Drought-Escape Behaviors of Aquatic Insects May Be Adaptations to Highly Variable Flow Regimes Characteristic of Desert Rivers
Total Page：16
File Type：pdf, Size：1020Kb
September 2008 Notes 399 DROUGHT-ESCAPE BEHAVIORS OF AQUATIC INSECTS MAY BE ADAPTATIONS TO HIGHLY VARIABLE FLOW REGIMES CHARACTERISTIC OF DESERT RIVERS DAVID A. LYTLE,,* JULIAN D. OLDEN, AND LAURA E. MCMULLEN Department of Zoology, Oregon State University, Corvallis, OR 97331 (DAL, LEM) School of Aquatic and Fishery Sciences, University of Washington, Seattle, WA 98195 ( JDO) *Correspondent: [email protected] ABSTRACT—We document how two species of desert aquatic insects use positive rheotaxis to escape drought in desert rivers. We observed ca. 3,600 adults of the long-toed water beetle Postelichus immsi (Coleoptera: Dryopidae) crawling upstream concurrent with upstream recession of surface water in the Santa Maria River, La Paz and Mohave counties, Arizona. At the same time, we observed larvae of the gray sanddragon Progomphus borealis (Odonata: Gomphidae) burrowing and swimming upstream in large densities (690 larvae/m2). Both taxa moved with sufficient speed to arrive at perennial reaches of the river before being overtaken by drought. RESUMEN—Se documenta co´mo dos especies de insectos acua´ticos de zonas dese´rticas usan el reotaxis positivo para escapar a la sequı´a. Se observaron alrededor de 3,600 adultos del escarabajo Postelichus immsi (Coleoptera: Dryopidae) arrastra´ndose rı´o arriba mientras que el agua superficial se retrasaba en el rı´o Santa Marı´a en los condados de La Paz y Mohave, Arizona. Paralelamente se observaron larvas de THE SOUTHWESTERN NATURALIST 53(3):399–402 400 The Southwestern Naturalist vol. 53, no. 3 Progomphus borealis (Odonata: Gomphidae) de gran densidad (690 larvas/m2) desplaza´ndose rı´o arriba por vı´as subterra´neas y nadando. Ambas especies viajaron con suficiente velocidad para alcanzar los tramos perennes del rı´o, evadiendo la sequı´a. The dynamic hydrology of desert streams had a fairly abrupt beginning and end, with provides a unique opportunity to observe how density of individuals being somewhat uniform organisms evolve strategies for surviving floods along the entire length. Individuals observed and droughts. While several examples of behav- immediately upstream of the column showed no iors to escape floods have been documented evidence of positive rheotaxis but were foraging (Meffe, 1984; Lytle and Smith, 2004; Lytle and actively in all directions. We did not observe White, 2007), less is known about behavior to whether these upstream individuals were eventu- escape drought. This study documents behaviors ally incorporated into the column. Postelichus that allow desert aquatic insects to escape immsi occurs on debris and under rocks in streams drought by moving to perennial reaches of a from western Texas to California (Brown, 1972), river. and can be the most abundant insect in some On 10 April 2007, we observed aquatic insects desert streams (Evans and Hogue, 2006). in a drying reach of the Santa Maria River, La Paz In tandem with the dryopid beetles, we and Mohave counties, Arizona (34u1893.80N, observed larvae of the gray sanddragon Progom- 113u25927.50W). About 1500 h on a clear day phus borealis (Odonata: Gomphidae) moving (high of 31uC recorded at Parker, Arizona), the upstream in large numbers (690 larvae/m2). lower end of a 3–4-km perennial reach was Sanddragon larvae burrowed to just underneath receding upstream at a rate of 0.16 cm/s, the surface of the substrate, so most larvae presumably due to high evapotranspirative de- remained hidden beneath sand as they moved. mand from riparian plants, including Salix good- However, numerous burrows left over from this dingii (Goodding willow), Populus fremontii (Fre- movement were apparent after the water receded mont cottonwood), and Tamarix ramosissima (salt (Fig. 1). We observed larvae moving upstream at cedar). Velocity of flow immediately above the 0.17 cm/s (60.06 cm/s; n 5 6). Although this recession point was 9.9 cm/s (95% C.I. 6 burrowing rate was slightly greater than the rate 1.0 cm/s; n 5 5 observations), whereas down- of stream recession of 0.16 cm/s, larvae occa- stream of this point the river became hyporheic sionally left the substrate to swim actively in short in deep sand. bursts of ca. 10 cm. This behavior increased their Beginning 11 m upstream of the recession average speed and allowed escape from the point, we observed thousands of adult long-toed recession point. Progomphus borealis occurs in water beetles Postelichus immsi (Coleoptera: Dryo- sandy shallows of rivers and streams throughout pidae) crawling upstream in a 37 m-long column the arid western regions of the United States and along the shallow river margins (Fig. 1). We Mexico (Needham et al., 2000). counted an average of 97 individuals/m of Other taxa that were present in small num- stream (634 individuals; n 5 12 counts), giving bers, so that we cannot be sure if they were an estimate of 3,600 individuals moving up- systematically moving upstream, include larvae of stream. All showed positive rheotaxis, in that the giant water bug Abedus herberti (Hemiptera: crawling always was oriented in the upstream Belostomatidae), adults of Berosus (Coleoptera: direction. Individuals moved at a rate of Hydrophilidae), tadpoles of the canyon treefrog 0.70 cm/s (60.03 cm/s; n 5 12), which was four Hyla arenicolor (Hylidae), longfin dace Rhinichthys times faster than rate of stream recession, chrysogaster (Cypriniformes: Cyprinidae), and suggesting that the behavior would allow escape mosquitofish Gambusia affinis (Cyprinodonti- from local drought (a deeper, more permanent formes: Poecilidae). Larvae of the American river reach was 80 m upstream from the reces- rubyspot damselfly Hetaerina americana (Odo- sion point). Most individuals traveled along the nata: Calopterygidae) were not observed moving stream margins in water ,1 cm deep, possibly to upstream but were seen stranded in drying areas, avoid the faster current. Individuals and small often underneath mats of algae where they groups (typically 2–3 individuals) occasionally might be protected from desiccation. became dislodged by the current and tumbled Some taxa used positive rheotaxis so long as some distance downstream. The moving column sufficient water current was available. Physa snails September 2008 Notes 401 FIG. 1—Left) column of long-toed water beetles (Postelichus immsi) moving upstream to avoid drought. Recession point is at top of figure. Burrows of gray sanddragons (Progomphus borealis) also are visible near the 11.5- cm tall vial. Right) closeup of Postelichus crawling upstream (flow travels from bottom to top of figure). (Pulmonata: Physidae) in the flowing active have recorded synchronous emergence of aquat- channel exhibited positive rheotaxis at a rate of ic beetles en masse from drying desert ponds 0.12 cm/s (60.06 cm/s; n 5 12 individuals), but (Zimmerman, 1959; Smith, 1973; Kingsley, individuals in isolated pools with no flow 1985). In the dytiscid beetles Rhantus gutticollis, remained beneath algae mats. Soldier fly larvae R. binotatis, and Eretes sticticus, mass emigration (Diptera: Stratiomyidae) and adult hydrophilid was preceded by loud, sustained vocalization, beetles exhibited this as well. The directional cue which raises the question of whether individuals provided by flowing water apparently is necessary can communicate a signal to emigrate (Smith, for upstream movement, because we never 1973; Kingsley, 1985). Although we did not observed individuals of any taxon moving sys- notice vocalizations during our observations of tematically upstream in still-water reaches. This is Postelichus and Progomphus, underwater vocaliza- congruent with the observation of Gray and tion remains a possibility. In the case of Rhantus Fisher (1981) that post-flood, upstream move- studied by Smith (1973), sound production and ment of aquatic insects generally is greater emigration were associated with elevated tem- during periods of higher discharge, although peratures, high density of individuals, and they noted little upstream movement of either P. absence of suitable bottom substrates; i.e., immsi (5Helichus immsi; Nelson, 1989) or P. proximate cues associated with drying ponds. borealis. During our observations, both rate of flow and The proximate cues used to initiate the water temperature were similar above and below observed behavior to escape drought by P. immsi the column, so there apparently was no physical and P. borealis are not known. Several studies cue to signal that the recession point was 402 The Southwestern Naturalist vol. 53, no. 3 approaching. Increased density of conspecifics LITERATURE CITED remains one possible cue, such that individuals are recruited into the dense column as it moves BROWN, H. P. 1972. Aquatic dryopoid beetles (Coleop- tera) of the United States. Biota of Freshwater upstream. It also is possible that individuals Ecosystems Identification Manual 6 (18050 ELD), caught at the drying end of the stream become Environmental Protection Agency, Washington, activated to move upstream. If the latter is D.C. occurring, then individuals at the front of the EVANS, A. V., AND J. N. HOGUE. 2006. Field guide to column should have crawled from the furthest beetles of California. University of California Press, distance downstream. Both of these hypotheses Berkeley. are testable by appropriate mark-release of GRAY, L. J., AND S. G. FISHER. 1981. Postflood recoloni- individuals and careful observation in the field. zation pathways of macroinvertebrates in a lowland Drought in desert rivers occurs at multiple Sonoran Desert stream. American Midland Natu- ralist 106:249–257. timescales: daily, as reaches dry locally due to KINGSLEY, K. J. 1985. Eretes sticticus (L.) (Coleoptera: evapotranspiration of riparian plants; seasonally, Dytiscidae): life history observations and an account as the water table drops due to low rainfall; and of a remarkable event of synchronous emigration episodically following floods, as river margins from a temporary desert pond. Coleopterists contract post-flood (Stanley et al., 1997). We Bulletin 39:7–10. have not observed synchronous upstream move- LYTLE, D. A., AND R. L. SMITH. 2004. Exaptation and flash ment of either taxon during routine benthic flood escape in the giant waterbugs.