IQ!UlHszllfllflllllLHUMHJMHMMMM| L; E This is to certify that the thesis entitled ABUNDANCE AND DISTRIBUTION OF THE CLADOCERAN ZOOPLANKTON Bosmina lonfiirostris, Eubosmina coregoni, Dgphnia galeata mendotae AND Ephnia retrocurva IN THE NEARSHORE WATERS OF LAKE MICHIGAN NEAR LUDINGTON, MICHIGAN presented by Joan Ellen Duffy has been accepted towards fulfillment of the requirements for MASTER OF SCIENCE degreein Fisheries and Wildlife (WAKE § Major professor Date February 15, 1980 0-7639 ’ -‘%\\l\\' I II‘ , .' . vin- momsrm umsm ‘ .wl.‘ I/WE _‘ A U e in . my. ‘9 ~-::a:,w< ;..._' OVERDUE FINES: 25¢ per day per item RETURNIMS LIBRARY MATERIALS: Place in book return to move charge from circulation records ABUNDANCE AND DISTRIBUTION OF THE CLADOCERAN ZOOPLANKTON Bosmina longirostris, Eubosmina coregoni, Daphnia galeata mendotae AND Daphnia retrocurva IN THE NEARSHORE WATERS OF LAKE MICHIGAN NEAR LUDINGTON, MICHIGAN BY Joan Ellen Duffy A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1980 ABSTRACT ABUNDANCE AND DISTRIBUTION OF THE CLADOCERAN ZOOPLANKTON Bosmina longirostris, Eubosmina coregoni, Daphnia galeata mendotae AND Daphnia retrocurva IN THE NEARSHORE WATERS OF LAKE MICHIGAN NEAR LUDINGTON, MICHIGAN by Joan Ellen Duffy The seasonal distribution of Bosmina longirostris, Eubosmina core- goni, Daphnia galeata mendotae and Daphnia retrocurva in Lake Michigan near Ludington, Michigan was studied in 1975-1977. The abundance of the four Cladocera was analyzed with respect to several physical and meteorological factors. Bosmina was the dominant Cladocera, reaching greatest densities in the summer and fall, and Eubosmina was most abundant in the fall. The Daphnia were common only in the summer and fall. Sta- tistical tests showed that there were no significant differences between species abundances at three stations in any year. The Cladocera were bimodally distributed in 1975 and 1976, and had monocyclic patterns‘of distribution in 1977, when abundances were generally reduced. The en- vironmental factors analyzed explained between 11.7 and 70.4 percent of the variation in abundance. Mean total zooplankton abundance was most frequently an important variable in explaining variance though wind direction, water temperature and air pressure were also important. ACKNOWLEDGEMENTS I would like to thank my major professor, Dr. Charles R. Liston, for providing this research Opportunity and for guidance, encouragement and advice throughout the study. Dr. Clarence D. McNabb provided encouragement, support and valuable advice throughout the study. Dr. John L. Gill helped develop and broaden my statistical background, and provided welcomed assistance with the quantitative aspects of the study. I thank Fred Koehler, Joe Bohr, Rick Ligman and Bob Anderson, my fellow graduate students who gave me much emotional support and who willingly shared their advice and knowledge. Dan Brazo, Greg Peterson, Rich O'Neal and Leo Yeck provided valuable field expertise while I was at our field station. Mary Whalen, Bill Richardson, Martha Hardy and Tom Green were very helpful and patient in compiling data for the study from many volumes of computer output. Dr. Stan Zarnoch provided computer time and many helpful comments on my statistics. Facilities for collecting and analyzing samples were provided mainly at the Michigan State University Great Lakes Research Laboratory, near Ludington, Michigan. Two people were very instrumental in my choice of careers. I thank Walt Duffy for his confidence in me, for introducing me to the study of 200plankton and for his aid in the analysis of some of the zooplankton samples. I thank Dr. Louis A. Helfrich for his supervision and instruction, and for encouraging me in further studies. ii iii Consumers Power and Detroit Edison Companies provided most of the funds for this study. Partial funding was provided by Michigan Agricul- ture Research Station. TABLE OF CONTENTS Page LIST OF TABLES ..................................................... V LIST OF FIGURES .................................................... Vi INTRODUCTION .......................... ......... .................... 1 DESCRIPTION OF SAMPLING AREA .............. ....... .................. 3 METHODSANDMTERIflS ..... ......OOOOOOOO ...... O ........ ... ...... 0.. Field MethOd O O O O O O O O I O ........... O ....... I O O O O O O O O O O O O O I O O O O O 0 Laboratory Methods ... ................. . ..... . ......... ........ Statistical Methods ................. ......................... . \DCDO‘ 0‘ RESULTS ................ ...... ... ..... .. ................. ........... 11 Dunnett-type test OOOOOOOOOOOOOOOOOOOOOO....OOOOOOOOOOOOOOOOOOO Seasonal Distribution ......... . ....... ..... ........... ... ..... Analysis of Variance ............. ........... .................. Stepwise Multiple Linear Regression ........ ................... DISCUSSION ........ ........... ............ ...... .... ..... ........... 32 SUMMARY ............... ....... ...................................... 38 LITERATURE CITED .......... ....... ..... ..... ................. ..... .. 40 APPENDIX ......................................... ........ .......... 43 iv Number Al A2- A37 LIST OF TABLES Page Zooplankton sampling dates in Lake Michigan in 1975- 1977000000.0000...0.000.000...............OOOOUOOOOOOOOOOOO 7 Results of Dunnett-type test of control versus treatment stations for four Cladocera in Lake Michigan, 1975- 1977. MSD-Iminimum significant differences (No. m‘3); DIFF'I actual differenc (No. m‘3); a! 0.05.... ..... 12 Results of the analyses of variance of abundance be- tween years for four Cladocera in Lake Michigan, 1975-1977.............OOOOOOOOOOOOOOOOOO......OOOOOOOOOOOOO22 Independent variables used in the stepwise multiple linear regression analyses of Cladoceran abundance in Lake MiChigan, 1975-19770 cccccccc cocoa-0.00000cocoa-00000000....24 Results of the multiple linear regression analysis of the abundance of Bosmina longirostris in Lake Michigan in 1975-19770000000 ..... cocoso.oooooooooooooooooo0.0000000. 26 Results of the multiple linear regression analyses of the abundance of Eubosmina coregoni in Lake Michigan in 1975-1977. ....... ..... ........ .......................... 27 Results of the multiple linear regression analyses of the abundance of Daphnia galeata mendotae in Lake Michigan in 1975-1977....... ..... ... ....................... 29 Results of the multiple linear regression analyses of the abundance Daphnia retrocurva in Lake Michigan in 1975-1977 ...... .... ..... ....... ............ ......... ..... .. 30 Values of variables used in multiple linear regression analyses. See Table 4 for description of variable names............ ....... 0.0.00.0.........OOOOIOOOOOOO0.0... Descriptive statistics of four Cladoceran species in Lake Michigan during 1975-1977 ........... . ................ . 45-80 Number LIST OF FIGURES Page Map and location of permanent sampling stations in Lake Michigan adjacent to the Consumers Power Pumped Storage Plant near Ludington, Michigan... ......... 5 Mean abundance for four species of Cladocera in Lake Michigan during 1975................................ 15 Mean abundance for four species of Cladocera in Lake Michigan during 1976......................... ...... . 17 Mean abundance for four species of Cladocera in Lake Michigan during 1977.... ...... ......... ...... . ...... 20 vi INTRODUCTION The Cladoceran zooplankton are an important part of the aquatic ecosystem in Lake Michigan but have received little detailed study until recently. This study was undertaken to quantify the occurrence of Cladoceran 200plankton species in a nearshore area of Lake Michigan during 1975 to 1977, and to determine to what extent the abundance of these species was affected by various physical and meteorological factors. Bosmina longirostris, Eubosmina coregoni, Daphnia galeata mendotae and Daphnia retrocurva were chosen because of their common occurrence in the Cladoceran zOOplankton. Environmental factors chosen for analysis were wind speed, wind direction, water temperature, air pressure, photoperiod, water turbidity, water transparency and mean total 200- plankton abundance. Early zooplankton studies in Lake Michigan were descriptive in nature and concentrated on taxonomy (Birge 1882, Forbes 1882, Ward 1896). Eddy (1927) was the first to obtain data on seasonal distribution of zooplankton in nearshore southern Lake Michigan. Ahlstrom (1936) conducted the first offshore zooplankton study. Several early studies investigated zooplankton in the water supplies of several major cities on Lake Michigan (Damman 1945, 1960; Williams 1962, 1966). Wells (1960) was the first to conductzaquantatitive study of the seasonal distribution of zooplankton in eastern Lake Michigan. He later noted a change in the species compo- sition in 1966 (Wells 1970) which he attributed to alewife predation. Most large zooplankton had declined in numbers while smaller zooplankton increased in numbers. After a dramatic decline in the alewife population in 1967 the zooplankton species composition began to shift back to its earlier structure. Gannon (1972) conducted a comprehensive study of seasonal distribution and abundance of zooplankton, and showed the effects of eutrophication on the zooplankton community. Roth and Stewart (1973) studied the zooplankton in southeastern Lake Michigan near the Donald C. Cook Nuclear Plant at Bridgeman, Michigan. The zooplankton community at the site of the present study was first studied by Duffy (1975). He reported the abundance and seasonal dis- tribution of the zooplankton, and investigated their vertical distribution. Duffy and Liston (1978) compared the zooplankton community of the Luding- ton Pumped Storage Reservoir to that of Lake Michigan at the control site of the present study. DESCRIPTION OF STUDY AREA The study area is the site of a current environmental study con- ducted by the Department of Fisheries and Wildlife, Michigan State Uni- versity, to determine the effects of the Ludington Pumped Storage Power Plant on the aquatic biota of Lake Michigan. The zooplankton population of the area has been monitored since 1972 as part of the overall environ— mental study (Duffy and Liston 1979). The area is 6.4 km (4.0 miles) south of Ludington, Michigan, adjacent to and south of the power plant (Figure 1). The impact stations are 0.8 km (0.5 miles) north of the breakwater of the power plant (station 5) and 0.8 km (0.5 miles) south of the breakwater (station 3; Figure 1). Both stations are 12 meters deep and have sand and gravel substrates. The control station is 4.8 km (3.0 miles) south of the breakwater, in an area considered unaffected by currents from the power plant. It is 12 meters deep and has a sandy substrate. Figure 1. Map and location of permanent sampling stations in Lake Michigan adjacent to the Consumers Power Pumped Storage Plant near Ludington, Michigan. 'I///// LUDINGTON / \\_ l Pen Heron”: Figure 1. SAMPLING STATIONS Ludington Pumped Storage P7019m Lalo A N h. ° ' z a: 9 Miss J n, g 5 ..... l. 0 “a: 3:33;; I - R E SERVOIR a 3 “.':‘:';i:':j:j:i: .... . 'b 2 EB L. ‘ § S Q s“ Q Q .. Q Bax: V "...- Lato L METHODS AND MATERIALS Field Methods Zooplankton samples were collected approximately biweekly in 1975 and 1976, and monthly in 1977 (Table 1). Duplicate samples were taken between 0700 hours and 1200 hours at depths of 1 meter, 4 meters and 12 meters, resulting in 6 samples per station. For each sample, 100 liters of water were pumped through a number 20 (64 micron) nylon plankton net (Tonolli 1971). A small volume of club soda was added to relax the animals and minimize distortion of taxonomic features (Gannon and Gannon 1975) and the sample was preserved in 10 percent formalin. The samples were allowed to settle one week, and were then concentrated to a volume of approximately 50 ml. The formalin was then replaced by 70 percent alcohol and a few mililiters of glycerin added to prevent the organisms from becoming brittle. Water temperature and water transparency were measured in the field at the time the plankton were sampled. Water temperature was‘measured with a YSI thermistor, and temperature was recorded to the nearest tenth degree Celsius. The water temperature value reported is the average of surface and bottom water temperatures. water transparency was measured with a secchi disc and recorded in meters. Water samples for turbidity measurements were collected at sampling time and returned to the laboratory. Wind direction, wind speed (knots) and air pressure (mm Hg) were obtained from daily readings made at the U.S. Coast Guard station at Table 1. Zooplankton sampling dates in Lake Michigan in 1975-1977. 1975 4-22 5-2 5-13 5-29 6-17 6-30 7-14 7-28 8-11 8-27 9-9 9-24 10-7 11-5 1977 4-18 5-17 6-13 7-21 8-18 9-13 10-22 Ludington, Michigan. A mean value for these parameters was calculated for the six hour period before the sampling time on each sampling date. Photoperiod values were calculated as hours between sunrise and sunset from observations made at Muskegon, Michigan, 85 km south of the sampling area . Laboratory Methods The zooplankton samples were examined using a binocular microscope (magnification 7-60X), a compound microscope (magnification 100-400X) and a chambered counting cell (Gannon 1971). Each sample was mixed gently with a magnetic stirrer and a subsample of 2-10 ml was removed with a wide-mouth syringe for identification and enumeration. Subsample size was gauged so as to count 100-150 of the common organisms. The Chi- square (X2) was used to test the randomness of the counting method, and the conditions of randomness were met (Duffy 1975). The four Cladocera were identified using keys by Brooks (1957) and Deevey and Deevey (1971). Counts were converted to numbers per cubic meter for the analyses. Total zooplankton were counted and their abundance per cubic meter calculated. The mean total zooplankton abundance per date was calculated as a measure of competition with the other zooplankton. The abundance of the species being examined was subtracted from the mean total zooplankton abundance on each date for the analyses. Turbidity was determined with a Hach model 2100A turbidimeter (Hach Chemical Co., .Ames, Iowa). Turbidity was recorded to the nearest tenth Formazin Turbidy Unit (FTU). Statistical Methods The data were transformed by several methods to determine which would satisfy the assumption of a normal distribution required by para- metric procedures; the log (y+1) transformation proved to be the most suitable for these data. For the hypothesis of no mean differences between densities of zooplankton at the control station (station 1) and the impact stations (stations 3 and 5) in each year, the untransformed data were tested using a Dunnett-type procedure for data with heterogeneity of variance. This t-like test is based on an experiment-wise Type I error rate because the comparisons are correlated (Gill 1978). The 95 percent minimum significant difference (MSD) for each comparison (station 1 vs. station 3; station 1 vs. station 5) for each species and year was calculated for comparison with the actual difference between the means. The MSD was calculated by (t )(sfi), where the t value is a percentage point aD/2,V from the student's t distribution with v degrees of freedom using GD 8 1-(0.95)1/m, m=3 stations, and s- is the standard deviation of the dif- D ference between means for the control station and an impact station. A one-way analysis of variance was conducted to test the hypothesis that there were no differences between years for each species. Based on the results of the Dunnett-type test, untransformed data for all stations for each species were grouped together for this analysis. Contrasts between years (1975 vs. 1976; 1975 vs. 1977; 1976 vs. 1977) were tested when the F-rations for the analysis proved to be significant at the 5 percent level. Stepwise multiple linear regression procedure was used to test for relationships between the dependent variable (abundance) and the set of 10 independent variables measured. The general model for the multiple linear regression procedure is y - Xb - e, where y is a matrix of the dependent variables, X is a matrix of independent variables, b is a matrix of regression parameters, and e is a matrix of error variables. It is assumed that the errors are normally and independently distributed with homogeneous variance for any set of values of the independent vari- ables «HJJ.1978). The stepwise method of multiple linear regression involves the re-examination of all variables in the model at each stage of analysis; variables already in the model may be rejected at a later stage. An F—statistic is calculated for each variable at each stage of the regression. A variable is accepted in the model if it is significant at the 10 percent level or better, and is rejected from the model if it falls below the 25 percent significance level. The errors ~ m .m> ~ m .m> g m .m> d m .m> d m .m> H mucoHoHMfim Hanuom I mmHn “Anna .ozv mucououwwv unmofiugcwfim aaafisae n am: 03mg a“ mumooomau uzom you mcoaumum ucoaumouu momuo> Houucoo mo umou umhuluuoccao mo muasmmm .N manna mmHo mm: hmHQ am: mmHa mm: mch cm: m>u=oouuou mecnmmn caucuses muuoamw mwcnmma “cowouoo mfiamonsm mfiuumoufiwcoa mafiEmom .mo.o a a “Amie .ozv .Athumaaa .amwaeofiz 13 was used to graph the species abundance. Figures 2, 3 and 4 show this mean seasonal abundance for the four Cladocera in Lake Michigan in 1975-1977. In 1975 Bosmina, D. galeata mendotae and D. retrocurva showed two peaks in abundance, in summer and fall, and showed a general increase in abundance at the end of the sampling period (Figure 2). Bosmina reached maximum abundance on 17 June (18,472 me) and had a second peak abundance on 24 September (14,078 m-3). Bosmina were least abundant in May (2 In"3 on 2 May and on 13 May). Eubosmina were not identified in the samples in 1975 until August, but showed patterns of abundance similar to Bosmina from that date on. They were most abundant in the samples at the end of the sampling period (8,129 m-3) on 5 November, but had an earlier peak on 24 September (3,257 m-B). The lowest abundance of EEEEET ‘mina recorded in 1975 was on 9 September (22 m-3). The Daphnia species were never as abundant as either Bosmina or Eubosmia in 1975. 2, galeata reached peak abundance on 27 August (1,209 m-3), and showed another peak in September (910 m"3 on 24 September). It was least abundant in the spring; on 22 April there were 3 m.—3 but 2, galeata disappeared in the samples until 29 May (4 m-3). '2. retrocurva was absent from the samples until 14 July, the time of their lowest abundance (7 m-3). They gradually increased in abundance, but showed a sudden decrease on 9 September, which was followed by their maximum of 1,063 m"3 on 24 September. All four Cladocera showed a marked decrease on 9 September (Figure 2), which was followed by an increase in abundance on the next sampling date. This is most likely due to a decrease in water temperature on 9 September, the result of an upwelling in Lake Michigan (see discussion). 14 Figure 2. Mean abundance of four Species of Cladocera in Lake Michigan in 1975. 24o“- 2o.o - nab- |2.0 '- 6g;- 3.0 "' 2.5- Aimdooco (namstOOO) 2.5” Figure 2. 15 Daphnia gdoara Inward Bataan “flour/s Eubosmina canyon! Daphnia re Ira curva 16 Figure 3. Mean abundance of four species of Cladocera in Lake Michigan in 1976. Mundane. (no. '3 1: I000) 24.0 - 20.0 - A IB.O"' 0 l2.0- 1 WI 4.0 3.5 r 2.5 '- 2.0 '- O.5 '- O Figure 3. 17 Demo galeata mcnabfa Bosmina Imp/rosin! Eubosmina canyon] Daphnia retrocurva 18 In 1976 a pattern of abundance similar to 1975 was seen for all species (Figure 3). Bosmina decreased after sampling was begun in the spring, and was lowest in abundance on 28 April (165 m-3). A peak was seen on 21 June (15,056 m-3) and again on 15 August, when the maximum was reached (22,948 m-3). .A slight.decrease in abundance was seen at the end of the sampling period. Eubosmina also showed a decrease in abun- dance immediately after the start of the sampling season; the lowest 3 on 25 April), but there were no numbers recorded were in April (11 m- Eubosmina in the samples on 7 June. Eubosmina had only one peak in abun- dance in 1976, on 15 September (2,261 m-3). There was also a slight increase in abundance of Eubosmina at the end of the sampling season. The Daphnia were low in abundance until the end of June, and did not show an increase in abundance at the end of the sampling period. ‘2, galeata were rareimlthe samples until June; the lowest numbers in the samples were on 12 May (2 m-B), but no 2, galeata were in the samples on 28 April, 25 May or 7 June. A peak of abundance was reached on 21 July (1,339 m-3) and the maximum was seen on 15 September (1,448 m-3). 2, retrocurva showed their lowest abundance on 28 April (2 m-3), then were absent from the samples in May. They reached a single peak in abundance on 21 July (439 m-3) but remained in low numbers throughout the year. In 1977 most of the species showed only a single peak in abundance, and were generally reduced in numbers (Figure 4). However, the frequency of sampling in 1977 was reduced to once a month, which could explain the differences seen. Bosmina were lowest in abundance in April (25 m-3 on 18 April). The maximum abundance was on 21 July (5,283 m -3) and this was followed in August by a similar abundance (4,576 in“3 on 18 August). 19 Figure 4. Mean abundance of four species of Cladocera in Lake Michigan in 1977. Am. (no :63. I000) amoL |6.0 +- 120" 6.02- 3.5“ 3.0 r 2.0 '- 0.5 '- 20 a Daphnia gdcalo mandala o Bosmina longz'rasrfl': A Eubosmina carogam’ 0 Daphnia retrocurva ‘9' Figure 4. 21 Eubosmina was not recorded in samples on either 18 April or 13 June, but was present in moderate numbers in May (234 In”3 on 17 May). The peak in Eubosmina abundance was on 21 July (760 m-3) and the lowest abundance was seen on 22 October (107 m-3). The Daphnia species were again rare in the spring. 2, galeata was not recorded in samples until 21 July, and although it was not abundant it showed its greatest density of 1977 on this date (286 m-3). The lowest abundance for D, galeata in 1977 was on 13 September (23 m-3). 2, retrocurva was not recorded in the samples until 13 June, and this was the lowest abundance of this species seen in 1977 (28 m-3). The maximum was on 21 July (1,960 m-3). Both Daphnia species showed slight increases in abundance at the end of the sampling period. In summary, similar bimodal patterns in abundance were seen in 1975 and 1976, but most species in 1977 showed a single peak in abundance. Bosmina abundance was similar in 1975 and 1976 but decreased in 1977 by several thousand organisms per cubic meter. Eubosmina abundance ap- peared to have decreased over the years from a maximum of over 8,000 per cubic meter in 1975 to a maximum of less than 1,000 per cubic meter in 1977. The highest numbers of Eubosmina were recorded in the fall of 1975 and 1976, and in the summer of 1977. Daphnia galeata mendotae decreased in abundance in 1977, although it was never as abundant in 1975 or 1976 as the two bosminids. Daphnia retrocurva showed an increase in abundance in 1977; it was comparatively low in abundance in 1975 and decreased in 1976. The analysis of abundance between years for each species (Table 3) showed that there were significant differences in abundance between years 22 Gas. dwfi. wao. mom. Hoo.v Hoo.v goo. moo. Hmw. mulch mnlmm chums mummy cooauom mummuuaou now mofiugaflnmnoum man. mm.o emo. o~.m moo.v sm.~m moo. wo.m Amwaaanmnoum oaumeum ooamfium> mo mwmzams< can mmm mum mom thUOHumH macsmmo omuomsoa uumodmw «gunman “cowouoo u:aEmonzm mfiuumouwmcoa mcaamom mowoomm .AAAH - mama .amwaeuaz mama :« mofiooam muooovmao unom MOM munch cooauon mocmvasnm mo oocmfium> mo mumhamsm am no muasmom .m manna 23 for Bosmina (p=.OO3), Eubosmina (p<.001), and Daphnia galeata mendotae (p'.026). Daphnia retrocurva was not significantly different in abundance between years (p'.595). Contrasts between years for Bosmina showed that 1977 was significantly different from both 1975 and 1976. There was no significant difference between abundance of Bosmina in 1975 and 1976. In 1975 Eubosmina was significantly different in abundance from both 1976 and 1977, but there was no difference in abundance of Eubosmina in 1976 and 1977. Abundance of Daphnia galeata was not greatly different between 1975 and 1976, or between 1975 and 1977, but was significantly different between 1976 and 1977. A total of 12 multiple linear regression analyses were performed on the data, one analysis for each combination of species and year. The data were combined for the three stations in the analyses because the Dunnett-type test showed there were no significant differences between stations in any of the years (Table 2). This increased the size of the data set for each analysis and provided more power to the tests. The vari- ables used in the analyses and their variable names are given in Table 4. The mean total zOOplankton variable appeared in seven of the multiple linear regression equations from the complete analyses, and in six of those from the partial analyses. All other factors appeared less frequently, but did appear in at least two of the regression equations for both the partial and complete analyses. Tables 5 through 8 present the significant variables in the multiple linear regression analyses of abundance of Bosmina, Eubosmina, D, galeata mendotae and D: retrocurva in 1975-1977. Included in these tables are the significance levels of the variables in the order included in the equations, the R2 values resulting from the addition of each of the variables, 24 Table 4. Independent variables used in the stepwise multiple linear regression analyses of Cladocera abundance in Lake Michigan, 1975-1977. VARIABLE VARIABLE NAME ‘UNITS Wind Direction WD deviation from North (1-16) Wind Speed WS knots Water Temperature WT oC Atmospheric Pressure PR mm Hg Photoperiod PH hours Turbidity TU FTU Water Transparency SC meters -3 Mean Total Zooplankton TZP No. m 25 the simple correlation of each significant variable with the dependent variable, the overall significance of the equation with the addition of each variable, and the standardized regression coefficients (Betas). The Beta values are the most valuable in explaining the importance of the variables includedixithe equations, as they are standardized to correct for differences in the units and variability of the variables. To stan- dardize the regression coefficients, the deviation for each variable is divided by the estimated standard deviation for that variable. The important variables explained a total of between 34.9 percent (1976) and 44.4 percent (1977) of the observed variation in the abundance of Bosmina (Table 5). Air pressure was the most important variable in 1975, and accounted for 16.7 percent of the total variation explained in the analyses. Water temperature was the most important factor in the 1976 analysis, explaining 19.4 percent of the variation, and was the second most important factor in the 1975 analysis. Wind speed explained the most variation (36.7 percent) in the 1977 analysis. The analysis of Eubosmina abundance produced equations that explained a total of between 11.7 percent (1977)unui70.4 percent (1975) of the observed variation (Table 6). Wind direction was the most important variable in the 1975 analysis, explaining 4.8 percent of the observed variation. In 1976 wind speed was the most important variable in the analysis, contributing 10.6 percent to the observed variation. Wind speed was also the second most important variable in the 1975 analysis. In 1975 only air pressure was significant in the analysis of Eubosmina abun- dance, explaining 11.7 percent of the variation. Table 5. Results of the multiple linear regression analyses of the abundance of Bosmina longirostris in Lake Michigan in 1975- 1977. TOTAL VAR* VAR 2 SIMPLE REGRESS. YEAR NAME SIGNIF R r SIGNIF BETA 1975 PR <.001 .167 -.409 <.001 -.673 (N-204) SC .003 .203 .197 <.001 .178 WT .005 .233 .262 <.001 .419 WD <.001 .291 -.081 <.001 .058 PH <.001 .380 .106 <.001 -.353 TZP <.001 .409 .055 <.001 -.209 1976 WT <.001 .194 .441 <.001 .657 (N=215) PH <.001 .274 .296 <.001 .475 WS <.001 .338 -.148 <.001 .402 TU .056 .349 .167 <.001 -.126 1977 WS <.001 .367 .606 <.001 .891 (N8106) TZP <.001 .444 .241 <.001 -.398 * See Table 4 for explanation of variable names. 27 Table 6. Results of the multiple linear regression analyses of the abundance of Eubosmina coregoni in Lake Michigan in 1975-1977. TOTAL VAR* VAR 2 SIMPLE REGRESS. YEAR NAME SIGNIF R r SIGNIF BETA (N'78) WD .010 .488 -.018 <.001 -1.713 NS <.001 .675 .205 <.001 1.530 TZP .010 .704 .016 <.001 -.244 1976 WS <.001 .106 .325 <.001 .558 (N=167) WT <.001 .198 .169 <.001 .453 PH <.001 .293 -.102 <.001 -.339 1977 PR .004 .117 .342 .004 .342 (N-70) * See Table 4 for explanation of variable names. 28 The regression equations for the analyses of the abundance of Daphnia galeata mendotae explained between 23.2 percent (1975) and 24.6 percent (1976) of the observed variation in abundance (Table 7). In 1975 water temperature was the most important variable in the analysis, contributing 9.5 percent to the observed variation. In 1976 wind direction was the most significant variable in the analysis, explaining 21.5 percent of the variation. There were no significant variables in the equation for the analysis in 1977. This could be attributable to the smaller sample size in this analysis (N=41) as well as to the inability of the available vari- ables to explain the variance of this species. In the analyses of the abundance of Daphnia retrocurva, the signi- ficant variables explained from 16.5 percent (1977) to 31.0 percent (1976) of the variance (Table 8). Air pressure was the most important variable in the analysis of 1975 abundance, explaining 11.6 percent of the variation. In 1976, wind direction was the most important variable in the analysis, explaining 9.4 percent of the variation, and photoperiod explained 7.8 percent of the variation. Both variables had nearly equal beta values, though opposite in sign. The variable mean total zooplankton was the only significant variable in the equation for the 1977 analysis, explaining 16.5 percent of the observed variation. This variable was also included in the equations for 1975 and 1976. In summary, the important variables in the 12 analyses explained between 11.7 and 70.4 percent (mean 31.8 percent) of the variation in abundance observed in the four Cladocera. Mean total zooplankton, the most frequently significant variable in the analyses, accounted for between 1.4 and 16.5 percent (mean 6.9 percent) of the variation explained when it appeared in an equation. Wind direction and water temperature were Table 7. Results of the multiple linear regression analyses of the abundance of Daphnia galeata mendotae in Lake Michigan in 1975-1977. TOTAL VAR* VAR 2 SIMPLE REGRESS. YEAR NAME SIGNIF R r SIGNIF BETA 1975 WT <.001 .095 .308 <.001 .443 (N=162) <.001 .232 -.242 <.001 -.394 1976 WD <.001 .215 -.463 <.001 -.456 (N-126) TU .026 .246 .196 <.001 .177 1977 No variables in the equation (N=41) * See Table 4 for explanation of variable names. 30 Table 8. Results of the multiple linear regression analyses of the abundance of Daphnia retrocurva in Lake Michigan in 1975-1977. TOTAL VAR* VAR 2 SIMPLE REGRESS. YEAR NAME SIGNIF R r SIGNIF BETA 975 PR <.001 .116 .340 <.001 .363 (N=114) WS .065 .142 .154 <.001 .147 TZP .045 .173 .110 <.001 .180 1976 WD .002 .094 -.306 .002 -.673 (N=102) PH .003 .171 .192 <.001 .674 TZP <.001 .280 -.062 <.001 -.516 SC .043 .310 .185 <.001 -.255 1977 TZP <.001 .165 .406 <.001 .406 (N-70) * See Table 4 for explanation of variable names. 31 ' most frequently the most important variable in the equations. Wind direction explained 4.8 to 21.5 percent (mean 11.8 percent) of the vari- ation in the equations when it was significant in the analyses. Like- wise, water temperature explained 1.0 to 44.0 percent (mean 14.1 percent) of the observed variation in equations it appeared in. DISCUSSION The four Cladocera showed a seasonal distribution that is typical of that reported in other studies. Duffy (1975) also found Bosmina to be the dominant Cladocera in 1974 in the same area of Lake Michigan. In 1974 Bosmina comprised 24 to 26 percent of the total zooplankton in July and August (up to 30 percent in the present study), although the abundance was not at its maximum then. The seasonal pattern of abun- dance for Bosmina in 1974 was essentially the same as in 1977 in the present study, but it reached greater densities in 1974 (up to 30,0001TTD. Similar densities were seen in the present study in 1975 and 1976. Eubosmina appeared earlier in samples in 1976 and 1977 than in 1974, but were most common at the same time of year. Eubosmina abundance was greater in 1975 and 1976 than reported by Duffy in 1974, but was similar in 1977. Daphnia retrocurva was the most abundant daphnid in 1974, but its densities exceeded those of Daphnia galeata only in 1977 in the present study, when a similar but later maximum was reported (approxi- mately 2,500 m-3). 2, galeata was more abundant in 1975 and 1976 than in 1974, but decreased in numbers again in 1977. D, galeata reached peak abundance earlier in all years of the present study than in 1974. 2. retrocurva was most abundant on a later date in 1975 than in 1974 and on earlier dates in 1976 and 1977 than 1974. The inshore stations sampled by Roth and Stewart (1973) in south- eastern Lake Michigan are comparable to the Ludington site. Again, 32 33 in 1972, Bosmina was the most common Cladocera. It had a bimodal seasonal distribution with peaks in July-August and again in October, as in 1975 of the present study. Abundance of Bosmina in August of 1972 reached 180,000 n-3, compared to means of up to 22,900 In"3 at the Ludington study site. Eubosmina appeared earlier in samples in 1976 and 1977 than in 1972, but showed a similar seasonal distribution. Eubosmina densities reported by Roth and Stewart are similar to those in 1975 in this study. The densities they reported for the Daphnia species are greater than seen in 1975 through 1977, and the seasonal abundance showed a monocyclic distribution. The increased abundance of the four Cladocera seen in this 1973 study could be due to the trophic status of the area of Lake Michigan studied by Roth and Stewart. They characterize their inshore station as more eutrophic than the offshore stations, and Beeton (1963) discusses the advanced trophic statecmfthat part of the lake. These Cladocera, especially Bosmina, are more common in eutr0phic waters. Evans and Hawkins (1977) studied the same area of Lake Michigan as Roth and Stewart. They found greater abundances of Bosmina in July of 1974, 1975 and 1976 than in the same month in the present study. The other species were not present in significant numbers at similar depths in their study in July. The decrease in abundance of all four Cladocera on 9 September 1975 was attributable to the upwelling in Lake Michigan evident on that date (Table Al.). It is not possible to tell the duration of the upwelling from the available data, but the water temperature averaged 7.90 C colder on that date than the previous sampling date. Upwellings are caused by easternly winds pushing warmer surface waters away from shore and bringing in colder waters from deeper in the lake. The wind direction was 34 predominantly northwesternly in direction on 8 September, shifting to east-northeast on 9 September. Upwellings in Lake Michigan have been documented by various authors (Carr g£._al. 1973; Liston.gg..al. 1974; Duffy 1975) and zooplankton abundances have been shown to be strongly influenced by water temperature (e.g. Hutchinson 1967; Duffy 1975). Diatoms, a principle food of zooplankton, are concentrated in the area of an upwelling (Hutchinson 1967) and their growth is augmented by an increase in nutrients in the epilimnion from nutrient rich waters brought up from the hypolimnion (Liston and Anderson 1979). The Cladocera responded to warming temperatures and probable increases in diatom abundance by increasing their abundance on the next sampling date (24 September; Figure 2). The difference in abundance of Bosmina in 1977 compared to the other years is due in part to the decreased sampling frequency in that year. Appauently the monthly sampling dates did not concur with periods of maximum Bosmina abundance; Bosmina densities in 1975 and 1976 were four times those of 1977. There were no major differences between the other factors measured in the study during the three years. Other factors most likely contributed to the observed decrease in abundance, but these are not readily discernable. The differences in abundance of Eubosmina in 1975 compared to the other years were most likely due to the lack of data on abundance in spring and early summer. The dis- tribution of Eubosmina in the spring and early summer of both 1976 and 1977 was similar. The differences in abundance of Daphnia galeata men- data in 1977 were again likely due to decreased sampling frequency. 35 In the present study the sampling frequency was reduced to monthly in 1977 because of time and personnel restrictions. The significant differences in abundance of some of the Cladocera in 1977 make it evident that monthly sampling can lead to underestimated abundances of zooplank- ton populations. A complex array of physical and biological factors determine zooplankton abundance and species life histories. Adult Clado- cera have the capacity to release a brood of young every few days to a week (Pennak 1978) and the detection of major peaks in abundance may be delayed by monthly sampling, or may be missed entirely if unfavorable environmental conditions should arise. The mean total zooplankton variable in the multiple linear regression analyses is essentially a measure of competition from other zooplankton at the time and place of sampling. It was calculated for the four species on every date for all years. This variable was included in the regression equations most frequently, although it rarely accounted for the most explained variance (mean of 7 percent of the explained variance). Mean total zooplankton was included in equations for Daphnia retrocurva for all three years of the present study. This was the only case where one or more variables was consistently important for a species across all years. Wind speed, water temperature, photoperiod and mean total zooplankton were important in analyses in two of the three years for Bosmina longirostris, but the other species had no variables that appeared in two analyses for more than one year. It is not possible to predict which variables are the most important in explaining the variation in abundance for a given species in a given year; the relative influence of the variables in this study varies from year to year. 36 Wind direction is important because of its influence on water flow in the lake. Its influence on zooplankton distribution and abundance can be seen in mechanical transport of organisms in induced water currents, or more indirectly by influencing water temperature in the lake. Pre- vailing motions of wind and water can cause important changes in the thermal structure of the lake (e.g., upwellings), which is positively correlated with overall patterns of zooplankton abundance (Patalas 1969). Wind speed works in conjunction with wind direction, and its influence is temporal. Water temperature influences zooplankton abundance directly by con- trolling growth rates and development (Hutchinson 1967), and thus pro- duction, and indirectly by stimulating the growth of phytoplankton. Duffy (1975) found that variation on water temperatures between years was most likely the major factor contributing to observed differences in abundance of zooplankton in Lake Michigan. Patalas (1969) found that temperature and its distribution in the water was a decisive factor governing zooplankton abundance in the spring and summer, although the relationship was less clear in the fall. Roth and Stewart (1973) found thermal stratification to be a major factor in determining zooplankton distribution. Photoperiod may be a controlling stimulus in the development of some Cladocera (Stross 1971; Wetzel 1975), and interacts with temperature in many situations. Photoperiod is an important stimulus in the vertical migration of zooplankton, which has been suggested to be a mechanism for predator avoidance and for optimal feeding and growth. Vertical migration can also influence zooplankton sampling. In the present study 37 photoperiod was frequently an important factor in explaining variation in Cladoceran abundance. Water transparency as measured by a secchi disc is highly influenced by turbidity in the water, and is closely correlated with percentage transmission of light (Wetzel 1975). Light transmission may affect growth rates of Cladocera directly (Jacobs 1962), and it acts indirectly to determine population success by affecting phytoplankton abundance (Wetzel 1975) and possibly efficiencies of growth (Buikema 1971). It is also a controlling factor in vertical migration (McNaught and Hasler 1964). Turbidity negatively affects the productivity of aquatic environments (Murphey 1962), through both its abiogenic and biogenic components (e.g., self-shading in phytoplankton; Wetzel 1975). However, water transparency and turbidity were not frequently important variables in this study, both appearing in fewer equations than any other variables. The effects of changes in atmospheric pressure on zooplankton has not been investigated thoroughly. Atmospheric pressure is one factor determining how much oxygen is dissolved in water, and thus indirectly influences zooplankton abundances. Air pressure was frequently an important variable in explaining the variation observed in Cladocera abundance. SUMMARY The seasonal distribution and abundance of four Cladocera at three stations in nearshore Lake Michigan (12 meter depth) near Ludington, Michigan were studied during 1975-1977. Samples were collected biweekly with a pump and net method. The species were analyzed using multiple linear regression techniques with wind direction, wind speed, water temperature, air pressure, photoperiod, water turbidity, water trans- parency and mean total 200plankton as factors in the analyses. A Dunnett-type test of abundance between stations showed no signi- ficant differences in any year. The Cladocera were generally bimodally distributed in 1975 and 1976, and had monocyclic patterns of distribution . in 1977. Bosmina longirostris was the dominant Cladocera, comprising 80-100 percent of the total Cladocera in some seasons. It reached abun- 3 3 dances of 18,472 m‘ in 1975 and 22,948 m' 3 in 1976, but only reached 5,283 m- in 1977. Eubosmina coregoni was most abundant in the fall. It appeared to have decreased in abundance over the years, from 8,129 m-3 in 1975 to 760 m-3 in 1977. Daphina spp. were common only in the summer 3 in 1976 but decreased in 1977. 2, retrocurva decreased from 1,063 m-3 in 1975 to and fall. ‘2. galeata mendotae reached 1,448 m- 3 to 286 m- 3 439 m- in 1976, but increased in abundance again to 1,960 m73 in 1977. An analysis of abundance between years for each species showed that only 2, retrocurva was not significantly different in abundance between any years. 38 39 The multiple linear regression analyses explained between 11.7 and 70.4 percent of the variation in abundance of the four Cladocera. |No variables were consistently important in explaining variance in all species or in all years. Mean total zooplankton was most frequently a signi- ficant variable in explaining variance, and accounted for between 1.4 and 16.5 percent of the variation. The variables most frequently important in the analyses were wind direction, whch explained 4.8 to 21.5 percent of the variation, and water temperature, which explained 1.0 to 44.0 percent of the variation. All variables appeared in at least two multiple regression equations. LITERATURE CITED Ahlstrom, E. H. 1936. The deep-water plankton of Lake Michigan, exclu- sive of the Crustacea. Trans. Amer. Microsc. Soc. 55: 286-299. Beeton, A. M. 1965. Eutrophication of the St. Lawrence Great Lakes. Limnol. Oceanogr. 10: 240-254. Birge, E. A. 1882. Notes on Crustacea in Chicago water supply with re- marks on the formation of the carapace. Chicago Med. J. and Exam., 43: 584-590. Brooks, J. L. 1957. The systematics of North American Daphnia. Mem. Conn. Acad. Arts and Sci. 13, 180 p. Buikema, A. L. 1971. Nonphotoperiodic light response of Daphnia. In: Symposium on ecology of the Cladocera. Trans. Amer. Micros. Soc. 90(1): 100-121. Carr, J. F., J. W. Moffett, and J. E. Gannon. 1973. Thermal characteris- tics of Lake Michigan, 1954-1955. U.S. Fish. Wildl. Serv., Fish. Bull. 69, 143 p. Damann, K. E. 1945. Plankton studies of Lake Michigan, I. Seventeen years of plankton data collected at Chicago, Illinois. Amer. Mid. Nat. 34: 769-796. ' . 1960. Plankton studies on Lake Michigan, II. Thirty- three years of plankton data collected at Chicago, Ill. Trans. Amer. Microsc. Soc. 79: 397-404. Deevey, E. S. and G. B. Deevey. 1971. The American species of Eubos- mina Seligo (Crustacea, Cladocera). Limnol. Oceanogr. 16: 201- 218. Duffy, J. E. and C. R. Liston. 1978. Comparison of zooplankton densities between Lake Michigan and the Ludington Pumped Storage Reservoir, 1973-1975. Mich. State Univ., Dept. Fish. Wildl., 1976 Ann. Rept. to Consumers Power Co., Vol. II, No. 2. 90 p. and . 1979. Final report of the zooplankton studies in nearshore Lake Michigan adjacent to the Ludington Pumped Storage Reservoir and at a control station, near Ludington, Michigan. 1979 Annual Rept. to Consumers Power, Mich. State Univ., Dept. Fish. Wildl., In progress. 4O 41 Duffy, W. G. 1975. The nearshore zooplankton of Lake Michigan adjacent to the Ludington Pumped-Storage Reservoir. Master's thesis, Dept. Fish. Wildl., Mich. State Univ. 135 p. Eddy, S. 1927. The plankton of Lake Michigan. Bull. Ill. State Div. Natur. Hist. Surv., 17(4): 203-232. Evans, M. S. and B. E. Hawkins. 1977. A multi-year comparison of summer zooplankton distributions in southeastern Lake Michigan. Pre- sentation to 1977 Int. Assoc. Great Lakes Res. Conf., Ann Arbor, MI. Forbes, S. A. 1882. On some Entomostraca of Lake Michigan and adjacent waters. Amer. Natur. 16: 537-542, 640-649. Gannon, J. E. 1971. Two counting cells for the enumeration of zoo- plankton micro-Crustacea. Trans. Amer. Micro. Soc. 90(4): 486-490. 1972. A contribution to the ecology of zooplankton Crus- tacea of Lake Michigan and Green Bay. Ph.D. thesis, Univ. Wisconsin, 257 p. and Cannon. 1975. Observations on the narcotization of Crustacean zooplankton. Crustaceana 28: 220-224. Gill, J. L. 1978. Design and analysis of experiments in the animal and medical sciences. Vol. I. The Iowa State Univ. Press; Ames, Iowa. 409 p. Hutchinson, G. E. 1967. A Treatise on Limnology. Vol. II. Introduction to lake biology and limnoplankton. John Wiley & Sons. New York. 1115 p. Jacobs, J. 1962. Light and turbulence as co-determinants of relative growth rates of cyclomorphic Daphnia. Int. Rev. ges. Hydrobiol. 47(1): 146-156. Liston, C. R., P. I. Tack and W. G. Duffy. 1974. A study of the effect of installing and operating a large pumped storage project on the shores of Lake Michigan near Ludington, Michigan. Vol. II. Lim- nological studies. 196 p. Consumers Power. Liston, C. R. and R. C. Anderson. 1979. A study of Lake Michigan's inshore diatoms near the Ludington Pumped-Storage Project during 1972-1976. 1978 Annual report to Consumers Power, Michigan State Univ., Dept. Fish. Wildl., In progress. McNaught, D. C. and A. D. Hasler. 1964. Rate of movement of populations of Daphnia in relation to changes in light intensity. J. Fish. Res. Bd. Canada 21: 291-318. 42 Murphey, G. I. 1962. Effects of mixing depth and turbidity on the pro- ductivity of freshwater impoundments. Trans. Amer. Fish. Soc. 91(1): 69—76. Nie, N. H., C. H. Hull, J. G. Jenkins, K. Steinbrenner and D. H. Bent. 1975. Statistical Package for the Social Sciences. Second Ed. McGraw Hill Co., New York. 675 p. Patalas, K. 1969. Composition and horizontal distribution of crustacean zooplankton in Lake Ontario. J. Fish. Res. Bd. Canada 26: 2135-2164. Pennak, R. W. 1978. Freshwater Invertebrates of the United States. Second edition. Wiley Interscience Pub., John Wiley & Sons; New York. 803 p. Roth, J. C. and J. A. Stewart. 1973. Nearshore zooplankton of south- eastern Lake Michigan, 1972. Proc. 16th Conf. Great Lakes Res., Int. Assoc. Great Lakes Res. 1973:. 132-142. Stross, R. G. 1971. Photoperiodism and diapause in Daphnia: a strategy for all seasons. In: Symposium on ecology of the Cladocera. Trans. Amer. Microsc. Soc. 90(1): 100-121. Tonolli, V. 1971. Methods of zooplankton collection. In: A manual on methods for the assessment of secondary productivity in fresh waters. Eds. W. T. Edmondson and G. G. Winberg. IBP Handbook No. 17. 358 p. Ward, H. B. 1896. A biological examination of Lake Michigan in the Traverse Bay region. Bull. Michigan Fish. Comm., No. 6, 99 p. Wells, L. 1960. Seasonal abundance and vertical movements of planktonic Crustacea in Lake Michigan. U.S. Fish. Wildl. Serv., Fish. Bull. 60: 343-369. 1970. Effects of alewife predation on zooplankton p0pulations in Lake Michigan. Limnol. Oceanogr. 15(4): 556-565. Wetzel, R. G. 1975. Limnology. W. B. Saunders Co., Philadelphia. 743 p. Williams, L. G. 1962. Plankton population dynamics. Nat. Water Qual. Netwk., U. S. Publ. Health Serv., Publ. No. 663, Suppl. 2, 90 p. . 1966. Dominant planktonic rotifers of major waterways of the United States. Limnol. Oceanogr. 11: 83-91. APPENDIX 43 Table A1. Values of variables used in multiple linear regression analyses. See Table 4 for description of variable names. Date WD WS PR PH TU SC 1 3 5 1 3‘ 5 1975 4-22 7 8 30.27 13:42 2.8 2.1 3.2 1.5 2.0 1.3 5-2 9 14 30.08 14:08 '1.4 2.3 2.4 3.5 3.0 1.6 5-13 1 8 30.40 14:33 2.9 1.2 1.7 2.5 3.0 3.0 5-29 2 5 29.88 15:03 1.0 1.0 1.1 2.7 2.5 2.6 6-17 9 16 29.69 15:19 1.3 1.3 1.1 3.0 5.0 4.5 6-30 5 6 30.26 15:17 1.2 0.8 1.0 5.5 6.0 4.3 7-14 3 8 30.06 15:03 1.1 1.0 1.1 4.0 4.0 5.0 7-28 11 6 29.96 14:38 3.5 2.6 2.6 2.5 2.0 2.0 8-11 9 11 29.89 14:06 1.9 1.7 2.0 3.8 3.8 3.8 8-27 13 16 30.30 13:24 2.3 3.0 2.9 3.2 3.5 3.5 9-9 5 5 30.37 12:48 1.8 4.3 4.6 5.0 1.9 1.8 9-24 3 8 30.31 12:05 2 4 3.8 3.5 3.5 1.8 --- 10-7 5 6 30.21 11:28 -—- 1.8 2.5 --- 6.6 3.1 11-5 10 13 30.18 10:09 2.5 1.9 2.4 3.0 3.0 4.0 1976 4-14 9 13 29.86 13:20 2.5 --- 3.4 3.5 4-28 16 7 30.28 13:58 5.0 3.0 1.4 2.4 5-12 1 8 29.98 14:31 4 6 3.0 1.4 2.4 5-25 1 8 30.04 14:56 2.5 2.2 3.8 3.2 6-7 10 5 30.11 15:13 1.0 0.7 4.2 5.2 6-21 10 5 30.01 15:20 1.2 --- 3.4 2.8 7-6 10 5 29.98 15:13 2.0 2.1 3.9 2.1 7-21 3 7 30.02 24:52 - --- 4.1 4.1 8-2 1 8 30.18 14:28 1.8 4.1 3.0 3.5 8-15 2 9 30.06 13:56 1.6 2.9 3.4 2.2 9-15 3 9 30.19 12:31 1.5 3.5 3.3 2.5 9-26 5 8 29.83 12:00 1.7 5.2 3.8 2.9 10-11 5 7 30.04 11:17 1.4 0.9 5.1 3.0 1977 4-18 6 6 29.91 13:31 1 5 --- 1.4 5.4 --- 5-17 9 9 29.95 14:41 2.8 1.7 2.1 4.0 3.5 6-13 3 8 30.07 15:18 2.1 2.0 1 4 3.7 5.6 7-21 9 11 30.03 14:52 1.8 1.9 3 3 4.5 4.9 8-18 16 10 30.04 13:49 1.9 1.6 2.1 5.5 5.5 9-13 5 8 29.78 12:37 2.9 2.7 3.6 3.0 2.2 10-22 15 8 27.79 10:46 1.4 1.6 2.5 2.8 --- Table A1. (cont'd.) 44 DATE 22* 1 1 3 5 1975 4-22 3.0 1,858.5 1,444.6 1,157.5 5-2 5.2 1,991.5 1,420.4 1,960. 5-13 5.4 3,335.5 2,978.9 ----- 5-29 12.0 19,915.7 26,534.7 19,087.0 6-17 13.6 30,312.3 33,382.7 38,661.2 6-30 16.0 14,444.o 9,071.3 18,019.5 7-14 12.3 8,530.5 12,858.0 8,992.2 7-28 20.4 9,191.9 7,301.7 7,760.6 8-11 19.1 4,620.2 5,621.1 3,987.6 8-27 20.1 2,486.8 3,217.1 4,469.4 9-9 12.3 2,208.0 2,976.2 3,426.1 9-24 14.7 13,465.0 16,030.1 7,155.6 10-7 -—-- ---- 5,888.6 6,210.2 11-5 11.0 6,860.1 11,374.9 4,953.4 1976 4-14 4.0 4.4 4.2 1,467.5 1,517.6 ,046.1 4-28 6.2 6.0 5.8 2,308.4 2,113.6 2,003.5 5-12 7.0 7.1 7.0 2,259.6 2,796.3 4,820.4 5-25 6.0 5.8 5.9 1,187.6 6,463.2 1,505.7 6-7 7.8 8.0 9.0 1,583.4 18,870.4 2,989.6 6-21 16.6 17.5 18.1 26,516.4 25,466.0 14,990.0 7-6 11.7 15.0 11.6 11,584.0 6,692.8 7,813.7 7-21 14.2 14.8 16.0 7,421.2 14,366.7 7,336.0 8-2 7.5 6.8 8.5 26,563.9 39,543.0 32,785.3 8-15 12.5 12.6 11.9 23,568.3 28,414.0 35,219.0 9-15 14.8 15.5 17.0 12,652.5 12,356.0 11,277.7 9-26 10.4 8.0 8.4 17,083.9 3,780 3 4,576.1 10—11 13.0 13.4 13.8 4,415.6 5,897 5 5,146.6 1977 4-18 3.0 3.2 1,772.7 2,034.5 848.0 5-17 11.8 12 6 4,507.8 5,389.3 4,438.5 6-13 5.5 6.0 5,194.7 7,276.7 4,349.5 7-21 22.5 22.5 13,672.7 13,672.7 27,592.7 8-18 14.6 14.0 16,553.6 23,553.2 14,829.9 9-13 17.9 18.0 6,787.8 4,799.6 5,211.3 10-22 9.0 9.2 5,950.0 5,955.0 2,693.3 *Abundance of the species being analysed was subtracted from this value to obtain the variable TZP for each date and year. 45 H.mam.a H.mma oa~.om1o a.amc.m ocuz oaeoe a.aam.m m.aaa.a oma.ma1aaa.a a.mmm.m m1aa 1 1 1 1 a1oa k.awa.c w.o~o.m oma.m~1eoa.m m.~ac.aa a~1a H.mm a.wa 33316 c.c~ a1m c.4m ~.oa ac~1o n.ca A~1m a.maa o.am maa1o o.ca~ Ha1m m.koc.a m.cm~.~ NHH.aH1omm ~.mcm.n m~1e m.maa.~ ~.Hao.H omm.m1oma.a m.~ac.m 431A c.aca.w a.amo.a cam.a~1aa~.m A.awm.oa om1c m.amm.m m.~am.s oa~.om1oac.~ a.oa~.aa aa1c ”.mcm m.ae~ och.a1oaa o.aoo.a a~1m 1 1 1 o «31m 1 1 1 o N1m 1 1 1 1 «~13 q<>mmezH 11mmmmmm1 _ mozfiumwuomoo .~< manna 46 a.~mm.a c.caa.a ooa.cm1o a.kam.m amuz .58. w.mmc.~ H.~om.a ooa.ma1oam.a A.Ha~.a A133 m.am~.m o.omc.H mmo.aa1caa m.m~a.m a1oa ~.maa.m o.aa~.a ock.wm1mac.aa m.~a~.m~ a~1a a.mc ~.am oH~1o a.ca a1a a.ac~ a.mma cam1o m.caa a~1m n.3om.m a.~m~.a om~.HH1amc h.aca.m 331m a.ac~.~ m.-a.a Nam.m1oma o.o~a.m m~1a o.a- m.acm occ.m1oo~.m ~.mam.a «H15 ~.mao.~ o.aao.a oqc.c1oma m.mma.m om1c o.aaa.aa c.aam.m ooa.cm1oam.aa m.m~m.oM A316 a.a~c c.~am oom.~1o~a ~.cma.a a~1m m.ma “.6 oa1o 4.6 ma1n k.~a «.6 mm1o m.a ~1m m.wa a.a ka1o o.o~ «~14 A<>mmezH 1mmmmmm11 mozcm Am1a .czv zfiuafiuomon .m< canoe 47 n.4oa.a m.acc oom.n~1o o.cao.a ccuz aches c.aha 4.4mm omm.~1ama N.~H¢.H m1aa H.oco.~ a.-o.a caa.m1aaa.a “.4Nm.s ~1oH m.aca.~ ~.Haa.a oa~.~a1-m.~ c.a~m.k a~1m ~.Hw N.oa ac~1o m.mm A1a c.~aa c.ma aam1~a m.c~ A~1w ~.o- m.aoa acm1c~a m.aam 331m o.a~m.m c.-a.a aca.ma1asa m.caa.s m~1k H.ch.a a.aam cma.c1mhm.a ~.coa.m 331a m.aca.a o.aac.a ooa.m~1cae.a ~.oom.oa om1c m.aa~.m m.awm.a oma.ma1oom.c m.mam.¢a aa1c «.AAA «.6mm cam.m1oaa.a a.aa~.~ a~1n 1 1 1 o ma1m 1 1 1 o ~1m 1 1 1 0 -1a A<>mmezH 11mmmmm1 muzfiumwuomoa .¢< manna 48 O" a.~mk.3 c.33o.3 ooc.3~1o . o.~co.m «~12 3mmHzH mommw mug Ania .23 2%.” make mozmnHmzoo Rom nm3umwuomon .m< canoe 49 6.666.3 6.636 666.6316 6.636.6 6612 36666 6.636.6 6.636.3 666.631666.6 6.663.63 6133 6.636 6.666 . 636.61636 6.666.3 6163 6.636.3 6.366 666.31336.3 6.666.6 3616 3.63 6.6 6316 6.63 616 6.666 6.663 66616 6.636 6616 1 1 1 1 3316 1 1 1 1 6616 1 1 1 1 3316 1 1 1 1 6616 1 1 1 1 6316 1 1 1 1 6616 1 1 1 1 6316 I 1 1 I NNIQ 36666626 66666 66266 2616 .626 26623 1.6666 6626636266 666 66662666 .m663 a6 m coaumum um cmwfinoaz 6664 :6 Haemouoo mcfismonsm 60 6666666666 o>6umfiuomon .o< 06669 67 50 0" 6.363 6.666 666.316 6.663.3 6612 36666 6.666.3 6.636 666.31666.3 6.363.6 6133 6.666 6.366 663.61366 6.366.3 6163 6.666 3.663 666.61366.3 6.366.6 3616 6.66 6.33 6616 6.33 616 3.636 6.666 666.3166 6.636 6616 3316 6616 3316 6616 6316 6616 6316 I l I OOOOOOOOO leq A<>MMHZH mommm muz6umfiuomoa .64 oases 51 cnz 6.66 6.36 663.316 3.666 3612 36666 6.636 6.663 666166 6.666 6133 1 1 1 1 6163 6.666 6.363 663.31636 6.666 3616 6.363 6.36 66616 6.663 616 6.363 6.636 666.316 6.666 6616 6.666 6.663 666.31663 6.666 3316 3.66 3.33 666166 6.666 6616 6.633 6.66 66316 6.366 3316 6.63 3.36 66316 6.66 6616 6.63 6.66 63316 6.66 6316 1 1 1 6 6616 1 1 1 6 6316 1 1 1 6 616 1 1 1 1 6613 36666623 law! 6626 661a . 626 26623 131316161311 6626636266 666 66662666 .6663 :6 3 :oHumum um cmmfinofiz 6664 66 66606665 mumofimw Managua 60 6666666666 o>auaauomoo .m4 oases 52 w.qm6 m.-q m.N56 w.mem 6.60 6.00m 6.666.3 m.mm6 6.w66 m.oo6 6.N6 A<>mMHzH mozmaHhZOU Noo o.~o n.6mm m.mm m.ahm m.om ~.m¢m w.6mm 6.00 o.mm o.mm mommm nm66m666669 ocnz AmmHzH mozmthzoo NC¢ ~.oq6 6.363 6.66 6.666 6.63 6.666.3 6.633 3.633 6.66 6.33 momma nm66m6uomma «mnz A6=oou6mu 66:666: 60 6666666666 «>66a666669 .66< 66666 55 w.mea m.Hom n.5m m.m¢m n.5H ~.m~m ¢.am~ c.¢- H.m¢ A<>mmHzH mozwnHmzoo Nam m.aw N.OOH “.ma m.mq¢ h.m a.~oa o.m~H n.0ma q.H~ momma QMusuouuou macnmma mo mowumwumum m>6quuummn .ozv z66626 66666 66266 6616 .626 266:6 . 6666 6626666266 666 66662666 .6666 :6 6 6066666 66 66w6£662 666A :6 66666666m606 6:6860m no 6666666666 6>66afiuumma .q6< 6666B 6666 6666 666 612 58 6 12 6 6 6.666.6 6.666.6 636.6616 6.666.6 6612 66666 6.666 6.666 636.61366 6.666 66166 6.666 6.666 63616 6.666 6616 6.666 6.666 66616 6.663 6616 6.663.6 6.666.6 666.661663.6 6.666.66 6616 6.366.6 6.666.6 666.661666.6 6.666.36 616 6.666.3 6.636.6 633.661666.6 6.666.66 6616 6.666.6 6.666.6 666.661666.6 6.666.6 616 6.666.6 6.666.3 636.661666.6 6.666.66 6616 3.66 6.63 6661666 6.666 6616 6.666 6.666 666166 6.666 6616 6.666 6.66 663166 6.666 6616 6.666 6.66 66316 6.666 6613 6.366 6.666 636.6166 6.366 3613 66>66626 66666 66266 661s..626 26666 6666 6626666266 666 66662666 .6666 :6 m 6066666 66 :6w6566z 6666 66 66666066w606 6665606 60 6666666666 6>666666669 .m6< 66669 S9 6.636.6 6.666 6.336 6.666 3.666.6 3.666.3 6.666.6 6.636.6 6.636.66 6.66 6.66 6.66 6.36 6.666 A<>MMHZH mozmathoo Noa :6 m cowumum um cmwwsuwz 6x64 :6 m6uumOMHmsoa mafiamom no 6666666666 m>6uawuumma 6.663.6 w.nmm 5.65 6.666 3.366.3 6.666.6 6.636.6 6.666 6.666.6 5.66 m.om 6.66 m.Hq «.mm momma ammMHzH mozmnHmzoo Nom 6 coauMum um awm6£o6z mxmq =6 acowmuou mcaamonsm mo 6666666666 m>wum66omma m.~ma w.om6 n.6m n.6ww n.Hm «.mwa «.moa 6.66 o.m o.c~ n.60m mommm GmmmHzH mozmnHmzoo Noo m coauMum um cowanuwz 03mg :6 “commune ncwamonam mo mufiumaumum u>6unfiuumma H.@HH 0.05 «.mam H.mNH «.moa m.mm w.ww w.ma m.mm “.0 m.omH mommm QmmMHzH mozmnHmzoo Nom m.mma N.oN mommm am6uawuomma ocuz AmMHzH mozmnHmzoo Nom m.mma w.¢~ m.¢~ m.mn¢ c.mo «.60: ~.mmN m.nm m.o N.oN momma nmwumauomma ocuz AmmHzH muzmQHmzoo Noa m.mma w.m~ m.qHH m.~m~ H.6Ho a.omq m.NmH q.nm m.mc mommm nm6umwuommn ocuz AmmezH mozmnHmzou new m coaumum um cmwwsuwz 03mg :6 N.OOH 0.6m n.6w o.nme m.¢~a o.om mommm nm6uaauowma cmnz AmMHzH muzmthzoo NO¢ .osmm :« 6 coauwum um unwanoaz 6364 :6 m>u=oouumu aficnmwa mo muaumfiuMum m>auafiuummn «.mc o.n mommm Qm6mez6 .-Jmmmmw- 66266 Am-a .626 26626 .4wmflw- 6626666266 666 66662666 .0636 :6 m coaumum um :mwfisuwz mxmq 56 m>u=oouumu macfimma mo moaumaumum m>6umauommo .qm< wanna 68 o.o« n.~m N.¢OH m.hHH q.ama H.MH A<>mmHzH mozmnHmZOU Noo .cm¢6 :6 m coaumum um :meSUAZ 0x64 :6 m>6=oouumu «assume we mafiumaumum m>auafiuomoa c.m~ o.oc N.vm m.wm c.0m mommm nm6umauumma .oN< 7O 6666 6666 666 6-2 6 6-26 6.666 6.666 666.66-6 6.666.6 66-2 66666 6.666 6.666 666.6-6 6.666 66-66 6.666 6.666 666.6-666 6.666.6 666-6 6.666 6.666 666.6-666.6 6.666.6 66-6 6.666.6 6.666.6 666.66-666.6 6.666.6 66-6 6.666.6 6.666 666.6-666.6 6.666.6 66-6 6.666 6.666 666-666 6.666 66-6 6.66 6.66 666-6 6.66 66-6 6<>66626 66666 66246 66-2 .626 26626 6666 mozmnHmzoo Nom Qm6umfiuomon .NN< 06669 71 6 6.666.6 6.666 666.6-6 6.666.6 66-2 66666 6.666 6.666 666.6-6 6.666 66-66 6.666 6.666 666-6 6.666 66-6 6.666.6 6.666 666.6-6 6.666.6 66-6 6.666.6 6.666.6 666.6-666.6 6.666.6 66-6 6.666.6 6.666 666.6-6 6.666.6 66-6 6.666 6.666 666-6 6.666 66-6 6.66 6.66 666-6 6.66 66-6 66566626 66666 66266 Am-a . 626 26666 66666 mozmaHmzoo Nom Qmauaauummn .mm4 66669 72 o.oo6 0.666 m.oo6 c.66m m.qmm o 6<>mmBzH mozmathcu Mom .6666 :6 6 coaumum um cmw63662 @666 :6 fiaowmuou mc6amonsm no 6666666666 m>666666606 6.0m 6.06N «.mn 6.66N m.mm6 o mommm 6666326666 666.6-6 666.6-6 666-666 666.6-6 . 6 6 666.6-6 mozmMHzH mozmnHmzou Nom ~.mmH o.m~ “.mmu 6.5ma ¢.qhm N.Hm mommm nmfi6afinumma oNuz A4809 NNIOH malm mHIw HNIN mate main walq MBmMBzH mozmonzoo nco .5566 :H n :0H6666 66 ammfinowz 6:64 :« ficowmuoo mafiamonsm «0 6666666666 6>a6awuomwa m.maa m.mma c.6wa m.mmm m.mo mommm nmmmHzH muzmQHmzoo Nco m 6066666 66 w.maa m.mmH mommm Qma6awuo6ma oo6.6-o muzmmazH mozmnHmzou NO¢ m :066666 66 m.mca m.mma o.oh o.m- mommm nmH6awuomma omm.6uo oo~.6uo omnuo omm.6no muzmmazH .udmmmmm. 6626666206 Rom nmfiumwuummo NHuz Ammezm mommm muzmm Anna .ozv zmmzH mama mozmnHmzou Mom nmu=oou666 wasnmmo mo 66%66fi6666 6>H6aw66669 .mm< magma 79 N.sam o.wnn c.0om H.mmm H.6mo o o o A<>mmHzH muzmaHmzoo Nom .mhmm 6H n 6066666 66 :6wmnomz 6364 cm 6>uaoou666 66::m6n 60 6066666666 6>H6mmuu6wn w.¢ma ~.onm “.mea m.m~¢ m.c¢m mommm nmmmazm mommm 665m Anus . 05 24626 6.63 mozmnHmzou Nom Gmusoou666 666666: 60 6066666666 6>66666666o .mm< 6H66H