<b>Supporting data - Length-dry mass relationships of aquatic insects: geographic and taxonomic variation in a digital database - Freshwater Biology</b>

Dataset belonging to Length-dry mass relationships of aquatic insects: geographic and taxonomic variation in a digital database in Freshwater BiologyAquatic insects are an abundant, yet declining, taxonomically heterogeneous group with special importance in both aquatic and terrestrial ecosystems. Accurate estimations of insect biomass during their aquatic stages are essential to advance our fundamental knowledge about insects, their roles in ecosystems and their vulnerability to human impact. However, assessing insect biomass from samples using classical drying techniques is time consuming and prohibits the use of samples for other analyses. A widely applied method is therefore to use length-dry mass power regressions to obtain dry mass (DM) from body lengths (BL) using literature-derived parameter values. However, application of this method relies on reliable and accessible parameter values, preferably matching the studied specimens taxonomically and geographically. Here, we aimed to increase parameter accessibility in the literature to (1) facilitate researchers in employing more appropriate length-mass regressions in their studies, (2) identify knowledge gaps that can direct future research towards unexplored regions and understudied taxonomic groups, and (3) visualise the relative contribution of geographic variation (differences among continents) and taxonomic variation (differences among families within each order) to regression lines. We compiled a parameter dataset based on 25 publications for eight insect orders with aquatic life stages: Coleoptera, Diptera, Ephemeroptera, Hemiptera, Megaloptera/Neuroptera, Odonata, Plecoptera and Trichoptera, and made the dataset available digitally. This parameter dataset is derived from over 15000 measured insects of 84 (sub)families and 233 genera from all continents, except Africa and Antarctica. We found parameter values to be widely available at the order level, but at the resolution of family and genus level values are missing for respectively 65% and 94% of the taxa.Identified knowledge gaps were the need for (1) more data on variation among families that is collected standardised within the same geographic regions, (2) targeted collections of data for different orders within the same study areas, to reveal variation among families and genera, and (3) careful reporting of the exact methodologies used, to identify variation introduced by methodological variation. Geographic and taxonomic variation is visually presented in figures for further interpretations. We conclude that power regressions can be a powerful method, but due to data-shortage at the genus or family taxonomic levels, order level regressions with less reliability are necessarily applied. By providing parameters in a new digital dataset, we hope to facilitate users in more efficient assessment of parameter availability for studied taxa in any geographic region. The identified knowledge gaps can be used to direct future research efforts. More accessible parameter data will facilitate more reliable assessments of aquatic insect biomass, and benefit future studies on this important and abundant group of organisms bridging aquatic and terrestrial ecosystems.

《淡水生物学》中水生昆虫体长-干重关系数据集：数字数据库中的地理与分类变异
水生昆虫是一类数量丰富但正逐渐减少的分类异质性类群，在水生和陆地生态系统中均具有特殊重要性。准确估计昆虫水生阶段的生物量，对于深化我们对昆虫的基础认知、其在生态系统中的作用及其对人类影响的脆弱性至关重要。然而，使用经典干燥技术从样本中评估昆虫生物量耗时较长，且会阻碍样本用于其他分析。因此，一种广泛应用的方法是利用文献推导的参数值，通过体长-干重幂回归从体长（BL）获取干重（DM）。但该方法的应用依赖于可靠且可获取的参数值，最好与研究样本的分类和地理特征相匹配。
本文旨在提升文献中参数的可获取性，以（1）帮助研究者在其研究中采用更合适的体长-干重回归模型；（2）识别知识空白，引导未来研究关注未探索区域和研究不足的分类群；（3）可视化地理变异（大陆间差异）和分类变异（各目内科间差异）对回归模型的相对贡献。
我们基于25篇文献，针对8个具有水生生活史阶段的昆虫目（鞘翅目、双翅目、蜉蝣目、半翅目、广翅目/脉翅目、蜻蜓目、襀翅目和毛翅目）构建了参数数据集，并将其以数字形式公开。该参数数据集来源于除非洲和南极洲外所有大陆的15000余只测量昆虫，涵盖84个（亚）科和233个属。
研究发现，目水平的参数值较为丰富，但在科和属水平上，分别有65%和94%的类群缺乏参数值。已识别的知识空白包括：（1）需在相同地理区域内标准化收集更多科间变异数据；（2）在同一研究区域内针对不同目开展定向数据收集，以揭示科和属间的变异；（3）详细报告所用的具体方法，以识别方法学变异引入的差异。地理和分类变异通过图表直观呈现，供进一步解读。
我们认为，幂回归是一种强有力的方法，但由于属或科分类水平的数据不足，不得不应用可靠性较低的目水平回归模型。通过在新的数字数据集中提供参数，我们希望帮助用户更高效地评估任何地理区域内研究类群的参数可获取性。已识别的知识空白可用于指导未来的研究方向。更易获取的参数数据将促进水生昆虫生物量的更可靠评估，并助力关于这一连接水生与陆地生态系统的重要且丰富类群的未来研究。