ecologia-oct2025

Población panmíctica de la polilla polinizadora Tegeticula baja (Lepidoptera: Prodoxidae) a lo largo de la distribución de sus plantas hospederas

Maria Clara Arteaga ^a, *, C. Rocio Alamo-Herrera ^a, Anna Darlene van der Heiden ^{a, b}, Nicole Sicaeros-Samaniego ^a

^a Centro de Investigación Científica y de Educación Superior de Ensenada, Departamento de Biología de la Conservación, 

Carr. Tijuana-Ensenada 3918, Zona Playitas, 22860 Ensenada, Baja California, Mexico

^b Uppsala University, Department of Medical Biochemistry and Microbiology, Hursagatan 3, 752 37, Uppsala, Sweden 

*Corresponding author: arteaga@cicese.mx (M.C. Arteaga)

Received: 14 March 2025; accepted: 24 July 2025

Abstract

Ecological interactions and demographic history shape the genetic diversity of populations. Tegeticula baja is the specialist pollinator of different Yucca hosts in the Baja California Peninsula, a region that experienced changes in habitat distribution during the Pleistocene. To assess the effects of host specificity and historical changes in habitat configuration, we i) analyzed the genetic structure of moth populations associated with 3 different Yucca plant species, ii) identified signatures of historical demographic changes, and iii) reconstructed the past potential distribution of T. baja at different periods. We genotyped the COI of 128 moths from 39 locations and estimated genetic diversity, population structure, and demographic history. We found an overall haplotype diversity of 0.708 and a nucleotide diversity of 0.0015. Moth populations associated with the 3 hosts exhibited similar diversity levels, with no evidence of genetic structure. These findings suggest that ecological associations with different host plants do not drive T. baja diversification. Instead, its demographic history has played a more significant role in shaping the levels and the distribution of the genetic diversity.

Keywords: Ecological interaction; Genetic diversity; Historical demography; Insect-plant association; Population structure; Yucca

Resumen

Las interacciones ecológicas y la historia demográfica moldean la diversidad genética de las poblaciones. Tegeticula baja es una polinizadora especialista de diferentes yuccas hospederas en la península de Baja California, una región que experimentó cambios en la distribución de los hábitats durante el Pleistoceno. Para evaluar los efectos de la especificidad del huésped y los cambios históricos en la configuración de los hábitats, i) analizamos la estructura genética de las poblaciones de polillas asociadas a distintos hospederos, ii) identificamos señales de cambios demográficos históricos y iii) reconstruimos su distribución potencial en el pasado. Genotipificamos el COI de 128 polillas de 39 localidades y estimamos la diversidad genética, la estructura poblacional y la historia demográfica. Encontramos una diversidad haplotípica global de 0.708 y una diversidad nucleotídica de 0.0015. Las poblaciones de polillas asociadas a las 3 especies de plantas mostraron niveles de diversidad similares, sin evidencia de estructura genética. Estos hallazgos sugieren que la asociación ecológica con diferentes plantas huésped no impulsa la diversificación de T. baja. En cambio, su historia demográfica ha desempeñado un papel más importante en la configuración de su diversidad genética.

Palabras clave: Interacciones ecológicas; Diversidad genética; Historia demográfica; Asociación planta-insecto; Estructura poblacional; Yucca

Introduction

The ecological interaction between pollinator insects and their host plants plays a crucial role in shaping the distribution and genetic diversity of the species involved, influencing processes such as natural selection, genetic drift, and gene flow among populations (Futuyma, 2000; Gloss et al., 2013, 2016). Additionally, other factors, such as geographic distance (e.g., Driscoe et al., 2019) and demographic history shaped by past climatic changes (e.g., Liu et al., 2016; Smith et al., 2011), can influence the amount of genetic variation within populations and its spatial distribution.

Moths of the genus Tegeticula Zeller are specialist pollinators of the genus Yucca Linnaeus, maintaining an obligate mutualism (Engelmann, 1872; Pellmyr, 2003). Generally, each Tegeticula species pollinates a single Yucca species. However, few moth species pollinate multiple Yucca species (Althoff et al., 2006, 2012). During the flowering period, adult moths emerge and mate within Yucca flowers. The female uses specialized mouthparts to collect pollen and transfer it to other flowers. Upon arrival, she inserts her ovipositor into the ovary to lay her eggs. After, she deposits the pollen onto the flower’s stigma, ensuring fertilization and fruit development, from which the larvae will feed on a small portion of the seeds (Engelmann, 1872; Pellmyr, 2003). This obligate interaction between yucca and yucca moths influences the gene flow, facilitating differentiation and diversification processes in both pollinators and host plants (Arteaga et al., 2020; Leebens-Mack & Pellmyr, 2004; Leebens-Mack et al., 1998) 

Yucca plants and yucca moths are distributed across North America (Pellmyr et al., 2008). This region has experienced changes in habitat distribution due to Pleistocene climatic fluctuations, which have impacted the genetic diversity and structure of both Yucca populations and their pollinators (Alemán et al., 2024; Arteaga et al., 2020; De la Rosa et al., 2020; Smith et al., 2011). In the Baja California Peninsula (BCP), Mexico, Tegeticula baja Pellmyr is an endemic moth species distributed from the central to the southern part of the peninsula. This moth pollinates 2 endemic Yucca species and their hybrid populations, which exhibit an allopatric distribution: Yucca valida Brandegee occurs in the arid ecosystems of the Central Desert and the northern Magdalena Plains, hybrid populations are found in the southern Magdalena Plains (Arteaga et al., 2020), and Y. capensis Lenz is located in the tropical dry forest of southern BCP (Lenz, 1998; Pellmyr et al., 2008; Turner et al., 1995).

The flowering phenology of these endemic Yucca species and their hybrid populations is asynchronous and influenced by water availability. Yucca valida blooms from April to July (Turner et al., 1995), hybrid populations flower in August and September, and Y. capensis blooms from September to October (Arteaga et al., 2015; Lenz, 1998). The asynchronous flowering limits the temporal coexistence of moths from different populations, as each group responds to the floral signals of its host plant. Additionally, these moths exhibit short-distance dispersal (Álamo-Herrera et al., 2022). Consequently, the temporal availability of floral resources and the limited dispersal of these pollinators may contribute to genetic structuring among populations across their distribution.

Given the obligate interaction between pollinators and plants, we hypothesize that distinct genetic lineages of T. baja are associated with each host Yucca species, influenced by geographic distances and asynchronous flowering. Additionally, considering historical habitat changes in the BCP (Dolby et al., 2015), we expect an impact on the species’ demographic history. Specifically, we i) examined the genetic structure among moth populations associated with each Yucca species and their hybrids, ii) identified signals of historical population changes, and iii) determined how the distribution of suitable conditions for the species has changed over time. This study will contribute to a better understanding of how ecological interactions and historical environmental changes shape the genetic diversity of pollinating moths in the Yucca-Tegeticula mutualism.

Materials and methods

We visited 75 localities where the endemic yuccas and their hybrids were found (Fig. 1A). We collected 3 to 5 fruits from at least 5 plants per locality. We dissected the fruits and examined them for the presence of T. baja larvae, which are typically found among the seeds. We stored the larvae in 96% ethanol and preserved them at -80 °C. We collected a total of 128 moth larvae from 39 of the 75 localities (Fig. 1A), including 16 sites of Yucca valida (N = 49), 8 of Y. capensis (N = 39), and 15 of hybrid plants (N = 40).

We used 20 mg of larval tissue for DNA extraction following the commercial kit “Qiagen DNeasy Blood & Tissue Kit” protocol. We amplified a fragment of the Cytochrome Oxidase subunit I (COI) marker using primers S1461 (5’-ACAATTTATCGCCTAAACTTCAGCC-3’) and A2302 (5’-CTACAAATCCTAATAATCCATTG-3’; Smith et al., 2009). The PCR mixture consisted of 5 μl of buffer (1X), 2 μl of MgCl (2 mM), 0.4 μl of dNTPs (0.6 mM), 0.2 μl of Taq polymerase (1U), 1 μl of each primer (0.4 mM each), 3 μl of DNA, and 12.4 μl of molecular-grade water, for a total reaction volume of 25 μl. Thermocycler conditions were: initial denaturation at 95 °C for 3 min, 35 cycles of denaturation at 94 °C for 30 s, annealing at 48 °C for 45 s, extension at 72 °C for 1 min, and a final elongation at 72 °C for 1 min. We verified the amplification quality using 1% agarose gels. PCR products were sequenced by SeqXcel (www.seqxcel.com). We also tested protocols for amplifying the nuclear EF1α gene and 9 nuclear microsatellites previously used in other species of the same genus (Drummond et al., 2010; Smith et al., 2008); however, amplifications were unsuccessful.

We visualized and edited COI sequences using MEGA X 4.0 and aligned them with MUSCLE (Tamura et al., 2007). We estimated the genetic diversity for the complete dataset and separately for individuals collected in locations from Y. valida, Y. capensis, and the hybrid Y. valida × Y. capensis using DNAsp (Rozas et al., 2003). We calculated the number of polymorphic sites (PS), number of haplotypes (H), haplotype diversity (h), and nucleotide diversity (π). To explore genealogical relationships, we constructed a haplotype network using the Median-Joining method in NETWORK 5.0 (Bandelt et al., 1999). Finally, to assess genetic structure among moths associated with different host plants, we implemented an analysis of molecular variance (AMOVA) using Arlequin (Excoffier et al., 2005).

We assessed the demographic history of T. baja using 3 methods. First, we conducted Tajima’s D test using DNAsp (Rozas et al., 2003). Negative values indicate population expansion, while positive values suggest population reduction (Nakamura et al., 2018). Second, we performed a mismatch distribution analysis in Arlequin, where an unimodal distribution suggests population expansion, while a multimodal distribution indicates a stable population size (Rogers & Harpending, 1992). Finally, we conducted a Bayesian Skyline Plot (BSP) analysis (Drummond et al., 2005) using BEAST 2.6.0 (Bouckaert et al., 2014). As input data, we used the commonly reported nucleotide substitution rate for COI in arthropods (1.77% divergence per lineage per million years; Papadopoulou et al., 2010), a strict molecular clock, and the HKY substitution model defined in JmodelTest2 (Darriba et al., 2012). We ran 100 million steps, sampling every 10,000 generations in the MCMC method. We calculated the effective sample size (ESS) value and constructed the BSP using TRACER 1.7 (Rambaut et al., 2018).

Table 1

Genetic diversity of Tegeticula baja based on the mitochondrial COI marker. The table includes the host plant, sample size (N), number of polymorphic sites (PS), number of haplotypes (H), haplotype diversity (h), and nucleotide diversity (π).

Host yucca plant	N	PS	H	h	π
Yucca valida	49	14	17	0.733	0.0016
Hybrid populations	40	9	10	0.750	0.0015
Yucca capensis	39	11	11	0.617	0.0012

We built species distribution models to assess changes in the geographic distribution of suitable conditions for Tegeticula baja and to support the interpretation of its genetic diversity and demographic history estimates. We considered 4 time periods: present (years 1970-2000), mid-Holocene (6,000 years ago), Last Glacial Maximum (22,000 years ago), and Last Interglacial (120,000 – 140,000 years ago). We constructed the models using the 75 sampling points obtained in this study, the 19 bioclimatic variables from the WorldClim 2.1 database (Fick & Hijmans, 2017), and the MAXENT software (Phillips et al., 2006), with 80% of the data used for training and 20% for validation. We configured 2,000 iterations and 10 replicates. The model evaluation included the area under the curve (AUC) and binomial probabilities, where an AUC > 0.9 reflects excellent predictive capacity. 

Results

The alignment of 128 sequences resulted in 767 bp with 27 polymorphic sites, defining 30 haplotypes (NCBI ID: PX127684-PX127713; Fig. 1B). The overall haplotype diversity was moderate (h = 0.708), and nucleotide diversity was low (π = 0.0015). Moth populations associated with the 3 host plants exhibited similar diversity levels (Table 1). The haplotype network indicated that haplotypes were closely related, and that the most abundant haplotype was widely distributed among the 3 moth populations pollinating different host plants (Fig. 1B). The AMOVA revealed that the variance among moths associated with different host plants was low and not significant (Fst = 0.011, p = 0.13). In contrast, most variation was found within populations. 

Since no signs of genetic structure were detected, demographic analyses and niche modeling were conducted considering all individuals as a panmictic population. Demographic analyses provided evidence of a historical population expansion. Tajima’s D test showed significant negative values (D = -2.27970, p < 0.01), and the mismatch analysis distribution was unimodal (Fig. 2A). Consistently, the BSP analysis suggested a population expansion beginning approximately 25,000 years ago (Fig. 2B). 

The species distribution models, with high predictive power (AUC > 0.9), revealed historical changes in the extent and distribution of environmentally suitable areas for T. baja (Fig. 3). During the Last Interglacial (120 ka), the distribution was limited to 2 areas, one in the central and other in the southern regions of the peninsula. During the Last Glacial Maximum (22 ka) and the mid-Holocene (6 ka), an expansion occurred in both regions. Finally, in the present period (1970-2000), the potential distribution of suitable conditions for these moths is observed to be continuously present along the western portion of the peninsula, connecting the central and southern regions.

Discussion

In the obligate mutualism between Yucca plants and Tegeticula moths, the distribution of feeding and oviposition resources provided by host plants determines the presence of moths in the landscape. In this study, we evaluated the genetic structure of Tegeticula baja populations associated with different Yucca species with allopatric distributions. Contrary to our expectations, we found a single panmictic population of pollinator moths throughout its geographic range. Additionally, we detected signals of historical demographic expansion. This suggests that the ecological association with different hosts is not driving the diversification of this species and that its historical demography has played a more relevant role in the distribution of its genetic diversity.

The genetic structure of a species is influenced by multiple factors, such as dispersal capacity, the intensity of ecological interactions, and the climatic history of the areas it inhabits (Driscoe et al., 2019; Futuyma, 2000; Smith et al., 2011). Specifically, T. baja has a limited dispersal distance per generation, and only 1 generation per year (~ 42 m; Álamo-Herrera et al., 2022); its host plants have a discontinuous distribution in the current landscape, and they also exhibit asynchronous flowering (Arteaga et al., 2020; Lenz, 1998; Turner et al., 1995). Together, these factors suggested that we could find genetic structure among moth populations associated with different Yucca species. However, we did not observe significant genetic differentiation. It is possible that the age of origin of its host plants and changes in the distribution of suitable habitat conditions could partially explain this pattern.

The divergence between the 2 endemic Yucca species of the BCP is estimated to have occurred approximately 500,000 years ago (Alemán et al., 2024). The formation of hybrid populations is even a more recent event, proposed to have occurred during the Pleistocene, around 21,000 years ago, when favorable climatic conditions allowed the co-occurrence of Y. valida and Y. capensis in the same region (Arteaga et al., 2020). This period of change in host plant distribution likely also influenced the distribution of the moth Tegeticula baja. Species distribution models and demographic analyses support this, indicating a population expansion beginning around 25,000 years ago, followed by a stabilization phase approximately 3,000 years ago. These historical changes in habitat configuration likely shaped the demographic history of the moth, reducing the potential for genetic divergence across its range due to alternating periods of population isolation and secondary contact. Although the current distribution of host plants is fragmented, the slow generational turnover of T. baja, with 1 generation per year, suggests that insufficient time has passed for genetic drift to produce detectable genetic structure.

The phylogeographic pattern of Tegeticula baja does not exhibit genetic structuring associated with host plant identity, which contrasts with that of other Tegeticula species, where genetic differentiation is correlated with either geographic distance or Yucca host species. For example, T. yucasella exhibits high genetic differentiation associated with geographic distance and interactions with different Yucca species (Leebens-Mack & Pellmyr, 2004). Similarly, T. maculata, the pollinator of Hesperoyucca Engelmann, exhibits genetic clades associated with the biogeographic history of its region (Althoff et al., 2007; Segraves & Pellmyr, 2001). Although all these studies, including ours, used the same mitochondrial marker (COI subunit), we did not detect genetic differentiation in our samples. These differences between our findings and previous reports may be related to the spatial scale, which is much larger in the study of T. yucasella, and to the time of origin and biogeographic history of the host plants in the case of T. maculata (Segraves & Pellmyr, 2001).

The levels of genetic diversity detected in the panmictic population of T. baja were moderate. In particular, moths associated with Y. valida exhibited a higher number of haplotypes, possibly due to their larger geographic range (Fig. 1A). The overall nucleotide diversity was low (π = 0.0015), falling below values reported for other species in the genus, such as T. antithetica and T. synthetica (π = 0.004 and 0.005, respectively; Smith et al., 2008). This pattern of low genetic diversity observed in T. baja may be related to the historical demographic growth detected in this species. A similar pattern was reported in the panmictic population of the parasitoid wasp Digonogastra sp., which interacts with 2 Tegeticula species in the BCP (π = 0.002; Álamo-Herrera et al., 2024). This supports the idea that the levels and distribution of genetic variation in these moths are more related to their historical demography than to their ecological interactions.

In conclusion, integrating genetic data with species distribution models allows us to understand how the climatic history of the BCP has influenced the distribution of genetic diversity and demographic changes in this species. Since the moths’ life cycle depends on Yucca flowering, which in turn responds to precipitation, climate change is likely to affect the population dynamics of these insects. Periods of extreme drought, such as those occurring in recent decades in the peninsula, may impact moth demography and exacerbate a population decline. Future studies focusing on the ecology and evolution of Prodoxus species associated with Yucca and Hesperoyucca in this region could enhance our understanding of the hidden diversity within this group and complement existing information on the northern species (Smith & Leebens-Mack, 2024). 

Acknowledgements

The authors are grateful to Lita Castañeda and Mario Salazar for their help with laboratory analysis, technical support, and assistance in the fieldwork. They also thank Alberto López Alemán for his valuable assistance in improving the English language of the manuscript. This study was supported financially by Secretaría de Ciencia, Humanidades, Tecnología e Innovación (Secihti) (CB-2014-01-238843, infra-2014-1-226339). The Rufford Foundation also provided financial support for a part of this study (RSG 13704-1) and the Jiji Foundation. The authors thank the Associate Editor and the anonymous reviewers for their valuable comments. The authors declare no conflicts of interest.

References

Álamo-Herrera, C. R., Arteaga, M. C., Bello-Bedoy, R., & Rosas-Pacheco, F. (2022). Pollen dispersal and genetic diversity of Yucca valida (Asparagaceae), a plant involved in an obligate pollination mutualism. Biological Journal of the Linnean Society, 136, 364–374. https://doi.org/10.1093/biolinnean/blac031

Álamo-Herrera, C. R., Arteaga, M. C., & Bello-Bedoy, R. (2024). Genetic diversity and phenotypic variation in a parasitoid wasp involved in the yucca – yucca moth interaction. Revista Mexicana de Biodiversidad, 95, e955461. https://doi.org/10.22201/ib.20078706e.2024.95.5461

Alemán, A., Arteaga, M. C., Gasca-Pineda, J., & Bello-Bedoy, R. (2024). Divergent lineages in a young species: the case of Datilillo (Yucca valida), a broadly distributed plant from the Baja California Peninsula. American Journal of Botany, 111, e16385. https://doi.org/10.1002/ajb2.16385

Althoff, D. M., Segraves, K. A., Leebens-Mack, J., & Pellmyr, O. (2006). Patterns of speciation in the yucca moths: parallel species radiations within the Tegeticula yuccasella species complex. Systematic Biology, 55, 398–410. https://doi.org/10.1080/10635150600697325

Althoff, D. M., Svensson, G. P., & Pellmyr, O. (2007). The influence of interaction type and feeding location on the phylogeographic structure of the yucca moth community associated with Hesperoyucca whipplei. Molecular Phylogenetics and Evolution, 43, 398–406. https://doi.org/10.1016/j.ympev.2006.10.015

Althoff, D. M., Segraves, K. A., Smith, C. I., Leebens-Mack, J., & Pellmyr, O. (2012). Geographic isolation trumps coevolution as a driver of yucca and yucca moth diversification. Molecular Phylogenetics and Evolution, 62, 898–906. https://doi.org/10.1016/j.ympev.2011.11.024

Arteaga, M. C., Bello-Bedoy, R., León-de la Luz, J. L., Delgadillo, J., & Domínguez, R. (2015). Phenotypic variation of flowering and vegetative morphological traits along the distribution for the endemic species Yucca capensis (Agavaceae). Botanical Sciences, 93, 765–770. https://doi.org/10.17129/botsci.214

Arteaga, M. C., Bello-Bedoy, R., & Gasca-Pineda, J. (2020). Hybridization between yuccas from Baja California: Genomic and environmental patterns. Frontiers in Plant Science, 11, 685. https://doi.org/10.3389/fpls.2020.00685

Bandelt, H. J., Forster, P., & Röhl, A. (1999). Median-joining networks for inferring intraspecific phylogenies. Molecular Biology and Evolution, 16, 37–48. https://doi.org/10.1093/oxfordjournals.molbev.a026036

Bouckaert, R., Heled, J., Kühnert, D., Vaughan, T., Wu, C. H., Xie, D. et al. (2014). BEAST 2: a software platform for Bayesian evolutionary analysis. Plos Computational Biology, 10, e1003537. https://doi.org/10.1371/journal.pcbi.1003537

Darriba, D., Taboada, G. L., Doallo, R., & Posada, D. (2012). jModelTest 2: more models, new heuristics and high-performance computing. Nature Methods, 9, 772. https://doi.org/10.1038/nmeth.2109

De la Rosa-Conroy, L., Gasca-Pineda, J., Bello-Bedoy, R., Eguiarte, L. E., & Arteaga, M. C. (2020). Genetic patterns and changes in availability of suitable habitat support a colonisation history of a North American perennial plant. Plant Biology, 22, 233–242. https://doi.org/10.1111/plb.13053

Dolby, G. A., Bennett, S. E. K., Lira-Noriega, A., Wilder, B. T., & Munguia-Vega, A. (2015). Assessing the geological and climatic forcing of biodiversity and evolution surrounding the Gulf of California. Journal of the Southwest, 57, 391–455.

Driscoe, A. L., Nice, C. C., Busbee, R. W., Hood, G. R., Egan, S. P., & Ott, J. R. (2019). Host plant associations and geography interact to shape diversification in a specialist insect herbivore. Molecular Ecology, 28, 4197–4211. https://doi.org/10.1111/mec.15220

Drummond, A. J., Rambaut, A., Shapiro, B. E. T. H., & Pybus, O. G. (2005). Bayesian coalescent inference of past population dynamics from molecular sequences. Molecular Biology and Evolution, 22, 1185–1192. https://doi.org/10.1093/molbev/msi103

Drummond, C. S., Xue, H. J., Yoder, J. B., & Pellmyr, O. (2010). Host-associated divergence and incipient speciation in the yucca moth Prodoxus coloradensis (Lepidoptera: Prodoxidae) on three species of host plants. Heredity, 105, 183–196. https://doi.org/10.1038/hdy.2009.154

Engelmann, G. (1872). The flower of yucca and its fertilization. Bulletin of the Torrey Botanical Club, 3, 33.

Excoffier, L., Laval, G., & Schneider, S. (2005). Arlequin (version 3.0): an integrated software package for population genetics data analysis. Evolutionary Bioinformatics Online, 2005, 47–50. https://doi.org/10.1177/117693430500100003

Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. International Journal of Climatology, 37, 4302–4315. https://doi.org/10.1002/joc.5086

Futuyma, D. J. (2000). Some current approaches to the evolution of plant–herbivore interactions. Plant Species Biology, 15, 1–9. https://doi.org/10.1046/j.1442-1984.2000.00029.x

Gloss, A. D., Dittrich, A. C. N., Goldman-Huertas, B., & Whiteman, N. K. (2013). Maintenance of genetic diversity through plant–herbivore interactions. Current Opinion in Plant Biology, 16, 443–450. https://doi.org/10.1016/j.pbi.2013.06.002

Gloss, A. D., Groen, S. C., & Whiteman, N. K. (2016). A genomic perspective on the generation and maintenance of genetic diversity in herbivorous insects. Annual Review of Ecology, Evolution, and Systematics, 47, 165–187. https://doi.org/10.1146/annurev-ecolsys-121415-032220

Leebens-Mack, J., Pellmyr, O., & Brock, M. (1998). Host specificity and the genetic structure of two yucca moth species in a yucca hybrid zone. Evolution, 52, 1376–1382. https://doi.org/10.1111/j.1558-5646.1998.tb02019.x

Leebens-Mack, J., & Pellmyr, O. (2004). Patterns of genetic structure among populations of an oligophagous pollinating Yucca moth (Tegeticula yuccasella). Journal of Heredity, 95, 127–135. https://doi.org/10.1093/jhered/esh025

Lenz, L. W. (1998). Yucca capensis (Agavaceae, Yuccoideae), a new species from Baja California Sur, Mexico. Cactus and Succulent Journal, 70, 289–296.

Liu, S., Jiang, N., Xue, D., Cheng, R., Qu, Y., Li, X. et al. (2016). Evolutionary history of Apocheima cinerarius (Lepidoptera: Geometridae), a female flightless moth in northern China. Zoologica Scripta, 45, 160–174. https://doi.org/10.1111/zsc.12147

Nakamura, H., Teshima, K., & Tachida, H. (2018). Effects of cyclic changes in population size on neutral genetic diversity. Ecology and evolution, 8, 9362–9371. https://doi.org/10.1002/ece3.4436

Papadopoulou, A., Anastasiou, I., & Vogler, A. P. (2010). Revisiting the insect mitochondrial molecular clock: the mid-Aegean trench calibration. Molecular Biology and Evolution, 27, 1659–1672. https://doi.org/10.1093/molbev/msq051

Pellmyr, O. (2003). Yuccas, yucca moths, and coevolution: a review. Annals of the Missouri Botanical Garden, 90, 35–55.

Pellmyr, O., Balcazar-Lara, M., Segraves, K. A., Althoff, D. M., & Littlefield, R. J. (2008). Phylogeny of the pollinating yucca moths, with revision of Mexican species (Tegeticula and Parategeticula; Lepidoptera, Prodoxidae). Zoological Journal of the Linnean Society, 152, 297–314. https://doi.org/10.1111/j.1096-3642.2007.00361.x

Phillips, S. J., Anderson, R. P., & Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190, 231–259. https://doi.org/10.1016/j.ecolmodel.2005.03.026

Rambaut, A., Drummond, A. J., Xie, D., Baele, G., & Suchard, M. A. (2018). Posterior summarization in Bayesian phylogenetics using Tracer 1.7. Systematic Biology, 67, 901–904. https://doi.org/10.1093/sysbio/syy032

Rogers, A. R., & Harpending, H. (1992). Population growth makes waves in the distribution of pairwise genetic differences. Molecular Biology and Evolution, 9, 552–569. https://doi.org/10.1093/oxfordjournals.molbev.a040727

Rozas, J., Sánchez-Del Barrio, J. C., Messeguer, X., & Rozas, R. (2003). DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics, 19, 2496–2497. https://doi.org/10.1093/bioinformatics/btg359

Segraves, K. A., & Pellmyr, O. (2001). Phylogeography of the yucca moth Tegeticula maculata: the role of historical biogeography in reconciling high genetic structure with limited speciation. Molecular Ecology, 10, 1247–1253. https://doi.org/10.1046/j.1365-294X.2001.01275.x

Smith, C. I., Godsoe, W. K., Tank, S., Yoder, J. B., & Pellmyr, O. (2008). Distinguishing coevolution from covicariance in an obligate pollination mutualism: asynchronous divergence in Joshua tree and its pollinators. Evolution, 62, 2676–2687. https://doi.org/10.1111/j.1558-5646.2008.00500.x

Smith, C. I., Drummond, C. S., Godsoe, W., Yoder, J. B., & Pellmyr, O. (2009). Host specificity and reproductive success of yucca moths (Tegeticula spp. Lepidoptera: Prodoxidae) mirror patterns of gene flow between host plant varieties of the Joshua tree (Yucca brevifolia: Agavaceae). Molecular Ecology, 18, 5218–5229. https://doi.org/10.1111/j.1365-294X.2009.04428.x

Smith, C. I., Tank, S., Godsoe, W., Levenick, J., Strand, E., Esque, T. et al. (2011). Comparative phylogeography of a coevolved community: concerted population expansions in Joshua trees and four yucca moths. Plos One, 6, e25628. https://doi.org/10.1371/journal.pone.0025628

Smith, C. I., & Leebens-Mack, J. H. (2024). 150 Years of coevolution research: evolution and ecology of yucca moths (Prodoxidae) and their hosts. Annual Review of Entomology, 69, 375–391. https://doi.org/10.1146/annurev-ento-022723-104346

Tamura, K., Dudley, J., Nei, M., & Kumar, S. (2007). MEGA4: molecular evolutionary genetics analysis (MEGA) software version 4.0. Molecular Biology and Evolution, 24, 1596–1599. https://doi.org/10.1093/molbev/msm092

Turner, R. M., Bowers, J. E., & Burgess, T. L. (1995). Sonoran Desert plants: an ecological atlas. Tucson: University of Arizona Press.

Ensambles de macroalgas en diferentes zonas arrecifales de una playa urbana tropical en Brasil

Caio Ceza da Silva-Nunes ^{a, c, d, e,} *, Edilene Maria dos Santos Pestana ^{b, c}, Cibele Conceição dos Santos ^{b, c}, Lorena Pedreira Conceição ^{a, c}, José Marcos de Castro Nunes ^{a, c} 

^a Universidade Estadual de Feira de Santana, Departamento de Ciências Biológicas, Programa de Pós-Graduação em Botânica, Av. Transnordestina s/n, Novo Horizonte, CEP 44036-900, Feira de Santana, Bahia, Brazil

^b Universidade Federal da Bahia, Instituto de Biologia, Programa de Pós-Graduação em Biodiversidade e Evolução, Rua Barão de Jeremoabo, 668 – Campus de Ondina, CEP: 40170-115, Salvador, Bahia, Brazil

^c Universidade Federal da Bahia, Instituto de Biologia, Laboratório de Algas marinhas, Rua Barão de Jeremoabo, 668 – Campus de Ondina, CEP: 40170-115, Salvador, Bahia, Brazil 

^d Universidade Estadual do Sudoeste da Bahia, Departamento de Ciências Exatas e Naturais, Programa de Pós-Graduação em Ciências Ambientais, Campus Universitário “Juvino Oliveira”, BR 415, Km 04 CEP: 45.700-000 Itapetinga, Bahia, Brazil

^e Colégio Estadual de Tempo Integral Antônio Batista, Rua Presidente Vargas, 83 CEP: 46380-076 Candiba, Bahia, Brazil

*Corresponding author: caiobio08@gmail.com (C.C. da Silva-Nunes)

Received: 26 February 2025; accepted: 01 August 2025

Abstract

Marine macroalgae are commonly used as bioindicators because they are sensitive to environmental changes. This study aims to verify the composition of macroalgae in the intertidal region in 3 reef zones on Itapuã beach, located in Salvador-Bahia, Brazil and which presents high tourist activity. The samples were obtained in July 2017, in the intertidal zone, in 3 reef regions: Protected Region (PR), Tide Pool Region (TP) and Frontal Region (FR). In each zone, 3 transects (20 m each) were placed, in which 5 squares measuring 20 x 20 cm were arranged at random points. Additionally, individuals were collected around each transect for qualitative analysis. Dry biomass was measured, and statistical tests were carried out to obtain diversity, equitability and similarity data. Fifty-two taxa were identified, 27 Rhodophyta, 16 Chlorophyta and 9 Phaeophyceae. TP had the highest species richness (34) and diversity; however, no significant differences were found in macroalgal biomass between the 3 reef zones analyzed. This study contributes to the understanding of the composition and structure of phytobenthic communities in intertidal regions of the Bahian coast.

Keywords: Bahia; Bioindicator; Brazil; Phytobenthos

Resumen

Las macroalgas marinas se utilizan comúnmente como bioindicadores porque son sensibles a los cambios ambientales. Este estudio tiene como objetivo verificar la composición de macroalgas en la zona intermareal en 3 regiones arrecifales de la playa de Itapuã, ubicada en Salvador-Bahía, Brasil y que tiene alta actividad turística. Las muestras fueron obtenidas en julio de 2017, en la zona intermareal, en 3 regiones arrecifales: región posterior (PR), pozas de marea (TP) y región frontal (FR). En cada región, se colocaron 3 transectos (de 20 m cada uno), donde se distribuyeron aleatoriamente 5 cuadrantes de 20 x 20cm. Además, se recolectaron individuos alrededor de cada transecto para análisis cualitativo. Se midió la biomasa seca y se realizaron pruebas estadísticas para obtener datos de diversidad, equitabilidad y similitud. Se identificaron 52 taxones: 27 Rhodophyta, 16 Chlorophyta y 9 Phaeophyceae. La TP presentó la mayor riqueza de especies (34) y diversidad; sin embargo, no se encontraron diferencias significativas en la biomasa de macroalgas entre las 3 regiones arrecifales. Este estudio contribuye a la comprensión de la composición y estructura de las comunidades fitobentónicas en zonas intermareales del litoral bahiano.

Palabras clave: Bahia; Bioindicador; Brasil; Fitobentos

Introduction

The State of Bahia has the longest coastline in Brazil, 1,103 km showing a great diversity of environments: sandy beaches, coral reefs, sandstone formations, rocky shores and mangroves (Nunes & Paula, 2004a). Furthermore, it is a region that presents a great diversity of substrates and geographical features, which is reflected in the diversity of marine flora (Nunes, 2005b). Nunes and Paula (2002) divided reef formations into 3 zones: Frontal Region (FR), Protected Region (PR) and Tide Pool Region (TP). The FR is a very hydrodynamic region, where direct collision of waves occurs. The PR is the region before the lagoon and after the reef top, it may have pools and is a region protected from direct wave collision. The TP is deep and may suffer greater wave action, forming pools on the reef top, or may suffer less or no wave action, forming pools on the reef plateau.

In recent decades, coastal areas have been undergoing an intense process of urban development, which has caused significant environmental pressures and impacts, especially in benthic communities in reef formations (Costa et al., 2012; Nascimento, 2013). In Salvador, these impacts come from urban expansion, real estate speculation, tourism, human activities, such as fishing and trampling, causing the degradation of the flora and a reduction in the richness and diversity of species, especially in the composition and structure of algal communities. 

Marine macroalgal communities have a great ecological role, being, together with microalgae, at the base of the food chain as primary producers, being a source of food for a large part of the marine fauna. The presence of macroalgae along the coast is responsible for softening the impact of waves on the sea coast, and also plays the role of habitat for other organisms, such as animals or as a substrate for algae (Nunes, 2010; Pedrine, 2013).

When these organisms are exposed to some human interference, they are sensitive to changes in their habitat, such as the increase in the concentration of organic matter in the water, which leads to an increase in the biomass of some algal groups, or even the depletion of some nutrient that causes the disappearance of a certain species, working as bioindicators of environmental quality. In addition, the degradation of this community’s structure favors the emergence and persistence of more resistant and opportunistic species, as well as the exclusion of more fragile species (Nascimento, 2013; Nunes, 2010).

Functional groups of algae are based on similarities in their morphological and anatomical characteristics, in addition to their ecological characteristics. Steneck and Dethier (1994) grouped algae into 7 categories, these being microalgae, filamentous, foliaceous, cylindrical-corticate, coriaceous, articulated calcareous and encrusting. In the model proposed by Steneck and Dethier (1994), it is highlighted that environments of high productivity and low disturbance present high biomass and diversity of morphofunctional groups, providing an abundance of coriaceous and cylindrical-cortical algae, as they have a relatively large size and a longer life cycle. 

Knowledge about macroalgae on the coast of Bahia has been expanded through taxonomic studies, with the metropolitan area of Salvador being one of the main regions of the State’s coast studied, for example: Altamirano and Nunes (1997), Amorin et al. (2006), Barreto et al. (2004), Macedo et al. (2009), Marins et al. (2008), Nunes (1998a, b, 1999, 2005a, b), Nunes and Guimarães (2008, 2009, 2010), Nunes and Paula (2000, 2001, 2002, 2004a, b, 2006), Nunes et al. (2005). 

Studies regarding the community structure of marine macroalgae have been carried out on the Brazilian coast. The majority are concentrated in the subtidal zone, namely, Amado Filho et al. (2003), Figueiredo et al. (2004), Villaça et al. (2010) in Rio de Janeiro, Oliveira-Carvalho et al. (2003) in Pernambuco, and Horta et al. (2008) in Santa Catarina. And in the intertidal zone, Barbosa et al. (2008) in Espírito Santo, and Muñoz and Pereira (1997) in Pernambuco. In Bahia, there are studies by Caires et al. (2013), Costa et al. (2012), and Ferreira et al. (2022), who carried out studies in the intertidal zone, and Costa Jr. et al. (2002) and Marins et al. (2008) in the subtidal zone. Although the coast of Bahia is the longest in Brazil, studies aimed to understand the structure of phytobenthic communities are scarce, creating a gap in the knowledge of these communities.

The present study analyzes the intertidal phytobenthic community in 3 reef zones of Itapuã beach, aiming to identify and compare differences in the composition and structure of the species between the studied areas, using the Feldmann and Cheney indices as bioindicators of local environmental characteristics.

Materials and methods

Itapuã beach is located in Salvador, Bahia in Brazil (Fig. 1), and corresponds to 2 large rocky bodies, and has coarse sand beaches, many reefs and rocky bollards (Nunes, 1998b). The collection was carried out on July 11, 2017, in the intertidal region, during low spring tide. The intertidal region was compartmentalized into the 3 regions proposed by Nunes and Paula (2002). Nunes (2010) was followed for the collection protocol. 

Three 20 m transects were placed in each zone and arranged parallel to the coastline. In each transect, 5 squares measuring 20 x 20 cm were previously drawn. All material contained in the square was removed using a spatula. For qualitative analysis, specimens were collected around each transect. The collected material was placed in plastic bags or polyethylene bottles, fixed with 4% formalin. All material from the collections was taken to the Marine Algae Laboratory (LAMAR), at the Biology Institute of the Federal University of Bahia, where it was analyzed using a stereomicroscope (Leica© – Zoom 2000) to separate and identify the epiphytic algae.

Taxon identification was done using an Olympus© – CX 22 microscope. Blades were assembled from freehand cuts, made with the aid of razor blades and box cutters. Calcareous algae were analyzed based on prior decalcification by immersing the samples in 0.6 M nitric acid. For specific identification, references commonly used in phycology were used (Dawes & Mathieson, 2008; Littler & Littler, 2000; Nunes, 2005b; Nunes & Guimarães, 2008; Nunes & Paula, 2000, 2001, 2002). For the arrangement of taxa, the taxonomic arrangement of Guiry and Guiry (2025) was mainly followed.

For the analysis of dry biomass, the infrageneric taxa were previously identified, separated and dried in an oven at 60 ºC for 48 hours and weighed until constant biomass was obtained. The results are presented in grams/m² of dry weight. The results for each reef zone were compared regarding the total number of taxa, and dry biomass using Shannon-Wiener diversity (H’) and Pielou equitability (J) indexes.

Using the macroalgae biomass data found in each reef zone, the Kruskal-Wallis test was performed to check whether there was a significant difference between the biomass recorded in the different zones. The Bray-Curtis similarity index was also calculated with the biomass data and Non-metric Multidimensional Scaling (nMDS) was performed using the transformed data. ANOSIM was performed for values in which there was a significant difference between the similarity results. The analyses were carried out using SigmaPlot 12 and Primer V6 software.

The species dominance index is used to assess how much one or a few species stand out in relation to the others within a biological community. It considers the proportion of individuals or the biomass of each species, allowing us to identify which of them exert the greatest influence on the structure and functioning of the ecosystem. The higher the value of the index, the greater the dominance, which indicates that a few species concentrate the majority of the individuals present. On the other hand, lower values reflect a more balanced and diverse community, with a lower concentration of individuals in a few species (Melo, 2008). 

Simpson’s dominance index was used, calculated as D = ∑ (ni​/N) 2, where ni is the number of individuals of a species and N is the total number of individuals of all species in the community. The value of D varies between 0 and 1, with values close to 1 indicating high dominance, that is, few species dominate the environment, while values close to 0 indicate greater diversity and balanced distribution among species.

The Feldmann index (obtained by dividing the number of species of Rhodophyta by that of Phaeophyceae [R/P]) and the Cheney index (adding the number of species of Rhodophyta to that of Chlorophyta, and dividing this value by the number of browns [R + C/ P]) (Figueiredo et al., 2008).

Results

Fifty-two taxa were identified, 27 (52) belonging to the Rhodophyta, 16 (31) Chlorophyta, and 9 (17) Phaeophyceae (Heterokontophyta). Table 1 presents the taxa, their distribution throughout the 3 zones of the reef and their respective morphotypes.

Among the zones, TP presented the greatest species richness (34) with Rhodophyta as the most representative group, 17 taxa, followed by Chlorophyta (11) and Phaeophyceae (6). Chaetomorpha minima, Gayliella dawsonii, Gracilaria ferox, Lejolisia mediterranea, Ptilothamnion speluncarum, and Stylonema alsidii were unique to this environment. PR was second in terms of species richness, such as Caulerpa chemnitzia, Caulerpa sertularioides, Cladophora corallinicola, Hypnea pseudomusciformis, Hypnea sp. 1, and Melanothamnus gorgoniae only occurred in this region. FR presented the lowest richness among the environments, such as Ceramium corniculatum, Codium taylori, Dictyopteris jamaicensis, Dictyopteris polypodioides, Sphacelaria tribuloides, and Wrangelia argus, all exclusive for the region. However, it is worth highlighting a greater representation of rhodophytes for the protected and FR. 

Among the species found, 16 occurred in all regions such as Amansia multifida, Amphiroa anastomosans, Anadyomene stellata, Bryopsis pennata, Colpomenia sinuosa, Crouania attenuata, Dictyopteris delicatula, Dictyosphaeria versluysii, Dictyota mertensii, Gelidiella acerosa, Halimeda opuntia, Jania pedunculata var. adhaerens, Palisada perforata, Phyllodictyon anastomosans, Ulva flexuosa, and Ulva rigida.

Regarding the morphotypes, among the 52 taxa found, 19 were filamentous, 17 cylindrical-corticates, 9 foliaceous, 3 articulated calcareous, 3 encrusting and 1 coriaceous. Cylindrical-corticate algae were well represented in the 3 regions, predominating in the PR, followed by foliaceous algae; in the TP, the filamentous morphotype was the one with the highest representation, followed by cylindrical-corticated and foliaceous ones, the same pattern was observed in the FR. 

The diversity of analysis based on the Shannon-Wiener index, showed that the TP is more diverse than the others, followed by the PR, and finally by the FR. Dominance values are inversely proportional to diversity values, which shows that in the FR, there is dominance of one or a few species over the others (Table 2).

In table 3, biomass data and the percentage of contribution of each macroalgae in the zone biomass are presented. The average total biomass of each zone was 108.01 g.m-2 in the FR, 48.83 g.m-2 in tide pool and 474.59 g.m-2 in the PR. The Kruskal-Wallis test showed that no significant differences were found in macroalgal biomass between the three reef zones analyzed (Kruskal-Wallis p = 0.869). However, the ordering of the sampling points according to the presence/absence of macroalgae (Fig. 2) shows the dispersion of samples in the 3 zones, mainly samples from the PR. The ANOSIM analysis (Table 4) shows that there was a significant difference between the FR and the TP, and between the FR and the PR, but there was no significant difference between the PR and the TP. The Feldmann and Cheney indices (Fig. 3) for the 3 reef regions studied were consistent with the flora of tropical and warm temperate regions. 

Table 1

List of taxa found in the reef zones on Itapuã beach and their respective morphotypes: PR = Protected Region; TP = Tide Pool; FR = Frontal Region. Morphotypes: FT = filamentous; FC = foliaceous; CC = cylindrical corticates; CR = coriaceous; AC = articulated calcareous; E = encrusting. Presence (+) and absence (–).

Taxa	PR	TP	FR	Morphotype
Chlorophyta
Ulvophyceae
Bryopsidales
Bryopsidaceae
Bryopsis pennata J.V. Lamouroux	+	+	+	FT
Caulerpaceae
Caulerpa chemnitzia (Esper) J.V. Lamouroux	+	–	–	CC
C. racemosa (Forsskål) J. Agardh	+	+	+	CC
C. sertularioides (S.G. Gmelin) M. Howe	+	–	–	CC
Codiaceae
Codium taylorii P.C. Silva	–	–	+	CC
Halimedaceae
Halimeda opuntia (Linnaeus) J.V. Lamouroux	+	+	+	AC
Ulva flexuosa Wulfen	+	+	+	FC
U. rigida C. Agardh	+	+	+	FC
Cladophorales
Anadyomenaceae
Anadyomene stellata (Wulfen) C. Agardh	+	+	+	FC
Boodleaceae
Phyllodictyon anastomosans (Harvey) Kraft & M.J. Wynne	+	+	+	FT
Cladophoraceae
Chaetomorpha minima Collins & Hervey	–	+	–	FT
Cladophora corallinicola Sonder	+	–	–	FT
C. laetevirens (Dillwyn) Kützing	–	+	+	FT
C. vagabunda (Linnaeus) Hoek	+	–	+	FT
Siphonocladaceae
Dictyosphaeria versluysii Weber Bosse	+	+	+	E
Valoniaceae
Valonia aegagropila C.Agardh	+	+	–	CC
Heterokontophyta
Phaeophyceae
Dictyotales
Dictyotaceae
Dictyopteris delicatula J.V. Lamouroux	+	+	+	FC
D. jamaicensis W.R. Taylor	–	–	+	FC
D. polypodioides (De Candolle) J.V. Lamouroux	–	–	+	FC
Dictyota mertensii (C.Martius) Kützing	+	+	+	FC
Table 1. Continued
Taxa	PR	TP	FR	Morphotype
Padina antillarum (Kützing) Piccone	+	+	–	FC
Spatoglossum schroederi (C. Agardh) Kützing	–	+	+	FC
Ectocarpales
Scytosiphonaceae
Colpomenia sinuosa (Mertens ex Roth) Derbès & Solier	+	+	+	E
Fucales
Sargassaceae
Sargassum polyceratium Montagne	–	+	+	CR
Sphacelariales
Sphacelariaceae
Sphacelaria tribuloides Meneghini	–	–	+	FT
Rhodophyta
Florideophyceae
Ceramiales
Callithamniaceae
Crouania attenuata (C. Agardh) J. Agardh	+	+	+	FT
Ceramiaceae
Ceramium corniculatum Montagne	–	–	+	FT
Centroceras clavulatum (C. Agardh) Montagne	+	–	–	FT
Gayliella dawsonii (A.B. Joly) Barros-Barreto & F.P. Gomes	–	+	–	FT
Ceramothamnion brasiliensis (A.B. Joly) M.J. Wynne & C.W. Schneider	+	–	+	FT
Rhodomelaceae
Alsidium triquetrum (S.G. Gmelin) Trevisan	–	+	–	CC
Amansia multifida J.V. Lamouroux	+	+	+	CC
Digenea simplex (Wulfen) C. Agardh	+	+	–	CC
Herposiphonia bipinnata M. Howe	–	+	–	FT
H. tenella (C. Agardh) Ambronn	–	+	–	FT
Melanothamnus gorgoniae (Harvey) Díaz-Tapia & Maggs	+	–	–	FT
Palisada perforata (Bory) K.W. Nam	+	+	+	CC
Wrangeliaceae
Lejolisia mediterranea Bornet	–	+	–	FT
Ptilothamnion speluncarum (Collins & Hervey) D.L. Ballantine & M.J. Wynne	–	+	–	FT
Wrangelia argus (Montagne) Montagne	–	–	+	FT
Corallinales
Corallinaceae incrustante
Jania pedunculata var. adhaerens (J.V. Lamouroux) A.S. Harvey, Woelkerling & Reviers	+	+	+	AC
Corallinaceae incrustante	+	–	+	E
Lithophyllaceae
Table 1. Continued
Taxa	PR	TP	FR	Morphotype
Amphiroa anastomosans Weber-van Bosse	+	+	+	AC
Gelidiales
Gelidiaceae
Gelidium capense (S.G. Gmelin) P.C. Silva	+	–	+	CC
Gelidiellaceae
Gelidiella acerosa (Forsskål) Feldmann & Hamel	+	+	+	CC
G. ligulata E.Y. Dawson	+	–	+	CC
Gigartinales
Cystocloniaceae
Hypnea pseudomusciformis Nauer, Cassano & M.C. Oliveira	+	–	–	CC
Hypnea sp.1	+	–	–	CC
Gigartinaceae
Chondracanthus acicularis (Roth) Fredericq	+	+	–	CC
Gracilariales
Gracilariaceae
Gracilaria ferox J. Agardh	–	+	–	CC
Nemaliales
Galaxauraceae
Galaxaura rugosa (J. Ellis & Solander) J.V. Lamouroux	–	+	+	CC
Stylonematophyceae
Stylonematales
Stylonemataceae
Stylonema alsidii (Zanardini) K.M. Drew	–	+	–	FT
Total	33	34	32

Discussion

Rhodophyta was the most representative group in the present study, followed by Chlorophyta and Phaeophyceae. This pattern has been observed in other studies on phytobenthic macroalgal communities as typical on the Brazilian coast (Braga et al., 2014; Costa et al., 2012; Ferreira et al., 2022). It should be noted that rhodophytes tend to dominate the environment in the absence of large brown algae in tropical environments, suggesting that there is competition between algae that use different strata in the phytobenthic community (Figueiredo et al., 2004). Furthermore, studies have confirmed the decline of Phaeophyceae at the same rate that Chlorophyta species have been increasing, which can be attributed to the negative effects of anthropogenic impact on the coastal environment (Oliveira & Qi, 2003; Scherner et al., 2013). Representatives of Phaeophyceae are more sensitive to pollutants (heavy metals and excess organic matter produced by sewage), which can even negatively affect the germination and cell division of these organisms (Kevekordes, 2001). Representatives of Chlorophyta, such as species of Ulva Linnaeus, have the capacity of benefit from contamination by pollutants, being considered opportunistic species (Scherner et al., 2012).

Table 2

Shannon-Wiener Diversity index, dominance and equitability of macroalgae in the intertidal region of Itapuã beach.

Region	Shannon-Wiener	Dominance	Equability (J’)
Frontal Region	3.466	0.0312	1
Tide Pool	3.526	0.0294	1
Protected Region	3.497	0.0303	1

The taxa that contributed most to the average biomass of macroalgae found on Itapuã beach were Amphiroa anastomosans, Digenea simplex, Gelidiella acerosa, and Sargassum polyceratium. It has been previously reported that algae with greater structural complexity, such as corticate algae, are better adapted to environments with greater hydrodynamics and luminosity, as this morphotype confers greater resistance to desiccation (Costa et al., 2012; Villaça et al., 2008).

Table 3

Biomass and percentage of contribution of macroalgae species in the constitution of biomass in each of the zones studied: FR = Frontal Region, TP = Tide Pool, and PR = Protected Region. Pi = Percentage of importance.

Taxa	FR (g.m-2)	Pi	TP (g.m-2)	Pi (%)	PR (g.m-2)	Pi (%)
Amphyroa anastomosans	70.07	64.87	15.76	32.28	31.59	6.66
Anadyomene stellata	1.41	1.31	1.76	3.61	0.00	0.00
Bryopsis pennata	0.00	0.00	0.20	0.42	0.00	0.00
Caulerpa racemosa	1.42	1.32	0.00	0.00	0.00	0.00
C. sertularioides	0.00	0.00	0.00	0.00	1.11	0.23
Centroceras clavulatum	0.00	0.00	0.00	0.00	0.25	0.05
Chondracanthus acicularis	0.00	0.00	0.00	0.00	0.00	0.00
Cladophora vagabunda	0.00	0.00	0.00	0.00	0.15	0.03
Colpomenia sinuosa	0.16	0.15	0.08	0.15	0.00	0.00
Dictyopteris delicatula	5.42	5.01	5.48	11.22	0.00	0.00
D. jamaicensis	0.00	0.00	0.00	0.00	0.00	0.00
Dictyosphaeria versluysii	0.67	0.62	0.25	0.51	0.09	0.02
Dictyota mertensii	0.00	0.00	0.00	0.00	1.13	0.24
Digenea simplex	0.00	0.00	0.97	1.98	430.92	90.80
Gelidiellaceae	0.00	0.00	2.00	4.09	0.00	0.00
Gelidium sp. 1	0.00	0.00	0.94	1.92	0.19	0.04
Gelidium sp. 2	0.00	0.00	0.00	0.00	0.09	0.02
Gelidiella acerosa	14.27	13.21	5.97	12.24	2.69	0.57
G. ligulata	0.71	0.66	0.00	0.00	0.00	0.00
Halimeda opuntia	0.31	0.29	2.14	4.38	0.48	0.10
Hypnea sp. 1	0.40	0.37	0.00	0.00	0.00	0.00
Hypnea sp. 2	0.00	0.00	0.00	0.00	0.00	0.00
Jania pedunculata var. adhaerens	7.60	7.03	0.00	0.00	0.27	0.06
Lobophora variegata	0.00	0.00	0.39	0.80	0.00	0.00
Padina antillarum	0.00	0.00	2.26	4.64	1.15	0.24
Palisada perforata	4.86	4.50	0.00	0.00	0.05	0.01
Phyllodictyon anastomosans	0.00	0.00	0.00	0.00	0.00	0.00
Sargassum polyceratium	0.00	0.00	9.69	19.85	0.00	0.00
Spatoglossum schroederi	0.10	0.09	0.00	0.00	0.00	0.00
Ulva rigida	0.63	0.58	0.83	1.69	3.55	0.75
Valonia aegagropila	0.00	0.00	0.11	0.22	0.86	0.18
Total	108.01		48.83		474.59

The difference found in macroalgae biomass between reef zones can be explained by the hydrodynamics to which they are exposed on the reef, being wave action the main responsible for the spatial distribution of the community on the reef. The movement of waves helps to obtain nutrients, favoring the productivity of macroalgae and thus increasing biomass, however this only applies to those algae that have adaptations to resist the impact of waves, as they are also responsible for removing macroalgae from the substrate (Diez et al., 2003; Hurd et al., 1996; Leigh et al., 1987).

The highest biomasses were found in FR and PR. The FR is constantly submerged and this favors the establishment of algae, as they are submerged for longer, suffer less desiccation and are less exposed to direct sunlight. The lowest biomass values were recorded in the TP, a region that retains water during low tide, however the algae are subjected to intense solar radiation, high salinity and water temperature. The negative effects of these stressors result in lower biomass and also there is less hard substrate for fixation (Costa et al., 2012; Villaça et al., 2010). 

Table 4

Results of ANOSIM tests for significance of differences between sample groups based on observed macroalgal species distribution data.

Reef regions	(Global R= 0.613;  significance level = 0.4%)
Region pairs	Statistical R	Level
FR, TP	0.815	10
FR, PR	0.667	10
TP, PR	0.481	20

The dominance of filamentous, cylindrical-corticate and foliaceous morphotypes was observed, corroborating the model by Orfanidis et al. (2001), which proposes that impacted environments should have a greater abundance of algae with intensely branched (corticated), laminar (foliaceous) and filamentous thallus, characterized by high growth rates and a short life cycle (annual). However, conserved environments would have an abundance of algae with thick thallus (coriaceous), articulated calcareous and encrusting, cacterized by low growth rates and a long life cycle (perennial), which reflected in the low representation of articulated, encrusting and coriaceous calcareous morphotypes. In a study carried out by Nascimento (2013) in Salvador and the North Coast of Bahia, the foliaceous and cylindrical-corticate morphotypes were more representative in both impacted and preserved environments, highlighting the predominance of these morphotypes.

The distribution of algae according to morphofunctional groups reflects the conditions of reef zones, since each taxa has its adaptive characteristics. Cylindrical-cortical algae were well represented in the 3 reef zones, with a predominance in the PR, as this morphology confers greater resistance to desiccation, helping these algae to establish themselves in different environments (Costa et al., 2012; Villaça et al., 2008). Tide pool region was dominated by filamentous algae, morphotype usually associated with benthic communities in early successional stages as they show rapid growth (Braga et al., 2014).

Regarding phytogeography, the Feldmann (F) and Cheney (C) indices found for the different regions of the Itapuã reef characterized the PR and the TP as tropical and the FR as warm temperate, according to the classification proposed by Horta et al. (2001). Factors such as the greater hydrodynamism in the FR may justify this classification in relation to its algal community, since the increase in the number of species of brown algae is noticeable as one approaches the FR. Bouzon et al. (2006), studying floristic and phytogeographic aspects of marine macroalgae in the Bays of Santa Catarina, attributed the variations in the Feldmann and Cheney indices to the exclusion of species of brown algae and the favoring of opportunistic species of red and green algae; in this study, lower values of these indices were recorded in the frontal zones and in the pool zone, where the richness of species of brown algae was greater.

Understanding the composition and structure of phytobenthic communities on the Brazilian coast, as well as on the coast of Bahia is extremely important. Despite the vast coastal extension, there are few studies referring to the algal community structure in intertidal zone as most studies focus on the subtidal zone. Also, most studies on the composition and structure of phytobenthic communities are restricted to the state of Rio de Janeiro. More studies should be carried out along the coast of Bahia to fill the gaps and alternating dry and rainy periods and/or comparing the reef zones of different beaches in the intertidal zone. The results suggest that the Itapuã reefs are in a good state of conservation.

Acknowledgements

CCSN, acknowledges the post-doctoral scholarship to the State University of Southwest Bahia (UESB) – Notice 266/2023.; CCSN, LPC, EMSP, to the Coordination for the Improvement of Higher Education Personnel – Brazil (CAPES), finance code 001; CCS, to the Bahia Research Support Foundation (FAPESB – T.O. B., No. BOL0416/2017) for a scholarship.; JMCN, to the National Council for Scientific and Technological Development (CNPq), Brazil for the research productivity fellowship (308261/2022-4).

References 

Altamirano, M., & Nunes, J. M. C. (1997). Contribuciones al macrofitobentos del Município de Camaçari (Bahia, Brasil). Acta Botanica Malacitana, 22, 211–215. https://doi.org/10.24310/abm.v22i0.8639 

Amado Filho, G. M., Barreto, M. B. B., Marins, B. V., Felix, C., & Reis, R. P. (2003). Estrutura das comunidades fitobentônicas do infralitoral da Baía de Sepetiba, RJ, Brasil. Revista Brasileira de Botânica, 26, 329–342. https://doi.org/10.1590/s0100-84042003000300006 

Amorin, P. R. R., Moura, C. W. N., & Moniz-Brito, K. L. (2008). Estudo morfotaxonômico das espécies de Halimeda, Penicillus e Udotea (Bryopsidales, Chlorophyta) do Recife de Franja da Ilha de Itaparica, Bahia. Anais do XI Congresso Brasileiro de Ficologia & Simpósio Latino-Americano sobre Algas Nocivas. Rio de Janeiro: Museu Nacional. Série Livros. 

Barbosa, S. O., Figueiredo, M. A. O., & Testa, V. (2008). Structure and dynamic of benthic communities dominated by macrophytes in praia de Jacaraípe, Espírito Santo, Brazil. Hoehnea, 35, 563–575. https://doi.org/10.1590/S2236-89062008000400008 

Barreto, M. B. B., Brasileiro, P. S., Nunes, J. M. C., & Amado-Filho, G. M. (2008). Algas marinhas bentônicas do sublitoral das formações recifais da Baía de Todos os Santos, BA – 1. Novas ocorrências. Hoehnea, 31, 321–330. 

Braga, A. C. S., Tâmega, F. T. S., Pedrini, A. G., & Muniz, R. A. (2014). Composição e estrutura da comunidade fitobentônica do infralitoral da praia de Itaipu, Niterói, Brasil: Subsídios para monitoramento e conservação. Iheringia. Série Botânica, 69, 267–276. https://isb.emnuvens.com.br/iheringia/article/view/90

Bouzon, J. L., Salles, J. P., Bouzon, Z., & Horta, P. A. (2006). Aspectos florísticos e fitogeográficos das macroalgas marinhas das baías da ilha de Santa Catarina, SC,Brasil. Insular, 35, 69–84. 

Caires, T. A., Costa, I. O., Jesus, P. B. D., Matos, M. R. B. D., Pereira-Filho, G. H., & Nunes, J. M. D. C. (2013). Evaluation of the stocks of Hypnea musciformis (Rhodophyta: Gigartinales) on two beaches in Bahia, Brazil. Brazilian Journal of Oceanography, 61, 65–71. https://doi.org/10.1590/S1679-87592013000100007 

Costa-Júnior, O. S., Attrill, M. J., Pedrini, A. G., & De Paula, J. C. (2002). Spatial and seasonal distribution of seaweeds on coral reefs from Southern Bahia, Brazil. Botanica Marina, 45, 346–355. https://doi.org/10.1515/BOT.2002.035 

Costa, I. O., Caires, T. A., Pereira-Filho, G. H., & Nunes, J. M. C. (2012). Macroalgas bentônicas associadas a bancos de Hypnea musciformis (Wulfen) J.V. Lamour. (Rhodophyta-Gigartinales) em duas praias do litoral baiano. Acta Botanica Brasilica, 26, 493–507. https://doi.org/10.1590/S0102-33062012000200025 

Dawes, C., & Mathieson, A. (2008). The seaweeds of Florida. Gainesville: University of Florida Press. 

Diez, I., Santolaria, A., & Gorostiaga, J. M. (2003). The relationship of environmental factors to the structure and distribution of subtidal seaweed vegetation of the western Basque coast (N Spain). Estuarine, Coastal and Shelf Science, 56, 1041–1054. https://doi.org/10.1016/S0272-7714(02)00301-3 

Ferreira, S. M. C., Lolis, L. A., Noga, P. M., Affe, H. M. J., & Nunes, J. M. C. (2022). A highly diverse phytobenthic community along a short coastal reef gradient in northeastern Brazil. Universitas Scientiarum, 27, 34–56. https://doi.org/10.11144/Javeriana.SC271.ahdp

Figueiredo, M. A. O., Barreto, M. B. B., & Reis, R. P. (2004). Caracterização das macroalgas nas comunidades marinhas da área de Proteção Ambiental de Cairuçú, Parati, RJ – Subsídios para futuros monitoramentos. Revista Brasileira de Botânica, 27, 11–17. https://doi.org/10.1590/S0100-84042004000100002

Figueiredo, M. A. O., Horta, P.A., Pedrini, A. G., & Nunes, J. M. C. (2008). Benthic marine algae of the coral reefs of Brazil: a literature review. Oecologia Brasiliensis, 12, 259–270. https://doi.org/10.4257/oeco.2008.1202.07 

Guiry, M. D., & Guiry, G. M. (2025). AlgaeBase. World-wide electronic publication, University of Galway. Retrieved on February 21, 2025 from: https://www.algaebase.org

Horta, P. A., Amancio, E., Coimbra, C. S., & Oliveira, E. C. (2001). Considerações sobre a distribuição e origem da flora de macroalgas marinhas brasileiras. Hoehnea, 28, 243–265. 

Horta, P. A., Salles, J. P., Bouzon, J., Scherner, F., Cabral, D., Bouzon, Z. L. et al. (2008). Composição e estrutura dos fitobentos do infralitoral da Reserva Biológica Marinha do Arvoredo, Santa Catarina, Brasil —implicações para a conservação. Oecologia Brasiliensis, 12, 51–57. https://doi.org/10.4257/oeco.2008.1202.06 

Hurd, C. L., Harrison, P. J., & Druehl, L. D. (1996). Effect of seawater velocity on inorganic nitrogen uptake by morphologically distinct forms of Macrocystis integrifolia from wave-sheltered and exposed sites. Marine Biology, 126, 205–214. https://doi.org/10.1007/BF00347445

Kevekordes, K. (2001). Toxicity tests using developmental stages of Hormosira banksii (Phaeophyta) identify ammonium as a damaging component of secondary treated sewage effluent discharged into Bass Strait, Victoria, Australia. Marine Ecology Progress Series, 219, 139–148. https://doi.org/10.3354/meps219139

Leigh, E. G., Paine, R. T., Quinn, J. F., & Suchanek, T. H. (1987). Wave energy and intertidal productivity. Proceedings of the National Academy of Sciences, 84, 1314–1318. https://doi.org/10.1073/pnas.84.5.1314 

Littler, D. S., & Littler, M. M. (2000). Caribbean reef plants. An identification guide to the reef plants of the Caribbean, Bahamas, Florida and Gulf of Mexico. Washington D.C.: OffShore Graphics.

Macedo, T. S., Varjão, A. S., Fernandes, L. L., Silva, D. F., Almeida, J. S., Messias, J. C. et al. (2009). Levantamento taxonômico e diversidade das macroalgas marinhas bentônicas da praia da Pituba, Salvador, Bahia. Revista Eletrônica de Biologia, 2, 29–39. 

Marins, B. V., Brasileiro, P. A., Barreto, M. B. B., Nunes, J. M. C., Yoneshigue-Valentin, Y., & Amado-Filho, G. M. (2008). Subtidal benthic marine algae of the Todos os Santos Bay, Bahia State, Brazil. Oecologia Brasiliensis, 12, 229–242. https://doi.org/10.4257/oeco.2008.1202.05 

Melo, A. S. (2008). O que ganhamos “confundindo” riqueza de espécies e equabilidade em um índice de diversidade? Biota Neotropica, 8, 21–27. https://doi.org/10.1590/s1676-06032008000300001

Muñoz, A. O. M., & Pereira, S. M. B. (1997). Caracterização quali-quantitativa das comunidades de macroalgas nas formações recifais da praia do Cupe Pernambuco, Brasil. Trabalhos Oceanográficos da Universidade Federal do Pernambuco, 25, 93–109. https://doi.org/10.5914/tropocean.v25i1.2731 

Nascimento, O. S. (2013). Impactos da urbanização sobre a estrutura, cobertura de morfotipos funcionais e heterogeneidade química das comunidades de algas
recifais (M.Sc. Thesis). Instituto de Biologia, Universidade Federal da Bahia, Salvador. 

Nunes, J. M. C. (1998a). Catálogo de algas marinhas bentônicas do Estado da Bahia. Acta Botanica Malacitana, 23, 5–21. https://doi.org/10.24310/abm.v23i0.8547 

Nunes, J. M. C. (1998b). Rodofíceas marinhas bentônicas da orla oceânica de Salvador, Estado da Bahia, Brasil. Insula, 49, 27–37. 

Nunes, J. M. C. (1999). Phaeophyta da Região Metropolitana de Salvador, Bahia, Brasil (M.Sc. Thesis). Instituto de Biociências, Universidade de São Paulo, São Paulo. 

Nunes, J. M. C. (2005a). A família Liagoraceae (Rhodophyta, Nemaliales) no Estado da Bahia, Brasil. Hoehnea, 32, 429–444. 

Nunes, J. M. C. (2005b). Rodofíceas marinhas bentônicas do Estado da Bahia, Brasil (Ph.D. Thesis). Instituto de Biociências, Universidade de São Paulo, São Paulo. 

Nunes, J. M. C. (2010). Taxonomia morfológica: metodologia de trabalho. In A. G. Pedrini (Org.), Macroalgas: uma introdução à Taxonomia (pp. 53–70). Rio de Janeiro, Rio de Janeiro: Technical Books. 

Nunes, J. M. C., & Guimarães, S. M. P. B. (2008). Novas referências de rodofíceas marinhas bentônicas para o litoral brasileiro. Biota Neotrópica, 8, 89–100. https://doi.org/10.1590/S1676-06032008000400008

Nunes, J. M. C., & Guimarães, S. M. P. B. (2009). Primeira referência de plantas gametofíticas em Spermothamnion nonatoi (Ceramiales, Rhodophyta). Rodriguésia, 60, 259–264. https://www.jstor.org/stable/23499987

Nunes, J. M. C., & Guimarães, S. M. P. B. (2010). Morfologia y toxonomía de Scinaia halliae (Scinaiaceae, Rhodophyta) em el litoral de Bahia y Espírito Santo, Brasil. Revista
de Biología Marina y Oceonografía, 45, 159–164. http://dx.doi.org/10.4067/S0718-19572010000100017

Nunes, J. M. C., & Paula, E. J. (2000). Estudos taxonômicos do gênero Padina Adanson (Dictyotaceae – Phaeophyta) no litoral do Estado da Bahia, Brasil. Acta Botanica Malaci-
tana, 26, 21–43. https://doi.org/10.24310/abm.v25i0.8470 

Nunes, J. M. C., & Paula, E. J. (2001). O gênero Dictyota Laumouroux (Dictyotaceae – Phaeophyta) no litoral do Estado da Bahia, Brasil. Acta Botanica Malacitana, 26, 5–18. https://doi.org/10.24310/abm.v26i0.7375 

Nunes, J. M. C., & Paula, E. J. (2002). Composição e distribuição das Phaeophyta nos Recifes da Região Metropolitana de Salvador, Bahia, Brasil. Iheringia, 57, 113–130. 

Nunes, J. M. C., & Paula, E. J. (2004a). Chnoosporaceae, Scytosiphonaceae, Sporochnaceae e Sphacelariaceae (Phaeophyta) no Estado da Bahia, Brasil. Biotemas, 17, 7–28. 

Nunes, J. M. C., & Paula, E. J. (2004b). Estudos taxonômicos de Ectocarpaceae e Ralfsiaceae (Phaeophyta) da Região Metropolitana de Salvador, Ba, Brasil. Acta Biologica Leopoldensia, 26, 37–50. 

Nunes, J. M. C., & Paula, E. J. (2006). O gênero Dictyopteris J.V.Lamour. (Dictyotaceae – Phaeophyta) no estado da Bahia, Brasil. Hidrobiológica, 16, 251–258. https://hidro
biologica.izt.uam.mx/index.php/revHidro/article/view/1036

Nunes, J. M. C., Santos, A. C. C., & Santana, L. C. (2005). Novas ocorrências de algas marinhas bentônicas para o Estado da Bahia, Brasil. Iheringia Série Botânica,60,
99–106. https://isb.emnuvens.com.br/iheringia/article/view/209

Oliveira, E. C., & Qi, Y. (2003). Decadal changes in a polluted bay as seen from its seaweed flora: the case of Santos Bay in Brazil. AMBIO: A Journal of the Human Environment, 32, 403–405. https://doi.org/10.1579/0044-7447-32.6.403 

Orfanidis, S., Panayotidis, P., & Stamatis, N. (2001). Ecological evaluation of transitional and coastal waters: a marine benthic macrophytes-based model. Mediterranean Marine Science, 2, 45–65. https://doi.org/10.12681/mms.266 

Pedrini, A. G. (2013). Macroalgas (ocrófitas multicelulares) marinhas do Brasil. Rio de Janeiro, Rio de Janeiro: Technical Books. 

Scherner, F., Barufi, J. B., & Horta, P. A. (2012). Photosynthetic response of two seaweed species along an urban pollution gradient: evidence of selection of pollution-tolerant species. Marine Pollution Bulletin, 64, 2380–2390. https://doi.org/10.1016/j.marpolbul.2012.08.012 

Scherner, F., Horta, P. A., De Oliveira, E. C., Simonassi, J. C., Hall-Spencer, J. M., Chow, F. et al. (2013). Coastal urbanization leads to remarkable seaweed species loss and community shifts along the SW Atlantic. Marine Pollution Bulletin, 76, 106–115. https://doi.org/10.1016/j.marpolbul.2013.09.019 

Steneck, R. S., & Dethier, M. N. (1994). A functional group approach to the structure of algal-dominated communities. Oikos, 69, 476-498. http://dx.doi.org/10.2307/3545860

Villaça, R., Yoneshigue-Valentin, Y., & Boudouresque, C. F. (2008). Estrutura da comunidade de macroalgas do infralitoral do lado exposto da ilha de Cabo Frio (Arraial do Cabo, RJ). Oecologia Brasiliensis, 12, 206–221. https://doi.org/10.4257/oeco.2008.1202.03 

Villaça, R., Carvalhal-Fonseca, A., Jensen, V. K., & Knoppers, B. (2010). Species composition and distribution of macroalgae on Atol das Rocas, Brazil, SW Atlantic. Botanica Marina, 53, 113–122. https://doi.org/10.1515/BOT.2010.013 

Zar, J. H. (2010). Biostatistical analysis. Upper Saddle River, New Jersey: Prentice Hall International.

Patrones de distribución de cerambícidos (Coleoptera: Cerambycidae) en México

Miguel Ortega-Huerta ^a, *, Felipe Noguera ^a, Arcelia Claudina Herrera-Solís ^b 

^a Universidad Nacional Autónoma de México, Instituto de Biología, Estación de Biología Chamela, Sede Colima, Carlos de la Madrid Béjar, s/n, Km 1.5, Colonia Centro, 28090 Colima, Colima, Mexico

^b Secretaría de Educación Pública, Programa de Telesecundaria, La Selva, 47750 Atotonilco el Alto, Jalisco, Mexico

*Corresponding author: maoh@ib.unam.mx (M. Ortega-Huerta)

Received: 31 December 2025; accepted: 20 August 2025

Abstract

This study constructed a georeferenced database of Cerambycidae species collected in Mexico to document their distributional patterns in the country. A sample of 24 species with a significant number of records was modeled to generate their potential distributions, applying a consensus approach. Four prediction algorithms were used: Maxent, Support Vector Machine, Generalized Linear Model, and Artificial Neural Networks. A total of 1,699 locations were obtained after applying cleaning and georeferencing procedures, resulting in 414 total number of species georeferenced. Species with ≥ 20 records included 9 genera and 24 species with 779 records; species with 5-20 records included 41 genera and 124 species with 1,072 records; species with < 5 records included 94 genera and 266 species with 512 records. Only species with ≥ 20 records were modeled. According to the Maxent algorithm, there were variables with high contribution percentages in predictions. Even though the most frequent values of the environmental (response) variables indicate which areas dominated the species distribution, the range of such values provides an estimate of the span of environmental values where species can occur. Too much taxonomic field work is needed to document the species diversity of Cerambycidae in Mexico. 

Keywords: Collection records; Cerambycidae; Species distribution model; Response variable 

Resumen

Para este estudio se elaboró una base de datos georreferenciados de especies de Cerambycidae rcolectadas en México para documentar sus patrones de distribución. Una muestra de 24 especies fue modelada para generar sus distribuciones potenciales mediante la aplicación de un enfoque de consenso. Se usaron 4 algoritmos de predicción: Maxent, Support Vector Machine, Generalized Linear Model y Artificial Neural Networks. Un total de 1,699 localidades fueron obtenidas después de aplicar procedimientos de limpiado y georreferenciación, lo que resultó en un total de 414 especies georreferenciadas. Especies con ≥ 20 registros incluyeron 9 géneros y 24 especies con 779 registros; especies con 5-20 registros incluyeron 41 géneros y 124 especies con 1,072 registros; especies con ˂ 5 registros incluyeron 94 géneros y 266 especies con 512 registros. Solamente las especies con ≥ 20 registros fueron modeladas. De acuerdo con el algoritmo Maxent, existieron variables con altos porcentajes de contribución en las predicciones. Los valores más frecuentes de las variables ambientales (respuesta) indicaron cuáles dominaron la distribución de especies y el rango de tales variables provee un estimado de la amplitud de valores ambientales, donde las especies pueden estar presentes. Hace falta mucho trabajo taxonómico de campo para documentar la diversidad de especies de Cerambycidae en México.

Palabras clave: Registros de colectas; Cerambycidae; Modelo de distribución; Variable de repuesta

Introduction

The family Cerambycidae (longhorn beetles) is one of the largest groups of the order Coleoptera, with approximately 35,000 described species, most of which are tropical or equatorial (Monné, 2005; Nearns et al., 2017). About 9,000 species have been described from Alaska to Argentina, and 1,621 species are recorded in Mexico (Bezark & Monné, 2023; Noguera, 2014). The species diversity in Mexico represents 18% of the American fauna and 4.6% of the global fauna (Noguera, 2014). Climate, host plant availability and food resources are the main factors determining Cerambycidae species’ occurrence. Species distribution and biogeographic information is limited, focused mainly on describing these fauna’s origin and lineage (Toledo & Corona, 2006). The main habitat types for Cerambycidae species in Mexico include tropical dry forest, pine, pine-oak, and oak forests, as well as tropical evergreen forest. 

The diversity of Cerambycidae is reflected in their color, body shape, and morphology of adults; these have a body size between ± 2.5 mm (Cyrtinus sp.) to over 17 cm (Titanus giganteus). Some species mimic ants (tribes Clytini and Tillomorphini), bees, wasps (Rhinotragini), and beetles (Lycid, Pteroplatini). Larvae are xylophagous and phytophagous, so they play an important role in helping decompose dead and nearly dead trees (Linsley, 1961).

When the 3 dimensions (identity, space, and time) included in biological inventories are integrated with spatial environmental data, it is possible to study a wide range of themes, such as ecology, evolution, and applications in agriculture and human health (Graham et al., 2004). Moreover, the use of information contained in scientific collections is considered fundamental in biogeographic research (Anderson & Martínez-Meyer, 2004).

Advances in statistical techniques, numerical analysis, machine learning algorithms, and geographic information systems (GIS) have been responsible for a significant increase in the elaboration and application of predicted species distribution models over the last few decades (Guisan & Zimmermann, 2000). The application of prediction algorithms and GIS makes it possible to obtain the probability of species presence in locations where there is a lack of species distribution information. This has been particularly useful in domains of ecosystem conservation and management, where it has been possible to identify and protect areas with high biological diversity, notwithstanding the lack or limited data for groups of species (Lobo et al., 2002; Zaniewski et al., 2002). Species distribution modeling is considered an interface between ecological theory and statistical modeling (Austin, 2002).

Many species distribution algorithms have been developed, which aim to improve the prediction of such models (Franklin, 2010). Considering the wide variety of prediction algorithms (Elith & Graham, 2009), a consensus of models has been adopted as an approach to generate more robust species distribution predictions (Araújo & New, 2007). Species distribution algorithms differ in various ways: selection of relevant prediction variables and their response behavior, definition of a fitted function for each variable, weighting of each variable contribution, possibility of prediction variables interaction, and prediction of geographic species occurrence patterns (Elith et al., 2006).

However, one of the main problems in obtaining species distribution predictions is that taxonomic studies of species are incomplete and lack uniformity across different regions. In fact, new species are discovered and recorded frequently (Lobo et al., 2002). Models have been developed that relate species distributions to climate variables for a wide range of taxonomic groups, including plants, insects, mammals, birds, reptiles, and amphibians, allowing for their comparative performance (Huntley et al., 2004). Even though there are many examples of modeling insect species distribution (e.g., Buse et al., 2007; Ballesteros-Mejia et al., 2013, 2017; Barredo et al., 2015; Crawford & Hoagland, 2010; D’Amen et al., 2015; Eickermann et al., 2023; Hassall, 2012; Jung et al., 2016; Lobo, 2016; Ma & Ma, 2023; Senay & Womer, 2019; Silva et al., 2016; Ulrichs & Hopper, 2008; Urbani et al., 2017; Watts & Worner, 2008), other taxonomic groups are preferred nevertheless the higher species diversity of the former.

This study’s main objective is to assemble a database of Cerambycidae species occurrence in the different natural regions of Mexico, based on the information included in biological inventories. Moreover, the study will generate potential habitat distribution models of Cerambycidae species with a significant number of occurrence data.

Materials and methods

The database of sites where Cerambycidae species occur was constructed by retrieving information from recent taxonomic studies. The database also included the results of surveys carried out since 1995, as part of the project Insecta of Tropical Dry Forest in Mexico. Priority information for building the database consisted of species’ taxonomic identity and location data. Biota v2.02 (Colwell, 1996) was the database management system used to store and organize the species information retrieved from both scientific literature and surveys conducted in Mexico’s various natural regions.

Species and locality information were retrieved from taxonomic studies by a group of 5 biology students. Due to logistical constraints, the database was divided into 2 subsets. Subset 1 comprised 5,473 records representing 170 species, which were located in 882 localities (190 locations were georeferenced). Subset 2 contained 5,052 records belonging to 268 species located in 1,093 localities (222 locations were geo-referenced).

The retrieved records included a species taxonomic hierarchy, which was scrubbed to eliminate duplicate records. In total, 1,945 localities were recorded with only 412 geo-referenced. Therefore, georeferencing location information was conducted using: ArcView (v3.2) and ArcMap (v10.0) geographic information systems; geographic data such as roads, localities, and political regionalization provided by INEGI and Conabio, Mexican government agencies; Gazetteers such as GEOLocate, JRC Fuzzy Gazetteer, Biogeomancer, INEGI’s geographic names; MaNIS/HerpNet/ORNIS Coordinate Calculator; Google Map and Geogle Earth.

Species distribution models were generated from those species with ≥ 20 records, which was only a small percentage (6%) of the total species georeferenced (414 species). Species presence records were imported into ArcMap, ensuring that no duplicates or misplaced sites, such as those located in the sea, were included.

Nineteen bioclimatic variables (Supplementary material: A1) were obtained from the WorldClim project (Fick & Hijmans, 2017; https://www.worldclim.org/data/worldclim21.html) while topographic variables (Supplementary material: A1) were obtained from the Hydro 1k dataset (https://www.usgs.gov/centers/eros/science/usgs-eros-archive-digital-elevation-hydro1k).

The bioclimatic variables were generated by interpolating monthly climate data obtained from meteorological stations between 1950 and 2000 (Hijmans et al., 2005). Based on the existing correlation among bioclimatic variables, a subset of variables was selected, avoiding correlations greater than 0.800 between variable pairs. Both the bioclimatic and topographic variables had a 1 × 1 km spatial resolution.

This study generated distribution models based on the application of a consensus approach throughout obtaining the median of 4 algorithms: Maximum Entropy (Maxent), Support Vector Machine (SVM), Generalized Linear Model (GLM), and Artificial Neural Networks (ANN). Maxent was applied independently (Phillips & Dudik, 2008) while the other 3 algorithms were applied by using the Modeco software (Guo & Liu, 2010). 

Maxent is a general-purpose machine learning method with a simple and precise mathematical formulation (Phillips et al., 2006). Maxent estimates the distribution (geographic range) of a species by finding the distribution that has maximum entropy (i.e., it is closest to the geographically uniform or most spread out) subject to constraints derived from environmental conditions at recorded occurrence locations (Phillips et al., 2017). Maxent is a general approach for presence-only modeling of species distributions (Phillips et al., 2006). Main parameters applied to generate Maxent models included: hinge, linear, and quadratic were the feature types used; 30% of samples were used for model validation; 3.0 was the regularization multiplier. 

The SVM are statistically based models rather than loose analogies with natural learning systems (Guo et al., 2005). SVM are not based on characteristics of statistical distributions so there is no theoretical requirement for observed data to be independent, overcoming the problem of autocorrelated observations. However, model performance will be affected by how well the observed data represent the range of environmental variables (Drake et al., 2006). Even though SVM are designed for positive and negative objects, normally negative data is not available and therefore we have a one-class dataset, which requires the separation of a target class from the rest of the feature space (Guo et al., 2005). Schölkopf (2001) developed an SVM of one class.

SVM uses a functional relationship named kernel to map data onto a new hyperspace in which complicated patterns can be more simply represented (Müller et al., 2001). SVM consists of projecting vectors into a high-dimensional feature space by means of a kernel, which makes possible the fitting of the optimal hyperplane that separates classes using an optimization function (Pouteau et al., 2012). The main SVM parameters are: SVM type = C-SVC; Kernel = radial basis function; degree = 3; gamma = 0.5; cost = 1. 

The variants of GLM are widely applied to generate species distribution models (Norberg et at., 2019). GLM is a linear regression method where a predictor is selected to be included or dropped from the considered set of predictors based on a predefined simplification method to minimize overfitting (Catalano et al., 2023). GLM use parametric functions such as linear or higher-degree polynomials to model the relationship between the response and predictive variables (Valavi et al., 2022). A link function transforms the scale of the dependent variable, then a GLM is able to relax the distribution and constancy of variances assumptions that are commonly required by traditional linear models (Guo & Liu, 2010). The GLM is commonly used to model dependent variables that are discrete distributions and are nonlinearly related to independent variables (Guisan et al., 2002). Logit was the link function to run the GLM.

ANNs extract linear combinations of the input variables as derived features and model the output as a nonlinear function of these derived features (Hastie et al., 2001). ANN utilizes intermediate nodes in what is referred to as a “hidden layer”, where each node contributes differentially with respect to the variables included in the model (Williams et al., 2009). ANN provides a flexible generalization of GLM and performs better than the latter when modelling nonlinear relationships (Lek et al., 1966). The BP-ANN parameters were, momentum = 0.3 and learning rate = 0.1

Compounded models were validated by applying the partial ROC test (Peterson et al., 2008). Partial ROC calculation has been proposed because of several advantages: it removes the emphasis on absence data, emphasizes the role of omission error when evaluating niche model predictivity and analyzes limited sector of the ROC space which are not directly relevant (Peterson et al., 2008). The NicheToolBox application (https://luismurao.github.io/GSoC/ntb_tutorial.html) was used to calculate the partial ROC statistics.

A portion of 30% of the total species presence samples was separated to be used as independent samples for model validation. The partial ROC test was run through 500 iterations to calculate the average of ROC statistics. After obtaining consensus distribution models, these were converted to binary models using a threshold of 0.50 across species for cross tabulating the response variables and to generate a richness model. The presence (≥ 0.5)/ absence (< 0.50) for the 24 species were summed to obtain a version of the richness model.

Results

A total of 1,699 locations with complete data were obtained after applying cleaning and georeferencing procedures; however, 246 locations lacked complete geographic information. The total number of species georeferenced was 414 (Supplementary material: A2). Most of data consisted of species with < 20 records (see Supplementary material: A2): species with ≥ 20 records included 9 genus and 24 species with 779 total records; species with 5-20 records included 41 genus and 124 species with 1,072 total records; species with < 5 records included 94 genus and 266 species with 512 total records (Fig. 1).

The groups of species with ≥ 20 and 5-19 records are distributed mostly on the Pacific slope within the states of Oaxaca and Jalisco (Table 1). On the other hand, states like Aguascalientes, Coahuila, México City, and Tabasco included only 1 species. In relation to the country’s natural regions, most records with the highest species presence were found in the tropical dry forest ecoregions (Fig. 2). 

In the case of the ≥ 20 group, the tropical forests included twice as many records (348) than the temperate forests (140 records). Despite such a difference, temperate forests showed only 2 fewer species (21 species) than the tropical dry forests (23 species). In the group 5-19 records per species, the differences in record numbers were more accentuated: tropical dry forests included 3 times records (543 for 114 species) than the temperate forests (174 for 63 species). Three natural regions for this group (5-19 records) included a similar number of records: semi-desert (170 records), temperate forests (174 records), and tropical rainforests (172 records). Finally, in this group (5-19 records), the mangrove biome included almost one record per species, 27 and 22, respectively.

Based on the correlation matrix among the prediction variables, the number of variables was reduced from 19 bioclimatic and 3 topographic to 9 and 3, respectively. All selected variables (Table 2) had correlation indexes < 0.80. Figure 3 and Table 3 show the contribution of each variable to the generation of distribution models by the Maxent algorithm.

Table 1

Number of Cerambycidae species by state in Mexico.

State	Species
Baja California	38
Baja California Sur	62
Campeche	8
Chiapas	92
Chihuahua	8
Coahuila	4
Colima	22
Mexico City	3
Durango	15
Guanajuato	2
Guerrero	67
Hidalgo	14
Jalisco	96
Estado de México	25
Michoacan	27
Morelos	45
Nayarit	49
Nuevo Leon	17
Oaxaca	104
Puebla	36
Queretaro	3
Quintana Roo	30
San Luis Potosi	24
Sinaloa	47
Sonora	16
Tabasco	3
Tamaulipas	22
Tlaxcala	1
Veracruz	66
Yucatan	33
Zacatecas	7

According to the maxent modelling, some variables showed almost the total contribution percentage in model prediction: For instance, the precipitation of driest month (wc_bio14) had 94% contribution in predicting Phaea acromela and 87% contribution in predicting Eburia brevispinis potential distributions. Similarly, precipitation seasonality (wc_bio15) contributed 79% and 76% for modeling Neocompsa puncticollis asperula and Psyrassa cylindricollis potential distributions, respectively. Other high contribution percentages included: precipitation of wettest month (wc_bio13) had 73% contribution predicting Lagocheirus binumeratus; wc_bio15 had 71% contribution predicting Neocompsa alacris; temperature annual range (wc_bio7) had 71% contribution predicting Lagocheirus araneiformis ypsilon; isothermality (wc_bio3) had 70% contribution predicting Tetraopes umbonatus; wc_bio14 had 61% contribution predicting Lagocheirus procerus; and elevation had 60% contribution predicting Tetraopes femoratus.

Table 2

Selected prediction variables.

Bioclimatic variables
	wc_bio1 = Annual Mean Temperature
	wc_bio2 = Mean Diurnal Range (Mean of monthly [max temp – min temp])
	wc_bio3 = Isothermality (BIO2/BIO7) (×100)
	wc_bio5 = Max Temperature of Warmest Month
	wc_bio7 = Temperature Annual Range (BIO5-BIO6)
	wc_bio8 = Mean Temperature of Wettest Quarter
	wc_bio13 = Precipitation of Wettest Month
	wc_bio14 = Precipitation of Driest Month
	wc_bio15 = Precipitation Seasonality (Coefficient of Variation)
Topographic variables
	Elevation
	Aspect
	Topographic Index

Considering prediction variables with high contribution percentages for multiple species models, Figure 3 and Table 3 show the most important variables: wc_bio15 (mean = 19%), wc_bio14 (mean = 17%), wc_bio7 (mean = 16%), wc_bio13 (mean = 15%), elevation (mean = 14%), and wc_bio3 (mean = 10%). On the other hand, those prediction variables that had low contribution percentages for fewer species included: Max temperature of warmest month (wc_bio5, mean = 0.09%), mean temperature of wettest quarter (wc_bio8, mean = 0.80%), aspect (mean = 1.06%), mean diurnal range (wc_bio2, mean = 1.07%), topoindex (mean = 2.3%), annual mean temperature (wc_bio1, mean=2.9).

The 4 prediction algorithms, Maximum Entropy (Maxent), Support Vector Machine (SVM), Generalized Linear Model (GLM), and Artificial Neural Networks (ANN) were applied to each of the 24 species that have ≥ 20 records. The models obtained consisted of probability approximations generated by both Modeco and Maxent (ClogLog). The different models for each species were combined by calculating the median value. Then, model accuracy was obtained by calculating the partial ROC test with 500 simulations. These results are shown in Table 4. 

In general, the modeled species exhibited high mean AUC ratios and high mean partial AUC values, indicating good model performance, as values of 2.0 and 1.0, respectively, represent a perfect model fit. Mean AUC ratios varied between 1.51 and 1.97, and partial AUC values varied between 0.75 and 0.98. Species with very high mean AUC ratios (> 1.9) included Lagocheirus binumeratus, Psyrassa cylindricollis, Psyrassa sthenias, Neocompsa alacris, Eburia nigrovittata, Euderces batesi. On the other hand, the species with the lowest mean AUC ratios (> 1.5 and < 1.7) were Susuacanga ulkei, Dylobolus rotundicollis, and Tetraopes discoideus. 

Based on the most important prediction variables identified by the Maxent algorithm, the response variables corresponding to each species model are shown in figures 4-9. For the elevation variable, there were species that preferred elevations between 0 and 50 m: Eburia laticollis, Susuacanga stigmatica, Psyrassa basicornis, and Susuacanga ulkei, which also showed significant predicted areas with elevations above 500 m. On the other hand, there were species selecting distribution areas at much higher elevations: Dylobolus rotundicollis and Tetraopes femoratus at 1,500-2,000 m, Tetraopes discoideus at 2,000-2,500 m, and Euderces auricaudus at 2,400-2,600 m (Fig. 4).

Isothermality (wc_bio3), which is an indicator of daily temperature variation with respect to annual temperature variation, had preferred values < 100 for species models built with this variable as important (35-70%): Highest preferred isothermality values were similar for different species models: Phaea tenuata (65), Euderces batesi (69), Tetraopes umbonatus (70), and Psyrassa sthenias (71) (Fig. 5).

According to Maxent, temperature annual ranges (wc_bio7) were also an important prediction variable for species with different preferring values: 3 species models (Susuacanga stigmatica, Lagocheirus araneiformis, and Psyrassa basicornis) showed temperature annual ranges preferred at 170-180 mm, while Psyrassa sthenias preferred the 18-19ᵒC range. Other species models showed preference for higher temperature annual ranges (Fig. 6): Euderces auricaudus (21-22 ᵒC), Sphaenothecus trilineatus (23-24 ᵒC) and Dylobolus rotundicollis (24-25 ᵒC).

Precipitation of the wettest month (wc_bio13) was another important prediction variable whose highest preferred values varied according to different species: Lagocheirus araneiformis, Dylobolus rotundicollis, Sphaenothecus trilineatus, and Tetraopes discoideus at 150-200 mm; Lagocheirus obsoletus, Mecas obereoides, Phaea tenuata at 200-250 mm; and Lagocheirus binumeratus at 300-350 mm (Fig. 7).

Table 3

Percentage of contribution of each variable in the elaboration of distribution models by the Maxtent algorithm.

Species	aspect	elevation	topoindex	wc_bio1	wc_bio2	wc_bio3	wc_bio5	wc_bio7	wc_bio8	wc_bio13	wc_bio14	wc_bio15
Eburia brevispinis	0.5	0	0	0	4.8	0	0	7.8	0	0	87	0
E. laticollis	0	27.4	0	25.7	0	0	0	1.9	0	0.3	11.3	33.4
E. nigrovittata	0	0	0	26.5	0	0	0	10.4	0	2	13.5	47.6
Susuacanga stigmatica	0	50.2	0	2.7	0	0	0	35.9	1.5	9.8	0	0
S. ulkei	0.1	25.8	0	0	0.1	2.2	0	0	13.8	0	44.9	13.2
Euderces auricaudus	0	30.7	0	0	0	15.3	0	53.1	0.5	0	0.4	0
E. batesi	0.5	24.2	0	0	0	35.1	1.8	19.8	0.3	15.4	3	0
Lagocheirus araneiformis ypsilon	0.3	0.1	5.5	0	1.7	0	0.1	70.8	0	21.1	0	0.4
L. binumeratus	7.9	0.7	0	0	0	0.3	0.1	17.5	0	73.4	0	0
L. obsoletus	0	1.2	2.9	1	0	6.1	0	15.3	0	46.4	16.2	10.9
L. procerus	0	4.7	0	1.4	6.5	0	0	1.8	0.4	13.1	60.8	11.4
Mecas obereoides	0	0	0	2.2	0	0	0	15.3	0	37.3	1.6	43.5
Dylobolus rotundicollis	0	27.1	10.3	0	0	0	0	20.3	1.1	22.6	18.5	0
Neocompsa alacris	0	1.9	10.7	5.3	0	0	0	5.6	0	3.6	2.1	70.7
N. puncticollis asperula	7	0	0	0	2	0	0	4.7	0	3.3	4.4	78.6
Phaea acromela	3.3	0	0	0	0	0.2	0	0	0	2.4	94.2	0
P. tenuata	2.4	6.3	0.5	0	0	38.5	0	3.4	0	43.4	2	3.5
Psyrassa basicornis	0	26.6	0	5.3	0.1	16.5	0	44.5	1.6	5.3	0.1	0
P. cylindricollis	1	9	0.6	0	0	0	0.1	3	0	7.8	2.1	76.3
P. sthenias	2.4	0.6	0	0	0	35.5	0	20.7	0	4.5	17.2	19.1
Sphaenothecus trilineatus	0	0	0.4	0.2	0	14.4	0	27	0	25.1	20.9	12
Tetraopes discoideus	0	41.2	10.5	0	0	16.5	0	1.2	0	24.3	2.3	4.1
T. femoratus	0	59.9	13.9	0	10.5	0	0	0	0	0	2.3	13.4
T. umbonatus	0	0.4	0	0	0	70.1	0	3.5	0	0	0	26
Mean	1.06	14.08	2.30	2.93	1.07	10.45	0.09	15.98	0.80	15.05	16.87	19.34
Median	0.00	3.30	0.00	0.00	0.00	0.10	0.00	9.10	0.00	6.55	2.65	11.15
Standard deviation	2.17	18.15	4.35	7.31	2.60	17.81	0.37	18.62	2.81	18.89	27.21	25.63

Table 4

Partial ROC results for modeled species.

Species	Mean AUC ratio	Mean partial AUC
Eburia brevispinis	1.884237	0.9420767
E. laticollis	1.857125	0.9285514
E. nigrovittata	1.939383	0.9696851
Susuacanga stigmatica	1.738087	0.869013
S. ulkei	1.512725	0.7563525
Euderces auricaudus	1.822893	0.9114264
E. batesi	1.972096	0.9860455
Lagocheirus araneiformis ypsilon	1.789459	0.8944192
L. binumeratus	1.909858	0.9549165
L. obsoletus	1.772241	0.886116
L. procerus	1.749532	0.8747619
Mecas obereoides	1.767577	0.8837588
Dylobolus rotundicollis	1.664294	0.8320284
Neocompsa alacris	1.938096	0.9690422
N. puncticollis asperula	1.883306	0.9416244
Phaea acromela	1.866931	0.9334454
P. tenuata	1.776351	0.888124
Psyrassa basicornis	1.871859	0.9359075
P. cylindricollis	1.919474	0.9597301
P. sthenias	1.921234	0.9606008
Sphaenothecus trilineatus	1.876181	0.9380483
Tetraopes discoideus	1.680489	0.8401879
T. femoratus	1.815506	0.907635
T. umbonatus	1.707098	0.8535488

The precipitation of the driest month (wc_bio14) was also an important prediction variable for which its response variable took the highest preferred values between 0 and 25-50 mm: Susuacanga ulkei and Lagocheirus procerus showed the highest preferred values of 0 mm, Dylobolus rotundicollis and Sphaenothecus trilineatus at 3 mm, and Eburia brevispinis and Phaea acromela at 25-50 mm (Fig. 8).

Finally, the precipitation seasonality (wc_bio15), which is the coefficient of variation of precipitation, was an important prediction variable whose preferred highest values (110) seem similar for this group of species (Fig. 9): Eburia laticollis, Eburia nigrovittata, Mecas obereoides, Neocompsa alacris, Neocompsa puncticollis asperula, and Psyrassa cylindricollis.

A composite map was generated by adding each of 24 binary species distribution models (Fig. 10). In general, species that highly concurred (16-21 spp.) in only 2 very confined areas, located in southern Sinaloa and southern Oaxaca. On the other hand, large areas with no species concurring were in northern Mexico (Fig. 10). This richness map was cross-tabulated with the ecoregion map to show the biomes associated with the different concurring intervals (Fig. 11). Each richness interval was tabulated for different ecoregions, showing that the highest range (16-21 spp.) corresponded to the tropical dry forest in 75% and the tropical humid forest in 24%. The second highest interval (11-15 spp.) corresponded again to the tropical dry forest (69%), but this time, the temperate mountains were in the second place with 17%, and the tropical humid forest with 14%. On the other hand, the areas with no species concurrence corresponded to the North American Desert (45%), followed by the Great Plains (21%), temperate mountains (12%), and semi-desert (11%). The tropical dry forest occupied 8% of such areas with no species concurrence, while the tropical humid forest occupied only 0.2%.

Discussion

Considering the original number of records included in the databases (> 10,000) along with the localities, both georeferenced and without geographic coordinates, and the number of species, there was the expectation to obtain a database with a significant number of species and records. However, there existed a very limited number of species (24) with enough records (≥ 20) to be used in modeling species potential distribution. In fact, the species with ≥ 5 and < 20 records were also limited (124), and the bulk of species (266) had < 5 records (Supplementary material: A2). It is evident that extensive taxonomic fieldwork is necessary to document the species diversity of Cerambycidae in Mexico. In this regard, and apparently supporting this pattern of records, 7.4% of the species recorded in Mexico have no locality records in the country, 45.5% have been recorded in only one state, and 16% in 2 states. This means that nearly 69% of the species have either a restricted distribution or are poorly studied in terms of their distribution (F. A. Noguera, unpublished data).

By mapping the geographic location of Cerambycidae species records in Mexico, it is noteworthy that the sampling intensity does not necessarily reflect the intensity of sampling, as the information included primarily corresponds to taxonomically studied groups rather than studies aimed at documenting the diversity of the different regions of the country, and sampling gaps are revealed by state and ecoregion across the country. In fact, the map somewhat confirms the current knowledge about the diversity of this group in the various states of the republic (Martínez-Hernández et al., 2024; Noguera, 2014; Noguera unpublished data). For example, in fact, 565 species have been recorded in Oaxaca, 485 in Veracruz, 435 in Chiapas, 397 in Jalisco, 283 in Guerrero, and 216 in Morelos —states with the highest number of records included in the study. In contrast, states such as Tlaxcala (10 species), Campeche (11), Aguascalientes (12), Tabasco (21), Zacatecas (26), Guanajuato (33), Coahuila (35), Mexico City (36), Nuevo León (69), and Chihuahua (73) recorded the fewest species in this study.
In the ≥ 20 records per species group, the most sampled biome was the tropical dry forest (348 records and 23 species), while the temperate forests included 21 species with 140 records. 

The 24 species distribution models primarily represent examples of this family’s species that are better sampled in the country. Different from relying on a single prediction algorithm, this study presents consensus models, generated from combining probability versions of 4 algorithms: Maxent, Artificial Neural Networks, Generalized Linear Model, and Support Vector Machine. The median was the statistic chosen for combining the 4 algorithms because of its characteristic of being a location parameter in contrast with the mean, which combines in-depth partial values, and is affected by outliers. Although the variations among models for the same species were evident, all 24 species distribution models showed good performance, as indicated by the AUC ratios and partial AUC values.

According to Maxent, prediction variables differed in importance for predicting species potential distribution models. The variables that had higher mean importance among species included precipitation seasonality (wc_bio15), precipitation of driest month (wc_bio14), temperature annual range (wc_bio7), precipitation of wettest month (wc_bio13), elevation, and isothermality (wc_bio3). As in this case, other studies on Cerambycidae, where the potential distribution of some species in this group was modeled, also showed that the predictive bioclimatic variables were diverse and contributed to the models to varying degrees. For example, for Psacothea hilaris, the predictive variables were precipitation of the warmest quarter (wc_bio18) and isothermality (wc_bio3) (Ruzzier et al., 2024); for Rosalia alpina, they were elevation and mean temperature of the driest quarter (wc_bio9) (Bosso et al., 2018); for Morimus asper, they were the maximum temperature during the warmest month (wc_bio5) and altitude (Kostova et al., 2023); for Batocera lineolata, they were maximum temperature in January, precipitation in July, and temperature seasonality (wc_bio4) (Li et al., 2020); for Xylotrechus arvicola, they were precipitation in October, mean maximum temperature in January, mean minimum temperature in July, mean maximum temperature, and mean minimum temperature in August (Felicísimo et al., 2021); for Monochamus carolinensis, they were precipitation of the warmest quarter (wc_bio18), precipitation seasonality (wc_bio15), precipitation of the coldest quarter (wc_bio19), mean diurnal range (wc_bio2), and minimum temperature of the coldest month (wc_bio5) (Zhao et al., 2023). This recorded variety in predictive environmental variables corresponds to the region where each of these studies was conducted. For P. hilaris, the main areas were Italy and the Mediterranean region; for R. alpina, it was Europe; for M. asper, it was Bulgaria; for B. lineolata, it was China; for X. arvicola, it was Spain; and for M. carolinensis, it was on a global scale.

The response variables obtained by cross-tabulating the species presence models with the environmental (prediction) variables identify the most favorable habitat conditions, according to the predicted distribution models. Even though the most frequent values of the environmental variables indicate which areas dominated the species distribution, the range of such values provides an estimate of the span of environmental values where species can occur. 

The 24 modeled species showed spatial correspondence, where the highest interval (16-21 species) is confined to restricted areas with tropical and humid tropical forests in the country. Larger species spatial correspondence areas corresponded to the lowest intervals (1-5 and 6-10 species), which were distributed in the tropical dry forest and temperate forests. It is worth mentioning that biomes such as the North American deserts and the Great Plains showed the highest proportion of areas with the absence of species. 

References

Anderson, R. P., & Martı́nez-Meyer, E. (2004). Modeling species’ geographic distributions for preliminary conservation assessments: an implementation with the spiny pocket mice (Heteromys) of Ecuador. Biological Conservation, 116, 167–179. https://doi-org.pbidi.unam.mx:2443/10.1016/S0006-3207(03)00187-3

Araújo, M. B., & New, M. (2007). Ensemble forecasting of species distributions. Trends in Ecology & Evolution, 22, 42–47. https://doi.org/10.1016/j.tree.2006.09.010 

Austin, M. P. (2002). Spatial prediction of species distribution: an interface between ecological theory and statistical modelling. Ecological Modelling, 157, 101–118. https://doi.org/10.1016/S0304-3800(02)00205-3 

Ballesteros-Mejia, L., Kitching, I. J., Jetz, W., Nagel, P., & Beck, J. (2013). Mapping the biodiversity of tropical insects: species richness and inventory completeness of African sphingid moths. Global Ecology and Biogeography, 22, 586–595. https://doi.org/10.1111/geb.12039 

Ballesteros-Mejia, L., Kitching, I. J., Jetz, W., & Beck, J. (2017). Putting insects on the map: near-global variation in sphingid moth richness along spatial and environmental gradients. Ecography, 40, 698–708. https://doi.org/10.1111/ecog.02438 

Barredo, J. I., Strona, G., De Rigo, D., Caudullo, G., Stancanelli, G., & San-Miguel-Ayanz, J. (2015). Assessing the potential distribution of insect pests: case studies on large pine weevil (Hylobius abietis L) and horse-chestnut leaf miner (Cameraria ohridella) under present and future climate conditions in European forests. EPPO Bulletin, 45, 273–281. https://doi.org/10.1111/epp.12208

Bezark, L. G., & Monné, M. A. (2013). Checklist of the Oxypeltidae, Vesperidae, Disteniidae and Cerambycidae, (Coleoptera) of the Western Hemisphere. Retrieved from: http://bezbycids.com/byciddb/checklists/WestHemiCerambycidae2024.pdf 

Bosso, L., Smeraldo, S., Rapuzzi, P., Sama, G., & Garonna, A. P. (2018). Nature protection areas of Europe are insufficient to preserve the threatened beetle Rosalia alpina (Coleoptera: Cerambycidae): evidence from species distribution models and conservation gap analysis. Ecological Entomology, 43, 192–203. https://doi.org/10.1111/een.12485

Buse, J., Schröder, B., & Assmann, T. (2007). Modelling habitat and spatial distribution of an endangered longhorn beetle-A case study for saproxylic insect conservation. Biological Conservation, 137, 372–381. https://doi.org/10.1016/j.biocon.2007.02.025 

Catalano, G. A., D’Urso, P. R., Maci, F., & Arcidiacono, C. (2023). Influence of parameters in SDM application on citrus presence in mediterranean area. Sustainability, 15, 7656. https://doi.org/10.3390/su15097656 

Crawford, P. H. C., & Hoagland, B. W. (2010). Using species distribution models to guide conservation at the state level: the endangered American burying beetle (Nicrophorus americanus) in Oklahoma. Journal of Insect Conservation, 14, 511–521. https://doi.org/10.1007/s10841-010-9280-8

Colwell, R. K. (1996). Biota: the biodiversity database manager. Sinauer Associates, Sunderland, Massachusetts. Systematic Biology, 46, 574–575. 

D’Amen, M., Pradervand, J. N., & Guisan, A. (2015). Predicting richness and composition in mountain insect communities at high resolution: a new test of the SESAM framework. Global Ecology and Biogeography, 24, 1443–1453. https://doi.org/10.1111/geb.12357

Drake, J. M., Randin, C., & Guisan, A. (2006). Modelling ecological niches with support vector machines. Journal of Applied Ecology, 43, 424–432. https://doi.org/10.1111/
j.1365-2664.2006.01141.x

Elith, J., Graham, C. H., Anderson, R. P., Dudík, M., Ferrier, S., Guisan, A. et al. (2006). Novel methods improve prediction of species’ distributions from occurrence data. Ecography,
29, 129–151. https://doi.org/10.1111/j.2006.0906-7590.04596.x

Eickermann, M., Junk, J., & Rapisarda, C. (2023). Climate change and insects. Insects, 14, 678. https://doi.org/10.3390/insects14080678 

Elith, J., & Graham, C. H. (2009). Do they? How do they? Why do they differ? On finding reasons for differing performances of species distribution models. Ecography, 32, 66–77. https://doi.org/10.1111/j.1600-0587.2008.05505.x 

Felicísimo, A. M., Armendáriz, I., & Alberdi, V. (2021). Modelling the potential effects of climate change in the distribution of Xylotrechus arvicola in Spain. Horticultural Science (Prague), 48, 38–46. https://doi.org/10.17221/85/2019-HORTSCI 

Fick, S. E., & Hijmans, R. J. (2017). WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas. International Journal of Climatology, 37, 4302–4315. https://doi.org/10.1002/joc.5086

Franklin, J. (2010). Mapping species distributions. Spatial inference and prediction. Cambridge UK: Cambridge University Press. https://doi.org/10.1017/CBO9780511810602 

Graham, C. H., Ferrier, S., Huettman, F., Moritz, C., & Peterson, A. T. (2004). New developments in museum-based informatics and applications in biodiversity analysis. Trends in Ecology & Evolution, 19, 497–503. https://doi.org/10.1016/j.tree.2004.07.006 

Guisan, A., & Zimmermann, N. E. (2000). Predictive habitat distribution models in ecology. Ecological Modelling, 135, 147–186. https://doi.org/10.1016/S0304-3800(00)00354-9

Guisan, A., Edwards, T. C., & Hastie, T. (2002). Generalized linear and generalized additive models in studies of species distributions: setting the scene. Ecological Modelling, 157, 89–100. https://doi.org/10.1016/S0304-3800(02)00204-1

Guo, Q., Kelly, M., & Graham, C. H. (2005). Support vector machines for predicting distribution of Sudden Oak Death in California. Ecological Modelling, 182, 75–90. https://doi.org/10.1016/j.ecolmodel.2004.07.012 

Guo, Q., & Liu, Y. (2010). ModEco: an integrated software package for ecological niche modeling. Ecography, 33, 637–642. https://doi.org/10.1111/j.1600-0587.2010.06416.x

Hassall, C. (2012). Predicting the distributions of under-recorded Odonata using species distribution models. Insect Conservation and Diversity, 5, 192–201. https://doi.org/10.1111/j.1752-4598.2011.00150.x

Hastie, T., Tibshirani, R., & Friedman, J. (2001). The elements of statistical learning: data mining, inference, and prediction.
New York: Springer. https://doi.org/10.1007/978-0-387-21606-5

Hijmans, R. J., Cameron, S. E., Parra, J. L., Jones, P. G., & Jarvis, A. (2005). Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology: A Journal of the Royal Meteorological Society, 25, 1965–1978. https://doi.org/10.1002/joc.1276 

Huntley, B., Green, R. E., Collingham, Y. C., Hill, J. K., Willis, S. G., Bartlein, P. J. et al. (2004). The performance of models relating species geographical distributions to climate is independent of trophic level. Ecology Letters, 7, 417–426. https://doi.org/10.1111/j.1461-0248.2004.00598.x

Jung, J. M., Lee, W. H., & Jung, S. (2016). Insect distribution in response to climate change based on a model: Review of function and use of CLIMEX. Entomological Research, 46, 223–235. https://doi.org/10.1111/1748-5967.12171

Kostova, R., Bekchiev, R., Popgeorgiev, G., & Kornilev, Y. V. (2023). First exhaustive distribution and habitat modelling of Morimus asper (Sulzer, 1776) sensu lato (Coleoptera, Cerambycidae) in Bulgaria. Nature Conservation, 53, 39–59. https://doi.org/10.3897/natureconservation.53.104243

Lek, S., Delacoste, M., Baran, P., Dimopoulos, I., Lauga, J., & Aulagnier, S. (1996). Application of neural networks to modelling nonlinear relationships in ecology. Ecological Modelling,90, 39–52. https://doi.org/10.1016/0304-3800(95)00142-5 

Li, A., Wang, J., Wang, R., Yang, H., Yang, W., Yang, C. et al. (2020). MaxEnt modeling to predict current and future distributions of Batocera lineolata (Coleoptera: Cerambycidae) under climate change in China. Écoscience, 27, 23–31. https://doi.org/10.1080/11956860.2019.1673604 

Linsley, E. G. (1961). The Cerambycidae of North America. Part I. Introduction. University of California Publications in Entomology, 18, 1–135.

Lobo, J. M. (2016). The use of occurrence data to predict the effects of climate change on insects. Current Opinion in Insect Science, 17, 62–68. https://doi.org/10.1016/j.cois.
2016.07.003 

Lobo, J. M., Lumaret, J. P., & Jay-Robert, P. (2002). Modelling the species richness distribution of French dung beetles (Coleoptera, Scarabaeidae) and delimiting the predictive capacity of different groups of explanatory variables. Global Ecology and Biogeography, 11, 265–277. https://doi.org/10.1046/j.1466-822X.2002.00291.x

Ma, G., & Ma, C. S. (2022). Potential distribution of invasive crop pests under climate change: incorporating mitigation responses of insects into prediction models. Current
Opinion in Insect Science, 49, 15–21. https://doi.org/10.1016/j.cois.2021.10.006 

Martínez-Hernández, J. G., Rös, M., Pérez-Flores, O., & Toledo-Hernández, V. H. (2024). Checklist of the Cerambycidae (Coleoptera: Chrysomeloidea) of Oaxaca, Mexico. Zootaxa, 5405, 185–208. https://doi.org/10.11646/zootaxa.5405.2.2

Monné, M. A. (2005). Catalogue of the Cerambycidae (Coleoptera) of the Neotropical region. Part I. Subfamily Cerambycinae. Zootaxa, 946, 17–65. https://doi.org/10.11646/zootaxa.946.1.1 

Müller, K. R., Mika, S., Tsuda, K., & Schölkopf, K. (2002). An introduction to Kernel-based learning algorithms. In Yu Hen Hu, & Jenq-Neng Hwang (Eds.), Handbook of neural network signal processing. Boca-Raton: CRC Press. https://doi.org/10.1201/9781315220413 

Nearns, E. H., Lord, N. P., Lingafelter, S. W., Santos, A., Miller, K. B., & Zaspel, J. M. (2017). LONGICORN ID. Retrieved from: https://cerambycids.com

Noguera, F. A. (2014). Biodiversidad de Cerambycidae (Coleoptera) en México. Revista Mexicana de Biodiversidad, 85 (Supl.), S290–S297. https://doi.org/10.7550/rmb.32966 

Norberg, A., Abrego, N., Blanchet F. G., Adler, F. R., Anderson, B. J., Anttila, J. et al. (2019). A comprehensive evaluation of predictive performance of 33 species distribution models at species and community levels. Ecological Monographs, 89, e01370. https://doi.org/10.1002/ecm.1370

Peterson, A. T., Papeş, M., & Soberón, J. (2008). Rethinking receiver operating characteristic analysis applications in ecological niche modeling. Ecological Modelling, 213, 63–72. https://doi.org/10.1016/j.ecolmodel.2007.11.008

Phillips, S. J., Anderson, R. P., & Schapire, R. E. (2006). Maximum entropy modeling of species geographic distributions. Ecological Modelling, 190, 231–259. https://doi.org/10.1016/j.ecolmodel.2005.03.026

Phillips, S. J., Anderson, R. P., Dudík, M., Schapire, R. E., & Blair, M. E. (2017). Opening the black box: an open-source release of Maxent. Ecography, 40, 887–893. https://doi.org/10.1111/ecog.03049

Phillips, S. J., & Dudík, M. (2008). Modeling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecography, 31, 161–175. https://doi.org/10.1111/j.0906-7590.2008.5203.x

Pouteau, R., Meyer, J. Y., Taputuarai, R., & Stoll, B. (2012). Support vector machines to map rare and endangered
native plants in Pacific islands forests. Ecological Informatics, 9, 37–46. https://doi.org/10.1016/j.ecoinf.2012.03.003 

Ruzzier, E., Lupi, D., Tirozzi, P., Dondina, O., Orioli V., Jucker, C. et al. (2024). A two‑step species distribution modeling to disentangle the effect of habitat and bioclimatic covariates on Psacothea hilaris, a potentially invasive species. Bio Invasions, 26, 1861–1881. https://doi.org/10.1007/s10530-024-03283-9

Schölkopf, B., Platt, J. C., Shawe-Taylor, J., Smola, A. J., & Williamson, R. C. (2001). Estimating the support of a high-dimensional distribution. Neural Computation, 13, 1443–1471. https://doi.org/10.1162/089976601750264965

Senay, S. D., & Worner, S. P. (2019). Multi-scenario species distribution modeling. Insects, 10, 65. https://doi.org/10.3390/insects10030065 

Silva, D. P., Aguiar, A. G., & Simião-Ferreira, J. (2016). Assessing the distribution and conservation status of a long-horned beetle with species distribution models. Journal of Insect Conservation, 20, 611–620. https://doi.org/10.1007/s10841-016-9892-8

Toledo, V. H., & Corona, A. M. (2006). Patrones de distribución de la familia Cerambycidae (coleóptera). In J. J. Morrone, & J. Llorente Bosques (Eds.), Componentes bióticos principales de la entomofauna mexicana, México (pp. 425–474). México D.F.: Las Prensas de Ciencias, UNAM.

Ulrichs, C., & Hopper, K. R. (2008). Predicting insect distributions from climate and habitat data. BioControl, 53, 881–894. https://doi.org/10.1007/s10526-007-9143-8

Urbani, F., D’alessandro, P., & Biondi, M. (2017). Using Maximum Entropy Modeling (MaxEnt) to predict future trends in the distribution of high altitude endemic insects in response to climate change. Bulletin of Insectology, 70, 189–200.

Valavi, R., Guillera-Arroita, G., Lahoz-Monfort, J. J., & Elith, J. (2022). Predictive performance of presence-only species distribution models: a benchmark study with reproducible code. Ecological Monographs, 92, e01486. https://doi.org/10.1002/ecm.1486 

Watts, M. J., & Worner, S. P. (2008). Comparing ensemble and cascaded neural networks that combine biotic and abiotic variables to predict insect species distribution. Ecological Informatics, 3, 354–366. https://doi.org/10.1016/j.ecoinf.2008.08.003 

Williams, J. N., Seo, C., Thorne, J., Nelson, J. K., Erwin, S., O’Brien, J. M. et al. (2009). Using species distribution models to predict new occurrences for rare plants. Diversity and Distributions, 15, 565–576. https://doi.org/10.1111/j.1472-4642.2009.00567.x

Zaniewski, A. E., Lehmann, A., & Overton, J. M. (2002). Predicting species spatial distributions using presence-only data: a case study of native New Zealand ferns. Ecological Modelling, 157, 261–280. https://doi.org/10.1016/S0304-3800(02)00199-0

Zhao, J., Zou, X., Yuan, F., Luo, Y., & Shi, J. (2023). Predicting the current and future distribution of Monochamus carolinensis (Coleoptera: Cerambycidae) based on the maximum entropy model. Pest Management Science, 79, 5393–5404. https://doi.org/10.1002/ps.7753