonality, expressed as Julian date, would influence green iguana detectability independent of the type of intervention, which could potentially affect CPUE interpretability and implications on localized nocturnal bulk removal.

Study site

Surveys were conducted in a contracted green iguana removal effort area along two unconnected ~2 km sections of the C-11 Canal on the eastern basin in Broward County, Florida, defined for this study as control and treatment populations (Figure 1). The selection of this area was based on presence and abundance of iguanas, personnel safety, access, and vegetation composition and structure. Personnel safety and access are important considerations when performing removal efforts at night in urban areas. Similar vegetation structure (trees bordering canals, see below) allowed for consistent removal techniques. We considered these two sections of the C-11 Canal as separate (non-interconnected) populations for the purpose of performing count surveys (which occurred during the same survey day) as they were spaced 438 m apart and were separated by a section of canal and a street with no vegetation and an absence of iguanas. However, this design did not allow us to account for immigration between sections or from surrounding areas in-between surveys (e.g., from T1 to T2). To evaluate the efficacy of using nocturnal bulk removal and CPUE, we felt it was important to evaluate an actual iguana removal effort rather than set up an artificial experiment, which would limit its application to real-world management efforts.

The C-11 Canal is divided into western and eastern basins at water control structure S13A and is managed and maintained by the South Florida Water Management District primarily for drainage purposes. Several smaller canals drain into the C-11 Canal from the north and south. The C-11 Canal on the eastern basin runs in a west-east direction between Griffin Road and Orange Drive in Davie, Florida. Residential areas and shopping centers border Griffin Road to the south, and residential areas and a county park border Orange Drive to the north. The C-11 Canal is immediately bordered by 10–15 m wide grassy verges on both sides. The grassy verge on the north side of C-11 Canal is managed by the town of Davie as a linear park. Patches of trees, predominantly pond apple (Annona glabra), oak (Quercus sp.), strangler fig (Ficus sp.), cocoplum (Chrysobalanus icaco), and umbrella tree (Schefflera actinophylla) are found along both sides of the C-11 Canal. Green iguanas are observed foraging and basking on the grassy verges during the day and resting in the trees at night.

Count surveys

We conducted 20 diurnal green iguana count surveys (five per period) by boat recording number of iguanas observed in both control and treatment populations. Surveys at control and treatment areas were done during the same day and in the same direction (first treatment then control). Because the iguanas were numerous, active, and often obscured by vegetation during the count surveys, we were limited to counting the iguanas and could not estimate size class. It also was not possible to sex the iguanas remotely from the boat during the count surveys. Surveys were conducted between 10:00 h and 17:00 h when iguanas were most active. We recorded survey dates, survey start and end times, general weather conditions – clear, partly cloudy, overcast, and scattered showers, and environmental data – air temperature (°C), relative humidity (%), and one-minute average wind speed (mph) – at survey start and end points. Environmental data were collected using a Kestrel 3000 weather-meter (Kestrel Instruments, Boothwyn, PA, USA).

We ran generalized linear mixed models (GLMM) with Poisson (link = log) and negative binomial (NB, link = log) distributions in R software version 4.5.1 (R Core Team, 2025) via ‘glmer’ and ‘glmmTMB’ functions (lme4 and glmmTMB packages; Bates et al., 2015; Brooks et al., 2017) to assess effects of environmental conditions (air temperature, relative humidity, and wind speed), cloud cover (clear, partly cloudy, overcast), and Julian survey date on iguana counts including population (control/treatment) as a random effect. Model selection was done based on AIC values (lowest), goodness of fit (estimated via Pearson χ² check Ben Bolker–style, 1 = good fit, < 0.7 = under disperse, > 2 = over disperse), and variance explained by model (via r,squaredGLMM function, MuMIn package; Bartoń, 2025). Relevance of variables was selected based on critical values (p-value < 0.05).

Removals

We performed 30 nocturnal removal surveys via boat and truck/foot in the treatment population over a three-month period between T0 and T1 (mid-February-mid-May 2018). We used Fenix HP15 900-lumen headlamps and a Fenix TK41C 1,000-lumen tactical flashlight (Fenix Lighting, Littleton, CO, USA) to locate iguanas in trees, then we attempted to remove all iguanas seen with an extendable pole, tongs, net, or by hand. To comply with Florida state laws on invasive wildlife and Prohibited species, we euthanized all iguanas immediately after removal according to American Veterinary Medical Association (2020) guidelines and approved by the University of Florida Institutional Animal Care and Use Committee. We then recorded snout-vent length (SVL; cm), total length (TL; cm), mass (kg), and sex of captured individuals. We evaluated removal trends and means/ranges of morphometric values by sex and size class, which followed the designations of Meshaka et al. (2007): hatchling < 10 cm SVL, juvenile ≥ 10 cm and ≤ 20 cm SVL, adult > 20 cm SVL. We used all individuals for SVL calculations, but we excluded 156 individuals (44 adult males, 79 adult females, 15 juvenile males, and 18 juvenile females) with incomplete or missing tails from the TL and mass calculations because they would have negatively biased them.

We calculated capture per unit effort (CPUE) per removal survey by dividing the total number of iguanas removed per survey by the respective effort (hr) spent and distance (km) traveled. We calculated capture probability as the proportion of iguanas removed from what was detected during each removal survey and evaluated trends in capture probability among surveys. We ran generalized linear models (GLM; glm function, stats package) using the same parametrization described for count surveys but without including population (control/treatment) as random effects as removals only occurred in the treatment section. We used number of iguanas removed as a response variable as the survey distance was always constant (2 km). Model selection was based on AIC values (lowest), goodness of fit (variance of errors = ∑ E i 2 N − P), and variance explained by model (null deviance – observed deviance/null deviance). Relevance of variables was selected based on critical values (p-value < 0.05).

Count surveys

We conducted 20 diurnal counts surveys totaling 47.53 hr. of survey effort and 84.0 km of survey distance. Mean survey effort was 1.19 ± 0.61 hr., and mean survey distance was 4.2 ± 0.01 km. We observed a total of 4,741 green iguanas with a mean of 119 ± 54 and a range of 87–454 total observations per survey. We observed a total of 2,666 iguanas in the control population with a mean of 133 ± 45 and a range of 49–216 observations per survey. We observed a total of 2,075 iguanas in the treatment population with a mean of 104 ± 57 and a range of 32–238 observations per survey. We observed the highest numbers of iguanas during the T0 sampling period, with 937 iguanas in the control population and 953 iguanas in the treatment population (Figure 2A). We observed the lowest counts of iguanas in the control population during the T3 sampling period with 421 iguanas, but we observed the lowest counts in the treatment population during the T1 sampling period (right after removal effort) with 328 iguanas.

Both GLMMs showed significant intercepts for fixed parameters. However, we selected the NB model as it showed best indications (lower AIC) of performance, (AIC = 422, c ^ = 1.0, variance explained 49%) compared to the Poisson model (AIC = 733, c ^ = 13.3, variance explained 93%). Julian date was the only variable that showed a significant (negative) effect on iguana diurnal counts (size estimate = -0.24, p-value = 0.006; Figure 2B), predicting a reduction of at least one iguana observed per day across the sampled year.

Removals

We conducted 30 nocturnal removal surveys totaling 157.8 hr. of survey effort and 159.9 km of survey distance. We conducted 16 removal surveys via boat (total effort = 38.96 km) and 14 surveys via vehicle/foot (total effort = 69.18 km) removing a total of 508 and 300 iguanas per method, respectively. We observed a total of 3,701 iguanas with a mean of 123 ± 59 observations per survey and a range of 51 to 251 observations per night survey (Figure 3A). Iguana night observations per month increased during February (n = 382) to May (n = 849), with the highest counts occurring during April (n = 940). We removed a total of 808 iguanas with a mean of 27 ± 22 and a range 3 to 106 removals per night survey. Overall, iguana removals per month decreased from February (n = 195) to early April, increased in late April (n = 228), and then fluctuated through May (n = 173), with the highest number of removals occurring during April (Figure 3A).

We removed 168 adult males, 299 adult females, 104 juvenile males, 233 juvenile females, and 3 iguanas of unknown sex (Table 1). We did not capture hatchlings in our study because it was outside of the hatching season for iguanas in Florida (between May and August, Krysko et al., 2007). When all size classes of iguanas were combined, 34% of iguanas removed were male, 66% were female, and < 1% were of unknown sex. We removed 339 juveniles and 469 adults, which comprised 42% and 58% of the total, respectively. We recorded a mean SVL of 22.5 ± 8.35 cm (range 10.7 - 48.5 cm), a mean TL of 72.7 ± 26.3 cm (range 17.4 – 138.3 cm), and a mean mass of 0.46 ± 0.49 kg (range 0.03 - 3.40 kg) for the removed iguanas (Table 1). There were no noticeable trends in mean iguana SVL or TL values during the removal surveys. Mean CPUE was 5.29 ± 3.32 iguanas/hr and 5.49 ± 3.79 iguanas/km (Figure 3B). CPUE per survey was variable but decreased from 13.93 iguanas/hr and 14.27/km during the first survey to 8.19 iguanas/hr and 10.89 iguanas/km during the last survey. Mean capture probability was 0.21 ± 0.09 and exhibited a range 0.05 to 0.51. Capture probability per survey decreased from 0.51 during the first survey to 0.21 during the last survey (Figure 3C).

Both GLMs showed significant intercepts. However, we selected the NB model as it demonstrated the best indications of performance (AIC = 250.10, c ^ = 1.43, variance explained 39.8%) compared to the Poisson model (AIC = 368.54, c ^ = 9.44, variance explained 49%). Mean air temperature and Julian date were the only variables with significant (positive and negative) effects on iguana removals, with size estimates of 0.34; p-value = 0.021 and -0.36; p-value = 0.002, respectively, predicting an increase in iguana captures as temperature increases (one iguana per degree Celsius within the three month-window) and a decrease in iguana captured as we get closer to the end of the three-month window.

Our results demonstrated that: 1) nocturnal removals were an effective bulk removal method for iguanas; 2) iguana counts were higher in the control population than in the treatment population following removals; and 3) iguana removal was effective in reducing iguana observations over a short period of time (three months after removal T2; Figure 2). However, it was also clear that by the final survey period (T3 = 12 months after removal) the number of iguanas observed were similar in both the treatment and control populations, showing that as time progressed natural factors also acted on the control population almost equaling the effect of removals. A more in-depth understanding of the population and spatial ecology of iguanas in South Florida would improve our ability to interpret these results as well as to help us to better understand how, when, and for how long, removal efforts should be applied to maximize efficiency. It is also important to notice the strong effect of seasonality (expressed as Julian date) on iguana counts (likely driven by detectability) independent of the type of intervention (Figures 2, 3), which may affect CPUE interpretability. While this does not imply that nocturnal bulk removal is not an effective method to remove iguanas, it stresses the need for use of more robust methods, that account for detectability to better understand how removal effort affects populations over time.

It is not unusual for detection to decrease during surveys because of an increase in wariness of the species being surveyed (Bayliss et al., 1986; Pacheco, 1996). However, we do not think that was likely for the diurnal count surveys. The count surveys were a low-disturbance method compared to the removal surveys. Despite conducting both count and removal surveys in the same area (i.e., treatment population), we conducted the counts at different times (day) than the removals (night). However, we did not account for other types of human activities such as mowing along the canal’s grassy verges, which occurred infrequently in the vicinity of our study area and could have affected detection. Yet, given the study site is a mosaic of canals, busy roadways, pedestrians, and housing developments, it is unlikely a disturbance such as mowing would cause a notable issue with detection. There also could have been natural factors such as predation and associated defensive behavior, which could have affected the sampled population during our study (Greene et al., 1978; Smith et al., 2006; Meshaka et al., 2007). However, urban wildlife populations, including lizards, often experience a reduced diversity of predators and demonstrate an increased ability for habituation, and recovery from antipredator responses, to consistent, predictable disturbances compared to populations inhabiting more natural habitats (Pellitteri-Rosa et al., 2017).

Despite their abundance and growing impact, we know very little about important aspects of the population ecology of invasive green iguanas in Florida, especially movements, habitat use (when not in trees), nesting, vital rates, or phenology (see Krysko et al., 2007; Ketterlin and Hardin, 2010; Claunch et al., 2025). Basic knowledge such as hatching sizes, growth rates, and population demographics is lacking for the species in Florida. Other iguanas may have come from adjacent sections of the canal into the treatment population as our removals created habitat vacancies, potentially leading to compensatory changes in population dynamics (Rivera-Milán and Haakonsson, 2020), or other opportunities that allowed them to successfully immigrate into the treatment section. Although iguanas did not move between the control population and the treatment population during standardized count surveys (as they were performed during the same day, back-to-back, always in the same direction), we cannot rule out that emigration from the control population into the treatment section could have contributed to the decrease in iguanas observed in the control section between T1 and T3. However, the extent of the decrease from 761 to 412, which was not accompanied by a corresponding increase in the treatment population suggests that other factors were at play. The mean home range size of invasive green iguanas in Florida is approximately 75% lower compared to iguanas in their native range, likely due to reduced predation pressure and availability of resources in urban environments (Campbell and Maple 2012). These factors which contribute to high iguana densities (626/km 2) in disturbed habitats (Meshaka et al., 2007) and may limit the need for ingress and egress within invasive populations, although the extent to which this may occur remains to be studied. Additionally, we also know little about habitat use by green iguanas in areas where they occur but there are no trees, and therefore where nocturnal bulk removal would not be possible.

Our results suggest that our removal effort had an immediate, but not lasting, effect on iguana abundance. However, it should be noted that we did not validate the effectiveness of our method to reduce the overall population size of iguanas. Nocturnal removals may serve to change the population structure over time due to differential removal of sex and size class demographics (Table 1). Yet, the potential impacts of these differential removal rates on South Florida iguana population biology are unknown. It is unclear whether the frequency and three-month period of our removals would have been sufficient to induce a change in the sampled iguana populations at the study site since we did not evaluate changes in iguana demographics during our follow up diurnal count surveys, as reliable identification of size classes was not permitted. Iguanas were numerous, active, and often obscured by vegetation which prohibited us from both counting (our priority) and estimating sizes. For similar reasons, and as we conducted count surveys from a boat, it was not possible to sex the iguanas. Although we do not know what proportion of the iguana population was removed during our surveys, we do know that the proportion of iguanas removed from what was detected (capture probability) was 0.21 ± 0.09 and exhibited a range of 0.05 to 0.51 (Figure 3C). Rivera-Milán and Haakonsson (2020) found that if harvest rates fell below 0.60 that an iguana population would increase rapidly due to the release of density dependent factors regulating a population and leading to an increase in survival, fecundity, or both; they caution that an incomplete knowledge of vital rates should lead to long term monitoring. Udell et al. (2022) found that a removal rate of 0.40 for invasive tegus would decrease the population and that if the removal rate fell to 0.30 the tegu population would increase. The relatively small proportion of iguanas removed from the treatment population in this study could explain the later increase in survey counts observed during T3.

We attributed the overall decrease in CPUE to a decrease in probability of capture. Although we removed 808 iguanas, corroborating the utility of nocturnal bulk removal, we observed multiple instances of adult iguanas perched beyond the reach of our extendable poles and in burrows along the canal banks. The nocturnal nature of our removals, as well as our proximity to vehicles, public walkways, and commercial and residential buildings rendered the use of potentially more effective methods such as firearms problematic. Our observed trends in CPUE may have also been related to seasonal activity patterns since we observed peak iguana observations and removals during mid-spring (April-May). Mid-spring exhibits favorable air and water temperatures, and it also falls within the breeding season for iguanas in southern Florida (Krysko et al., 2007). The decreasing trend in probability of capture could have been associated with an increase in wariness with increasing removal surveys (Webb and Messel, 1979).

The questionable reliability of using CPUE to inform management is concerning, yet CPUE is frequently used to evaluate invasive species removal efforts (e.g. Giddens et al., 2014; Acre et al., 2024). Our results demonstrate the need to clearly define management objectives, with careful consideration as to methodology and duration, and to select appropriate metrics and modeling tools to achieve desired management outcomes (Macnab, 1983; Ferraro and Pattanayak, 2006; Howe and Milner‐Gulland, 2012; Adams et al., 2014). It is important to note that using CPUE to guide management is not inherently unreliable, but requires validation. For example, Nafus et al. (2025) found that CPUE obtained during visual surveys for brown tree snakes was a consistent indicator of abundance but only when surveys occurred over the majority of the landscape where the population was being estimated. Farris et al. (2021) found that N -mixture model abundance estimates matched encounter rates 76% of the time for American alligators (Alligator mississippiensis) during night-time eyeshine counts. Data from Farris et al. (2021) had been collected via an established, adaptive sampling protocol, which sought to minimize potential sampling biases. This, as well as the taxa studied, may have contributed to the relatively high level of agreement between these different metrics.

We recommend performing abundance surveys independently of removal efforts to reduce reliance on CPUE as demonstrated in this study. We also advise that abundance surveys be performed in a manner such that detection probability can be explicitly calculated, for example using N -mixture models (Farris et al., 2021). However, we recognize that the opportunity to perform pre- and post-removal surveys to evaluate removal effectiveness is not always available due to limited resources. For the latter situation, we recommend that modeling, monitoring, and possibly additional removal efforts occur after the initial treatment to evaluate removal (McCann et al., 2013; Rivera-Milán and Haakonsson, 2020). It is possible that continued persistent attempts at nocturnal bulk removal may eventually serve to reduce the overall size of the iguana population (Smith et al., 2007). We expect this to be especially true if removal and monitoring are continuous, persistent, occur over a larger area, and occur at targeted times such as during the breeding season when nesting females can be removed along with their eggs, and during May-August when hatchling iguanas emerge (Krysko et al., 2007) and can be removed in large numbers. Supplementing nocturnal removal by targeting iguanas in burrows, as well as destruction of habitat and nesting sites, may make removal efforts more effective and further aid population reduction. While testing the utility of supplemental efforts was beyond the scope of this study, a multifaceted approach to iguana control, incorporating multiple methodologies (i.e., removal, exclusion, and habitat alteration) are likely necessary to ensure management goals are reached.

We recommend further research on the biology of iguanas in Florida paired with long-term population monitoring as this knowledge gap hinders inference from removal efforts, and therefore development of effective management strategies. In addition, an increased use of statistical modeling is needed to improve our ability to control iguana populations, as well as to understand impacts of removal (Krauss et al., 2014; Rivera-Milán and Haakonsson, 2020; Udell et al., 2022; Claunch et al., 2025). While our study allowed us to better evaluate our invasive reptile removal efforts along an urban canal in southern Florida, greater context and interpretation of our efforts are lacking due to insufficient knowledge of invasive green iguana population biology. Closing this information gap is essential to develop and validate control methods to effectively manage this well-established and problematic invader in Florida. This is a common problem with invasive reptiles in Florida (and we suspect for other taxa and locations) where, with the exception of Burmese pythons (Python bivittatus; see Guzy et al., 2023), relatively little is known about basic biology of most invasive reptiles (e.g., Argentine black and white tegus and Nile monitors).

Lack of a long-term response of the iguana populations assessed in our study to removal efforts raises the question: is removal worth it if it is not sustained? Removal efforts are a management action but not necessarily a management objective (Rivera-Milán and Haakonsson, 2020). Management objectives could be population reduction, containment, mitigation of impacts on native species, prevention of damage to infrastructure (e.g., seawalls and levees) or agriculture, or minimizing risks to human health and safety. Removing iguanas, or any other invasive reptile, in and of itself is not a demonstration of success. Success should be determined by achieving management objectives such as population reduction while ascertaining the costs and benefits accrued in doing so. If management objectives are not thought to be achieved or if the cost in doing so may outweigh the potential benefit, then we urge caution and careful consideration before continuing management action, particularly when actions are continued with no revision to frequency, duration, or magnitude.

Nocturnal removals were an effective bulk removal method in reducing iguana observations over a short period of time (three months) and resulted in a large number (800) of iguanas removed.

By the final survey period (T3 = 12 months after removal) the number of iguanas observed were similar in both the treatment and control populations, showing that as time progressed natural factors also acted on the control population almost equaling the effect of removals.

A more in-depth understanding of the population and spatial ecology of iguanas in South Florida would improve our ability to interpret these results as well as to aid understanding as to how, when, and for how long, removal efforts should be applied to maximize efficiency. We recommend that further research and modeling on the biology of iguanas in Florida be paired with long-term population monitoring.

We attributed the overall decrease in CPUE to a decrease in probability of capture which could have been a result of an increase in wariness of iguanas over time with increasing removal surveys.

Using CPUE to evaluate removal efforts is not necessarily unreliable but requires validation. We recommend performing monitoring surveys independently of removal efforts to reduce reliance on CPUE as done in this study. We also recommend that monitoring surveys be conducted in a manner such that detection probability can be explicitly calculated.

Continued, persistent attempts at nocturnal bulk removal could serve to reduce the overall iguana population size, especially if removal and monitoring were continuous, sustained, occurred over a larger area, and occurred at targeted times such as during the breeding season to target reproductive females and nests, and during May-August when hatchling iguanas emerge in large numbers.

Removal efforts are a management action but not necessarily a management objective (Rivera-Milán and Haakonsson, 2020). Management objectives could be population reduction, containment, prevention of impacts on native species, prevention of damage to infrastructure (e.g., seawalls and levees) or agriculture, or minimizing risks to human health and safety.