Thorax temperature and niche characteristics as predictors of abundance of Amazonian Odonata

1 1 Thorax temperature and niche characteristics as predictors of abundance of Amazonian Odonata 2 3 Lenize Batista Calvão1,2*, Ana Paula J. Faria1,2, Carina Kaory Sasahara de Paiva2, José Max Barbosa 4 Oliveira-Junior3, Javier Muzón4, Alex Córdoba-Aguillar*5, Leandro Juen1,2 5 6 1Programa de Pós-Graduação em Ecologia, Universidade Federal do Pará, Belém, Pará, Brazil 7 2Laboratório de Ecologia e Conservação (LABECO), Instituto de Ciências Biológicas (ICB), Universidade 8 Federal do Pará (UFPA), Belém, Pará, Brazil 9 3 Instituto de Ciências e Tecnologia das Águas (ICTA), Universidade Federal do Oeste do Pará (UFOPA), 10 Santarém, Pará, Brazil 11 4 Laboratorio de Biodiversidad y Genética Ambiental (BioGeA), Universidad Nacional de Avellaneda, 12 Avellaneda, Buenos Aires, Argentina 13 5 Departamento de Ecología Evolutiva, Instituto de Ecología, Universidad Nacional Autónoma de México, 14 , Ciudad Universitaria, Mexico City, Mexico 15 16 *Corresponding authors: 17 E-mail: [email protected] [email protected] .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 2 18 Abstract 19 Environmental architecture and body temperature drive the distribution of ectothermic species, 20 especially those with specific ecophysiological requirements or narrow ecological niches. In this study, 21 we evaluated the connection between thorax temperature and niche specialization concerning the 22 abundance and species contribution to the beta diversity of adult Odonata in Amazonian streams, 23 employing the Species Contribution to Beta Diversity (SCBD). Our hypotheses were (i) Odonata 24 species’ thorax temperature is positively correlated with both morphology (thorax width) and air 25 temperature, and (ii) the thorax temperature of the Odonata assemblage serves as a more influential 26 predictor than niche specialization in determining species abundance and SCBD. We sampled 46 27 streams in an anthropized landscape in the Northeastern and Southeastern regions of Pará state, Brazil. 28 Notably, niche breadth emerged as the variable influencing the abundance and SCBD of the Odonata 29 assemblage. Niche position is a predictor for Odonata SCBD and not suborders, and predictor for 30 abundance, except for Anisoptera. Both suborders exhibited a negative relationship between abundance 31 and thoracic temperature. In summary, our results underscore the necessity of considering both niche 32 and ecophysiological predictors to comprehensively assess the Odonata assemblage in Amazonian 33 streams. This holistic approach has implications for conservation efforts and bioassessment practices, 34 offering valuable insights into the collective response of Odonata as a group. 35 36 Introduction 37 One aim in ecology is to understand how species assemblages distribute as a function of each 38 species’ requirements. In this regard, ecological niche breadth and position are key predictors for the local 39 abundance [1–4]. The specific link that underlies species abundance and such predictors is explained 40 according to two models. The hypothesis that links niche breadth/tolerance, predicts that populations that 41 can remain viable in a wide range of environmental conditions have greater abundance. In this context, they 42 can be characterized as generalists (occur in a wide range of environmental conditions) and specialists 43 (occur in a more restricted range). Conversely, the hypothesis for the niche position predicts that non- 44 marginal species, those that occur in a larger availability of habitats, tend to be more abundant(11). 45 Generalist species, endowed with greater niche breadth, can occur in a wider range of 46 environmental conditions, leading to a smaller contribution of species for Beta diversity (SCBD) than with 47 the contribution of specialist species [10]. Accordingly, niche position can influence SCBD, as species in 48 marginal habitats use more specific environmental conditions than non-marginal species [10]. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 3 49 Simultaneously, the relationship of functional traits (e.g., thermoregulatory ability governed by body size) 50 with SCBD may occur if traits confer adaptations in the species, influencing their abundance and/or 51 occupancy at the site [2, 4]. 52 Odonata are aquatic insects whose thermal performance determines their distribution both at the 53 micro- [12] and macro-scale [13]. In this context, the body size and behavior of adult odonates are 54 associated with their thermoregulation strategies, categorizing species as thermal conformers, heliothermic 55 or endothermic [12]. These thermoregulatory strategies are correlated with body size, as heat exchange with 56 the environment occurs based on the surface/volume ratio [14]. Consequently, small percher species (e.g., 57 most zygopterans) generally exhibit thermal conformers or heliothermic, relying on air temperature to start 58 their activities [12]. These species can regulate heat loss by adjusting their body posture in response to light 59 [15–16] and by selecting habitats that support their thermal requirements [14, 17]. Conversely, larger 60 species of Odonata can inhabit open areas with reduced canopy cover [12] as they are heliothermic [18], 61 benefiting from direct sunlight exposure. Finally, larger species (e.g., mostly Anisoptera) are predominantly 62 endothermic, enabling them to overcome the limitations faced by thermal conformers, as they can internally 63 control heat loss or production through their thorax muscles [14]. 64 Ecophysiological requirements have frequently been crucial predictors for odonate abundance 65 or distribution in tropical streams [18, 19, 20, 21]. This correlation is primarily supported by the impact of 66 changes in vegetation cover and air temperature, both acting as environmental filters influencing odonate 67 community composition [21]. For example, the absence of vegetation and reduced shade favor most 68 anisopteran species due to their reliance on increased light input into the stream [12, 18, 20, 21, 22]. 69 Conversely, these environmental conditions are unsuitable for most Zygoptera or smaller species, as they 70 typically prefer habitats with a more consistent temperature and may not be able to thermoregulate 71 effectively in altered streams [20, 21, 22]. 72 In the present study, we aimed to evaluate the importance of thoracic temperature and niche 73 specialization on the abundance and contribution of each species to the SCBD of adult Odonata in 74 Amazonian streams. For this, we had the following predictions: (i) thoracic temperature of Odonata species 75 differs between suborders and correlates positively with morphology (thorax width) and air temperature. 76 Thus, larger species (Anisoptera) may exhibit heliothermic behaviors, leading to thoracic temperature 77 higher than air temperature. Conversely, smaller species (Zygoptera) tend to thermoregulate in response to 78 air temperature, resulting in thorax temperature similar to the ambient environment; (ii) the thoracic .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 4 79 temperature of the Odonata assemblage, as well as its suborders (Zygoptera and Anisoptera), serves as a 80 more important predictor than niche characteristics in predicting the abundance and SCBD of the species 81 given that ecophysiological traits play a crucial role in Odonata habitat selection (Fig 1). 82 Fig 1. Schematic drawing of hypotheses and images of Odonata were adapted from Stehr, F.W and 83 Tennessen, K.J. 84 Material and methods 85 Study area 86 The study was conducted in 46 streams across four municipalities in the Northeastern region 87 (Tomé-açu, Ipixuna do Pará, Concórdia do Pará and Acará) and three municipalities in the Southeastern 88 region (Paragominas, Canaã dos Carajás and Parauapebas) of Pará state, Brazil, covering basins ranging 89 from the Tocantins-Araguaia River and Capim River (Fig 2, S1 Table). Northeastern Pará is characterized, 90 according to the Köppen classification, by a tropical rainforest climate (Af) and tropical monsoon climate 91 (Am) [23], with temperature ranging from 22°C to 34°C (minimum: 22°C to 23°C; maximum: 30°C to 92 34°C). Southeastern Pará has a Savanna climate, classified as a tropical climate with a dry season (Aw) 93 [23, 24, 25]. The municipalities of Canaã dos Carajás and Parauapebas, located in Serra dos Carajás area, 94 are notable for their landscape, mainly due to their elevation, which ranges between 500 and 700 m a.s.l. 95 [24]. The average monthly temperature in this area varies from approximately 25°C to 26°C, with an annual 96 rainfall of 2.033 millimeters [24]. 97 Fig 2. Study area showing the 46 streams distributed in the Northeastern and Southeastern of Pará, Brazil. 98 Images of Odonata were adapted from Stehr, F.W and Tennessen, K.J. 99 Insect sampling 100 The 46 sampled streams ranged from 1st to 3rd order, with an average width of 2.8 meters and a 101 depth of 29.9 centimeters, according to Strahler [26] classification. Sampling periods were performed 102 between July 2017 and October 2018, consistently during the low precipitation period. Adult odonates were 103 collected only once stream, in 20 segments spaced 5m apart, distributed along a continuous 150-meter 104 longitudinal stretch in each stream [further details in 20, 27, 28]. Specimens were collected on stream banks 105 for one hour, always between 11:00 and 12:00 hrs (the peak activity period for adult odonates), at 106 temperatures above 19°C, using a 40cm diameter and 65cm in length entomological net [21, 29]. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 5 107 Standardization of sampling effort and climatic conditions was necessary due to the organisms' 108 thermoregulatory ability based on solar incidence [30], ensuring the presence of all ecophysiological groups 109 [21]. 110 The collected specimens were placed in tracing paper envelopes and subsequently fixed in 111 acetone P.A. (Propanone) for 12 hours for Zygoptera and 48-72 hours for Anisoptera. Species identification 112 was conducted to the species level using specialized taxonomic keys [31–35]. The biological material was 113 deposited in the Collection of Aquatic Insects at the Laboratory of Ecology and Conservation, Federal 114 University of Pará (UFPA), University Campus of Belém, Pará, Brazil [36]. 115 Measurements of physiological traits 116 The thoracic temperature of males was captured using an entomological net and quickly 117 measured within a maximum of 10 seconds to avoid physical damage and alteration in body temperature. 118 We used only male individuals because species identification is possible in this sex. Holding the wings, we 119 measured the body temperature (°C) of each individual using a Testo-805 infrared thermometer (accuracy 120 ± 1°C [-2 to +40°C], resolution 0.1°C [-9.9 to +199.9°C], and reaction time <1s). The recording was 121 performed by pointing the thermometer beam at the center of the right side of the thorax, at a distance of 122 five centimeters. 123 Thorax width was measured using a digital caliper (accuracy = 0.02 mm) for males of each species. 124 For species with multiple individuals, we used the averages, while for those with only one individual, we 125 recorded the absolute value. 126 Environmental variables in streams 127 For each stream, four environmental variables were measured: depth (cm), channel width (m), 128 Habitat Integrity Index (HII) and air temperature (in ºC). Depth was measured at three points of the stream: 129 left and right of the bank, and central region. The average of these measurements was used as a measure of 130 depth in each sampling unit. The HII was composed by 12 items that describe the degree of habitat integrity 131 in the stream: land use pattern beyond the riparian zone; width of riparian forest; completeness of riparian 132 forest, vegetation of riparian zone within 10 m of channel,; retention devices and sediment in the channel; 133 river bank structure; bank undercutting, stream bottom; distribution of riffles and pools; characteristics of 134 aquatic vegetation and detritus [37]. Each item presented four to six alternatives corresponding to the 135 observed condition related to habitat integrity. We transformed each item value to produce the HII, which .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 6 136 ranges from 0 (altered stream) to 1 (preserved stream), according to the habitat integrity conditions found 137 in each stream [37]. Note that this HII has been widely used in studies that have assessed the environmental 138 conditions of streams and their relationship with aquatic insect diversity [38–40]. Air temperature was 139 measured using a Testo-805 infrared thermometer. After measuring the thorax temperature of the 140 specimens, the thermometer was positioned in the microhabitat where each individual was collected. 141 Data analysis 142 To calculate niche breadth and position (i.e., predictor variables) of the species, we used the 143 Outlying Mean Index (OMI [11]) which is based on the following environmental variables: depth, channel 144 width, and HII. The OMI analysis calculates the distance between the average environmental conditions 145 used by the species (centroid) and the average environmental conditions of the sampled sites (hyperspace) 146 [11,41]. Species abundance data were logarithmized and environmental variables standardized. The 147 marginality significance (OMI) of the species was evaluated using the Monte-Carlo permutation test with 148 1000 permutations [11]. Thus, we obtained the breadth (environmental tolerance) and position (marginality) 149 of the ecological niche for each species. We carried out PCoA for Odonata composition visualization using 150 Hellinger and Bray-Curtis. 151 The ecological uniqueness of species (SCBD) was measured from the total variation of the 152 assemblage (Total Beta Diversity - BD Total) between streams. For this analysis, we submitted the Odonata 153 composition matrix to the Hellinger transformation, which was used to measure the Sum of Total Squares 154 (SSTotal). Then, we divided the SS Total by the number of sampled streams (n-1) and obtained the BD Total of 155 the assemblage that was partitioned into ecological uniqueness of species (SCBD) [9]. For more details on 156 the calculation of BDTotal and SCBD, see [9]. Finally, species with higher SCBD values had a higher relative 157 contribution to beta diversity [9]. 158 We performed the analysis of the relationship between the variables from the generalized linear 159 mixed models (GLMMs) and species as random factors. Thus, to assess (i) the difference in thoracic 160 temperature of Anisoptera and Zygoptera, the relationship between the interaction of air temperature and 161 thorax width on Odonata thoracic temperature, and the relationship between the difference of air and 162 thoracic temperature (Tair-Tth [ºC]) and thorax width for Anisoptera and Zygoptera, with Gaussian 163 distribution. The analyzed models contained each individual as a sample unit. 164 To evaluate (ii) the effect of thoracic temperature, niche breadth and position (predictor variables 165 standardized) on the abundance (response variable) of Odonata species, we carried out a GLMM and .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 7 166 suborder as a random factor, using the negative binomial distribution due to data overdispersion [42]. For 167 this model we used the Log linkage function. The analyzed models contained species as a sample unit. For 168 Suborders, we carried out a GLM (Negative Binomial). We performed the visual validation of the models 169 using the simulated envelope [43]. To assess the effect of thoracic temperature, niche breadth and position 170 (predictor variables standardized) on Odonata and suborders SCBD (response variable), we used a Beta 171 Regression [44]. This analysis is more suitable for the response variable (SCBD) distributed between values 172 from 0 to 1 [44]. The binding function used in this model was logit. 173 All analyses were performed using the R software [45] using MASS [46], hnp [43], betareg [44], 174 and vegan packages [47]. 175 Results 176 Odonata assemblage 177 859 specimens were collected, belonging to 56 species: 22 anisopterans and 34 zygopterans (S2 178 Table). 179 Thoracic temperature and size, and their relation with ambient temperature 180 Anisopterans showed greater temperature and thorax width than zygopterans. On average, 181 Anisoptera species have 5°C more than Zygoptera. Air temperature was higher 2°C in the places where 182 Anisoptera species were sampled (Table 1, Table 2 and Fig 3). Both thoracic width and air temperature 183 affect thoracic temperature, and (Table 2). 184 Table 1. Mean and standard deviation (SD) values of thorax temperature, air temperature, and thorax width 185 for Zygoptera and Anisoptera sampled in Brazilian Amazon streams. Anisoptera Thorax width (cm) Thorax temperature (ºC) Air temperature (ºC) Mean 3.439 33.867 31.636 SD 1.465 3.677 2.013 Zygoptera Mean 1.663 29.510 30.113 SD 0.446 1.603 1.327 186 187 Table 2. Results of the GLMM (Gaussian distribution) (species random effect) evaluating the relationship 188 between thorax temperature of Odonata (response variable) and suborders and interaction of air temperature 189 and thorax width. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 8 Value Std.Error DF t-value p-value Intercept 29.800 0.257 521 116.042 <0.001 Suborders 1.761 0.501 54 3.517 <0.001 Thorax width (cm) 0.820 0.171 521 4.809 <0.001 Air temperature (ºC) 1.159 0.096 521 12.020 <0.001 Thorax width *Air temperature 0.288 0.086 521 3.361 <0.001 190 Significant p values appear in bold. 191 192 Fig 3. Thoracic temperature of Anisoptera (Dark orange) and Zygoptera (Cyan). Black dots are outliers. 193 Anisopterans presented an average difference of air and thoracic temperature up to 8ºC above air 194 temperature and 6ºC below air temperature, showing a negative relationship with thorax width (Table 3 and 195 Fig 4). For Zygoptera, the same pattern was observed with an average difference of 3ºC above and 4ºC 196 below air temperature (Table 3 and Fig 4).Table 3. GLMM (Gaussian distribution) (species random effect) 197 results evaluating the difference between air and thorax temperature (Tair-Tth – variable response) with 198 thorax width of Anisoptera and Zygoptera (predictor). Anisoptera Value Std.Error DF t-value p-value Intercept 0.753 1.034 192 0.728 0.468 Thorax width (cm) -0.908 0.290 192 -3.132 0.002 Zygoptera Value Std.Error DF t-value p-value Intercept 2.237 0.578 330 3.867 0.000 Thorax width (cm) -0.983 0.341 330 -2.886 0.004 199 Significant p value appears in bold. 200 Fig 4. Relation of the difference between air and thorax temperature (Tair-Tth (ºC) response variable) and 201 thorax width (cm), for Anisoptera (Dark orange) dots bellow dotted line are species that have the thorax 202 temperature above air temperature. Zygoptera (Cyan). Images of Odonata from Stehr, F.W and Tennessen, 203 K.J. 204 Stream structure and odonate assemblage 205 There was a relation between environmental variables (depth, channel width and HII) and 206 Odonata species (p = 0.010) (OMI analysis global test). The variable that contributed the most to the first 207 sorting axis was HII (0.88), followed by width (-0.22) and depth (0.03). This variable is essential to assess 208 environmental integrity and demonstrates that Odonata species change composition with more intact 209 streams (HII above 0.7) and with multiple anthropic activities (Fig 5). Streams with greater habitat integrity 210 have a greater number of species of Anisoptera and Zygoptera that are only collected in these environments 211 (S3 Table). 212 Fig 5. Ordination (PCoA) of Odonata species and habitat integrity index (HII): green dots are HII above 213 0.7 and Coral dots are streams with multiple anthropic activities (HII bellow 0.7). 214 Niche and abundance of Odonata .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 9 215 The most abundant species of Zygoptera were C. rutilans, M. aenea and E. metallica. For 216 Anisoptera were F. amazonica, E. basalis and P. lais. The average the niche breadth was 0.221 (standard 217 deviation ± 0.289) and the position 2.219 (±1.94). Species with greater niche breadth were M. cupraea, H. 218 silvarum and M. aenea (Zygoptera) and E. fusca, O. walkeri and O abbreviata (Anisoptera). For niche 219 positions were H. auripennis, A. luteum and H. icterops (Zygoptera) and D. obscura, B. herbida and E. 220 cannacrioides (Anisoptera). 221 Niche breadth and position were the most important predictors of Odonata abundance and 222 suborders (Table 4 and Fig 6). Except niche position for Anisoptera. Only when evaluating suborders 223 separately, thorax temperature emerges as important predictor for them. 224 Table 4. GLMM results (negative binomial distribution) for Odonata (Suborder as a random effect), and 225 GLM for suborders evaluating the relationship of species abundance with niche breadth, position and thorax 226 temperature. Odonata Value Std.Error DF t-value p-value Intercept 2.232 0.188 51 11.878 <0.001 Niche position (NP) -0.503 0.241 51 -2.085 0.042 Niche breadth (NB) 0.856 0.189 51 4.529 |z|) Intercept 2.434 0.257 9.482 |z|) Intercept 2.569 0.285 9.006 <0.001 Niche position (NP) -0.886 0.310 -2.854 0.004 Niche breadth (NB) 0.866 0.175 4.963 <0.001 Thorax temperature (ºC) 0.868 0.421 2.064 0.039 227 228 Significant p values appear in bold. 229 On average, SCBD was 0.017 (standard deviation ± 0.026). Species with greater SCBD were 230 C. rutilans, H. indeprensa and M. aenea for the suborder Zygoptera and E. basalis, O. abbreviata and F. 231 amazonia for the suborder Anisoptera. When evaluating predictor variables for Odonata's SCBD, niche 232 breadth and position emerged as important predictor. When evaluating predictor variables for suborders 233 SCBD, niche breadth only emerged as predictor. (Table 5 and Fig 6). .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 10 234 Table 5. Beta regression results between SCBD of Odonata (suborders shown separately) and niche 235 breadth, position, and thorax temperature. Odonata Estimate Std. Error z Value Pr(>|z|) Intercept -4.174 0.150 - 27.860 <0.001 Niche position (NP) -0.290 0.136 -2.128 0.033 Niche breadth (NB) 0.404 0.090 4.481 |z|) Intercept -4.393 0.232 - 18.962 |z|) Intercept -4.053 0.260 - 15.606 <0.001 Niche position (NP) -0.371 0.213 -1.743 0.081 Niche breadth (NB) 0.384 0.108 3.547 <0.001 Thorax temperature (ºC) 0.058 0.328 0.178 0.859 236 Significant p values appear in bold. 237 Fig 6. Significant relationships of Abundance and SCBD of Odonata and the suborders and niche breadth, 238 position and thorax temperature. Odonata (Black dots), Anisoptera (Dark orange) and Zygoptera (Cyan). 239 Images of Odonata from Stehr, F.W. 240 Discussion 241 Niche breadth was an important predictor for abundance and SCBD for Odonata and its suborders. 242 Niche position also is important for Odonata abundance and SCBD, and Zygoptera abundance. Contrary 243 our expected hypothesis, the thoracic temperature was a good predictor only for abundance and when 244 evaluated suborders separately. Research shows that species with greater niche breadth are more abundant 245 and tend to present greater regional occupation [6]. Furthermore, these species, often display a more 246 generalist or tolerant behavior towards diverse environmental conditions, which may contribute to their 247 reduced vulnerability. [50]. In Odonata, this macroecological pattern may be due to the generalist species 248 being able to persist in degraded environments [51], resulting in a broader distribution range [52]. In our 249 study the families Libellulidae and Coenagrionidae, are more abundant and both are diverse families at a 250 continental scale which, which may explain their wide distribution [53–54]. Some species of 251 Coenagrionidae are associated to degraded environments, within this family E. metallica (associated with 252 streams with environmental change Faria et al. 2021) and N. luzmarinas uch as E. fusca, and O. walkeri .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 11 253 [55–56]. In addition, other families, M. cupraea, H. silvarum and M. aenea also presented greater niche 254 breadth. Species with lower abundance and smaller niche breadth tend to be sensitive to environmental 255 changes, which make them less likely to persist when the physical characteristics of streams change [29]. 256 The analysis shows that Odonata and the suborders separately, species with greater niche breadth 257 had a greater contribution to beta diversity, which is opposite to expected [6,50]. Possibly more habitat 258 generalist species and adults that are active dispersers and can survive in a wide range of environmental 259 conditions play an important role in governing SCBD. Previous study show that the pattern of beta diversity 260 of an assemblage is also adequately described by the common species, according to [58]. Only for Odonata 261 and Zygoptera we found niche position negatively associated with abundance e and com Odonata SCBD, 262 indicating that species, when very abundant, occur in non-marginal environments and occurs in more 263 common habitats in the region. However, we highlight here that species such as C. rutilans and M. aenea 264 (both contribute to beta diversity and have high niche breadth) in Amazonian streams seem to have limits 265 to occur in different intensities of multiple impacts, as these species can disappear with impacts that lead to 266 drastic changes in the streams (Faria et al. 2021). Contrary to our hypothesis, niche characteristics appear 267 more important for contribution of these species to beta diversity than body temperature. On the other hand, 268 for species abundance, thorax temperature is an important predictor, only when evaluated suborders 269 separately. In fact, much previous work has demonstrated how environmental filters affect Odonata as a 270 whole as microhabitat structure (Bank angle, woo in the stream bed), physicals and chemical, and canopy 271 cover dossel (Brito et al., 2021) and regional variables surroundings streams (Luke et al. 2017). 272 Ecofisiological traits as body size and thorax temperature, therefore, provides additional insights into the 273 patterns of suborders abundance and contribute to metacommunity dynamics due to its thermoregulatory 274 restrictions [18]. There is negative relationship between thoracic temperature and abundance. In general, 275 Anisoptera can heat their bodies through the heliothermic or even endothermic ability. These 276 thermoregulatory abilities are crucial for Odonata distribution [18]. Furthermore, heliothermic anisopterans 277 can benefit from habitats with reduced riparian vegetation and greater sunlight input and can be quite 278 abundant in these areas [21]. Conversely, larger species of Anisoptera (e.g., Gomphidae) with higher body 279 temperature may be endothermic. These species live inside forests and are often found in streams only when 280 they arrive for mating and oviposition [59]. These ecological and behavioral traits, make this species with 281 higher thoracic temperatures are more difficult to collect, at the same time more sensitive to loss of riparian .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 12 282 vegetation. Zygoptera with higher temperatures are large individuals such as H. auripennis and A. dives 283 and had lower abundance. 284 Thoracic temperature differs between suborders and is related to thorax width and air 285 temperature. Anisopterans presented on average thoracic temperature up to 8ºC above air temperature and 286 6ºC bellow, showing a negative relationship with thorax width. For Zygoptera, the thorax temperature is 287 maintained 3ºC above and 4ºC below air temperature. Differences in the behavior of these groups may 288 help explain this pattern. Anisoptera tend to be heliothermic or endothermic fliers and maintain the thoracic 289 temperature above that of the air. One of the ways to control heat loss is to alter the circulation of 290 hemolymph between the thorax and abdomen [15], posture adjustment (Corbet, 1999). Zygoptera can 291 maintain the temperature of the thorax closer to the air, previous studies demonstrate that this difference 292 can be up to 1ºC (Shelly, 1982). They can be conformers or in some cases and the largest being heliothermic, 293 or even a continuum that may exist between these groups (Corbet and May, 2018) and future studies can 294 better investigate these categories. 295 Finally, niche characteristics may be important for the distribution of Odonata. ecophysiological 296 traits also was important for Anisoptera and Zygpoptera abundance. May [15] suggests also that climate, 297 body size and behavior are essential for maintaining the body temperature of Odonata. Changes in streams 298 due of anthropic activities alter microclimatic patterns such as air temperature, fundamental for the 299 physiological processes of Odonata species, leading to a change in the composition of species in these 300 environments [38]. We have demonstrated that adult odonate species composition varies in relation to 301 habitat integrity. We therefore suggest that their monitoring would provide a good indicator of riparian zone 302 quality considering niche characteristics and their thermorregulation habilities. 303 304 Acknowledgments 305 We thank Ana Luisa Fares, Ana Luiza Andrade, Erlane José Cunha, Fernando Geraldo de Carvalho, for 306 helping us with the biological sampling. 307 References 308 1. Siqueira T, Bini LM, Cianciaruso MV, Roque FO, Trivinho-Strixino S. The role of niche measures in 309 explaining the abundance–distribution relationship in tropical lotic chironomids. Hydrobiologia. 2009; 636: 310 163. https://doi.org/10.1007/s10750-009-9945-z .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 13 311 2. Heino J, Grönroos M. Untangling the relationships among regional occupancy, species traits, and niche 312 characteristics in stream invertebrates. Ecol Evol. 2014; 4: 1931–1942. https://doi.org/10.1002/ece3.1076 313 3. Rocha MP, Bini LM, Siqueira T, Hjort J, Grönroos M, Lindholm M, et al. Predicting occupancy and 314 abundance by niche position, niche breadth and body size in stream organisms. Oecologia. 2018; 186: 205– 315 216. https://dx.doi.org/ 10.1007/s00442-017-3988-z 316 4. Vilmi A, Tolonen KT, Karjalainen SM, Heino J. Niche position drives interspecific variation in 317 occupancy and abundance in a highly-connected lake system. Ecol Indic. 2019; 99: 159–166. 318 https://doi.org/10.1016/j.ecolind.2018.12.029 319 5. Hutchinson GE. Concluding remarks. Cold Spring Harb Symp Quant Biol. 1957; 22: 415–427 320 6. Brown JH. On the relationship between abundance and distribution of species. Am Nat. 1984; 124: 255– 321 279. https://doi.org/10.1086/284267 322 7. Hanski I, Kouki J, Halkka A. Three explanations of the positive relationship between distribution and 323 abundance of species. In: Ricklefs RE, Schluter D, editors. Species diversity in ecological communities: 324 historical and geographical perspectives. Chicago: University of Chicago Press; 1993. pp 108–116. 325 8. Venier LA, Fahrig L. Habitat availability causes the species abundance-distribution relationship. Oikos. 326 1996; 76: 564–570. https://dx.doi.org/ 10.2307/3546349 327 9. Legendre P, De Cáceres M. Beta diversity as the variance of community data: dissimilarity coefficients 328 and partitioning. Ecol Lett. 2013; 16: 951–963. https://doi.org/10.1111/ele.12141 329 10. Heino J, Grönroos M. Exploring species and site contributions to beta diversity in stream insect 330 assemblagesOecologia. 2016; 183: 151–160. https://doi.org/10.1007/s00442-016-3754-7 331 11. Dolédec S, Chessel D, Gimaret-Carpentier C. Niche separation in community analysis: a new method. 332 Ecology. 2000; 81: 2914–2927. https://doi.org/10.1890/0012-9658(2000)081[2914:NSICAA]2.0.CO;2 333 12. Castillo-Pérez EU, Suárez-Tovar CM, González-Tokman D,Schondube JE, Córdoba-Aguilar A. Insect 334 thermal limits in warm and perturbed habitats: Dragonflies and damselflies as study cases. J Therm Biol. 335 2022; 103: 103164. https://doi.org/10.1016/j.jtherbio.2021.103164 336 13. Rocha-Ortega M, Rodríguez P, Bried J, Abbott J, Córdoba-Aguilar A. Why do bugs perish? Range size 337 and local vulnerability traits as surrogates of Odonata extinction risk. Proc R Soc B. 2020; 287: 20192645. 338 https://doi.org/10.1098/rspb.2019.2645 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 14 339 14. Corbet PS, May ML. Fliers and perchers among Odonata: dichotomy or multidimensional continuum? 340 A provisional reappraisal. Int J Odonatol. 2008; 11: 155–171. 341 https://doi.org/10.1080/13887890.2008.9748320 342 15. May ML. Thermoregulation and adaptation to temperature in dragonflies (Odonata: Anisoptera). Ecol 343 Monogr. 1976; 46: 1–32 344 16. May ML. Thermoregulation and reproductive activity in tropical dragonflies of the genus Micrathyria. 345 Ecology. 1977; 58: 787–798 346 17. De Marco P. The Amazonian Campina dragonfly assemblage: patterns in microhabitat use and 347 behaviour in a foraging habitat (Anisoptera). Odonatologica. 1998; 27: 239–248 348 18. De Marco Júnior P, Batista JD, Cabette HSR. Community assembly of adult odonates in tropical 349 streams: an ecophysiological hypothesis. PLoS One. 2015; 10: 10:e0123023. 350 https://doi.org/10.1371/journal.pone.0123023 351 19. Calvão LB, Vital MVC, Juen L, Lima-Filho GF, Oliveira-Junior JMB, Pinto NS, et al. 352 Thermoregulation and microhabitat choice in Erythrodiplax latimaculata Ris males (Anisoptera: 353 Libellulidae). Odonatologica. 2013; 42: 97–108. 354 20. Calvão LB, Juen L, de Oliveira-Junior JMB, Batista JD, De Marco Júnior P. Land use modifies Odonata 355 diversity in streams of the Brazilian Cerrado. J Insect Conserv. 2018; 22: 675–685. 356 https://doi.org/10.1007/s10841-018-0093-5 357 21. Oliveira-Junior JMB, Dias-Silva K, Teodósio MA, Juen L. The Response of Neotropical dragonflies 358 (Insecta: Odonata) to local and regional abiotic factors in small streams of the Amazon. Insects. 2019; 10: 359 446. https://doi.org/10.3390/insects10120446 360 22. Luke SH, Dow RA, Butler S, Vun Khen C, Aldridge DC, Foster WA, et al. The impacts of habitat 361 disturbance on adult and larval dragonflies (Odonata) in rainforest streams in Sabah, Malaysian Borneo. 362 Freshw Biol. 2017; 62: 491–506. http://dx.doi.org/10.1111/fwb.12880 363 23. de Andrade VMS, Cordeiro IMCC, Schwartz G, Rangel-Vasconcelos LGT, de Assis Oliveira F. 364 Considerações sobre clima e aspectos edafoclimáticos da mesorregião Nordeste paraense. In: Cordeiro 365 IMCC, Rangel-Vasconcelos LGT, Schwartz G, de Assis Oliveira F, editors. Nordeste Paraense: Panorama 366 geral e uso sustentável das florestas secundárias. Belém: EDUFRA; 2017. pp 59–96. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 15 367 24. Viana PL, de Oliveira Mota NF, dos Santos Bragança Gil A, Salino A, Zappi DC, Harley RM, et al. 368 Flora das cangas da Serra dos Carajás, Pará, Brasil: história, área de estudos e metodologia. Rodriguésia. 369 2016; 67: 1107–1124. https://doi.org/10.1590/2175-7860201667501 370 25. ICMBio, Instituto Chico Mendes de Conservação da Biodiversidade. Plano de Manejo da Floresta 371 Nacional de Carajás (diagnóstico). 2016 [cited 10 August 2022]. https://www.gov.br/icmbio/pt-br 372 26. Strahler AN. Quantitative analysis of watershed geomorphology. Eos. 1957; 38: 913–920. 373 https://doi.org/10.1029/TR038i006p00913 374 27. Juen L, Cunha EJ, Carvalho FG, Ferreira MC, Begot TO, Andrade AL, et al. Effects of oil palm 375 plantations on the habitat structure and biota of streams in Eastern Amazon. River Res Appl. 2016; 32: 376 2081–2094. https://doi.org/10.1002/rra.3050 377 28. Brito JP, Carvalho FG, Juen L. Response of the Zygopteran Community (Odonata: Insecta) to Change 378 in Environmental Integrity Driven by Urbanization in Eastern Amazonian Streams. Ecologies. 2021; 2: 379 150–163. https://doi.org/10.3390/ecologies2010008 380 29. Calvão LB, Nogueira DS, de Assis Montag LF, Lopes MA, Juen L. Are Odonata communities impacted 381 by conventional or reduced impact logging?. For Ecol Manage. 2016; 382: 143–150. 382 http://dx.doi.org/10.1016/j.foreco.2016.10.013 383 30. Corbet PS. Dragonflies Behavior and ecology of Odonata. UK: University of Edinburgh Scotland; 1999. 384 31. Lencioni FAA. The Damselflies of Brazil: An Illustrated Guide The Non Coenagrionidae Families. São 385 Paulo: All Print Editora; 2005. 386 32. Lencioni FAA. The Damselflies of Brasil: An Illustrated Guide Coenagrionidae. São Paulo: All Print 387 Editora; 2006. 388 33. Garrison RW, von Ellenrieder N, Louton JA. Dragonfly Genera of the New World: an Illustrated and 389 Annotated Key to the Anisoptera. Baltimore: The Johns Hopkins University Press; 2006. 390 34. Garrison RW, von Ellenrieder N, Louton JA. Damselfly Genera of the New World: an Illustrated and 391 Annotated Key to the Zygoptera. Baltimore: The Johns Hopkins University Press; 2010. 392 35. Pessacq P. Synopsis of Epipleoneura (Zygoptera, Coenagrionidae, ‘‘Protoneuridae”), with emphasis on 393 its Brazilian species. Zootaxa. 2014; 3872: 201–234. https://doi.org/10.11646/zootaxa.3872.3.1 394 36. Mendoza-Penagos CC, Vinagre SF, Miranda-Filho J, Cruz A, Carvalho FG, Oliveira-Junior JMB, et 395 al. Colecciones científicas en Latinoamérica: La colección de adultos de Odonata del Laboratorio de 396 Ecología y Conservación (LABECO) de la UFPA. Hetaerina. 2022; 4: 20–22. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 16 397 37. Nessimian JL, Venticinque EM, Zuanon J, De Marco Jr P, Gordo M, Fidelis L, et al. Land use, habitat 398 integrity, and aquatic insect assemblages in Central Amazonian streams. Hydrobiologia. 2008; 614: 117– 399 131. https://dx.doi.org/ 10.1007/s10750-008-9441-x 400 38. Oliveira-Junior JMB, De Marco Junior P, Dias-Silva K, Leitão RP, Leal CG, Pompeu P S, et al. Effects 401 of human disturbance and riparian conditions on Odonata (Insecta) assemblages in eastern Amazon basin 402 streams. Limnologica. 2017; 66: 31–39. http://dx.doi.org/doi:10.1016/j.limno.2017.04.007 403 39. Miguel TB, Oliveira-Junior JMB, Ligeiro R, Juen L. Odonata (Insecta) as a tool for the biomonitoring 404 of environmental quality. Ecol Indic. 2017; 81: 555–566. http://dx.doi.org/10.1016/j.ecolind.2017.06.010 405 40. Brasil LS, de Lima EL, Spigoloni ZA, Ribeiro-Brasil DRG, Juen L. The habitat integrity index and 406 aquatic insect communities in tropical streams: A meta-analysis. Ecol Indic. 2020; 116: 106495. 407 https://doi.org/10.1016/j.ecolind.2020.106495 408 41. Mykrä H, Heino J. Decreased habitat specialization in macroinvertebrate assemblages in 409 anthropogenically disturbed streams. Ecol Complex. 2017; 31: 181–188. 410 http://dx.doi.org/10.1016/j.ecocom.2017.07.002 411 42. Zuur AF, Ieno EN, Walker NJ, Saveliev AA, Smith GM. Mixed effects models and extensions in 412 ecology with R. New York: Springer Science & Business Media; 2009. 413 43. Moral RA, Hinde J, Demétrio CGB. Half-normal plots and Overdispersed models in R: the hnp package. 414 J Stat Softw. 2017; 81: 1–23. https://doi.org/10.18637/jss.v081.i10 415 44. Cribari-Neto F, Zeileis A. Beta Regression in R. J Stat Softw. 2010; 34: 1–24. 416 https://doi.org/10.18637/jss.v034.i02 417 45. R Development Core Team. R: A language and environment for statistical computing. R Foundation 418 for Statistical Computing, Vienna, Austria. 2018 [cited 10 August 2022]. https://www.R-project.org/ 419 46. Venables WN, Ripley BD. Modern Applied Statistics with S. New York: Springer; 2002. 420 47. Oksanen J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, et al. vegan: Community 421 Ecology Package. R package version 2.5–7. 2020 [cited 10 August 2022]. https://CRAN.R- 422 project.org/package=vegan 423 48. ANA, Agência Nacional de Águas e Saneamento Básico. 2020 [cited 10 August 2022]. 424 https://www.ana.gov.br/ 425 49. Ribeiro Jr JW, Siqueira T, Brejão GL, Zipkin EF. Effects of agriculture and topography on tropical 426 amphibian species and communities. Ecol Appl. 2018; 28: 1554–1564. https://doi.org/10.1002/eap.1741 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 17 427 50. Slatyer RA, Hirst M, Sexton JP. Niche breadth predicts geographical range size: a general ecological 428 pattern. Ecol Lett. 2013; 16: 1104–1114. https://dx.doi.org/ 10.1111/ele.12140 429 51. Sahlén G. Specialist vs. generalists in the Odonata - the importance of forest environments in the 430 formation of diverse species pools. In: Rivera AC, editor. Forest and dragonflies: fourth WDA International 431 Symposium of Odonatology, Pontevedra (Spain). Sofia-Moscow: Pensoft Publishers; 2005. pp 153–179. 432 52. Alves‐Martins F, Calatayud J, Medina NG, De Marco P, Juen L, Hortal J. Drivers of regional and local 433 diversity of Amazonian stream Odonata. Insect Conserv Divers. 2019; 12: 251–261. 434 https://doi.org/10.1111/icad.12327 435 53. Conrad KF, Wilson KH, Whitfield K, Harvey IF, Thomas CJ, Sherratt TN. Characteristics of dispersing 436 Ischnura elegans and Coenagrion puella (Odonata): age, sex, size, morph and ectoparasitism. Ecography. 437 2002; 25: 439–445. https://doi.org/10.1034/j.1600-0587.2002.250406.x 438 54. Dijkstra K-DB, Kalkman VJ. Phylogeny, classification and taxonomy of European dragonflies and 439 damselflies (Odonata): a review. Org Divers Evol. 2012; 12: 209–227. https://doi.org/10.1007/s13127-012- 440 0080-8 441 55. Oliveira‐Junior JMB, Shimano Y, Gardner TA, Hughes RM, Marco Júnior P, Juen L. Neotropical 442 dragonflies (Insecta: Odonata) as indicators of ecological condition of small streams in the eastern Amazon. 443 Austral Ecol. 2015; 40: 733–744. https://doi.org/10.1111/aec.12242 444 56. Faria APJ, Paiva CKS, Calvão LB, Cruz GM, Juen L. Response of aquatic insects to an environmental 445 gradient in Amazonian streams. Environ Monit Assess. 2021; 193: 1–12. https://doi.org/10.1007/s10661- 446 021-09553-6 447 57. Gaston KJ, Borges PAV, He F, Gaspar C. Abundance, spatial variance and occupancy: arthropod 448 species distribution in the Azores. J Anim Ecol. 2006; 75: 646–656. https://doi.org/10.1111/j.1365- 449 2656.2006.01085.x 450 58. Heino J, Soininen J. Are common species sufficient in describing turnover in aquatic metacommunities 451 along environmental and spatial gradients?. Limnol Oceanogr. 2010; 55: 2397–2402. 452 https://doi.org/10.4319/lo.2010.55.6.2397 453 59. Neiss UG, Hamada N. Ordem Odonata (odous = dente; gnatha = mandíbula). In: Hamada N, Nessimian 454 JL, Querino RB, editors. Insetos aquáticos na Amazônia brasileira: taxonomia, biologia e ecologia. Manaus: 455 Editora do INPA; 2014. pp 217–282. .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint 18 456 60. Misof B. Diversity of Anisoptera (Odonata): Infering speciation processes from patterns of 457 morphological diversity. Zoology. 2002; 105: 355–365. https://doi.org/10.1078/0944-2006-00076 458 61. Hykel M, Růžičková J, Dolný A. Perch selection in Sympetrum species (Odonata: Libellulidae): 459 importance of vegetation structure and composition. Ecol Entomol. 2020; 45: 90–96. 460 https://doi.org/10.1111/een.12778 461 62. Calvão LB, Siqueira T, Faria APJ, Paiva CKS, Juen L. Correlates of Odonata species composition in 462 Amazonian streams depend on dissimilarity coefficient and oviposition strategy. Ecol Entomol. 2022; 47: 463 998–1010. https://doi.org/10.1111/een.13188 464 63. Birnie-Gauvin K, Lynch AJ, Franklin PA, Reid AJ, Landsman SJ, Tickner D et al. The RACE for 465 freshwater biodiversity: Essential actions to create the social context for meaningful conservation. Conserv 466 Sci Pract. 2023; 5: e12911. https://doi.org/10. 1111/csp2.12911 467 64. Harvey JA, Tougeron K, Gols R, Heinen R, Abarca M, Abram PK et al. Scientists' warning on climate 468 change and insects. Ecol Monogr. 2023; 93: e1553. https://doi.org/10.1002/ecm.1553 469 65. Hill GM, Kawahara AY, Daniels JC, Bateman CC, Scheffers BR. Climate change effects on animal 470 ecology: butterflies and moths as a case study. Biol Rev. 2021; 2113–2126. https://doi: 10.1111/brv.12746 .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint .CC-BY 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted September 16, 2024. ; https://doi.org/10.1101/2024.09.14.613059doi: bioRxiv preprint