Digital Techniques for Monitoring Cerambyx cerdo in SAC “Gravine di Matera”: Population Size Insights

Study Areas

The study areas were selected based on the distribution of habitats suitable for Cerambyx cerdo and by identifying old oak specimens with signs of the species’ presence. The woodland formations within the “Gravine di Matera” SAC (Special Area of Conservation) are residual compared to those present until the last century, as fires and deforestation for cereal cultivation over the years have degraded them. Due to the extensive territory suitable for the species (over 1000 ha), two sampling areas were chosen, corresponding to the main woodland formations within the ZSC “Gravine di Matera”, named “Bosco di Lucignano” and “Bosco del Comune”. The decision to divide the sampling area into two sub-areas was driven by logistical challenges associated with sampling activities across the park area. Two monitoring sessions were conducted, one in 2019 and the other in 2021. The methods employed in this study focused on sampling adults, which are highly active seasonally. Therefore, sampling was conducted within a specific season period, estimated based on literature data. The first monitoring activity (2019) of Cerambyx cerdo took place from June 10, 2019, to July 7, 2019. The best time of year for monitoring C. cerdo adults is from late May to early August [12]. The survey initially targeted “Bosco di Lucignano”, followed by “Bosco del Comune”. Each survey lasted a total of 14 days per study area. To ensure representative data collection, five traps were positioned within both “Bosco di Lucignano” and “Bosco del Comune” were positioned at five different locations. The traps deployed in the “Bosco di Lucignano” area (Table 1) were activated on Quercus pubescens in the “Parco dei Monaci” locality, two on Quercus ilex in the “Murgia Sant’ Andrea” and “Vallone del Prete” localities, and two on Quercus trojana in the “Murgia Sant’ Andrea” and “Madonna del Giglio” localities (Table 1, Figure 1). Subsequently, the traps were also installed at the “Bosco del Comune” site. They were placed on Quercus pubescens at “Jazzo di Temparossa”, on Quercus ilex at “Vallone Conca d’Aglio”, on Quercus pubescens at “Volta Cornale”, on Quercus pubescens at “Jazzo del Comune” and “Bosco del Comune” (Table 1, Figure 1).

Click to enlarge

Click to enlarge

Trap Design and Capture-Mark-Recapture Protocol

To assess the abundance of Cerambyx cerdo, a Capture- Mark-Recapture (CMR) approach was employed targeting emerging adults. Adults of this species are easily sampled by the utilisation of aerial traps, baited with a mixture of wine and sugar which have been proven to be the most effective survey technique [13]. We developed a novel trap design based on the model described by De Zan LR, et al. [12] with the intent of simplifying the construction procedure (Figure 3). The traps were constructed using 25 cm high and 20 cm in diameter polyvinyl-chloride (PVC) tubes, with a total volume of 7853.9 cubic centimeters, commonly used in the construction industry. A metal mesh was affixed inside the tube to prevent captured insects from escaping into the attractant. At the upper part of the trap, a funnel with a 4 cm entrance hole was installed to allow beetles to enter while preventing their escape. Traps were active from dusk until the following afternoon, as C. cerdo is predominantly nocturnal. They were checked daily from Monday to Friday (Figure 3). Captured individuals were marked on the elytra with common nail polish (Figure 4), which has proven to be an effective way to mark coleopters [13]. Everyone was marked with colored nail polish, with one or more dots, unique to each trap, to facilitate the identification of recaptures. Captured C. cerdo individuals were released onto the trunk of the tree where they were trapped following marking. Traps proved highly efficient in keeping captured individuals alive, with no recorded deaths inside the traps, and all released individuals were in excellent condition. The bait consisted of a fermenting liquid attractant.

Literature suggests that trap effectiveness increases with the fermentation level of the liquid; therefore, preparation of the mixture was necessary several days before its use. Following the formula proposed by De Zan LR, et al. [12], each litre of liquid comprised two equal parts of red and white wine and 200 grams of white sugar. Preparation of this mixture was recommended to occur seven days before use. Each trap contained approximately 15 cl of liquid attractant, which was not replaced throughout the session. In cases of evaporation, the container was refilled with the same bait.

Click to enlarge

Click to enlarge

Innovation in Monitoring Activities

Innovative instruments were employed to facilitate the management of collected data during monitoring activities.

In the first year of monitoring (2019), innovation was implemented in the data storage and processing phases through the utilization of a shape file capable of calculating the parameters of the Schnabel function. In the second year of monitoring (2021), innovation extended to both survey and data management phases with the adoption of a dedicated software system called “Great cAPPricorn” (Figure 5). This software, developed by the Murgia Materana Regional Park specifically for monitoring Cerambyx cerdo, was utilized for the first time in the present study. All field data necessary for population estimation and documentation of survey activities related to Cerambyx cerdo were collected using the “Great cAPPricorn” Android application. This app allows for georeferencing of trap installation points along with textual and multimedia information, such as survey data, photos, or bait used. Furthermore, the software is capable of processing population estimates using the Schnabel method. The Android application guides users in applying the methodology outlined in the “Manual for monitoring species and habitats of Community interest (Directive 92/43/EEC) in Italy: animal species,” published by ISPRA, Manuals and Guidelines Series 141/2016 [14]. The use of the app in the field streamlines post-survey activities such as data archiving and processing. The monitoring system archives all entered data in dedicated databases and processes survey data to calculate population estimates of the species.

Additionally, the software generates detailed reports summarizing survey activities and data processing performed.

Click to enlarge

App Functionality for Data Processing

The incorporation of advanced data processing functionalities within our application (Great cAPPricorn) plays a pivotal role in enhancing the precision and efficiency of ecological studies, as demonstrated through our investigation into the population dynamics of C. cerdo in a specified study area. One of the main features of the application is the estimation of population size. To estimate the size of the C. cerdo population in the study area, we utilized the Schnabel Estimator function (Equation 1) widely employed in Capture-Mark-Recapture studies where more than two capturing events are carried out [14].

$$N = \frac{\sum (M_{t-1} \times C_t)}{\sum R_t + 1}$$

Equation 1: The population size $N$ is estimated using the Schnabel function where $C_t$ stands for the individuals captured at the sampling event $t$, $R_t$ indicates the individuals recaptured at sampling event $t$ and $M_t$ (t-1) are the individuals that were captured and marked at the previous sampling event $t-1$. We estimated population sizes for every single trap, assuming they were independent populations, and we then extended the population size to each one of the two major sampling areas, the "Bosco del Comune" and "Bosco di Lucignano" sites. Moreover, the App’s functionality extends to the calculation of confidence intervals for the mean population size indicator. The variance can be calculated as:

$$V \arctan \left( \frac{1}{N} \right) = \frac{\sum R_t}{\left( \sum (C_t M_t) \right)^2}$$

This formula gives the sampling uncertainty (variance) for the inverse of $N$ (1/N). To calculate the confidence interval around $N$, one way is to first calculate a confidence interval on the inverse of $N$ using the above formula, and apply the following formula:

$$\frac{1}{N} \pm 1,965 \times \sqrt{V \arctan \left( \frac{1}{N} \right)}$$

To calculate the confidence interval for $N$, take the inverse of the lower and upper bounds (limits) of the confidence interval calculated above. In addition to population size estimation, the App empowers researchers to perform sex ratio estimations—a crucial aspect in ecological studies. The sex ratio was calculated as recommended by Wilson K, et al. [15], as the proportion of adult males over the whole adult population. A strong sexual dimorphism is present in this species and is possible to easily recognize males as they have antennae that protrude when fully extended, beyond the tip of the abdomen. In the present study, the sex ratio, following Wilson K, et al. [15], was expressed as:

$$Sex \text{ ratio} = \frac{Males}{Males + Female}$$

Estimating Population Size (2019)

During the monitoring conducted at two sampling sites in 2019, a total of 78 specimens of Cerambyx cerdo were captured over 14 sampling days, exhibiting a balanced sex distribution with 37 females and 41 males, yielding a sex ratio of 0.5. All C. cerdo individuals were promptly marked and released back into their natural habitat. The population estimate (Table 3) at the "Bosco del Comune" site stood at 116.30, with a lower confidence interval of 69.3 and an upper confidence interval of 260.7. Meanwhile, at the "Bosco di Lucignano" site, the population estimate was 23.8, with a lower confidence interval of 12.3 and an upper confidence interval of 87.1. It is noteworthy that the confidence intervals are substantial relative to the calculated estimate, likely due to the limited number of recaptures recorded during monitoring.

Traps placed in various locations gave different catch results. Specifically, traps L1 and L5, sited on Quercus trojana, captured 3 and 2 males and 1 female of C. cerdo, respectively. In the "Bosco del Comune" locale, trap C3 reported 11 captures, including 3 males and 8 females of C. cerdo, alongside various specimens of Stictoleptura cordigera. Conversely, in the "Murgia Sant’Andrea" vicinity, traps L2 and L4 documented a significant presence of Vespa crabro, Cetonia aurata, and Proatetia morio, but no C. cerdo captures were recorded.

Estimating Population Size (2021)

During the study period, we observed significant variations in the population of Cerambyx cerdo in the two sampling sites, "Bosco di Lucignano" and "Bosco del Comune". In the "Bosco di Lucignano", we noted a substantial increase in the population of C. cerdo, with a total of 59 individuals captured. On the other hand, in the "Bosco del Comune", we did not observe significant changes in the population of C. cerdo during the study period, with a total of 53 individuals captured. It is also interesting to note the differences in sex composition between the two sampling sites. In "Bosco di Lucignano", we observed a higher number of individuals captured compared to "Bosco del Comune", with a relatively equal distribution between males and females. In contrast, in "Bosco del Comune", the proportion of males captured was lower compared to females. Overall, our results suggest that there are significant variations in the population of C. cerdo between the two sampling sites. The increase observed in “Bosco di Lucignano” may indicate the presence of more favourable environmental conditions or recovery of populations after any disturbances, while the stability in “Bosco del Comune” may require further investigation to fully understand the factors influencing the population of C. cerdo in that area. The population estimate (Table 3) in the “Bosco di Lucignano” is 216.8, with a lower confidence interval of 116.39 and an upper confidence interval of 660.4. Meanwhile, the population estimate (Table 3) in the “Bosco del Comune” is 141.7, with a lower confidence interval of

77.3 and an upper confidence interval of 379.6. Across both sampling sites, a total of 112 beetles (34 females and 78 males) were captured during the 14-day sampling period, resulting in a sex ratio of 0.6. All individuals of C. cerdo were promptly marked and released back into their natural habitat. These findings indicate fluctuations in population size between 2019 and 2021, with notable differences observed between the two sampling sites. The increase in population size in the “Bosco di Lucignano” area suggests potentially favourable conditions for C. cerdo in that habitat. However, further monitoring is necessary to assess the long- term trends and factors influencing population dynamics.

Comparative Analysis of Monitoring 2019 and 2021

In the “Bosco di Lucignano” area, our study used strategically positioned traps to measure the population changes of Cerambyx cerdo. The results from the two monitoring years, 2019 and 2021, showed different capture patterns across various locations. Notably, trap L2 in the “Murgia Sant’Andrea” site which had no captures of the target species in 2019, had the highest capture rate in 2021 (Table 2). This suggests a significant change in the distribution or abundance of C. cerdo within this specific area over the two years.

In contrast, the “Vallone del Prete” area consistently showed a notable capture rate in both monitoring years. However, the recapture rate during the 2021 monitoring period was lower compared to 2019, indicating potential fluctuations in the movement or behaviour of the population within this area. The population estimates for the two sites were 141.71 in “Bosco del Comune” and 216.83 in “Bosco di Lucignano” (Table 3). These estimates underline substantial variations in the population size of C. cerdo between the two years of monitoring, particularly evident in the “Bosco

di Lucignano” area. These findings highlight the dynamic nature of insect populations and emphasize the importance of continuous monitoring to understand and potentially mitigate fluctuations in population size and distribution.

Further investigation into the factors driving these changes is needed to inform effective conservation strategies for C. cerdo in the studied regions.