Phenological patterns of neotropical lotic chironomids: Is emergence constrained by environmental factors? TADEU SIQUEIRA, 1,2 * FABIO DE OLIVEIRA ROQUE 3 AND SUSANA TRIVINHO-STRIXINO 2 1 Programa de Pós-graduação em Ecologia e Recursos Naturais, Universidade Federal de São Carlos, São Carlos, SP, Brazil (Email: tadeucapuzzo@yahoo.com.br), 2 Laboratório de Entomologia Aquática, Departamento de Hidrobiologia, Universidade Federal de São Carlos, and 3 Universidade Federal da Grande Dourados, Faculdade de Ciências Biológicas e Ambientais, 79825-070, Dourados, MS, Brazil Abstract In this paper we analyzed the emergence phenology of a highly diverse chironomid assemblage to test for association between emergence and some environmental variables and for the presence of synchrony in emergence. We used a time series of 48 weekly samples from a tropical low order forested stream (south-eastern Brazil) to describe how this assemblage varied in an intra-annual scale. An eigenvector-based filtering approach was adapted to create temporal variables that could be used in our multiple regression analyses, trying to overcome the problems of temporal autocorrelation. Emergence of the Chironomidae, two subfamilies, concordant species, and of dominant species was not related to rainfall, temperature, moon phase or photoperiod.Taxonomic composition and species richness did not change across time. The number of emerging individuals of the subfamily Orthocla- diinae was significantly related to temperature and to temporal filters. The inclusion of the temporal filters into the analyses almost eliminated autocorrelation in the regression residuals. We detected interspecific synchrony in a group of species, but an absence of trends and periodicity in chironomid emergence, which was not related to environmental variables. This suggests that unknown factors, differing from those known to control emergence in temperate regions, operate in the tropics. The erratic behaviour of the analyzed series raises the question of whether chaotic dynamics may generate this variability. Key words: aquatic insects, phenology, synchrony, time series, tropical streams. INTRODUCTION Temporal variations in ecological communities (sea- sonal, cyclic, or chaotic) depend on complex interac- tions between extrinsic forces and intrinsic dynamics (Bjørnstad & Grenfell 2001). Factors contributing to this variability operate at different scales, ranging from regional-level factors such as climate and geology (e.g. Lytle 2001) to habitat-level factors acting on indi- vidual taxa (e.g. Armitage et al. 1995; Peckarsky et al. 2001). Phenology, the study of the relationship between temporal variation on biological phenomena and envi- ronmental factors, particularly climate (Scott & Epstein 1987), is fundamental to understand commu- nity dynamics (Stiles 1977; Williams et al. 1999). Moreover, in the context of global climate change, phenological studies have received increased attention from biosphere modellers (e.g. Chuine et al. 2000) to conservation biologists (e.g. Terborgh 1986), since there is evidence that ecological and population pro- cesses are affected by climatic fluctuations (Stenseth et al. 2002). Several taxa of aquatic insects, especially those from northern temperate zones, show evident emergence patterns driven by specific environmental conditions (Resh & Rosenberg 1984). Among the Chironomidae (Diptera), which is one of the most diverse groups of organisms in freshwater environments (Cranston 1995), earlier studies posited that latitude, or its most related variable such as temperature, has a constrain- ing influence on emergence period and synchronicity (e.g. Corbet 1964). Species often emerge at particular periods of the year and generalizations may apply to higher taxonomic groupings like tribes and even subfamilies (Armitage 1995). For example, temperate streams have been shown to exhibit a seasonal sequence in adult emergence from winter to autumn (Coffman 1973). Additionally, but in a spatial context, there is even some evidence for spatial ‘zonation’ of species with altitude, where cold-adapted assemblages from high altitude can be distinguished from those inhabiting warmer lowlands (see Rossaro 1991). In summary, the general idea is that these patterns are *Corresponding author. Accepted for publication February 2008. Austral Ecology (2008) 33, 902–910 © 2008 The Authors doi:10.1111/j.1442-9993.2008.01885.x Journal compilation © 2008 Ecological Society of Australia