Length-dry mass regressions for Leptonema (Trichoptera, Hydropsychidae) larvae in a Neotropical headwater stream

Abstract: Aim The objectives of this study were to evaluate which allometric measurements of Leptonema larvae are most suitable in order to develop mathematical equations to describe biomass relationships for the population of this taxon in a reference condition headwater stream. Methods We measured four body dimensions (body length, interocular distance, horizontal width of cephalic capsule and vertical width of the cephalic capsule) of 65 Leptonema larvae, which were collected in February 2022, in the Taboões spring, Serra do Rola Moça State Park, Minas Gerais, using a Surber sampler. For the determination of allometric measurements, each individual was photographed under a dissecting stereomicroscope (Leica M80) equipped with a digital camera. Each photographed specimen's length was measured using the Motic Image Plus 2.0 software. After measuring the linear body dimension and direct measurement of the biomass, we used these values to calculate the length-mass mathematical equations. To the equations use power models: DM = a Lb, where a/b are constants, DM is the dry mass, L is the linear body dimension. Results Among body dimensions of Leptonema larvae, body length showed the greatest range of variation, with values ranging from 4.03 to 25.28 mm, followed by head capsule vertical width (0.51 - 2.69 mm), head capsule horizontal width (0.55 - 2.22 mm) and interocular distance (0.24 - 1.88 mm). Our results show that body length provided the best-fitting equation for estimating biomass (R2 = 0.90). However, we observed a close fit between the other allometric measures, including high coefficients of determination, head capsule horizontal width (R2 = 0.85), interocular distance (R2 = 0.82), head capsule vertical width (R2 = 0.78). Conclusions These results will be useful in providing the best allometric measurement and equations to estimate the biomass of Leptonema larvae from the tropics.

Acta Limnologica Brasiliensia, 2023, vol. 35, e5 Mährlein et al., 2016). Furthermore, length-body mass relationships should be taxon-specific, as there may be a different relationship for each taxon (Baumgärtner & Rothhaupt, 2003;Becker et al., 2009). Therefore, studies on the biomass of aquatic insects should consider the specificity of each region.
Aquatic insects of the Hydropsychidae family are among the most diverse and abundant groups of freshwater ecosystems (Pes, 2005), with relevance to ecological processes, such as nutrient cycling and energy flow (Balachandran et al., 2012). Among the genera of this family, Leptonema is widely distributed in the tropics, comprising a significant proportion of the invertebrate biomass of tropical streams (Muñoz-Quesada, 1999). Leptonema larvae preferentially occur in rocky bottom habitats with strong water currents (Buss et al., 2004), where they filter small particles of organic matter in the water column (Gholizadeh & Heydarzadeh, 2020), forming an important link in the transfer of energy between the trophic chains. Thus, the structure of a Leptonema population can be a useful tool for understanding different ecological issues of headwater streams, including thermodynamic indicators (e.g., Linares et al., 2018;, fluvial mesocosm studies assessing drift movements (Calapez et al., 2017).
In this study, the main objectives were to evaluate which allometric measurements of the Leptonema body larvae are most suitable in order to develop mathematical equations to describe biomass relationships for the population of this taxon in a headwater stream. We want to identify which allometric measurements (body length, interocular distance, horizontal width of the cephalic capsule and vertical width of the

Introduction
The biomass of aquatic macroinvertebrates is an important metric for estimating energetic processes in lotic ecosystems, such as trophic relationships between functional feeding groups, population growth rates, secondary production of communities (Benke et al., 1999;Gjoni et al., 2022) and energy balance of ecosystems (Steele et al., 2007). However, directly weighing each organism is an impractical and often error-prone process, especially in terms of weighing dry mass or ash-free dry mass (Edwards et al., 2009). To avoid such problems, some studies suggest to estimate biomass indirectly, using length-mass equations (González et al., 2002;Horta et al., 2006;Edwards et al., 2009).
Indirect equation-based methods are faster and more efficient and have the advantage of keeping the taxon preserved for future evaluations (e.g., molecular analysis; Towers et al., 1994), without causing loss of the organism in the drying process (Mährlein et al., 2016). Thus, it is usual to establish length-mass relationships of taxa using the available literature (Benke et al., 1999;González et al., 2002;Martins et al., 2014). However, it is important that these data are used with caution, as they may not take into account the environmental and geographic variations of the studied area (Méthot et al., 2012), overestimating the true body mass.
cephalic capsule) of Leptonema larvae show higher correlation to biomass. We expect that body length is a better predictor of biomass, because it has a wide measurement range between allometric measurements.

Study area
Leptonema larvae were sampled in the Taboões headwater stream (20°03'38 "S -44°03'03" W), located in the Serra do Rola Moça State Park (PESRM), Minas Gerais state, Southeastern Brazil. The PESRM covers an area of 3,942 hectares and is located in a transition area between the Atlantic Forest and Cerrado (Reis & Machado, 2019), in the Rio São Francisco River basin. The Taboões stream is a reference site for human water supply with waters of excellent quality (special quality, Brazilian water classification by Brasil, 2000), 250 L/s discharge. According to the Köppen system, the climate is classified as Cwb (altitude tropical), with rainy summers and dry winters (Brandão et al., 1997). The average annual precipitation varies between 1,300-2,100 mm and the average temperature between 18º-21º C (Meyer et al., 2004).

Sampling and laboratory procedures
In the Taboões stream, a 50 meter transect was established, and the sediment substrate at 22 sampling points was collected using a Surber sampler (area of 0.9 m 2 and mesh of 250μm). Each sample was immediately sorted on a white tray, and all Leptonema larvae were individually deposited in 2 ml eppendorf-type microtubes, without the addition of preservatives and properly identified. The material was packed in a thermal box with ice and taken to the laboratory.

Length-mass equation calculations
In the laboratory, Leptonema specimens were identified (Pes, 2005) under a stereomicroscope, and then each individual was photographed under a dissecting stereomicroscope (Leica M80 model) equipped with a digital camera (Leica IC 80 HD model). For the determination of allometric measurements (Benke et al., 1999;Mährlein et al., 2016), four measurements of linear body length were chosen as a predictor of biomass: (i) body length; (ii) interocular distance; (iii) horizontal width of the cephalic capsule and; (iv) vertical width of the cephalic capsule (Figure 1). To determine body length, the distance from the anterior section of the head to the posterior section of the last abdominal segment was measured. Interocular distance was measured as the minimum distance between the eyes. For the horizontal width of the cephalic capsule, we measure the widest section of the head. Vertical width of the cephalic capsule was measured from the anterior part of the head to the beginning of the pronotum. Each photographed specimen's length was measured using the Motic Image Plus 2.0 software. Measurements of length and mass were performed with unbroken individuals, containing all appendages.
The measured individuals were placed individually in pre-weighed porcelain crucibles, dried in an oven at 60°C for 48 h (Becker et al., 2009), allowed to cool in a desiccator and their dry mass measured on a balance with ± 0.001 g accuracy. Subsequently, to estimate the ash weight, the individuals were incinerated in a muffle furnace at 550°C for 4 hours, and their ash mass was measured by the same procedure.
After the direct measurement of biomass, we used these values to calculate the length-mass equations for each of the measurements of linear body length (body length, interocular distance, horizontal width of the cephalic capsule, vertical width of the cephalic capsule). We also calculated a Pearson correlation between each measurement of linear body length and biomass. The power model was calculated for the four body dimensions of Leptonema larvae, using the least squares method. The adjustment of the equations was performed by the coefficient of determination (R 2 ) and the level of significance (p < 0.01) obtained by a Generalized Linear Model (GLM) using a Gaussian distribution. All calculations were made using the R software (R Core Team, 2015). Body length measurements and biomass measurements of the 65 Leptonema larvae were used for statistical analysis. To arrive at the equations that determine the length-mass relationship, we used models that predict mass as a power function of a linear dimension.

Results
Among body dimensions of Leptonema larvae, body length showed the greatest range of variation, with values ranging from 4.03 to 25.28 mm, followed by head capsule vertical width (0.51 -2.69 mm), head capsule horizontal width (0.55 -2.22 mm) and interocular distance (0.24 -1.88 mm) (Table 1).
Regression analyses show the length-to-mass relationship for body length, interocular distance, horizontal width of the cephalic capsule and horizontal width of the cephalic capsule of Leptonema larvae. Length-mass curves for the Leptonema larvae, using linear and logarithmic scales, of the power function ( Figure 2). The equations generated for each linear body length are listed in Table 2. All allometric body dimensions of Leptonema larvae showed significance relation for biomass (p < 0.01). Body length showed the best fit to estimate biomass, followed by horizontal head size, interocular distance, and vertical head size (Table 2).

Discussion
Our results showed that the power models presented a high correlation coefficient, explaining 78% to 90% of the variation in biomass of Leptonema larvae as a function of the allometric measurements used (body length, interocular distance, horizontal head size and vertical head size). In fact, our results are in line with other studies in the tropical region (e.g., Becker et al., 2009;Silva et al., 2010). This reinforces that power models for length-mass of aquatic macroinvertebrates provide satisfactory results for the relationship between body dimensions and biomass of freshwater invertebrates, including Leptonema larvae.
Although all length-mass relationships were significant across the entire range of allometric dimensions, body length was the best predictor, explaining up to 90% of the biomass variation. The result of this study supports our hypothesis that body length provides a better estimate of biomass for Leptonema larvae. Because it has a wider measurement range, body length is a measure often used to estimate insect larvae biomass (e.g., Genkai-Kato & Miyasaka, 2007;Mährlein et al., 2016). Similar results were found by Martins et al. (2014), who found that body length was the best biomass predictor for a population of Table 1. Range, mean, standard deviation (SD), coefficient of variation (CV = (SD / mean)*100, in 100%) and number of observations (N) for body dimensions and body mass of Leptonema larvae from Taboões stream. Phylloicus elektoros (Calamoceratidae, Trichoptera) in the Brazilian Central Amazon. Allometric measurements for intraocular distance, vertical head size, and horizontal head size also showed high correlation with biomass. These body dimensions were also used by Cressa (1999) and Becker et al. (2009), due to sclerotized linear dimensions such as cephalic capsule width and pronotum length being less subject to distortions, breaks and deformations under individual manipulations (Johnston & Cunjak, 1999;Becker et al., 2009), when compared to body length. Becker (2005), studying the life cycle of Agapetus fuscipes (Trichoptera) in a stream in Germany, found that pronotum length is a reliable measure for different larval instars of the species. However, in our study, adjusted regression of sclerotized structures provided a lower fit than body length.
Preservation of invertebrates in ethanol or formaldehyde, often indispensable due to the amount and time required to process the collected samples (Nolte, 1990;Dekanová et al., 2022), can lead to the loss of more than 50% of their biomass (Silva et al., 2010). Preservatives substances can dissolve lipids that are present in the larval body, mainly during the first three weeks after conservation, reducing biomass estimates (Wetzel et al., 2005;Benke & Huryn, 2010). In our study, Leptonema larvae were kept at low temperature (-20°C), without preservatives and all measurements were performed on unharmed individuals, which allowed a concise and safe determination of the relation between the biomass and the four body dimensions studied.

Conclusions
In conclusion, it was observed that the body length presented the best fit to estimate the Leptonema biomass, corroborating our initial hypothesis. However, all other allometric measurements studied provided good estimates for the biomass of this taxon. The power model described the length-mass relationships well and may be a good choice for studies with other aquatic insect species. Likewise, our results also reinforce the need for more length-mass studies of aquatic insects in the tropical region. We believe that, in order to obtain more reliable results, data on the lengthmass relationship should be obtained based on the population of the studied region. Although we may have more than one species for Leptonema larvae, we assume that multiple species differences are not relevant for genera biomass equations. As the larvae are so similar that only specialists could identify them if present, we can safely assume that they are similar enough for not have significant differences in the measurements used in this study. Therefore, the results presented here may be useful to determine the biomass of Leptonema larvae from the Neotropical Savanna. It is important to highlight that the length-dry mass equations can contribute to approaches with thermodynamic indicators for macroinvertebrate assemblages and will serve as a basis for future experimental studies in a mesocosm system, in order to test the Leptonema response to multiple pressures, such as flow change of water and oxygen depletion in streams of the Brazilian neotropical biome.