SEARCH

SEARCH BY CITATION

Keywords:

  • Catchment land use;
  • dispersal;
  • insect;
  • meteorological effects;
  • streams

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Abstract. 1. Dispersal of adult aquatic insects between streams may have important consequences for local and regional population dynamics, but little is known about how dispersal is affected by weather conditions.

2. The influence of meteorological variables on flight activity of adult stoneflies (Plecoptera: Leuctridae, Nemouridae, and Chloroperlidae) was investigated using Malaise traps adjacent to three upland streams in the Plynlimon area of mid Wales, U.K.

3. Numbers of adult stoneflies captured weekly in the traps were related positively to air temperature and related negatively to wind speed. Meteorological conditions during daylight showed stronger relationships with flight activity than did conditions at night.

4. There was inter-site variation in the strength of weather effects on stonefly flight. Wind speed was significant at only one site, which had higher average wind speed than the other sites.

5. Annual variation in weather conditions during adult flight periods may result in varying extent of dispersal between sites, influencing community dynamics over a wide area.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Dispersal by winged adult stages may enable stream-dwelling insects, such as stoneflies (Plecoptera), to circumvent terrestrial barriers between adjacent freshwater habitats. The movement of adults between habitats allows colonisation of new or previously disturbed habitats, and may have important consequences for local and regional population dynamics and inter-population gene flow (Whiles & Wallace, 1992; Palmer et al., 1996; Bunn & Hughes, 1997; Griffith et al., 1998).

Most adult aquatic insects are short lived relative to the duration of the larval stages and hence have a limited temporal window in which to disperse. The meteorological conditions experienced over the duration of adult life may exert an important influence on the extent of dispersal. Studies of terrestrial insects have demonstrated the effects of weather conditions, and particularly the effects of air temperature, on flight activity (Williams, 1940, 1961; Taylor, 1963; Johnson, 1969; Abrol, 1991; Peng et al., 1992). If, as seems likely, adult aquatic insects are affected in similar ways, variation in weather conditions within and between years may influence the dispersal and survival of adults. This may in turn affect the structure and dynamics of larval communities within streams (Zwick, 1990; Palmer et al., 1996; Harrison & Hildrew, 1998).

Only comparatively recently has inter-population dispersal by adult aquatic insects been examined (Collier & Smith, 1998; Griffith et al., 1998; Petersen et al., 1999; Briers et al., in press), and only one study (Waringer, 1991) has examined the effects of weather conditions on the flight of adult aquatic insects. In the work reported here, the influence of meteorological conditions on the flight activity of adult stoneflies (Plecoptera) was examined at three adjacent upland streams in mid Wales, U.K.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Field site

The study was carried out at three streams [Afon Hafren (52°28.5′N, 3°42.2′W), Afon Hore (52°28.0′N, 3°42.9′W), and Afon Gwy (52°27.2′N, 3°43.9′W)], which are headwaters of the Rivers Severn and Wye in the Plynlimon area, mid Wales, U.K. (Fig. 1). At the point of study, the streams are 3–4 m wide, with mixed substrata of gravel, cobbles, and bedrock (Upper Ordovician and Silurian mudstones, shales, and grits). All the streams have similar patterns of discharge and physico-chemical characteristics (Gee & Smith, 1995; Hudson et al., 1997). The invertebrate communities of the streams are dominated numerically by larval stoneflies, particularly Leuctridae (Gee & Smith, 1995, 1997; Monteith & Evans, 2000).

image

Figure 1. Map of the study streams showing the location of Malaise trap transects (solid bars crossing streams) and weather stations (•). Letters A and B refer to the designation of the different weather stations; see text for details. The extent of the transect has been exaggerated for clarity. The dashed line indicates the approximate extent of coniferous plantation forestry.

Download figure to PowerPoint

The three streams have contrasting catchment land-use. Approximately 50% of the catchment of the Afon Hafren, including the study area, is covered by mature Sitka spruce Picea sitchensis (Bong) Carrière and Norway spruce Picea abies Karst., which are planted up to the edge of the stream. This plantation was established between 1937 and 1964 (Neal et al., 1990). Similar plantation forest surrounds the majority of the Afon Hore, but in the study area it was clear-felled between 1985 and 1989 (Gee & Smith, 1997; Roberts & Crane, 1997). The area has since been replanted with a similar mix of forest, although a 50–60 m wide buffer strip either side of the stream has not been replanted with conifers. Riparian vegetation in the buffer strip consists of a mixture of grasses (mainly Festuca sp.), heather Calluna vulgaris (L.) Hull, bilberry Vaccinium myrtillis L., and scattered trees (mainly rowan Sorbus aucuparia L.). The catchment of the Afon Gwy is open, sheep-grazed moorland, supporting a plant community dominated by Nardus stricta L. and Festuca sp. on the drier slopes and Molinia caerulea (L.) Moench and Juncus spp. on the wetter valley bottom.

Stonefly sampling methods

Adult insects were captured in double-headed Malaise traps (Malaise, 1937). The two sides of the Malaise trap were divided by a panel of mesh that allowed insects trapped in each side to be collected separately in bottles at the head of the trap. The long axis of each Malaise trap was parallel to the stream edge. Thus, each side of the trap captured insects moving either away from or towards the stream. Captured insects were killed and preserved in 50% industrial methylated spirits. At each stream, eight traps were placed in a transect perpendicular to the river at approximately logarithmic intervals (2.7, 7.4, 20.1, 54.6 m) away from the channel on each bank for the purposes of studying dispersal of adult stoneflies away from the stream (Briers et al., in press). For the analyses presented here, the catches in all traps at each site were combined to give an estimate of overall abundance. The traps were emptied weekly from 24 May to 1 November 2000. Adult stoneflies captured in the traps were identified to species using Hynes (1977), sexed and counted in the laboratory, then preserved in 70% industrial methylated sprits.

Meteorological data

Meteorological data were obtained from two automatic weather stations that are part of a larger network established at the Plynlimon site (Hudson et al., 1997). Instrumentation for recording meteorological conditions is located ≈ 1.5 m above ground level at each site. One of the stations was located adjacent to the study site on the Gwy (A) and the other next to the site on the Hafren (B) (Fig. 1). Although the second station was near the Hafren study site, it was located outside the mature forest in a small clearing with vegetation similar to that in the buffer strips of the Hore. Each station recorded hourly measurements of wind speed (m s−1) and direction (0–360°), rainfall (mm), air temperature (°C), and relative humidity (%).

Statistical methods

The number of insects captured in a trap per unit time is a function of the local population size multiplied by its level of activity (Williams, 1940; Johnson, 1969). Analysis of the effects of meteorological factors on insect activity therefore has to account for seasonal changes in insect abundance that are independent of meteorological conditions. If the influence of seasonal changes in abundance can be removed from catches of insects, the remaining variation should be a reflection of insect activity.

Many adult stoneflies exhibit a unimodal pattern of seasonal abundance that approximates a normal (Gaussian) distribution (Harper & Pilon, 1970; Harper, 1973; Singh et al., 1984; Zwick, 1990; Petersen et al., 1999). Therefore, in order to factor out seasonal variation in abundance, Gaussian curves were fitted to frequency distributions of catches throughout the sampling period using non-linear regression in SigmaPlot 4.01 (SPSS Inc., 1997). The equation of the curve fitted was:

  • image

where x is the week of sampling, week 1 is the first week in which the species was captured in the traps, and y is the frequency of catching the species in the traps. The parameters μ and σ are the mean and standard deviation of the fitted curve respectively.

Some studies of the seasonal abundance of adult stoneflies have found evidence of protandry, i.e. emergence of males prior to females (e.g. Singh et al., 1984). In the species studied here, there was no evidence of protandry (Briers et al., in press) so males and females were combined for analysis. Residuals from the fitted Gaussian curve, i.e. more or fewer individuals captured in the traps than expected, were used as a measure of stonefly flight activity. The residuals were regressed against weekly mean meteorological variables (wind speed, rainfall, air temperature, relative humidity) using stepwise multiple linear regression (P for entry of variables into regression model = 0.05). There was some evidence of multicollinearity between the meteorological variables but this was not severe (Pearson correlation coefficients, P > 0.05) so all variables were retained for entry into the regression models. Meteorological data from station A were used for catches at the Gwy and from station B for the Hore and Hafren sites. Separate analyses were carried out for day and night periods (defined by sunrise and sunset times for Aberystwyth, 25 km to the west of the study site) in order to test whether there was any difference in the predictive power of variables for these two periods that could indicate that flight activity occurs more commonly during the day or night.

The direction of the wind relative to the orientation of the transect of Malaise traps may influence the relative catches in traps on the two opposing banks of each stream or in the two sides of each trap. The difference between the combined weekly catches from traps on each bank divided by the total abundance (both banks combined) was calculated to give a measure of the relative catch on each side of the transect. The difference between catches in each trap side (for all traps at each site combined), again divided by the total catch, was calculated as a measure of relative catch on each side of the traps. These measures were regressed separately on the difference between the orientation of each transect and the weekly mean angle of the wind using linear-circular regression (Fisher, 1993) to test for an effect of wind direction.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Details of the species and numbers of stoneflies captured in the Malaise traps were given by Briers et al. (in press). Here only the five most abundant species [Amphinemura sulcicollis Stephens, Leuctra fusca (L.), L. inermis Kempny, L. nigra (Olivier), and Siphonoperla torrentium (Pictet)], which together made up ≈ 75% of the catch, are analysed. Other species were not sufficiently abundant to warrant analysis. A summary of the meteorological conditions over the period of study is shown in Fig. 2. Weather conditions at both stations generally showed a similar pattern of seasonal variation. Mean air temperatures and wind speeds were higher during the day than at night (paired t-test of weekly means, air temperature: station A t23 = 10.77, P < 0.001, station B t23 = 12.11, P < 0.001; wind speed: station A t23 = 10.96, P < 0.001, station B t23 = 8.12, P < 0.001), and relative humidity was lower during the day than at night (station A t23 = 11.05, P < 0.001, station B t23 = 16.39, P < 0.001). Other variables did not show a consistent pattern of variation between day and night. Wind speed was markedly higher at station A than at station B (paired t-test of weekly mean wind speeds, day t23 = 16.35, P < 0.001, night t23 = 15.11, P < 0.001). A consistent shift in wind direction between day and night was found at station B but not at station A, possibly due to the influence of the adjacent forest on wind circulation. There was no evidence of differences in variability of weather conditions between night and day with the exception of relative humidity, which was more variable during the day than at night (variance ratio test of weekly means, station A F23,23 = 6.12, P < 0.001, station B F23,23 = 7.09, P < 0.001).

image

Figure 2. Summary of meteorological variables recorded at weather stations over the study period. Night means have been offset from day means to improve clarity. Figures shown are mean ± 1 SD.

Download figure to PowerPoint

Fit of Gaussian curves

Gaussian curves (Fig. 3) were fitted to the seasonal patterns of abundance of all species at all sites with the exception of A. sulcicollis at the Hore and L. fusca at the Hafren. At the Hore, the abundance of A. sulcicollis did not show a marked peak and the Gaussian function therefore did not converge to a stable fit. Leuctra fusca was present at only low abundance at the Hafren and insufficient numbers were present to enable the function to be fitted. A summary of the parameters and fits of the Gaussian curves is given in Table 1.

image

Figure 3. Seasonal frequency of adult stoneflies captured in Malaise traps. (a) A. sulcicollis , (b) L. fusca , (c) L. inermis , (d) L. nigra , (e) S. torrentium . The solid lines are Gaussian curves fitted to the data using non-linear regression. See Table 1 for details of curve parameters. For L. inermis and L. nigra , adults were present before the beginning of the sampling period but only by 1 or 2 weeks at the most and at low abundance. For A. sulcicollis at the Hore and L. fusca at the Hafren, no curve was fitted (see text for details).

Download figure to PowerPoint

Table 1.  Parameters and fit of Gaussian curves to stonefly seasonal frequency patterns.
SpeciesSiteµSEµσSEσr2P
A. sulcicollisHafren7.560.261.870.260.78<0.001
HoreNo stable fit     
Gwy6.470.171.720.170.87<0.001
L. fuscaHafrenLow abundance; model not fitted     
Hore7.470.302.630.310.79<0.001
Gwy3.240.071.530.070.89<0.001
L. inermisHafren3.220.051.530.090.78<0.001
Hore4.550.753.440.840.630.002
Gwy3.821.373.921.470.640.049
L. nigraHafren5.041.027.641.330.88<0.001
Hore6.052.746.770.720.68<0.001
Gwy1.640.291.770.270.79<0.001
S. torrentiumHafren7.660.412.820.480.680.001
Hore7.600.212.470.210.89<0.001
Gwy9.120.252.470.250.79<0.001

Influence of meteorological conditions on activity

Results of the stepwise regressions of meteorological factors on residuals from the Gaussian curves are shown in Table 2 for day data and Table 3 for night data. For the day analyses, air temperature was a significant predictor of stonefly activity in all except two regression models, in which no meteorological variables were entered. In all cases, air temperature had positive regression coefficients (b; Table 2), indicating that higher temperatures were related to greater activity of adult stoneflies. Day wind speed was the only other meteorological variable that was a significant predictor of stonefly activity (Table 2). Wind speed was related inversely to stonefly activity (negative b; Table 2).

Table 2.  Results of stepwise multiple regression of weekly mean meteorological variables during the day (sunrise–sunset) against stonefly flight activity (residuals from Gaussian curve). Variables are listed in the order in which they entered the model. * P  < 0.05, ** P  < 0.01, *** P  < 0.001.
SpeciesSiteDetails of variables enteredSignificance of final regression model
VariablebSEtAdjusted r2d.f.F
A. sulcicollisHafrenConstant−9.174.192.190.291,105.54*
  Air temperature0.750.322.35*   
 GwyConstant−42.1220.622.040.301,147.29*
  Air temperature4.341.602.70*   
L. fuscaHoreConstant22.744.535.02***0.411,1119.77***
  Air temperature1.630.364.45***   
 GwyConstant15.179.530.750.422,1313.43***
  Wind speed−7.943.322.40*   
  Air temperature2.561.082.37*   
L. inermisHafrenNo variables entered      
 HoreConstant−110.8729.833.72**0.441,1312.18**
  Air temperature7.792.233.49**   
 GwyConstant−44.8311.293.97**0.481,1016.74**
  Air temperature  3.673.670.894.09**  
L. nigraHafrenNo variables entered      
 HoreConstant60.7413.334.55***0.531,1417.83***
  Air temperature  4.394.391.044.22***  
 GwyConstant123.3920.885.91***0.592,1135.38***
  Air temperature10.421.437.27***   
  Wind speed−32.779.393.49**   
S. torrentiumHafrenConstant6.252.292.73*0.281,115.59*
  Air temperature0.410.172.37*   
 HoreConstant45.267.506.04***0.491,1434.88***
  Air temperature3.310.565.91***   
 GwyConstant−16.965.140.480.602,1116.83***
  Air temperature3.871.672.23*   
  Wind speed−13.965.992.33*   
Table 3.  Results of stepwise multiple regression of weekly mean meteorological variables during the night (sunset–sunrise) against stonefly flight activity (residuals from Gaussian curve). Variables are listed in the order in which they entered the model. * P  < 0.05, ** P  < 0.01, *** P  < 0.001.
SpeciesSiteDetails of variables enteredSignificance of final regression model
VariablebSEtAdjusted r2d.f.F
A. sulcicollisHafrenNo variables entered      
GwyNo variables entered      
L. fuscaHoreConstant15.735.682.77*0.261,115.27*
Air temperature1.390.612.29*   
GwyConstant55.936.958.05***0.421,1426.05***
Wind speed−15.082.955.10***   
L. inermisHafrenNo variables entered      
HoreConstant−64.4826.352.45*0.211,134.80*
Air temperature5.912.692.19*   
GwyConstant−33.3210.423.20**0.381,1011.14**
Air temperature3.521.053.33**   
L. nigraHafrenNo variables entered      
HoreConstant46.6310.964.25***0.381,1414.98**
Air temperature4.371.133.87**   
GwyConstant104.6824.084.35***0.431,1215.76**
Air temperature9.432.383.97**   
S. torrentiumHafrenNo variables entered      
HoreConstant−28.107.333.83**0.461,1413.79**
Air temperature2.800.753.71**   
GwyConstant−9.044.850.510.531,1223.41***
Wind speed−19.474.674.16**   
Air temperature3.971.123.56**   

Regression models of stonefly activity using night meteorological data also included air temperature and wind speed as the only significant variables (Table 3). The direction (positive or negative) of influence of the meteorological variables was consistent with the day analyses, however fewer models were a significant fit [11 (day) versus eight (night); Tables 2 and 3], and where both the day and night analyses gave a significant fit, the proportion of variation explained (r2) was significantly lower for the night analyses (paired t-test, t8 = 4.15, P < 0.01).

There was some evidence of interspecific variation in the strength of relationships between flight activity and weather variables, as indicated by differences in the proportion of variation explained by the regression models for different species (Tables 2 and 3). For both day and night analyses, wind speed was a significant factor only at the Gwy site. This may be a reflection of the different exposure of the sites, given the higher mean wind speed at weather station A (adjacent to the Gwy) than at station B (at the Hafren).

The mean angle of wind did not affect the relative catch of adult stoneflies at traps on each bank of the streams or on each side of the traps (linear-circular regression, all models P > 0.05). Variability in weekly mean wind direction was low at both weather stations (station A, range of weekly mean angles, day = 210–298°, night = 213–302°; station B, range, day = 188–218°, night = 188–256°) and the high degree of consistency in wind direction may have contributed to the lack of observed effect.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Results of this study suggest that air temperature is the primary factor influencing the flight activity of adult stoneflies, with higher temperatures resulting in greater flight activity. This is consistent with the study of other insect species (Taylor, 1963; Johnson, 1969; Peng et al., 1992), including aquatic groups such as Trichoptera (Waringer, 1991). There has been anecdotal evidence of the importance of temperature on the flight of adult stoneflies (Hynes, 1976), but this is the first study to document variability in flight activity in the field in relation to air temperature. In small streams, there is a close relationship between air and water temperatures (e.g. Crisp, 1997). Changes in water temperature are thought to trigger emergence of adult stoneflies (Singh et al., 1984; Gregory et al., 2000). It is therefore possible that part of the observed relationship between flight activity and air temperature is in fact a reflection of changes in the numbers of stoneflies emerging from the streams. Weekly counts of adult stoneflies collected in 10 emergence traps placed randomly in the study reaches (R. A. Briers et al., unpublished) enabled this to be tested for two species (L. inermis and S. torrentium) that were sufficiently abundant in the emergence traps to analyse. Weekly emergence counts were included along with meteorological variables in multiple regression analyses (see methods). In neither case was the number of adults emerging a significant predictor of flight activity. Therefore the results do appear to represent the effect of air temperature on flight activity rather than the emergence of adults.

Wind speed was the only other meteorological variable that appeared to influence stonefly flight, with higher wind speeds being associated with lower flight activity. High wind speeds may influence the catching efficiency of the Malaise traps, by deforming their shape or changing airflow around the traps. In a study of forest macrolepidoptera, however, Butler et al. (1999) found that Malaise trap catches were influenced less by weather conditions than were blacklight traps, and at the wind speeds experienced at the study sites (Fig. 2) variation in trap efficiency due to the effects of wind is unlikely to have influenced the results. The relationship between wind speed and flight activity was only significant at the Gwy. This is likely to be a reflection of the greater exposure (resulting in higher wind speeds) of this site relative to the other sites.

Regression models of flight activity in relation to weather conditions were significant in most cases, but the proportion of variation explained by the models was modest (Tables 2 and 3). This may be because the Malaise traps were sampled weekly. Summarising meteorological variables as weekly means disguises possibly significant short-term variations in weather conditions. Similar analyses carried out using maximum and minimum values of meteorological variables did not show improved predictive power. In a study of flight activity of adult Trichoptera (Waringer, 1991), daily catches showed consistently strong relationships between flight activity and meteorological variables (primarily air temperature), suggesting that adult insects respond quickly to changes in environmental conditions, however Waringer (1991) failed to account for seasonal variation in abundance independent of weather conditions. Therefore at least part of his results may simply reflect the phenology of the species studied. If collections had been made with greater frequency in this study, stronger relationships might have been evident. For collection intervals shorter than a day, diel patterns of activity may obscure variation in response to weather conditions (Johnson, 1969).

Although similar variables were included in the regression models for day and night periods, some differences were found between them. While day and night conditions were generally highly correlated, night analyses produced fewer significant models and generally lower r2 values. Brink (1949) and Hynes (1976) have suggested that the flight of adult stoneflies occurs mainly during the day, and it is possible that this is why day weather variables are associated more strongly with adult activity. In order to confirm this suggestion, however, it would be necessary to carry out separate trapping for the day and night periods to compare the relative activity of adult stoneflies at different times of day. With the exception of relative humidity, which was not a significant predictor of flight activity (see results and Fig. 2), there was no evidence of reduced variability of conditions during the night, which might have explained the differences between day and night analyses. Confirmation of the proposed restriction of flight activity to daylight hours will require more detailed studies of diurnal variation in flight activity.

The strength of the relationships between weather conditions and flight activity showed some inter-site variation. The Hafren study site showed the least consistent patterns, with only half of the day analyses and none of the night analyses giving a significant result (Tables 2 and 3). In contrast, between 80 and 100% of regression models were significant for the day and night analyses at the Hore and Gwy sites. Despite the proximity of weather station B to the Hafren study site, it was located outside the mature plantation forest where the Malaise traps were situated. The density and composition of forestry have a strong influence on meteorological conditions experienced within forested areas (Chen et al., 1993; Brosofske et al., 1997) so the weather conditions recorded by station B may not have been a good reflection of conditions within the forest. This may contribute to the relatively poor match between recorded weather conditions and adult stonefly flight activity at this site. Although the weather conditions recorded at station B were used for the Hore site, the station is located ≈ 1 km from the trapping area. It is therefore possible that the weather conditions experienced at the Hore site departed significantly from the conditions at station B. Comparison of data from stations A and B, however, indicated that weather conditions were generally highly correlated (mean wind speed, day r22 = 0.647, night r22 = 0.665; mean air temperature, day r22 = 0.926, night r22 = 0.883, all P < 0.001; see also Fig. 1). Given also that station B was located in vegetation similar to that at the Hore site (see methods), the use of station B for this site would seem to be justified.

Wind direction did not appear to influence the relative catches of adult stoneflies on each stream bank or on each side of the traps, but the high degree of consistency in wind direction may have limited the scope for detecting a directional effect. Given that the wind came from a consistent direction, the totals on the leeward bank or on the windward side of traps might have been expected to be higher. Previous analyses gave no evidence for such an effect (Briers et al., in press). Flight direction and hence trap catches are most likely to be influenced in strong winds. Here, flight activity was shown to be related inversely to wind strength. Therefore in strong winds, the number of insects in flight, and likely to be blown in the direction of the wind, is reduced. This may explain the lack of difference in the number of individuals trapped on each bank.

The influence of meteorological conditions on the flight activity of adult stoneflies may have important implications for local and regional population dynamics and community composition. Given that adult stoneflies have a limited temporal window in which to disperse, the extent of dispersal will be sensitive to the meteorological conditions during the flight period. While this study examined flight activity of adult stoneflies over relatively short distances away from the stream channel, infrequent longer distance movements are also likely to be highly dependent on weather conditions (Johnson, 1969) and hence the extent of inter-stream dispersal may also be influenced by similar factors. Interspecific variation in the strengths of the relationships between flight activity and meteorological conditions suggests that dispersal of some species may be more susceptible to meteorological influences than other species. The emergence and flight periods of the species studied were well defined and occurred in a predictable temporal sequence (Hynes, 1977; Petersen et al., 1999; see also Fig. 3). Interspecific variation in the strength of relationships between flight activity and weather may therefore in part reflect different meteorological conditions experienced by species emerging at different times of the year.

As well as interspecific differences, there was also variation in the strength of meteorological effects between sites. Such differences are likely to be the result of local variation in meteorological conditions, such as higher wind speeds at the Gwy. On a broader scale, the influence of meteorological conditions on the dispersal of adult aquatic insects may vary depending on the altitude or exposure of sites. The sites at which this study was undertaken are all in an upland area (≈ 400 m a.s.l.), which experiences more extremes of wind speed (e.g. gales) and lower average air temperatures than lowland areas (Goudie & Brunsden, 1994). Therefore, variation in the extent of dispersal of adult aquatic insects due to weather may be greater at upland sites than at lowland sites.

Dispersal between populations in adjacent sites can synchronise fluctuations in abundance over a wide area (Heino et al., 1997; Paradis et al., 1999). Variation in the extent of dispersal due to annual variation in weather conditions may therefore influence population fluctuations of larvae in adjacent streams. Bradley and Ormerod (2001) demonstrated synchronous variation in invertebrate community persistence across eight upland streams, which appeared to be linked to the North Atlantic Oscillation. Persistence was high in negative phases of the North Atlantic Oscillation (dry, cold winters) and lower during positive phases (warm, wet winters). Because the North Atlantic Oscillation primarily influences winter weather, it is unlikely to have a direct influence on flight activity and dispersal, which are restricted largely to spring and summer. Nevertheless, rates of larval growth and time of emergence can be influenced by the thermal regime of the stream (e.g. Gregory et al., 2000). Thus, in the warmer conditions that predominate during positive phases of the North Atlantic Oscillation, larval winter growth may be accelerated, leading to earlier emergence when weather conditions may not be so favourable for adult dispersal, disrupting synchronisation of population fluctuations at adjacent sites.

In conclusion, this study provides evidence for the influence of meteorological conditions on the flight activity of adult stoneflies. The influence of weather on insect flight has long been appreciated for terrestrial species (e.g. Williams, 1940). For freshwater insect communities, however, attention has focused on the effects of within- and between-year variability in stream physical-chemical conditions (e.g. McElravy et al., 1989; Resh & Rosenberg, 1989). Recent studies (e.g. Petersen et al., 1999) have begun to examine the role of adult dispersal, and the importance of adult stages in determining community composition and dynamics should not be underestimated (Zwick, 1990).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We would like to thank Iain Barber and two anonymous referees for comments on the manuscript and Dave Bradley for discussions about the influence of the NAO on stream communities. Access to field sites was kindly granted by the Forestry Commission and Mr Simon Bennett-Evans. Ken Blyth and Dave Biggin of CEH Wallingford provided meteorological data and Simon Grant of CEH Bangor gave assistance with queries regarding the weather stations. This work was supported by NERC grant GR3/12114.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Abrol, D.P. (1991) Path analysis of environmental factors influencing daily flight activity of Apis dorsata F. Acta Oecologia, 12, 819824.
  • Bradley, D.C. & Ormerod, S.J. (2001) Community persistence among stream invertebrates tracks the North Atlantic Oscillation. Journal of Animal Ecology, 70, 987996.
  • Briers, R.A., Cariss, H.M. & Gee, J.H.R. (in press) Dispersal of adult stoneflies (Plecoptera) from upland streams draining catchments with contrasting land-use. Archiv für Hydrobiologie.
  • Brink, P. (1949) Studies of Swedish stoneflies. Opuscular Entomologica, Suppl. XI, 1250.
  • Brosofske, K.D., Chen, J.Q., Naiman, R.J. & Franklin, J.F. (1997) Harvesting effects on microclimatic gradients from small streams to uplands in western Washington. Ecological Applications, 7, 11881200.
  • Bunn, S.E. & Hughes, J.M. (1997) Dispersal and recruitment in streams: evidence from genetic studies. Journal of the North American Benthological Society, 16, 338346.
  • Butler, L., Kondo, C., Barrows, E.M. & Townsend, E.C. (1999) Effects of weather conditions and trap types on sampling for richness and abundance of forest macrolepidoptera. Environmental Entomology, 28, 795811.
  • Chen, J.Q., Franklin, J.F. & Spies, T.A. (1993) Contrasting microclimates among clear-cut, edge, and interior of old-growth Douglas fir forest. Agricultural and Forest Meteorology, 63, 219237.
  • Collier, K.J. & Smith, B.J. (1998) Dispersal of adult caddisflies (Trichoptera) into forests alongside three New Zealand streams. Hydrobiologia, 361, 5365.
  • Crisp, D.T. (1997) Water temperature of Plynlimon streams. Hydrology and Earth Systems Sciences, 1, 535540.
  • Fisher, N.I. (1993) Statistical Analysis of Circular Data. Cambridge University Press, Cambridge.
  • Gee, J.H.R. & Smith, B.D. (1995) Impact of forest clear felling on stream invertebrates. R&D Project Record 270/13/ST National Rivers Authority, Bristol.
  • Gee, J.H.R. & Smith, B.D. (1997) Benthic invertebrates in the headwaters of the Wye and Severn: effects of forestry and clear-felling. Hydrology and Earth Systems Sciences, 1, 549556.
  • Goudie, A.S. & Brunsden, D. (1994) The Environment of the British Isles: an Atlas. Oxford University Press, Oxford.
  • Gregory, J.S., Beesley, S.S. & Van Kirk, R.W. (2000) Effect of springtime water temperature on the time of emergence and size of Pteronarcys californica in the Henry's Fork catchment, Idaho, U.S.A. Freshwater Biology, 45, 7583.
  • Griffith, M.B., Barrows, E.M. & Perry, S.A. (1998) Lateral dispersal of adult aquatic insects (Plecoptera, Trichoptera) following emergence from headwater streams in forested Appalachian catchments. Annals of the Entomological Society of America, 91, 195201.
  • Harper, P.P. (1973) Life histories of Nemoridae and Leuctridae in Southern Ontario (Plecoptera). Hydrobiologia, 41, 309356.
  • Harper, P.P. & Pilon, J.-G. (1970) Annual patterns of emergence of some Quebec stoneflies (Insecta: Plecoptera). Canadian Journal of Zoology, 48, 681694.
  • Harrison, S.S.C. & Hildrew, A.G. (1998) Distribution dynamics of epilithic insects in a lake littoral. Archiv für Hydrobiologie, 143, 275293.
  • Heino, M., Kaitala, V., Ranta, E. & Lindstrom, J. (1997) Synchronous dynamics and rates of extinction in spatially structured populations. Proceedings of the Royal Society of London, Series B Biological Sciences, 264, 481486.
  • Hudson, J.A., Gilman, K. & Calder, I.R. (1997) Land use and water issues in the uplands with reference to the Plynlimon study. Hydrology and Earth Systems Sciences, 1, 389397.
  • Hynes, H.B.N. (1976) Biology of Plecoptera. Annual Review of Entomology, 21, 135153.
  • Hynes, H.B.N. (1977) Adults and nymphs of British stoneflies (Plecoptera) a key. Scientific Publication of the Freshwater Biological Association no. 17. Freshwater Biological Association, Windermere, Cumbria.
  • Johnson, C.G. (1969) Migration and Dispersal of Insects by Flight. Methuen, London.
  • Malaise, R. (1937) A new insect trap. Entomologisk Tidskrift, 38, 148160.
  • McElravy, E.P., Lamberti, G.A. & Resh, V.H. (1989) Year-to-year variation in the aquatic macroinvertebrate fauna of a northern California stream. Journal of the North American Benthological Society, 8, 5163.
  • Monteith, D.T. & Evans, C.D. (2000) UK Acid Waters Monitoring Network: 10 Years Report. ENSIS Publishing, London.
  • Neal, C., Smith, C.J., Walls, J., Billingham, P., Hill, S. & Neal, M. (1990) Hydrogeochemical variations in Hafren forest stream waters, Mid-Wales. Journal of Hydrology, 116, 185200.
  • Palmer, M.A., Allan, J.D. & Butman, C.A. (1996) Dispersal as a regional process affecting the local dynamics of marine and stream benthic invertebrates. Trends in Ecology and Evolution, 11, 322326.
  • Paradis, E., Baillie, S.R., Sutherland, W.J. & Gregory, R.D. (1999) Dispersal and spatial scale affect synchrony in spatial population dynamics. Ecology Letters, 2, 114120.
  • Peng, R.K., Fletcher, C.R. & Sutton, S.L. (1992) The effect of microclimate on flying dipterans. International Journal of Biometeorology, 36, 6976.
  • Petersen, I., Winterbottom, J.H., Orton, S., Friberg, N., Hildrew, A.G., Spiers, D.C. et al. (1999) Emergence and lateral dispersal of adult Plecoptera and Trichoptera from Broadstone Stream, U.K. Freshwater Biology, 42, 401416.
  • Resh, V.H. & Rosenberg, D.M. (1989) Spatial-temporal variability and the study of aquatic insects. Canadian Entomologist, 121, 941963.
  • Roberts, G. & Crane, S.B. (1997) The effects of clear-felling established forestry on stream-flow losses from the Hore sub-catchment. Hydrology and Earth Systems Sciences, 1, 477482.
  • Singh, M.P., Smith, S.M. & Harrison, A.D. (1984) Emergence patterns of the stoneflies (Plecoptera) from a wooded stream in Southern Ontario. Aquatic Insects, 6, 233243.
  • SPSS Inc. (1997) SigmaPlot for Windows, Version 4.01. SPSS Inc. , Chicago, Illinois.
  • Taylor, L.R. (1963) Analysis of the effect of temperature on insects in flight. Journal of Animal Ecology, 32, 99117.
  • Waringer, J.A. (1991) Phenology and the influence of meteorological parameters on the catching success of light-trapping for Trichoptera. Freshwater Biology, 25, 307319.
  • Whiles, M.R. & Wallace, J.B. (1992) 1st-year benthic recovery of a headwater stream following a 3-year insecticide-induced disturbance. Freshwater Biology, 28, 8191.
  • Williams, C.B. (1940) An analysis of four years captures of insects in a light trap. Part II The effect of weather conditions on insect activity; and the estimation and forecasting of changes in the insect population. Transactions of the Royal Entomological Society of London, 90, 227306.
  • Williams, C.B. (1961) Studies in the effect of weather conditions on the activity and abundance of insect populations. Philosophical Transactions of the Royal Society of London (B), 244, 331378.
  • Zwick, P. (1990) Emergence, maturation and upstream oviposition flights of Plecoptera from the Brietenbach, with notes on the adult phase as a possible control of stream insect populations. Hydrobiologia, 194, 207233.

Accepted 7 October 2002