Effect of spatial and temporal urban isolation on the genetic diversity, acoustic variation, and morphological characteristics of an urban survivor bird species
Luis Cueva1,2*, Eric J. Fuchs1, Gilbert Barrantes1, Ruth Madrigal-Brenes1 & Luis Sandoval1
1Universidad de Costa Rica, Escuela de Biología, Ciudad Universitaria, Rodrigo Facio, San Pedro de Montes de Oca, San José, 11501-2060
2Programa de Posgrado en Biología, Universidad de Costa Rica, Escuela de Biología, Sistema de Estudios de Posgrado, Ciudad Universitaria, Rodrigo Facio, San Pedro de Montes de Oca, San José, 11501-2060
*Contact information: Luis Cueva, San Pedro, San José 11501-2060, +506 62784357,luisfer_cueva@hotmail.com
Abstract
Urbanization alters ecosystems, fragmenting natural habitats, and hence, increasing isolation between populations. Therefore, a reduction in gene flow among isolated populations is expected with greater distance and time since fragmentation. Changes in the structure, density, or community composition in the remaining habitats often result in species‘ modifications of vocalizations and morphological traits. However, the relationship between genetic, vocalizations and morphological divergence in urban areas over time remains poorly understood. We analyzed ten years of genetic, acoustic, and morphological data from isolated populations of the white-eared ground-sparrow. We used seven microsatellites (SSRs), recorded, and measured five acoustic traits, and measured six morphological traits, and compared them over a period of ten years. We found an increase in inbreeding, increase in song duration, number of elements, and frequency of maximum amplitude, but a reduction in female body size and changes in male beak. However, we only identified a significant correlation between genetic diversity and the acoustic characteristics of song.
Keywords: Melozone leucotis, Costa Rica, genetic structure, Bioacoustic
Introduction
Urban development results in rapid environmental changes (Grimm et al., 2008; Johnson & Munshi-South, 2017) which transform natural habitats into small patches of vegetation surrounded by a matrix of man built constructions such as buildings and roads. These fragments with depauperate resources are also frequently highly contaminated by noise, lights, or solid debris (Fahrig & Rytwinski, 2009; Biamonte et al., 2011). This reduction in natural habitats isolates populations, as urban development restricts or eliminates the movement of individuals across populations (Lynch & Baker, 1994; Crooks et al., 2004), lowering gene flow and, consequently, the genetic diversity of populations (Lynch & Baker, 1994; Johnson & Munshi-South, 2017). In birds, changes in behavior (e.g., vocalizations, nest construction, predator responses) and morphology (e.g., bill and body size) have also frequently been reported as a consequence of different selective pressures acting on isolated populations (e.g., sexual selection, predation, environment) (Slabbekoorn & Smith, 2002; Brumm, 2004; Warren et al., 2006; Foster et al., 2008; Laiolo & Arroyo, 2011; Luther & Derryberry, 2012; Sandoval et al., 2014; Geffroy et al., 2020; Corrales-Moya et al., 2021; Méndez et al., 2021)
The increase in distance between populations and the effect of drift and inbreeding in populations of small size are the primary causes of the loss of genetic diversity within populations and the increase in spatial genetic structure in urban isolated bird populations (Keller & Waller, 2002; Whitlock, 2004; Johnson & Munshi-South, 2017; Miles et al., 2019). In wren-tits Chamaea fasciata , significant genetic structure was found when populations were separated by a highway and a strip of residential development, in southern California (Delaney et al., 2010b). In the song sparrow Melospiza melodia, researchers found lower genetic diversity in populations where urban development had been extensive and had occurred over long periods of time (Unfried et al., 2013). These results indicate that urbanization has the potential to lower genetic diversity within fragmented populations inside urbanized matrices and increase the geographic genetic structure of bird populations.
The observed patterns of genetic structure caused by urbanization are analogous to differences in songs reported for several bird species: white-bellied short wingBrachypteryx major , rufous-and-white wrens Thryophilus rufalbus , and wedge-tailed sabrewing Campylopterus pampa . (Shaw & Lesnick, 2009; Gonzalez et al., 2011; Purushotham & Robin, 2016; Graham et al., 2017). Variation or polymorphism in the acoustic traits of songs among populations occurs in response to different habitat characteristics (e.g., vegetation density or noise levels), and they are aimed at improving sound transmission and communication across different environments (Morton, 1975; Boncoraglio & Saino, 2007; Ey & Fischer, 2009; Hardt & Benedict, 2021). Acoustic variation may also occur via female selection, differences in female choice among populations may lead to selective changes in song characteristics by males (Searcy & Andersson, 1986; Laiolo & Arroyo, 2011). Acoustic variations may also be the result of acoustic drift, which occurs when a small group of birds randomly develops distinct repertoires. (Baker & Cunningham, 1985; Wilkins et al., 2013). For example, in bird species that inhabit both urban and rural environments, songs from urban populations had higher frequencies than populations from birds inhabiting rural environments to cope with anthropogenic noise (Hu & Cardoso, 2009; Luther & Derryberry, 2012; Méndez et al., 2021). Differences in song traits may have evolved through acoustic drift as in populations of Darwin’s finches (B. R. Grant & Grant, 1996). These changes in songs can be maintained or increase over time, because songs are either learned or transmitted over generations (Boyd & Richerson, 1985; Warren et al., 2006). Hence, the correlation between genetic and vocal polymorphism may arise due to the relationship between song pattern diversity and the survival and reproductive success of individuals, as well as the presence of distinct genotypes within each population (Danielson-François et al., 2006; Irwin et al., 2008).
Variation in song acoustic characteristics (e.g., frequency and duration), have also been related with the body and beak size of individuals (Podos, 2001, 2010), because larger birds and larger beaks tend to produce lower frequency songs (Wallschläger, 1980; Bradbury & Vehrencamp, 1998; Fletcher & Tarnopolsky, 1999; Slabbekoorn & Smith, 2002a; Brumm, 2009 ). For ex ample, in purple-crowned fairy-wrensMalurus coronatus , barn swallows Hirundo rustica , and black swans Cygnus atratus; individuals with large body sizes had lower song frequencies (Galeotti et al., 1997; Patel et al., 2010; Hall et al., 2013). In House finches Haemorhous mexicanus from Arizona, urban individuals have longer and deeper beaks than rural birds and emit songs with slower trills and wider frequency range (Badyaev et al., 2008).
White-eared ground-sparrows, Melozone leucotis, inhabit the central valley of Costa Rica in isolated populations embedded inside the greater urban area (Sandoval et al., 2015; Juárez et al., 2020; Rodríguez-Bardía et al., 2022). Previous work on this species reported that urbanization limits the local movement of individuals and gene flow between populations, which resulted in significant genetic structuring among Costa Rican populations (Rodríguez-Bardía et al., 2022). Previous research has also identified changes in song dialects among different populations (Sandoval et al., 2014, 2016; Bonilla-Badilla 2021). Additionally, it has been observed that differences in the frequency and duration of songs among populations are associated with levels of anthropogenic noise. Specifically, males in populations exposed to higher levels of noise tend to exhibit an increase in both the minimum frequency and duration of their songs. (Sandoval et al., 2014, 2015, 2016; Juárez et al., 2020; Méndez et al., 2021; Rodríguez-Bardía et al., 2022). These studies suggest that genetic and cultural traits are under intense selective pressures imposed by urbanized environments. However, it is still unknown how these traits change over time and how they track population dynamics.
The main objective of this study is to describe the changes in genetic diversity, vocalization, morphology, and their relationship, over a 10-year period in populations of white-eared ground-sparrow isolated by urban development in Costa Rica. We hypothesize that differences in genetic diversity, song acoustics, and morphology will correlate with the length of time communities have been separated, as population divergence increases with isolation.
Material and Methods
We conducted this study in four populations of white-eared ground-sparrows in Costa Rica: (1) Estación Biológica Monteverde (MTV), Puntarenas province (10° 18ʹ N, 84° 48ʹ W, 1600 m), characterized by coffee plantations and large forest patches that connect with more extensive mature forests (Rodríguez-Bardía et al., 2022) without an urban isolation effect yet. (2) Getsemani (HDA), Heredia province (10 ° 01’N, 84 ° 05’W, 1350 m); it is a non-urbanized site, with secondary forest patches and abandoned coffee plantations, and dense thickets with little human presence (Juárez et al. 2020). Urbanization started to expand from the edges of the study area since 2000. (3) Jardín Botánico Lankester (JBL), Cartago province (9 ° 50’N, 83 ° 53’W, 1370 m); it includes secondary forests, gardens, and buildings, with little disturbance and human presence, but immerse in a urban matrix that had rapidly extended since 1990 (Juárez et al. 2020). (4) The Universidad de Costa Rica (UCR), San José province (09 ° 56’N, 84 ° 05’W, 1200 m); is a highly urbanized site, with a secondary forest reserve surrounded by buildings, open areas and gardens, and it is exposed to intense human disturbance (Juárez et al., 2020). Urban development began before the 1900’s but most fragmentation occurred after the 1970´s (Biamonte et al., 2011).
Historical and current sampling
Historical acoustic and morphological data was collected in 2011 and 2012 and was deposited in the Laboratorio de Ecología Urbana y Comunicación Animal (LEUCA), Universidad de Costa Rica. These data are part of a long-term research focus on analyzing the effect of urbanization on behavior, morphology, and genetic diversity of bird populations. To obtain historic genetic diversity estimates, we used blood samples collected in 2011 – 2012. We will refer for the historical 2011 – 2012-year sampling as “period 1”.
Current acoustic and morphological data was collected from 2018 to 2022. We recorded each banded pair using a Marantz PMD661 digital recorder and a Sennheiser ME66/K6 unidirectional microphone. The recordings were stored in WAV format with a sample rate of 44.1 kHz and a precision of 24 bits. All recording were conducted between 4:55 - 6:00 h when this species is most vocally active (Sandoval et al., 2016). We captured individuals using mist nets (12 x 2.5m, 15 mm mesh eye) inside each pair territory. We provided each captured individual with a unique numbered metal ring and a color combination of plastic rings. From each bird, we obtained 10 µL of blood from the brachial vein and stored it in 95% ethanol or Lysis buffer for molecular analyses (Seutin et al., 1991; Rodríguez-Bardía et al., 2022). We will refer for the period 2018 – 2022-year sampling as “period 2”.
All procedures were conducted in accordance with Costa Rican law. The Research Committee of the Biology School and the Animal Care Committee (ACC) of the University of Costa Rica granted to LS, the principal investigator, research permits and animal care protocols.
DNA extraction and SSR amplification
We extracted DNA from blood samples using the DNeasy blood and tissue Kit (Qiagen Inc., Valencia, CA, USA) following the manufacturer’s protocol. We used seven primers: Mme2, Mme7, Mme8, Asµ15, Escµ6, Gf01 y Gf05 (Petren, 1998; Jeffery et al., 2001; Bulgin et al., 2003) which have been previously shown to be polymorphic and amplified well for this species (Rodríguez-Bardía et al., 2022). We followed the Qiagen Multiplex master kit ((Qiagen Inc., Valencia, CA) procedure to amplify the aforementioned SSRs. All markers were amplified through Polymerase Chain Reactions (PCR) using 10 μL reaction, containing 2 μL of 0.4 µM primer mix, 1.5 μL of 20 ng of template DNA, 5 μL of Multiplex Master Kit (Qiagen) for mixed primers and Top Taq Master Kit (Qiagen) for Gf01 and 1.5 μL of water nuclease free (Qiagen). We followed the PCR thermal profiles and mixes described in Rodriguez et al. (2022). The PCR was performed in a Veriti™ thermocycler (Applied Biosystems, Foster City, CA, USA). Capillary electrophoresis was performed on a 3500 genetic analyzer (Applied Biosystems) using Hi-Di™ formamide and GeneScan™ 500 LIZ™ dye size standard (Applied Biosystems). Multilocus genotypes were scored using GeneMarker 1.91 (SoftGenetics, State College, PA, USA).
Acoustic analyses
We analyzed all songs using Raven pro 1.6 software (Cornell Lab of Ornithology, Ithaca, NY, USA). We classified each song according to its structure in different types following the classification of the sound library for white-eared ground-sparrow available in LEUCA and used in previous studies (Sandoval et al., 2014, 2015, 2016; Méndez et al., 2020). In addition, we measured the following acoustic characteristics: duration (s), minimum frequency (kHz), maximum frequency (kHz), frequency of maximum amplitude (kHz), and the number of elements. We used different windows to obtain the measurements: the spectrogram window was used to identify and classify sounds, the power spectrum was used to measure frequency characteristics with a threshold of -30dB relative to vocalization peak, and the oscillogram to measure the temporal characteristics of sounds. Spectrograms were constructed using a Hann window with a size of 512 samples, 3 dB bandwidth of 124 Hz; a Time grid with an overlap of 50% and hop size of 5.80 ms; and a frequency grid spacing with DFT size of 512 samples and grip spacing of 86.1 Hz.
Morphometric measurements
We measured six morphological traits: tarsus length (from the intertarsal joint to the middle of the foot sole), tail length, wing chord length (unflattened), exposed culmen length (from the tip of the beak to base of skull), beak width (at the beak gape), and beak depth (measured at right angles to the point on the lower mandible where the feathers end) following the protocol described in Sandoval & Mennill, (2013). We used a dial caliper (model: SPI Plastic Caliper 150mm, AVINET, NY, USA) to get the bill and tarsus measurements, a metal wing ruler (model: WING15ECON, AVINET, NY, USA) to measure wing chord and tail length.
Statistical analyses