UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phytoplankton ecology in a high arctic polynya Butler, Joanne Elizabeth 1985

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-UBC_1985_A7 B87.pdf [ 3.74MB ]
Metadata
JSON: 831-1.0062952.json
JSON-LD: 831-1.0062952-ld.json
RDF/XML (Pretty): 831-1.0062952-rdf.xml
RDF/JSON: 831-1.0062952-rdf.json
Turtle: 831-1.0062952-turtle.txt
N-Triples: 831-1.0062952-rdf-ntriples.txt
Original Record: 831-1.0062952-source.json
Full Text
831-1.0062952-fulltext.txt
Citation
831-1.0062952.ris

Full Text

PHYTOPLANKTON ECOLOGY IN A HIGH ARCTIC POLYNYA BY JOANNE ELIZABETH BUTLER B . S c , The University of Alberta, 1979 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF APPLIED SCIENCE in THE FACULTY OF GRADUATE STUDIES (Civi l Engineering Dept., Environmental Engineering Group) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March 1985 @ Joanne Elizabeth Butler, 1985 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make i t freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Civil Engineering The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date 15 March 1985 ) E - 6 ( 2 / 7 9 ) ABSTRACT Primary production was studied in Fram Sound, part of the Hell Gate-Cardigan Strait polynya, from June to August, 1982. Primary production rates, phytoplankton biomass (chlorophyll a) , and water transparency were measured and used in conjunction with modelled solar radiation values to numerically model primary production during this time. The major phytoplankton nutrients were also measured. Early season chlorophyll a concentrations were low, and the increased l ight avai labi l i ty due to reduced ice cover in this area did not appear to enhance early season production. Chlorophyll concentrations peaked twice; the f i r s t peak occured on 20 July and the second on 14 August. The mean primary production rate and phytoplankton -2 -1 -2 biomass were 998 mg Cm .d and 72 mg chl.m . This production rate is higher than that measured in other High Arctic areas. Nitrogen, phosphorus and s i l i c a were essentially homogeneously distributed during the sampling period and these concentrations varied l i t t l e from June to August except during 5 days in late August, when they decreased by half then returned to previous levels. Supervisors: - i i -TABLE OF CONTENTS ABSTRACT i i ACKNOWLEDGEMENTS v i i INTRODUCTION 1 MATERIALS AND METHODS 8 Study Area • 8 Sampling 13 Chlorophyll a 16 Nutrient Chemistry 16 Physical Oceanography 17 Primary Production 19 Photosynthesis-Light experiments 19 Model components 27 Stat ist ica l treatment of data 29 RESULTS 34 Chlorophyll a 34 Nutrient Chemistry 39 Physical Oceanography 56 Primary Production 61 DISCUSSION 75 Chlorophyll a 75 Nutrient Chemistry 75 Physical Oceanography 78 Primary Production 79 Conclusion 81 LITERATURE CITED 83 APPENDIX A. Solar Radiation Models 86 APPENDIX B. Conversion factors for solar radiation measurements 102 - i i i -LIST OF FIGURES Figure page 1. Map of recurring polynyas in the Canadian Arct ic 3 2. Map of the study area 10 3. Map of ice conditions in the Hell Gate-Cardigan Strai t polynya . . 12 4. Map of locations sampled in Fram Sound from 30 May to 17 August, 1982 15 5. Schematic diagram of the incubator used to measure the rate of primary production 21 6. Longitudinal and cross-sectional schematic diagram of the incubator used to measure the rate of primary production 23 7. A Photosynthesis-Irradiance curve, showing alpha, the slope of l ight- l imited photosynthesis, I. , the irradiance at the onset of l ight saturation and P m, the rate of primary production at light saturation 26 _o 8. Chlorophyll a concentrations (mg.m ) in the water column from June to August, 1982 (mean+2 standard deviations) 36 9. Isopleth of chlorophyll a concentrations in the top 50 m from June to August, 1982 38 10. S i l i c a concentrations (umol.L -*) in the water column from June to August, 1982 (mean+2 standard deviations) 41 11. Isopleth of s i l i c a concentrations (umol.L *) in the top 50 m from June to August, 1982 43 12. Isopleth of s i l i c a concentrations normalized to the maximum daily concentration 45 13. Particulate nitrogen in the water column (umol.L -''') from June to August, 1982 (mean+2 standard deviations) 48 14. Total nitrogen concentration (umol.L -*) in the water column from June to August, 1982 (mean+2 standard deviations) 50 15. Total phosphorus concentrations (umol.L -*) in the water column from June to August, 1982 (mean+2 standard deviations) 52 16. Particulate phosphorus concentrations (umol.L )in the water column from June to August, 1982 (x+2 standard deviations) 54 17. The change in extinction coefficients from June to August 1982 . . 58 18. The euphotic depth (the depth to which 0.5% of the surface l ight penetrated) from June to August, 1982 60 19. Alpha values (mg C/(mg chl.E.m )) obtained from photosynthesis-irradiance experiments 64 20. Frequency diagram of alpha values 66 B -1 21. P i values (mg C/(mg chl.h )) obtained from photosynthesis-irradiance experiments 68 • 22. Frequency diagram of P m values 70 _2 23. Integrated primary production (mg Cm ) calculated with three l ight regimes from June to August, 1982 73 LIST OF TABLES Table page 1. Descriptive s ta t is t ics for chlorophyll a and nutrients for days when more than one location was sampled 31-33 2. Results of the stepwise regression of chlorophyll a with dissolved nutrients and date 55 _o 3. Descriptive s ta t is t ics for alpha (mg C/(mg chl.Einstein.m )) and Pbm (mg C/(mg chl.h l ) ) 62 4. Primary productivity calculated from 1 June to 17 August, 1982, using three l ight regimes (maximum, cloud-corrected, and minimum solar radiation) 72 ACKNOWLEDGEMENTS My sincere thanks to a l l the people who have helped tranform my desire to learn about the Arctic from a project idea into the real i ty of a finished thesis. The Environmental Engineering Group of the C iv i l Engineering Department at the University of Br i t ish Columbia (UBC) gave me a place to start . Financial support was i n i t i a l l y provided by the National Science and Engineering Research Council and later by a UBC Graduate Fellowship. Prof. Jim Atwater and Dr. B i l l Oldham of this Department allowed me the opportunity to pursue a project of my choice. Jim Atwater's support and f l e x i b i l i t y as my supervisor at UBC enhanced a l l phases of my learning process in this program. Dr. Buster Welch of the Freshwater Institute, Winnipeg, agreed to supervise and f inancia l ly support my project while I was s t i l l at UBC, gave me the freedom to learn on my own, and many times helped me find the path when I was lost in the woods. I have benefited greatly from his wi l l ingly shared enthusiasm and knowledge of the Arct ic and science. Gerry Prach and Al Smith of the Canadian Wildlife Service (CWS) provided some log is t ic support at Cape Vera during the 1982 f i e ld season. Many thanks to Tim Si ferd , who provided support and invaluable assisstance in the f i e ld and to Alex Dzubin (CWS) for his encouragement at Cape Vera. The Polar Continental Shelf Project (Resolute Bay), headed by George Hobson, of the Department of Energy, Mines and Resources (EMR) generously supplied log is t ic support and generators. The helpful suggestions and solar radiation l i terature given by Dr. Bea Alt (EMR Ottawa) were greatly appreciated. To the many gifted and kind people at the Freshwater Institute, Winnipeg, who have helped during the data analysis and writing phases of this thesis I offer my heartfelt thanks. Particular thanks to Dr. Everett Fee, who provided the numerical model I used to calculate primary production and who graciously answered countless software questions; to John Legault, who expedited the equipment for my f i e ld work and helped in many ways; and to Ron Lypka, who gave computer advice and helped with the solar radiation model program. Mike Stainton and the Analytical Chemistry Section analysed the nitrogen and phosphorus nutrient samples, Hedy Kling identif ied phytoplankton samples, Er ic Marshall and the Library staff readily accessed even the most obscure l i terature, and Andries Blouw and the Graphics Dept. wi l l ingly and s k i l l f u l l y transformed rough maps and diagrams into informative f igures. Dr. Jim Reist provided s ta t is t ica l advice and Kathleen Martin-Bergmann made several helpful suggestions on ear l ier drafts of this manuscript. Martin Bergmann also gave many helpful suggestions along the way. I am grateful to Gordon Koshinsky and Buster Welch for providing the opportunity for uninterrupted writing last summer. For the moral support, encouragement and patience of Rob Walker during the data analysis and writing stages of this thesis I can only say thank-you very much and i t ' s f inished! INTRODUCTION Otto Sverdrup described his f i r s t view of the open water of Hell Gate (March 1900) as an area where " great pressure hummocks were dr i f t ing along at t e r r i f i c speed in the violent whirlpool caused by the strong t idal current", and that "none of us had ever seen waters so utterly impossible to navigate as the sound here" (in Taylor 1955). Polynyas are areas of water, surrounded by sea ice , where ice cover is reduced or absent for a l l or most of the year. Annually occuring polynyas found in the same location are called recurring polynyas and are distributed throughout the Canadian Arctic (Fig. 1). They range in size from the Cambridge Fiord polynya, south of Pond Inlet, with a 2 surface area of less than 1 km , to the North Water, estimated to be 100,000 km2 (Dunbar 1981). These areas of open water are formed and maintained by various combinations of wind, strong set (permanent) currents, t idal currents and upwellings of warmer deep water, and Dunbar (1981) has comprehensively reviewed several theories on the interaction of these factors. Wind is the most important factor in the formation and maintenance of polynyas. -1-Figure 1. Map of the recurring polynyas in the Canadian Arct ic (from Smith and Rigby 1981). -2 -3-Offshore winds remove newly formed ice from the lee shores and induce upwellings. The formation of sea ice increases the sa l in i ty of the surrounding water by freezing out solutes. When new ice is continually formed (which occurs in areas with strong prevailing winds) the sal in i ty of surface water increases until i t is denser than the water below, and subsequently sinks. When the deep water is warmer than the surface water, this sinking, called haline convection, results in a vertical exchange of water by displacement and heat is brought to the surface. The North Water, Cape Bathurst polynya and the Baffin Island coastal flaw lead have underlying layers of warmer water. It is postulated that these areas are formed and maintained by this combination of strong prevailing winds, haline convection and warm deep water (Dunbar 1981). Ice can also be mechanically removed by strong t idal currents. Dunbar (1981) suggests that this mechanism is probably the most dominant force maintaining the polynyas in Cumberland Sound and Frobisher Bay, where the t idal ranges reach 7.6 m and 13.1 m respectively (Sail ing Directions-Arctic Canada 1982). A combination of set and t idal currents help maintain the Hell Gate-Cardigan Stra i t , Penny Strait and Queens Channel polynyas (Smith and Rigby 1981). The biological importance of polynyas is attested to by the large number of marine birds and mammals associated with them, although few quantitative studies have been conducted to investigate their ecological significance (St ir l ing 1981). The Cape Bathurst polynya in the Beaufort Sea serves as a spring staging area for migrating beluga (Delphinapterus  leucas) and seabirds, as an overwintering area for subadult ringed seals 4-(Phoca hispida) and bearded seals (Erignathus barbatus). Lancaster Sound is a major feeding and breeding area for 2 to 3 mi l l ion marine birds and approximately 40,000 beluga and narwal (Monodon monoceros) annually, many of which congregate in the area between Devon and Bylot Islands from early spring to late autumn (Milne and Smiley 1978). The Fram Sound area supports a large colony (-10,000 pairs) of Northern Fulmars (Fulmaris g lac ia l is ) which nest near Cape Vera at the south end of the polynya from early May to late September, and the morthernmost known colony of Common Eiders (Somateria mollissima boreal is) which nest on St. Helenas Island in Fram Sound (Prach et a l . unpub. data). Walrus (Odobenus rosemarus), ringed seal , bearded seal and polar bear (Ursus  maritimus) were frequently seen during aerial surveys of the Hell Gate-Cardigan Strait polynya (pers.obs.). Approximately 100 walrus overwinter in the Penny Strait area each year (St i r l ing pers. comm.). A combination of factors probably accounts for the supposed high biological productivity associated with polynyas and nearby areas; the presence of open water for most or a l l of the year increases opportunities for marine birds and mammals to feed, and the fact that wi ldl i fe congregates in such areas suggests that food is indeed available. Offshore o i l and gas production and the imminent need for year-round shipping threaten to disturb several polynyas, since the absence of sea-ice for a l l or most of the year make these areas attractive shipping corridors. The Cape Bathurst polynya and Lancaster Sound are on the proposed tanker route for the year-round transportation of o i l and gas to southern markets (Beaufort Sea Hydrocarbon Production and Transportation Proposal 1984). -5-Hell Gate-Cardigan Strait and Fram Sound, a recurring polynya in the Canadian High Arctic, is the site of a long term ecological study begun by the Canadian Wildlife Service (CWS, Edmonton) in 1980 to provide baseline data on the biological significance of this area to marine wildlife. The purpose of the current study was to measure primary productivity at Fram Sound and incorporate these data into the CWS database. The contribution of early season production in Arctic polynyas has not previously been studied; the effects of reduced ice cover with the onset of polar day suggests primary production may proceed earlier than in surrounding ice-covered areas. The base of the marine food chain is mainly phytoplankton, single-celled photosynthetic organisms which synthesize organic carbon from carbon dioxide (CO )^ and water in the presence of light. This is refered to as primary production, and the rate of primary production, called primary productivity, is indicative of the production of higher trophic levels (Parsons et al. 1979). Primary productivity is commonly measured with a radioactive (^C) bioassay. A known concentration or activity of radioactive carbon is added to several water samples, and each sample is incubated at one of a series of light levels for a known time (Steeman-Nielsen 1952). Chlorophyll a, a photosynthetic pigment extracted from concentrated phytoplankton, is used as an index of phytoplankton biomass, and ^C-uptake rates are generally normalized to chlorophyll a. The amount 14 of C taken up by the phytoplankton is used to determine a light response at various light levels, and these measurements are in turn used to estimate the rate of primary production as a function of light -6 -in the sea. The quality and quantity of solar radiation change as the l ight penetrates the water column. The longer wavelengths are attenuated near the surface, usually within the top 10 m. The quantity of photosynthetically active radiation (PAR, 400-700 nm waveband) decreases exponentially with depth and strongly affects the rate of primary production. Measurements of submarine l ight , chlorophyll a 14 concentrations and C-uptake at various depths over time provide the necessary information to model primary productivity for the area being studied. Nitrogen, phosphorus and s i l i c a are phytoplankton nutrients, and their concentrations are often measured in conjunction with primary productivity studies to reveal i f nutrients, part icular ly nitrate or s i l i c a in the sea, may be l imiting productivity rather than l ight . Concentrations of these essential nutrients frequently change seasonally and with depth, depending on local and large scale hydrological conditions. Specif ic objectives of this study were: 1) to measure primary productivity (using * 4 C ) , chlorophyll a concentrations, water transparency and incoming solar radiation from June to August, 1982, 2) to model primary productivity in the Fram Sound area using the numerical method of Fee (1984), and 3) to measure the spatial and temporal distr ibution of nutrients. These data wi l l then be incorporated in the CWS Hell Gate-Cardigan Strai t Polynya Project database. -7-MATERIALS AND METHODS Study Area Hell Gate and Cardigan Strait are narrow channels through which water from Norwegian Bay flows into Fram Sound before entering the west end of Jones Sound. These two channels and Fram Sound form the Hell Gate-Cardigan Strait polynya, located approximately between 76°20'-76°50'N latitude and 89°30 1 -90°00 1 W longitude (Fig. 2). This polynya is maintained by a combination of strong set and t idal currents (Smith and Rigby 1981). The set or permanent current, which flows south from Norwegian Bay, ranges from 1.5 to 3.5 knots (Pilot of Arct ic Canada 1978). Strong winds at Cape Vera up to 70 knots (pers. obs.) probably influence the extent of ice formation and the movement of ice in Fram Sound. The maximum extent of open water in the polynya occurs in May, June and July. In September the bays and fiords freeze with new ice , and in October and November the straits are • clogged with older ice from Norwegian Bay (Smith and Rigby 1981). Open water is generally present in early December, and from then until spring the extent of open water varies (Fig. 3). Fram Sound was the area sampled for the present study. The depth of Fram Sound ranges from 150-200 meters mid-channel to a shallow water -8-Figure 2. Locator map of the study area. Norwegian Bay flows into Jones Sound through Hell Gate, Cardigan Strait and Fram Sound. Cape Vera was the base camp for the present study. -9-Figure 3. Map of ice conditions in the Hell Gate-Cardigan Strai t polynya. (from Smith and Rigby 1981). -11-large black areas indicate open water to 1 /10 ice cover thin line leading into thicker line indicates a crack/lead — single thin line indicates division between ice cover categories L indicates an open lead and is inserted in some cases where the lead is not obvious 1 2 / 1 0 - 5 / 1 0 ice cover 2 6 / 1 0 - 7 / 1 0 ice cover 2+ 8/10 ice cover 3 9 / 1 0 - 1 0 / 1 0 i c e c o v e r F fast ice N new ice O old ice M multiyear ice A 1st-yearice <A less than 1st-year ice S 2nd-yearice U unknown G grey ice W grey-white ice margin (about 1 km) of less than 50 meters along the southwestern shore. This shallow zone extends to approximately 5 km off the t ip of Cape Vera. Tides in this area are semidiurnal; at the Bay of Woe the maximum and average t idal ranges are 3 m and 1.3 m respectively (P. Davies, Canadian Hydrographic Service, pers. comm.). Sampling Fram Sound was sampled from 30 May to 17 August 1982. Sampling locations varied as necessitated by ice and weather conditions; strong winds and moving pack ice often restricted sampling to nearshore locations (Fig 4). When conditions permitted, more than one location was sampled on the same day to test for homogeneity of the water mass. Two locations were sampled on July 5 (Locations E,F) July 20 (L,M), August 2 (R,S), August 5 (T,U) and August 16 (X,Y). Three locations were sampled on August 17 (Z,AA,BB). An opaque polyvinyl chloride (PVC) 2 L Van Dorn bottle was used to col lect discrete samples, which were usually taken from 0,2,5,10,and 20 m. At locations where the maximum depth exceeded 20 m and weather conditions permitted, sampling was extended to 30,40,50 and 75 m. Samples were transferred to opaque 2 L polyethylene Nalgene bottles in the shade to avoid l ight shock, and kept cold and dark until analysed. Samples were usually processed within 2 h of co l lect ion. Subsamples were analysed in the f i e ld laboratory for chlorophyll a and s i l i c a concentrations; dissolved and suspended nutrients subsamples were preserved for later analysis at the Freshwater Institute (FWI). Radioactive carbon (* 4C) uptake experiments were conducted in the f i e l d -13-Figure 4. Map of the locations sampled in Fram Sound from 30 May to 17 August, 1982. Sampling was often restricted to nearshore locations because of high winds and moving pack ice . -14-I I—' F R A M SOUND Bay of Woe • F Walrus Fiord • Z A *.B »L .MI # j • ast. »R Helena Is. «X V W O H •AA Cape Vera 10 Km lab and the concentration of C was measured in Winnipeg. Chlorophyll a One hundred and seventy-two samples were analysed for phytoplankton chlorophyll a concentrations. Aliquots were vacuum f i l te red on Whatman GF/C glass f iber f i l t e r s (4.25 cm) at 8 psi until almost dry. The f i l t e r s were folded and chlorophyll a s ta t ica l ly extracted with 10 ml of 99.5% acetone for approximately 24 h in the dark at 0-5°C. Chlorophyll was measured fluorometrically using a Turner Model III fluorometer, calibrated according to the spectrophotometric method of Stainton et al.(1977). Chlorophyll is usually extracted in 90% acetone; to determine whether there was a quantitative difference in the chlorophyll concentrations extracted by these two concentrations of acetone, experimental extractions were done. These experimental extractions did not show a signif icant difference in the extraction ef f ic iencies of these two concentrations of acetone (one-way analysis of variance, p<0.05). Volumes f i l te red were 5-6 1 at each depth on 30 May and 14 June, 2.5 1 on 14 June and 2 July, and 0.5 1 for the remaining sampling period. Subsamples (125 ml) were preserved with Lugol's solution and 10 % formalin for microscopic analysis on 2, 12 and 28 July and 2, 14 and 17 August. Nutrient Chemistry S i l i c a - 1 6 -Soluble reactive s i l icon was measured in the f i e ld laboratory according to the manual colorimetric technique of Stainton et a l . (1977). Duplicate measurements were made and the mean value reported. Nitrogen and Phosphorus Particulate nutrients (N,P) were collected on preignited (500°C for 16 h) Whatman GF/C f i l t e r s , and the same volume was f i l te red for these samples as was used for chlorophyll determinations at each locat ion. Particulate nitrogen samples were vacuum f i l te red to dryness, placed in plastic petri dishes and dessicated in the dark with a s i l i c a gel dessicant. Samples were analysed at the FWI by an automated combustion technique (Stainton et a l . 1977). Particulate, phosphorus samples were also f i l tered to dryness. The f i l t e r s were placed in 20 ml Pyrex screw cap vials previously rinsed with f i l t r a t e , and later analysed at the FWI by the method of Stainton et a l . (1977). The f i l t r a t e from the particulate phosphorus f i l t r a t i o n was retained for dissolved macronutrient analysis. Total dissolved nitrogen (TDN) and phosphorus (TDP) samples were preserved with 100 ul of 4N Ultrex sulphuric acid and later analysed according to an automated photocombustion technique (Stainton et a l . 1977). Physical Oceanography At each location, submarine l ight (Photosynthetically Active Radiation,PAR) was measured using a LICOR LI-185A quantum meter with a Lambda f la t -p late cosine-corrected quantum sensor. Surface l ight (in air) was recorded, and measurements were made at 1 m intervals to 27 m. - 1 7 -These water transparency measurements, recorded under constant sky conditions whenever possible, were taken to provide data for the calculation of the extinction coefficient (k). Light is attenuated exponentially with depth in the sea, and the intensity at any depth can be calculated from the following equation (Raymont 1980) when the extinction (or attenuation) coefficient is known: where 1^=1ight intensity at depth d IQ=surface light intensity k=extinction coefficient d=depth These data are then used to calculate the l ight intensity at different depths by a numerical primary production model (Fee 1984). Water temperature was recorded to the nearest 0.1°C using a bead thermistor probe and digi ta l multimeter at lm intervals. Conductivity samples were heated to 25°C in a water bath and conductivity measured using a Radiometer conductivity meter. Thirty samples were retained to determine sal in i ty with an Autosal Laboratory Salinometer Model 8400, using UNESCO International Oceanographic tables. Sal ini ty was l inearly regressed with conductivity and the regression 2 equation (sal=0.79*cond-6.81, r =0.97) was used to calculate sal in i ty values from conductivity measurements. -18-Temperature and sal in i ty data were used to calculate density according to the revised SCOR/UNESCO equations (International Oceanographic Tables 1972). Primary Production Photosynthesis-Light experiments Microalgal photosynthetic rates were measured with * 4C-uptake in a shore-based incubator. The laboratory was darkened for the duration of the sample preparation and incubation period. This reduced light-shock to the samples during preparation and minimized stray l ight entering the incubator. The samples were held in a water bath at in-si tu temperatures (-1°C) prior to preparation for incubation, and were inverted several times to thoroughly mix the contents before being poured into the reagent bottles. Four depths from the sampling pro f i le , usually 2, 5, 10, and 20 m, were used for the photosynthesis-light experiments. Six pairs of 60 ml glass reagent bottles were f i l l e d to overflowing with water from each depth and then injected with 0.5 ml stock solution of 30 14 uC/ml NaH C0^ in s ter i le sal ine. Five pairs of bottles were placed in each of f ive compartments in an incubator (Figs. 5,6 ;described in detail by Shearer et a l . 1985); the sixth pair was darkened with aluminum f o i l as controls. The samples were incubated for 4 h in circulating seawater maintained at -1°C. A 400 Watt Sylvania metalarc lamp illuminated the incubator; this l ight source closely approximated clear sky wavelengths in the PAR waveband (McCree 1972). Light levels in the incubator compartments were measured with the same quantum meter and sensor used for submarine light measurements. These incubator -19-Figure 5. Schematic diagram of the incubator used to measure the rate of primary production. Photosynthetically active radiation was measured in the incubator with a f la t plate cosine-corrected sensor prior to and after the 4 h incubation, and later corrected for the backscattering of l ight in the incubator, (modified from Shearer et a l . 1985). -20-GEARMOTOR Figure 6. Longitudinal and cross-sectional schematic diagram of the incubator used to measure the rate of primary production. Mounted samples rotated on plexiglass disks to mix the samples during the incubation, (modified from Shearer et a l . 1985) -22-LONGITUDINAL SECTION A-A 1 CROSS SECTION B-B 1 measurements were multiplied by empirical correction factors (1.84, 1.75, 1.66, 1.61, 1.60 from highest to lowest l ight chambers) to correct for the backscattering of l ight in the incubator (Shearer et a l . 1985). After incubation, radioactive carbon uptake was assayed on 8 ml aliquots placed in 20 ml sc in t i l l a t ion vials by the ac id i f icat ion and bubbling method (Schindler et a l . 1972). After the bubbling procedure 0.1 ml of NCS, a tissue solubi l izer , was added to each sample with 10 ml of Beckman Ready-Solv MP f luor. Samples were counted with a Beckman LS 7500 l iquid sc in t i l l a t ion counter at the Freshwater Institute. p Alpha and P m The incubator data, which consisted of the photosynthetic response of the algae at 5 l ight levels, were used to calculate two parameters for each Photosynthesis-Irradiance curve; alpha, the slope of D l ight-l imited photosynthesis, and P m, the rate of primary production at light saturation, both normalized to chlorophyll (Fig. 7). These calculations were done with a non-linear curve f i t t i ng technique developed by Fee (1984), using a modified version of the hyperbolic tangent function recommended by Jassby and Piatt (1976). From these p values, the mean alpha and P m were calculated and subsequently used in a computer program which models primary production (Fee 1984). Primary Production Model To simulate primary production, this numerical model requires alpha • and P i values, water transparency and chlorophyll measurements, and solar radiation data. Production was calculated at intervals of 30 -24-Figure 7. A Photosynthesis-Irradiance curve, showing alpha, the slope of l ight- l imited photosynthesis, I. , the irradiance at the onset of l ight saturation, and P m, tne rate of primary production at l ight saturation. Alpha and P m were simultaneously determined with a non-linear least squares computer program developed by Fee (1984). •25-minutes at 11 depths for each day. The absolute l ight value at each depth was calculated from solar radiation data and water transparency D prof i les , and used to calculate primary production from alpha and P i values according to the following equations from Fee (1984): for I<Ik/20 for Ik/20<I<2*Ik for 2*Ik<I where a=alpha P =production per unit of chlorophyll I=irradiance Ik=PBm/alpha I'=(I-Ik/20) P =0 P B =a*r ( l -a*I7(4*P B m) PB=PBm These production prof i les are then integrated by this model to yie ld _2 daily production estimates (mg Cm ). Model Components Chlorophyll a Chlorophyll data were l inearly interpolated for the depths required for the production calculat ion. If the data did not extend to the euphotic depth, I l inearly regressed each chlorophyll prof i le and used the extrapolated values down to the euphotic depth. -27-Water Transparency Daily water transparency prof i les were extrapolated by the program by l inearly regressing the l o g ^ of the normalized submarine l ight measurements from 3 m to the maximun depth sampled. Data were normalized to the daily surface (0 m) reading. Rather than using the measurement taken just below the surface of the water (which is subject to error due to scattering of l ight by waves), the 0 m value was calculated by dividing the corresponding air measurement by 1.34, the immersion correction factor for the quantum sensor. In general, l ight extinction in the top 3 m was not log l inear, and l ight measurements shallower than 3 m were not included in the regression. The slope of the regression line is called the extinction coefficient (k). The regression l ine was extended to define the depth of the euphotic zone (Zp£ypj)» defined here as the depth to which 0.5% of the surface l ight penetrated, which was calculated by: Z E U p i =log 1 0 0.005- intercept /s lope using the slope and intercept from the linear regression. Chlorophyll and water transparency prof i les were l inear ly interpolated to estimate values between sampling days, and the f i r s t and last sampling values in a l l input datasets were used to calculate production outside the sampling period. -28-Solar Radiation Incoming solar radiation (insolation) data was modelled because empirical data was unavailable due to a malfuctioning str ip chart recorder during the sampling period. Three l ight regimes were modelled for the sampling period. These regimes were used to calculate three estimates of total production as a function of insolat ion; maximum (cloudless), cloud cover corrected (based on twice daily on-site meteorological observations), and minimum (assuming continuous 10/10 cloud cover). The model is described in detail in Appendix A. Stat is t ica l Treatment of Data To test for the homogeneity of the water mass, more than one location was sampled on days when weather conditions permitted. These data (chlorophyll a and nutrients) were tested in a two-way analysis of variance (depth, location) for each sampling date. The results of this test showed there were no signif icant trends with depth or location (p > 0.05) i . e . that the water column appeared to be well mixed. For each variable measured a coefficient of variation (the standard deviation divided by the mean) was calculated for each date and averaged for the 6 replicate sampling days to indicate var iabi l i ty associated with the measurements (Table 1). Because the water column was well mixed, mean water column concentrations of the measured variables were reported. These variables were expressed as the mean + 2 standard deviations. Standard error was not reported, since sample sizes varied. The variables measured during the sampling period were tested to determine i f changes in concentration or value increased, decreased or -29-showed no change with time. This was done using the General Linear Models (GLM) procedure from the Stat ist ical Analysis System's (SAS) l ibrary, which computes a linear regression (Freund and L i t t e l l 1981). The null hypothesis was that there was no change in the concentration of chlorophyll, etc. with time. A stat is t ica l significance was recorded when the probability of rejecting the null hypothesis when i t was true was 5% ( i . e . when alpha was <= 0.05). The actual probabiliy is reported for each test . To test whether the concentrations of the nutrients measured were related to chlorophyll a concentrations, the same s ta t is t ica l procedure (GLM) was used as above. Chlorophyll was the dependant variable, and the null hypothesis was that changes in nutrient concentration had no relationship to changes in chlorophyll a concentration, or that the slope of the regression line was zero. A stepwise regression was done using the STEPREG procedure from the SAS library to determine the measured variable which accounted for the most var iabi l i ty in chlorophyll concentration. This was done to indicate a possible l imiting factor in the growth of phytoplankton. -30-Table 1. Descriptive s ta t is t ics for chlorophyll a and nutrients for days when more than one location was sampled. This was done to indicate the var iab i l i ty of measurements made at different depths at two or more locations. a) Chlorophyll a date (1982) locations sampled mean s (mg.nT td . dev. 3 ) C.V.(%) (sd/mean) S .E . n depths sampled (m) , 5 Jul E,F 1.93 0.48 24.9 0.15 10 0,2,5,10,20 20 Jul L,M 2.87 0.75 26.1 0.24 10 0,2,5,10,20 2 Aug R,S 1.54 0.21 13.6 0.06 12 0,2,5,7.5,10,15 5 Aug T,U 1.07 0.14 13.1 0.04 10 0,2,5,10,20 16 Aug X,Y 3.12 0.22 7.05 0.07 10 0,2,5,10,15 17 Aug Z.AA.BB 2.01 0.40 19.9 0.10 15 0,2,5,10,20 x =17.4 b) S i l i c a date (1982) locations sampled mean std. .dev. (umol.L l ) C.V.(%) (sd/mean) S .E . n depths sampled (m) 5 Jul E,F 19.5 0.95 4.87 0.30 10 0,2,5,10,20 20 Jul L,M 22.8 1.15 5.04 0.36 10 0,2,5,10,20 2 Aug R,S 17.7 1.88 10.6 0.54 12 0,2,5,7.5,10,15 5 Aug T,U 19.7 0.58 2.94 0.18 10 0,2,5,10,20 16 Aug X,Y 12.4 0.35 2.82 0.11 10 0,2,5,10,15 17 Aug Z,AA,BB 16.3 1.39 8.53 0.36 15 0,2,5,10,20 x = 5.80% -31-c) Total dissolved nitrogen (TDN) date (1982) locations sampled mean std. ,dev. (umol.L i ) C.V.(%) (sd/mean) S . E . n depths sampled (m) 5 Jul E,F 35.5 2.16 6.08 0.68 10 0,2,5,10,20 20 Jul UM 34.2 4.44 13.0 1.40 10 0,2,5,10,20 2 Aug R,S 34.4 5.34 15.5 1.54 12 0,2,5,7.5,10,15 5 Aug T,U 32.5 3.96 12.2 1.25 10 0,2,5,10,20 16 Aug X,Y 24.4 1.54 6.31 0.49 10 0,2,5,10,15 17 Aug Z,AA,BB 29.8 2.19 7.35 0.56 15 0,2,5,10,20 x = 10.1% d) Total dissolved phosphorus (TDP) date locations mean std. ,dev. C.V.(%) S . E . n depths sampled (1982) sampled (umol.L ) (sd/mean) (m) 5 Jul E,F, 2.18 0.08 3.67 0.02 10 0,2,5,10,20 20 Jul L,M 2.21 0.14 6.33 0.04 10 0,2,5,10,20 2 Aug R,S 2.07 0.11 5.31 0.04 12 0,2,5,7.5,10,15 5 Aug T.U 2.17 0.08 3.69 0.02 10 0,2,5,10,20 16 Aug X.Y 1.70 0.05 2.94 0.02 10 0,2,5,10,15 17 Aug Z,AA,BB 1.95 0.06 3.08 0.02 15 0,2,5,10,20 x = 4.17% 32-e) Particulate nitrogen (PN) date locations mean std.^dev. C.V.(%) S .E . n depths sampled (1982) sampled (umol.L ) (sd/mean) (m) 5 Jul E,F 1.65 0.46 27.9 0.15 10 0,2,5,10,20 20 Jul L,M 1.41 0.58 41.1 0.19 10 0,2,5,10,20 2 Aug R,S 1.33 0.38 28.6 0.11 12 0,2,5,7.5,10,15 5 Aug T,U 0.94 0.20 21.3 0.06 10 0,2,5,10,20 16 Aug X,Y 2.81 0.61 21.7 0.20 10 0,2,5,10,15 17 Aug Z,AA,BB 1.49 0.74 49.7 0.19 15 0,2,5,10,20 x = 31.7% f) particulate phosphorus (PP) date locations (1982) sampled mean (umol std. dev. •L ) C.V.(%) (sd/mean) S .E . n depths sampled (m) 5 Jul E,F 0.12 0.03 25.0 0.08 10 0,2,5,10,20 20 Jul L,M 0.13 0.04 30.1 0.01 10 0,2,5,10,20 2 Aug R,S 0.10 0.03 30.0 0.01 12 0,2,5,7.5,10,15 5 Aug T,U 0.07 0.03 42.9 0.01 10 0,2,5,10,20 16 Aug X,Y 0.20 0.02 12.2 0.01 10 0,2,5,10,15 17 Aug Z.AA.BB 0.12 0.04 30.0 0.01 15 0,2,5,10,20 X = 28.4% -33-RESULTS Chlorophyll a Chlorophyll a concentrations ranged from 0.09 to 3.70 mg.m and there was a s ta t is t ica l ly signif icant increase in chlorophyll concentration during the sampling period (p<0.0001, r =0.31). From 30 May to 22 June the mean water column chlorophyll concentration was very _3 low, 0.1+0.02 mg m (x+2sd). The concentration increased to 1.39+0.30 _3 mg m by July 5 and for the next four weeks fluctuated about a mean of _3 2.38 mg m (Fig. 8). Chlorophyll concentrations dropped in early August but peaked again sharply on August 14, reaching the maximum seasonal _3 mean concentration of 3.70+0.50 mg m . Early and late season chlorophyll concentrations were relat ively constant with depth; there was some strat i f icat ion in mid-summer, with chlorophyll maxima at or near the surface (Fig. 9). Diatoms dominated samples taken from the 25-35 % light levels , consistently accounting for more than 90 % of the number of ce l ls present. Frag i l la r ia oceanica was numerically the most abundant diatom, and members of the genera Thai losiosira , Nitzschia, Peridinium, and a group of phytoflagellates were also present. -34-Figure 8. Chlorophyll a concentrations (mg.m )in the water column from June to August, 1982. The mean concentration was calculated for each sampling day and the mean + 2 standard deviations graphed. -35-1982 Figure 9. Isopleth of chlorophyll a concentrations in the top 50 meters from June to August, 1982. There was a sl ight s t ra t i f ica t ion of chlorophyll a in mid-summer, with higher concentrations near the surface. -37-CHLOROPHYLL mg m"3 J U N E JULY AUGUST Nutrient Chemistry S i l i c a Soluble reactive s i l icon concentrations ranged from 11.6 to 24.8 umol.L - ^, with a mean concentration of 19.2+6.1 umol.L -''' (Fig. 10). From 30 May to 9 August, concentrations ranged from 17.7 to 22.8 umol.L-''" but marked changes occured from 9-17 August (Fig. 11). From 9 August to 14 August the mean s i l i c a concentration decreased from 21.9 to 11.7 umol.L -*", then increased back to 18.4 umol.L-''" by 17 August. There was a s ta t is t i ca l l y signif icant decrease in the concentration of s i l i c a with time (p<0.0001, r 2=0.24). With the exception of 2 and 17 August, s i l i c a concentrations were relat ively homogeneous with depth. There was l i t t l e vert ical s t rat i f icat ion during the sampling period. To compare relative concentrations over time, s i l i c a concentrations were normalized to the maximum concentration for each day (Si ). With x max7 the exception of 17 July 2 August and 17 August, s i l i c a concentrations were generally 90-95% of S i m a x » and the maximum concentration was usually at the maximum depth sampled (Fig. 12). Nitrogen and Phosphorus The f i r s t nitrogen and phosphorus samples were collected on 22 June. Low chlorophyll concentrations prior to and including 22 June suggest these macronutrients were also close to winter levels at this time. The overall mean concentration of total nitrogen was 35 umol.L -*" (range 26-42 umol.L - ' ' '). Total nitrogen (TN) was calculated by summing -39-Figure 10. S i l i c a concentrations (umol.L ) in the water column from June to August, 1982. The mean concentration was calculated for each sampling day and graphed as the mean + 2 standard deviations. -40-Figure 11. Isopleth of s i l i c a concentrations (umol.L ) from June to August, 1982. S i l i c a concentrations varied l i t t l e with depth or during the sampling period. -42-S I L I C A -u . r r . o l -L " 1 0 22-4 19-6 20-9 19^ 4 18-2 21^ 8 19-4 • 18-9 i • 1 r22-0 • 19-4 f — ' ( < 2 0 ) I9-7/ 22^ -6 19-4 • /20 -9 / • 2 3 5 • 2Q-3 lg-8 17-5 15-8 15-8 21-4 11-6 12-2 14-6 4=» CO 10+ E 20 + Q-° 30 40 50 + 21- 9 • 2I;9 22- 0 • 22-2 22-1 222 22-7 CD 2(J-6 — I 21-6 25-0 24-8 2 l ; 6 I I ll;6 12.4 j , - . 2I;9 I I ll;7 12-,5 14-a I M;6 / 22;l 22-3. 22-4* A 2 0 A ' H - e 134(15) is-] • I I I I I \ 20-7 20-8 I I 1 10 A U G U S T I3J /I6-2 I / / 20.0 / / I !@>! 22-0 20 J U N E 30 10 20 31 J U L Y 20 Figure 12. Isopleth of s i l i c a concentrations normalized to the maximum daily concentration. This isopleth shows that although s i l i c a concentrations did not generally change much in the top 50 meters, the maximum concentration was usually at the maximum depth sampled. -44-~I 1 1 ' 1 1 1 1 1 1 1 1 h-20 30 10 20 31 10 20 JUNE JULY AUGUST total dissolved nitrogen, about 95% of TN, and particulate nitrogen (Fig. 13), about 5% of TN (Fig. 14). There was a s t a t i s t i c a l l y signif icant decrease in the total nitrogen concentration with time (p<0.001, r =0.37). A linear regression of chlorophyll with particulate nitrogen showed there was a s ta t i s t i ca l l y signif icant increase in chlorophyll concentration with increasing particulate nitrogen concentration, and this model accounted for 74% of the variance in 2 particulate nitrogen (p<0.0001, r =0.74). The seasonal pattern of phosphorus concentrations was similar to that of nitrogen. The mean concentration of total phosphorus was 2.13+0.27 umol.L" 1 (range 1.71 to 3.26 umol.L" 1 ) (Fig. 15). Total dissolved phosphorus averaged approximately 95% of total phosphorus, and ranged from 1.68 to 3.04 u m o l . L - 1 . There was a s t a t i s t i c a l l y s igni f icant 2 decrease in total phosphorus with time (p<0.0001, r =0.39). A l inear regression of chlorophyll with particulate phosphorus showed there was a s ta t is t ica l ly signif icant increase in chlorophyll concentration with 2 increasing particulate phosphorus concentrations (p<0.0001, r=0.73) . The mean concentration of particulate phosphorus was 0.12 umol .L - 1 (F ig. 16). To test for the dependency of chlorophyll a on nutrients, a stepwise multiple regression analysis of chlorophyll a with dissolved nutrients and date was computed (Table 2). This regression showed that 39% of the variation in chlorophyll a concentration was attributable to variations in s i l i c a concentration and that date and total dissolved nitrogen (TDN) and total dissolved phosphorus (TDP) concentrations were not important to the regression. Particulate nitrogen and phosphorus -46-Figure 13. Particulate nitrogen concentrations in the water column from June to August, 1982. The mean concentration was calculated for each sampling day, and the mean + 2 standard deviations were graphed. •47-1982 Figure 14. Total nitrogen concentrations in the water column from June to August, 1982. The mean concentration was calculated for each sampling day, and the mean + 2 standard deviations were graphed. -49-1982 Figure 15. Total phosphorus concentrations in the water column from June to August, 1982 (mean + 2 standard deviations). -51-19B2 Figure 16. Particulate phosphorus concentrations in the water column from June to August, 1982 (mean + 2 standard deviations). -53-[OLUTY) N0I1VU1N30N03 SndOHdSOHd 31VinOllUVd -54-Table 2. Results of stepwise regression of chlorophyll a with dissolved nutrients and date. Changes in s i l i c a concentration accounted for 39% of the var iabi l i ty in chlorophyll a concentrations, and changes in time, TDN and TDP had l i t t l e effect on changes in chlorophyll a concentration. Step Variable R2 Overall F 1 S i l i c a 0.387 51.09 2 date 0.408 27.58 3 TDP 0.418 18.88 4 TDN 0.418 14.04 -55-were not included in the regression because of the l ike ly autocorrelation between these variables and chlorophyll a. PHYSICAL OCEANOGRAPHY Water transparency Extinction coefficients ranged from -0.0377 m -* on June 14 to -0.0908 n f 1 on August 14, with a mean of -0.0566 n f 1 (Fig. 17). There was a sl ight but s ta t i s t i ca l l y signif icant decrease (p<0.0187) in extinction coeff icients (k) (increase in turbidity) with increasing chlorophyll concentration, although this model accounted for only 30% of the variation in k. Euphotic zone The mean depth of the euphotic zone was 39 m (range 22 to 53 m) (Fig. 18), and there was a s ta t is t ica l ly signif icant decrease in the depth of 0.5% surface light penetration with time (p<0.0349, r 2=0.25). A linear regression of euphotic depths with chlorophyll concentration showed a s ta t i s t i ca l l y signif icant decrease in the depth of the euphotic 2 zone with increasing chlorophyll concentrations (p<0.0144, r =0.32). Chlorophyll concentrations accounted for only 32 % of the variation in euphotic depths; l inear regressions of particulate nitrogen and phosphorus with euphotic depth showed that while particulate nitrogen concentrations were signif icant ly related to euphotic depth (p<0.0414, 2 r =0.30), chlorophyll concentrations accounted for s l ight ly more variat ion. Particulate phosphorus concentrations were not s igni f icant ly (p<0.05) related to euphotic depth. -56-Figure 17. The change in extinction coefficients from June to August, 1982. -57-i i i r—i i I I I | I I I I I I I i I | i — i — i — i — i — i — i — i — i — | — i — i — i i — i — i — i — i — i — | — r — l — i — i — i — i — i — i — i — | — i — i — i — i — i — i — i — • — i — | — i — i — i — i — i i—i—i—i—p 10 JUNE 20 JUNE 30 JUNE 10 JULY 20 JULY 30 JULY 10 AUG 20 AUG 1982 Figure 18. The euphotic depth, defined as the depth to which 0.5% of the surface l ight penetrates, from June to August, 1982. This depth was greatest (54 m) early in the season when the chlorophyll a concentrations were low, and was 22 m on 14 August during the second, and largest, phytoplankton bloom. -59-1982 Temperature, Sal in i ty and Density Water temperature and sa l in i ty showed no trends; they varied l i t t l e with depth or time. The mean water column temperature was -1 .0+0 .6°C , and ranged from -1.9 to 0 .1°C. The mean sal in i ty and density of the water column from 0 to 20 meters was 32.05+1.39 ppt and 25.76+1.11 kg.m respectively. PRIMARY PRODUCTION p Alpha and P m Al l the incubator data from the light saturation experiments were • used to calculate alpha and P m (Table 3). Alpha, the slope of l ight- l imited photosynthesis, ranged from 3.04 _2 to 25.94 mg C/(mg chl.Einstein.m ) during the sampling period, and the _2 overall mean was 8.54 mg C/(mg chl.E.m )(Fig. 19). A frequency diagram of these values showed that 90% were between 3 and 14 mgC/(mg -2 B chl.Einstein.m ) (Fig. 20). P m, the maximum rate of photosynthesis per unit of chlorophyll , ranged from 1.44 to 4.01 mg C/(mg chl.h - *"); the mean PBm was 2.73 mg C/(mg c h l . h - 1 ) (Figs. 21,22). p The mean values of alpha and P m were used in primary production calculations because linear regressions of these parameters with depth and time showed no s ta t is t i ca l l y signif icant depth or seasonal trends (p<0.05). Primary Productivity Primary productivity was calculated for the three l ight regimes modelled (Appendix A) from 1 June to 17 August, 1982 (Table 4). -61-Table 3. Descriptive stat is t ics for alpha and P m. Parameter Mean Std dev S .E . Median n alpha 8.54 5.56 0.98 6.70 32 PBm 2.73 0.71 0.13 2.56 32 -62-Figure 19. Alpha values (mg C/(mg Chl.E .m )) obtained from photosynthesis-light experiments. 63-ALPHA (mg C / (mg chl Einstein m - 2 ) ) UJ i — • o o p D _L (0 o _1_ ro D I • • cn 03 O D I ' l l I I I I I I—I—1—I—I ro D _ i «-. C r -<: 00 ro u ] c r oo OJ D <-, c r -< <X> O > c o o o o ro D > c cn Figure 20. A frequency diagram of alpha values. -65-Figure 21. P m values (mg C/(mg chl.h )) obtained from photosynthesis-light experiments. - 6 7 -o 3. 5CH 1 3. 00-_c IE o O E 2. 50H O o 2 2. ooH m CL 1. 50-i . ooH 1 — r -10 JULY — i — | — r — 20 JULY o o o o o o 0 o o -I 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 p 30 JULY 10 AUG 20 AUG 1982 Figure 22. Frequency diagram of P m values. -69-Early season productivity averaged approximately 0.09 g -2 -1 Cm .day . With cloud-corrected solar radiation, the two peaks of -2 -1 productivity during the sampling period were 2.5 g Cm .day on 20 -2 -1 July and 2.3 g Cm .day on 13 August. The plots of production vs time using the three l ight regimes (Fig. 23) show the highest peak of productivity occured on 20 July . I Figure 23. Integrated primary production (mg Cm ) calculated with maximum, cloud-cover corrected, and minimum solar radiation values from June to August, 1982. 72-g C / m 2 riPIXPROD g C / m 2 CCPROD g C / m 2 M I N P R O D 1 9 8 2 Table 4: Primary productivity calculated from 1 June to 17 August, 1982, using maximum, cloud-corrected, and minimum solar radiation -2 B values. (alpha=8.54 mg C/(mg chl.Einstein.m , P m=2.73 mg C /(mg c h l . h " 1 ) . g C m " 2 Solar Radiation Maximum(cloudless) Cloud-corrected Minimum(10/10 overcast) 106 77.9 51.5 -74-DISCUSSION Chlorophyll a From the end of May until late June, the concentration of chlorophyll was low (0.09 mg chl m ) and distributed evenly in the top 50 m. Early season concentrations were within the range of winter values at Resolute Bay, 0.001 to 0.1 mg chl.m" 3 (Welch and Kalff 1975) and somewhat lower than those at Frobisher Bay in mid-June 1963, 0.3 mg ch l . m"3(Grainger 1979). Chlorophyll concentrations peaked twice, with the f i r s t peak occuring from 20-30 July and the second peak, larger than the f i r s t and of much shorter duration, occuring between 9-15 August. The overall -2 -3 mean of 72 mg chl m (1.85 mg chl.m ) is comparable to values reported for northern regions such as Lancaster Sound (Borstad and Gower 1985) Baffin Bay (Harrison et a l . 1982) and Resolute Bay (Welch and Kalff 1975). Nutrients The concentrations of soluble reactive s i l i c a , total dissolved nitrogen (TDN) and total dissolved phosphorus (TDP) were highest early in the season (late June) when chlorophyll concentrations were low (0.09 _3 mg chl.m ). There were s ta t is t ica l ly signif icant decreases in the -75-concentrations of these three nutrients during the summer and a s ta t is t ica l ly signif icant increase in chlorophyll concentration. • This suggests that nutrients were being consumed by phytoplankton and were appearing as particulate nitrogen and phosphorus, which were posit ively correlated with chlorophyll concentration. The stepwise regression showed that the depletion of s i l i c a was greater than the depletion of TDN and TDP, but i t is unlikely that s i l i c a concentrations were ever l imit ing, since the largest change in the concentration of s i l i c a was during the second phytoplankton bloom, and i t only decreased by half during that time. The low nutrient concentrations which develop in surface waters during or after bloom conditions in such areas as Frobisher Bay (Grainger 1975) and Baffin Bay (Harrison et a l . 1982) did not occur in Fram Sound. This suggests that nutrients were being injected into the system, probably by the upwelling of deeper water. Vertical prof i les of temperature, sa l in i ty and nitrogen, phosphorus and s i l i c a concentrations indicate that the water column was well-mixed for the entire sampling period. S i l i c a Early season s i l i c a concentrations were about 20 u m o l . L - 1 , s l ight ly lower than the spring values of 25 to 30 umol .L - 1 recorded in Jones Sound from 1961-1963 (Apollonio 1976b). The most geographically comparable s i l i c a data are from an Arctic cruise in August and September, 1977; a mean s i l i c a concentration in Fram Sound of 22 umol .L - 1 from 50 to 200 m with vertical strat i f icat ion in the top 50 m ranging from 10.8 umol .L - 1 near the surface to 20 umol .L - 1 at 50 m was -76-reported (Jones and Coote 1980). The lowest s i l i c a concentrations I measured, 11.7 and 12.6 umol .L - 1 , occured on 14 August and 16 August respectively during the second phytoplankton bloom. These concentrations are s t i l l considerably higher than other workers have measured in late summer, for example 2-5.2 umol .L - 1 in the top 10 m in late July in Jones Sound (Apollonio 1976b), and 2.87 umol .L - 1 in the Baffin Bay mixed-layer in late August to mid-September (Harrison et a l . 1982). These data indicate that s i l i c a is probably re-supplied by local upwelling conditions in Fram Sound. Slight vertical s t ra t i f ic ta ion was evident on the last sampling day (17 August 1982). The data of Jones and Coote (1980) suggests that st rat i f icat ion may occur, although s i l i c a depletion probably does not proceed in Fram Sound to the low concentrations of other Arctic areas sampled to date. Nitrogen and Phosphorus Total nitrogen concentrations ranged from 26 to 42 u m o l . L - 1 . Because most nitrogen measurements in Arctic waters are of nitrate (N0^ ) and ammonia (NH^+), comparisons with geographically similar nitrogen data are not possible. The only available nitrate data for Fram Sound was measured in August and September 1977, and two locations were sampled. Nitrate concentrations were somewhat s t ra t i f i ed , with surface concentrations of 10 and 14 umol .L - 1 respectively (Jones and Coote 1980). These concentrations were comparable to those reported for Lancaster Sound (12 umol .L - 1 , Jones and Coote 1980) and considerably higher than those found in the Baffin Bay mixed layer (0.15 u m o l . L - 1 , Harrison et a l . 1982). -77-Particulate nitrogen and chlorophyll concentrations showed similar trends during the season. A linear regression of these two variables showed chlorophyll accounted for 74% of the variance in particulate nitrogen concentrations suggesting that particulate nitrogen can serve as an index of phytoplankton biomass in the absence of chlorophyll data. If the same water mass was sampled for the entire sampling period the total nitrogen concentrations should remain essential ly constant, but the concentration of total nitrogen decreased during this time. This is l ike ly due to either depletion of particulate nitrogen by sinking or zooplankton grazing, or to water masses passing the sampling area with different concentrations of total nitrogen. Phosphorus, l ike nitrogen, was measured as total dissolved phosphorus (TDP) and particulate phosphorus (PP); measurements of phosphate (PO^ ) cannot be extracted from TDP and PP values without additional data. Late summer phosphate data from Fram Sound showed that phosphate was relat ively homogeneous to 150 m with a mean concentration of 1.6 umol .L - 1 (Jones and Coote 1980). This concentration is comparable to late summer Baffin Bay mixed layers (1.2 u m o l . L - 1 , Harrison et a l . 1982) and Lancaster Sound (1.7 umol .L - 1 , Jones and Coote 1980). PHYSICAL OCEANOGRAPHY The euphotic zone is defined as the depth to which 1.0% or 0.5% of the surface light penetrates. The mean depth of the euphotic zone during the sampling period was 39 m (defined using 0.5%), which is comparable to the depth of 34 m for Baffin Bay in late summer (Harrison -78-et a l . 1982). The l inear regression of euphotic depth with chlorophyll a showed a s t a t i s t i c a l l y signif icant inverse relat ionship, although chlorophyll accounted for only 32% of the variance. This suggests that other factors are affecting water transparency, such as the seasonal bloom of zooplankton associated with changes in chlorophyll concentration, or that the 'noise' associated with low chlorophyll measurements obscured the relat ionship. The temperature and sa l in i ty data indicated that the water column was well mixed to the maximum depth sampled throughout the sampling period. This is consistent with the vertical prof i les of nutrient and chlorophyll data, which were also homogeneously distr ibuted. The low sal ini ty surface layer reported to develop in other Arct ic areas was not observed in Fram Sound. The water from Norwegian Bay was well mixed by the action of the set and t idal currents in Hell Gate, Cardigan Strai t and Fram Sound, and this vert ical instabi l i ty due to mechanical forces prevented the formation of a low sal in i ty surface of melt-water. PRIMARY PRODUCTION Primary production and phytoplankton biomass data are sparse for the eastern Canadian Arct ic (Harrison et a l . 1982, Borstad and Gower 1985, Welch and Kalff 1975, Apollonio 1976b). In Baffin Bay, phytoplankton biomass (chlorophyll a) and primary production rates , -2 -3 -2 -1 averaged 57 mg chl.m (1.26 mg.m ) and 227 mg Cm .d during the summer of 1978; large chlorophyll maxima (about 6 times greater than surface concentrations) were consistantly measured at or near the bottom of the euphotic zone (Harrison et a l . 1982). Although nitrate -79-concentrations were low in the euphotic zone, Harrison et a l . (1982) detected no apparent signs that nutrients limited production and concluded that nutrient concentrations may not be as important as previously believed in controll ing Arctic primary production (Dunbar 1968). Strong subsurface chlorophyll maxima were also present in Jones Sound just below the pycnocline in August, 1979, where chlorophyll was -3 -3 up to 18 mg.m and the average concentration was 1.97 mg chl.m (69 mg _p chl.m ) (Borstad and Gower 1985). Phytoplankton biomass reached a -2 -3 maximum concentration of 100 mg chl.m (15 mg.m ) in Resolute Bay during August 1972, and the total annual production was estimated at 45 -2 -1 g Cm .yr (Welch and Kalff 1975). The only winter chlorophyll measurements to date in the Lancaster Sound area were also taken at Resolute Bay and were below 0.1 mg chl.m (Welch and Kalff 1975). In Frobisher Bay, about 1500 km southeast of Resolute Bay, total annual -2 -1 production was approximately 40 and 70 g Cm .yr in 1968 and 1969; the difference in production between these two years was associated with seasonal differences in sea ice cover and hence l ight ava i lab i l i ty (Grainger 1975). Chlorophyll concentrations were somewhat higher in Frobisher Bay than Resolute Bay in August, ranging from 10-200 mg _2 chl.m in 1969, and nitrate, which became depleted in August, was thought to l imit primary production at that time (Grainger 1975). The average production rate during the sampling period of the present study, calculated with cloud-cover corrected insolat ion, was 998 -2 -1 -3 -1 mg Cm .d (26 mg Cm .d ). This is a considerably higher rate of production than reported by other workers for Arct ic waters, although there are no comparable primary production rates available for Arct ic -80-polynyas (Dunbar 1981). I calculated primary production rates for maximum and minimum solar radiation to provide a range of rates relating to insolat ion. These -2 -1 -2 -1 rates were 1.5 g Cm .d (maximum) and 0.76 g Cm .d (minimum), and the cloud-cover corrected and minimum solar radiation production rates were 72 % and 45% of the maximum insolation respectively. These calculations assume that the slope of l ight- l imited photosynthesis (alpha) on a P-I curve does not change with changes in ambient l ight and that nutrients are not l imit ing, i .e . that l ight is the only l imit ing factor. The production rates reported by other workers for Arct ic waters were determined by a variety of techniques; comparisons between photosynthetic rates measured by in-s i tu incubations or shore-based incubations with different incubator designs, using 1 4C-uptake or oxygen evolution methods, are probably not entirely va l id . Recent preliminary work comparing two incubator designs yielded somewhat different photosynthetic rates (Welch et a l . unpub. data). The high rate of production I calculated for Fram Sound was largely due to the magnitude of alpha and P m, which were 8.54 mg C/(mg -2 -1 chl.E.m ) and 2.73 mg C/(mg chl.h ) respectively. The only available D comparative alpha and P m data for Arctic waters, 2.97 mg C/(mg -2 -1 chl.E.m ) and 1.22 mg C/(mg chl.h ), are considerably lower than these values (Piatt et a l . 1982). These measurements were made in Baffin Bay and the water samples were taken from the 50% light leve l . These data are not s t r ic t l y comparable with those of the present study, since different incubator designs and incubation techniques were used. I -81-incubated 60 ml samples on rotating clear acryl ic disks at 5 l ight levels, whereas Piatt et a l . (1982) s ta t ica l ly incubated 1 ml samples at 50 light levels . The effect of different incubation techniques on the P outcome of alpha and P m is currently being investigated (Brian Irwin, pers comm). The lowest values of alpha and P m measured in Fram Sound, 3.04 mg -2 -1 C/ (mg chl.E.m ) and 1.44 mg C/(mg chl h ), were higher than the mean values from the 50 % l ight level in Baffin Bay, which were 2.97 mg C/(mg chl.E.m" 2) and 1.22 mg C/(mg chl .h" 1 ) (P iat t et a l . 1982). The difference between the alpha values in the two locations could be due to physiological differences between the two algal populations, differences in experimental design and techique, or a combination of both. A high alpha value indicates the phytoplankton have a high 'photosynthetic effeciency' . Further work must be done to determine i f phytoplankton in such a turbulant system as Hell Gate-Cardigan Stra i t Polynya and Fram Sound are more photosynthetically ef f ic ient than those from more ver t ica l ly stable locations. Conclusion Fram Sound is part of the Hell Gate-Cardigan Strai t polynya, an area kept relat ively ice-free year round by the combined effects of strong set and t idal currents and high winds. The results of the present study show the water column was well-mixed in this area; p nutrients were continuously supplied and alpha and P m values were higher than in less turbulant areas, where nutrients become depleted in the stable surface layer after a phytoplankton bloom. Chlorophyll a -82-concentrations were generally homogeneously distributed in Fram Sound, whereas in other Arct ic areas subsurface chlorophyll maxima usually develop. The high primary productivity measured in the study area suggests that secondary productivity may also be higher here and/or downstream of this and other polynyas in summer. -83-LITERATURE CITED Apollonio, S. 1976b. Primary production in the Canadian Arct ic Archipelago, 1961-1963. Unpublished manuscript. Bigelow Laboratory for Ocean Sciences. Contribution No. 76016. 46 p. Beaufort Sea Hydrocarbon Production and Transportation Proposal. 1984. Report to the Environmental Assessment Panel. Minister of Supply and Services Canada. Borstad, G.A. and J .F .R . Gower. 1985. Phytoplankton chlorophyll distr ibution in the eastern Canadian Arct ic . Arctic 37(3):224-233. Dahlgren, L. 1974. Solar radiation climate near sea level in the Canadian Arct ic Archipelago (Arctic Institute of North America, Devon Island Expedition 1961-1962). Ph.D. thesis, Uppsala University, Sweden. Dunbar, M. J . 1968. Ecological developements in polar regions; a study in evolution. Prentice-Hall,Inc..Englewood C l i f f s , N . J . 119 p. Dunbar, M. J . 1981. Physical causes and biological significance of polynyas and other open water in sea ice . | n : I. S t i r l ing and H. Cleator (ed.) Polynyas in the Canadian Arc t ic , Occasional paper 45, Canadian Wildl i fe Service. Fee, E . J . 1984. Freshwater Institute Primary Peoduction Model User's Guide. Can. Tech. Rep. F ish. Aquat. S c i . 1328. 36 p. Freund, R.J . and R.C. L i t t e l l . 1981. SAS for linear models: a guide to the ANOVA and GLM procedures. SAS Institute Inc. Grainger, E.H. 1975. A marine ecology study in Frobisher Bay, Arct ic Canada. In: Cameron, T.W.M. and L.W. Bi l l ingsley (ed.) Energy Flow-Its Biological Dimensions, A Summary of the IBP in Canada 1964-1974. 338 p. Grainger, E.H. 1979. Primary production in Frobisher Bay, Arct ic Canada. In: M.J. Dunbar (ed.) Marine Production Mechanisms International Biological Programme 20. Harrison, W.G., T. Piat t , and B. Irwin. 1982. Primary production and nutrient assimilation by natural phytoplankton populations of the eastern Canadian Arc t ic . Can. Journ. F ish. Aquat. S c i . 39:335-345. International Oceanographic Tables Volume 1. 1972. Published joint ly by the National Institute of Oceanography of Graeat Brit ian and the United Nations Educational, Sc ient i f ic and Cultural Organization (UNESCO). Jassby, A . D . , and T. Piatt . 1976. Mathematical Formulation of the relationship between photosynthesis and l ight for phytoplankton. -83-Limnology and Oceanography 24(4):540-547. Jerlov, N.G. and E. Steemann Nielsen (ed). 1974. Optical Aspects of Oceanography. Academic Press, London and New York. Jones, E .P . , and A.R. Coote. 1980. Nutrient distr ibutions in the Canadian Archipelago: indicators of summer water mass and flow character ict ics. Can. Journ. F ish. Aquat. S c i . 37:589-599. Le Grand, Y. 1968. Light, Colour and Vis ion. Chapman and Hall Ltd. London. Second Edit ion. 564 p. Luning, K. 1981. Light, p 326-355. In: The Biology of Seaweeds. C.S. Lobban and and M.J. Wynne (ed). The University of Cal i fornia Press, Berkeley. (Botanical Monographs-Vol 17.) McCree, K .J . 1972. Test of current definit ions of photosynthetically active radiation against leaf photosynthesis data. Agr ic . Meteor. 10:443-453. Milne, A.R. and B.D. Smiley. 1978. Offshore Dr i l l ing in Lancaster Sound: Possible Environmental Hazards. Dept. of Supply and Services Canada. Parsons, T .R. , M. Takahashi, and B. Hargrave. 1977. Biological Oceanographic Processes. Pergammon Press, 332 p. Piatt , T . , W.G. Harrison, B. Irwin, E.P. Home, and C L . Gallegos. 1982. Photosynthesis and photoadaptation of marine phytoplankton in the Arc t ic . Deep Sea Research 29:1159-1170. Raymont, J . E . G . 1980. Plankton and productivity in the oceans. Volume 1- Phytoplankton. Second edit ion. Pergammon Press. 489 p. Sail ing Directions, Arct ic Canada. 1982. Vol . 1, Third edit ion. Dept. Fisheries and Oceans. Supply and Services Canada. Schindler, D.W., R.V. Schmidt, and R.A. Reid. 1972. Acid i f icat ion and bubbling as an a l ternat ive . , to f i l t r a t i o n in determining phytoplankton production by the C method. J . F ish. Res. Board Can. 29:1627-1631. Shearer, J . A . , E.R. DeBruyn, D.R. DeClerq, D.W. Schindler, and E . J . Fee. 1985. Manual of Phytoplankton Primary Production Methodology. Can Tech. Rep. F ish. Aquat. Sci.1341. Smith, M., and B. Rigby. 1981. Distribution of polynyas in the Canadian Arctic. In I. S t i r l ing and H. Cleator (ed.) Polynyas in the Canadian Arc t ic , Occas pap. 45, Canadian Wildl i fe Service. Stainton, M.P., M.J. Capel, and F . A . J . Armstrong. 1977. The Chemical Analysis of Fresh Water, Second Edit ion. Can. F ish. Mar. Sery. Misc. -84-Spec. Publ. 25, 180 p. Steeman Nielsen, E. 1952. The use of radiocactive carbon (C-14) for measuring organic production in the sea. J . Cons. Explor. Mer 18:117-140. S t i r l i n g , I. 1981. The biological importance of polynyas in the Canadian Arc t i c . Arct ic 33:303-315. Szeicz, G. 1974. Solar radiation for plant growth. J . Applied Ecol . 11:617-636. Taylor, A. 1964. Geographical Discovery and Exploration in the Queen Elizabeth Islands. Memoir 3, Geographical Branch, Mines and Technical Survey, Ottawa. 172 p. Topham, D. R. ,R.G. Perkin, S.D. Smith, R.J . Anderson, and G. Den Hartog. 1983. An investigation of a polynya in the Canadian Archipelago,I, Introduction and Oceanography. Journal of Geophysical Research. Vol.88, No. C5, pp 2888-2899. Vollenweider, R.A. 1969. [Ed] A Manual on methods for measuring primary production in aquatic environments. IBP Handbook No.12, Blackwell Sc ient i f ic Publications, Oxford and Edinburgh. 213 p. Vowinckel, E. and S. Orvig. 1962. Relation between solar radiation income and cloud type in the Arc t ic . Jour. Appl. Meteor. 1:552-559. Welch, H .E . , and J . Kal f f . 1975. Marine metabolism at Resolute Bay, N.W.T.. p 69-75.In Proceedings of the Circumpolar conference on Northern Ecology. National Research Council , Ottawa. -85-APPENDIX A SOLAR RADIATION MODELS Dahlgren (1974) made numerous measurements and observations of the solar radiation climate at Truelove Lowland (Devon Base Camp), Devon Island, N.W.T. (75°40'N lat . 84°32'W long.) in 1961 and 1962. From these measurements and observations he derived the atmospheric coefficients necessary to calculate normal values of global radiation with a clear sky. He used direct solar radiation measurements made under cloudless, cloudy (less than 10/10 cover), and dense overcast conditions to derive empirical cloud cover correction factors and to ver i f iy and supplement the insolation (incoming solar radiation) values computed for different sky conditions. I used these data and meteorological (cloud cover) observations recorded from May to August, 1982 to calculate insolation at Cape Vera (76°14'N l a t . 9 8 ° 131W long.) under cloudless, cloudy and dense overcast conditions. These three datasets were then used in the primary production model to calculate primary production. Assumptions of the Model Dahlgren's data were collected twenty years ago, but the proximity of Truelove Lowland to Cape Vera and the geographical s imi lar i t ies between these locations rendered these data most suitable for estimating -86-insolation at Cape Vera (Fig. A . l ) . To use his data, I made the following assumptions: 1) Dahlgren's data represent ' typ ica l ' meteorological conditions at Truelove Lowland in 1961 and 1962, 2) the atmospheric conditions have not changed signif icantly in the High Arct ic since 1961/62, and 3) the summer of 1982 at Cape Vera was ' typ ica l ' meteorologically. Cape Vera is approximately 5° longitude west of Truelove Lowland, which means local noon (the time when the sun is highest in the sky) occurs about 20 minutes later at this location than at Truelove Lowland. This difference was considered negligible and was not corrected for in the model. Dahlgren's data col lect ion and treatment I wil l br ief ly describe the aspects of Dahlgren's work pertinent to the solar radiation model, followed by how I used these data to estimate insolation at Cape Vera. Dahlgren measured instantaneous clear sky radiation simultaneously with f i l t e r measurements made with an Angstrom pyrheliometer to determine the transmission properties of the a i r . He measured these properties, Angstrom's turbidity coeff ic ient , Linke's turbidity factor and the amount of precipitable water in the atmosphere, at different times of the day from February to October to characterize the atmospheric conditions on a daily and seasonal basis. Knowing these values, and having data for the intensity of direct solar radiation in an ideal atmosphere (an atmosphere without water vapor and aerosol particles) from other workers, Dahlgren calculated direct , clear sky -87-Figure A . l . Map of Jones Sound, showing the proximity of Truelove Lowland, where Dahlgren (1974) measured solar radiation, with Cape Vera. -8 8-insolation as a function of solar al t i tude. He also calculated diffuse radiation as a function of solar altitude and turbidi ty, both with high albedo (snow-covered ground) and with a lower albedo (ground bare of snow). Global radiation (g^), which consists of direct and diffuse radiation, was calculated from the following equation: gQ=I sin h + d -2 -1 where g^instantaneous global clear sky radiation (meal cm min ) I i n t e n s i t y of direct solar radiation h =solar altitude d =diffuse radiation From these data, Dahlgren constructed monthly curves of global radiation as a function of solar altitude from February to October. He also made 470 measurements of the global clear sky radiation during these months, and they were usually in close agreement with his calculated values. These curves were necessary in order to calculate the total amount of global clear sky radiation received at this location each day. Dahlgren computed daily totals of global clear sky insolation for every f i f t h day from 28 January to 14 November, during which period the sun was above the horizon for part or a l l of the day. This was done by calculating the solar altitude for even hours before and after local noon and, from the monthly curves of g n versus solar a l t i tude, -90-determining the global clear sky insolation at that time. These values were graphed; they defined the shape of the diurnal curve. Daily -2 -1 insolation with a clear sky (G Q ,cal.cm . day ) was then calculated by graphic integration (Table A . l ) . Cloud cover Dahlgren also measured instantaneous solar radiation with a dense overcast (10/10 dense cloud cover), denoted g ^ Q , in which case the global radiation consisted only of diffuse radiat ion. He made these measurements with different cloud types and calculated relat ive inso la t ion ,g^ /gg , as a function of time and cloud type. From these values he developed curves relating the effect of cloud type and month on the amount of insolation recieved and also meaned a l l the cloud type values for each month to estimate a monthly mean value of relat ive insolation for a l l cloud types (Fig. A.2) . Estimation of insolation at Cape Vera The purpose of estimating incoming solar radiation at Cape Vera was to provide mean half-hourly instantaneous insolation values from 1 June to 17 August. These data were then input to a primary production model (Fee, 1984) to calculate primary productivity at the south end of the Hell Gate-Cardigan Strai t Polynya during this time. Clear sky model The following steps were taken to compute the clear sky insolat ion: 1) The data from Table 1 were l inearly interpolated to provide insolation values from 00°° to 12 0 0 h for the missing days. -91-"Bible A J . Global radiation with a clear sky at actual solar distance. a global radiation (mcal/cm min) with a clear sky 0 Q = daily insolation (ly/day) with a clear sky (daily total of global radiation) D a daily insolation (ly/day) from a clear sky (daily total of diffuse radiation) Daily insolation Hour angle a g Q mcal/cm min Sun on g Q =» 0 GQ D D/GQ True solar horizon time -12 00 01 02 03 04 05 06 07 08 09 10 11 12 h m h m ly/day ly/day % Jan 28 0 Feb 8 14 8 10 25 17 15 49 43 20 86 77 25 137 122 Mar 1 166 152 5 203 188 10 261 246 15 316 301 20 373 358 25 427 415 Apr 1 506 489 5 546 532 10 595 583 15 639 628 20 683 669 25 722 709 May 1 769 754 5 800 788 10 836 821 15 865 850 20 893 876 25 915 900 <Jun 1 910 896 5 919 907 10 913 905 15 924 911 21 925 913 25 923 911 5 20 4 49 14 1 86 43 7 115 64 18 1 146 88 31 5 203 137 64 16 258 192 111 36 7 314 251 163 73 18 371 305 218 121 39 443 376 284 184 38 487 412 321 219 116 541 471 3eo 272 165 588 522 430 326 214 630 567 480 374 261 671 610 529 422 312 715 652 570 470 358 746 682 595 494 391 782 717 629 532 425 811 746 622 563 460 839 774 691 591 491 860 799 713 618. 514 859 798 722 625 526 869 809 731 637 538 367 306 726 634 535 873 812 733 638 543 375 314 736 643 547 373 812 734 642 545 8 26 4 45 11 78 26 6 113 50 17 5 153 77 32 12 5 204 1 1 4 54 26 14 255 161 93 53 32 285 192 124 77 51 323 231 158 107 76 358 264 187 133 105 386 291 215 160 128 412 319 242 186 150 427 339 259 205 169 440 350 273 218 184 439 355 281 227 191 44"? 360 288 234 198 451 366 291 236 203 453 365 290 235 202 0 00 0 00 2 46 1 12 2 59 2 16 \3 36 3 01 4 10 3 40 4 40 4 08 5 06 4 35 5 31 5 07 6 01 5 38 6 32 6 09 7 04 6 40 7 37 7 24 8 28 7 52 8 59 8 28 9 46 9 06 10 57 9 52 11 02 28 43 67 94 118 139 160 172 130 133 192 0 0 -2.0 2.0 100 4.1 4.0 98 10.7 7.7 72 22.0 12.8 58 38.9 19.2 49 51.9 25.2 49 .66.9 31.2 47 95.9 39.5 41 127.5 49.0 38 163.9 58.8 36 202.8 68.5 34 257.0 83.0 32 289.6 91.4 31 338.2 101.9 .30 385.5 112.6 29 432.6 123.7 29 484.5 134.8 28 542.1 147.1 27 581.5 154.0 27 630.8 163.6 26 674.8 170.4 25 713.9 176.5 25 748.5 181.8 24 ^763.3 160.8 21 700.4 162.3 21 784.5 136.6 17 796.0 137.3 17 798.6 137.3 17 797-4 137.3 17 Table A. I. (continued) telly Insolation Hour angle » Sun on g Q • 0 C D J/0 True solar aorlaon time -12 00 01 02 03 04 05 06 07 08 09 10 11 12 h D b m ly/day ly/day % J u l 1 903 894 853 793 714 624 527 431 349 277 224 190 179 769.4 133.3 17 5 882 868 834 774 694 605 509 415 335 264 212 180 168 745.5 129.0 17 10 868 855 820 759 680 588 495 401 319 249 197 162 153 723.8 127.0 18 15 854 840 802 741 665 572 476 382 299 229 178 146 135 698.6 125.2 18 20 835 822 783 723 645 552 456 363 282 212 161 127 116 671.6 122.2 18 25 812 797 759 699 621 528 431 340 258 188 136 107 98 638.1 119.8 19 Aug 1 777 764 724 663 583 488 390 300 218 148 102 74 66 584.8 116.5 20 5 756 742 704 638 560 463 365 274 192 128 82 54 51 552.4 116.0 21 10 722 706 666 602 520 426 333 239 161 97 54 33 29 505.0 109.0 22 15 684 669 630 566 487 389 298 207 129 68 33 20 16 461.2 102.1 22 20 641 628 589 525 444 350 255 168 96 42 19 7 5 10 24 413.5 94.0 23 25 600 584 544 480 399 309 216 132 61 23 5 1 0 9 30 366.0 85.8 23 Sep 1 540 526 489 420 343 250 158 78 27 7 8 32 9 54 307.5 73.6 24 5 505 491 450 388 305 216 124 51 17 2 8 06 9 15 275.2 66.1 24 10 456 442 402 342 259 171 87 30 6 7 31 8 34 235.8 60.1 26 15 403 390 354 293 213 125 51 13 6 59 7 58 197.3 52.9 27 20 355 344 301 242 164 85 30 4 6 28 7 24 161.8 46.7 29 25 301 288 250 190 118 51 12 5 57 7 03 127.5 40.4 32 Oct 1 244 234 194 138 73 25 3 5 23 6 16 94.6 33.4 35 5 211 197 158 106 48 12 4 56 5 50 75.1 28.4 38 10 156 143 110 63 25 4 4 24 5 20 50.7 22.3 44 15 111 102 71 35 11 3 50 4 50 33.0 16.9 51 20 74 64 42 19 2 3 14 4 15 19.7 12.4 63 25 44 38 23 7 2 32 3 47 10.6 8.4 79 30 24 20 9 1 1 36 3 12 4.9 4.6 94 Nov 3 14 8 2 0 00 2 46 2.0 2.0 100 14 0 0 00 0 0 -Figure A . 2 . Curves relating the effect of cloud type and month on the amount of incoming solar radiation received at Truelove Lowland, N.W.T.. 94-II m i v v v i v u v i i i i x x 2) A program from the IMSL l ibrary , IQHSCU, computed the coeff icients for a set of cubic polynomials, which were then used by another program from this l ibrary , DCSQDU, to interpolate between the hourly data points to calculate half-hourly radiation values. 3) The area between these half-hourly points was then integrated (with DCSQDU) and divided by 30 (min) to calculate the mean half-hour -2 -1 radiation value in meal.cm .min , resulting in insolation values corresponding to 1 2 1 5 , 1 2 4 5 , 1 3 1 5 e t c . . 4) A program,DCSQDU, was then used to integrate the total area under the curve defined by the half-hour values to compare with Dahlgren's calculated G Q values. The result of the integration was divided by -2 -2 1000 (to convert meal.cm to cal.cm , which is a langley) and multiplied by 2 to arrive at a daily to ta l . The integrated insolation values computed by the above method agreed within 1% of Dahlgren's calculated G Q values. 5) These mean half-hourly data were then multiplied by 0.0956 to -2 -1 -2 -1 convert the data from meal.cm .min to mE.m .min resulting in Photosynthetically Active Radiation (PAR) data (see Appendix B). 6) The sun was above the horizon continuously from 1 June to 17 August at this location, and the solar altitude from 24 0 0 to 12 0 0 h is the mirror image of the solar altitude from 12 0 0 to 2 4 0 0 h during this time. The shape of each diurnal curve was thus defined by -96-graphing half-hourly insolation values from 24 to 12 h followed by the same values graphed from 12 0 0 to 2 4 0 0 h, assuming an unobstructed horizon. The result of steps 1 to 6 provided global radiation data for a continuously clear sky from June 1 to August 17, the maximum amount of insolation possible during this time. To estimate insolation under cloudy and overcast sky conditions, two more datasets were computed. Cloud cover corrected model Cloud cover observations were made at Cape Vera as part of the twice daily aviation weather reports transmitted to Resolute Bay. These observations consisted of cloud type and the amount of sky covered by each type, and were made by three observers during the summer. They are the only cloud data available for June 1 to August 17 1982, since Landsat and NOAA sate l l i te imagery was unusable for this purpose. I used Dahlgren1s monthly mean relative insolation ( g 1 0 / g 0 ) values to calculate the relative insolation from these twice daily (07°° and 19 0 0h) cloud observations as follows: 1) The monthly values of g^ /gg were graphed, manually curve-f i t ted and a cloud cover correction factor (cccf) recorded for each day from June to August. 2) The total cloud cover (cc) from each f i e ld observation was calculated by summing the amount of sky covered by each type of -97-cloud, for example, 3/10 stratus, 5/10 altocumulus and 1/10 cirrus gave 9/10 total cloud cover. 3) To calculate the daily insolation under 'actual' conditions, G' was calculated for each observation : G'=cc*cccf*G 0+(l-cc*G Q) where G'= actual insolation with corresponding cloud cover (langleys/day) cc= cloud cover (in tenths) cccf= cloud cover correction factor (step 1) Gn= daily global radiation with a clear sky The G'njQOh a n d G'igoOh v a l u e s w e r e averaged for each day. (Although several equations have been proposed to calculate daily insolation at actual cloudiness, (Vollenweider 1969, Vowinckel and Orvig 1962) most rely on empirical constants untested in the Arc t ic . The equation used here was suggested by B. Alt (pers.comm.), and was the most suitable for the data available.) 4) A ratio of G'/Gg was calculated for each day. 5) Each hourly value from Dahlgren's original clear sky insolation dataset (after interpolation between days) was then multiplied by the corresponding daily ratio of G ' / G n . -98-6) Steps 2 to 7 from the clear sky model were then repeated with this cloud cover corrected dataset. Integration of the area under the daily curves produced the calculated G1 value within 1%. Dahlgren calculated monthly means of relat ive insolation G/G Q based on actual cloudiness; his values for June and July (the means of 1961/62) were 0.743 and 0.651 respectively. The modelled mean cloud cover corrected data values of relat ive insolation at Cape Vera were 0.524 and 0.660 for June and July. Densely Overcast Model Insolation was also calculated for a continuous 10/10 cloud cover from June to August using Dahlgren's data to estimate the minimun amount of insolation avai lable. This calculation was the same as that used in the cloud cover model except the following equation was used in step 3: G 1 ( f c c c f * G 0 where G^g = insolation with 10/10 cloud cover cccf = dai ly cloud cover correction factor Gg = daily global radiation with a clear sky The calculation of G^g was done once a day. The three insolation datasets are presented graphically (Fig. A.3). -99-Figure A.3. Incoming solar radiation calculated at Cape Vera with cloudless, cloud-cover corrected, and 10/10 overcast sky conditions. -100-SOLAR RADIATION CLIMATES FOR CAPE VERA, NWT 8 0 0 -7 0 0 -1 0 0 -01JUN82 01JUL82 LEGENDi C • CLEAR SKY 01AUGB2 CLOUD CORRECTED 01SEP82 DENSE OVERCAST APPENDIX B CONVERSION FACTORS FOR SOLAR RADIATION MEASUREMENTS The wavelengths of . solar radiant energy reaching the earth's surface range from approximately 290 to 3000 nm, and include near ultraviolet (290-380 nm), ' l i g h t ' , or radiation v is ib le to the human eye (380-760 nm), and infra-red (760-3000 nm) (Luning 1981). Fi f ty percent of the total energy of this spectral range is from the waveband 400-700 nm (Szeicz 1974). Irradiance is the radiant flux received by unit area per unit time, and i t can be measured in terms of either energy or quanta. Energy measurements are of the total incident energy of the spectral range from ultraviolet to infra-red. Quanta are fundamental units of energy, and the energy content of these units dif fers with wavelength (or frequency). Quanta in the v is ible range are called photons. Shorter wavelengths quanta possess more energy than longer wavelengths quanta, although longer wavelengths contain more quanta (Jerlov and Steeman Nielssen 1974). Quanta measurements are of the number of incident quanta of a specified waveband. Since the photobiochemical process is dependant on the number of quanta absorbed and not on their energy level (because one quantum cannot excite more than one molecule regardless of i t ' s energy level ) , quanta measurements are often used in primary production work. —102— Units of Measurement -1 -2 -1 Energy is usually measured in joules sec m (1 joule sec =1 watt), which are SI (Systeme International d'Unites) units, or calories per unit area per unit time which are often used in meteorology. A calorie is the amount of heat necessary to raise the temperature of 1 gram of water from 14.5 to 15.5°C, and the conversion of calories to joules is the factor 4.184, which is the specif ic heat of water at 15°C (LeGrand, 1968). Quanta are measured in Einsteins (E), where 1 Einstein 23 is 6.02*10 quanta (Avogadro's number). Conversion Factors Because measurements of quanta are most appropriate for primary 14 production work, and the light for the C incubation experiments was -2 -1 -2 measured in uE m sec , Dahlgren's solar radiation data (in meal cm min - 1 ) had to be converted to quanta units. I assumed that 50% of the total energy Dahlgren measured was photosynthetically active radiation (PAR), and used 4.184 to convert calories to joules. Energy _2 1 cal.cm =1 langley (ly) -1 -1 -2 -? 1 ly .sec =4.184 joules.sec .cm = 4.184 watts.cm -1 -1 -2 1 ly.min =4.184 joules.min .cm =(4.184/60)joules.sec - 1 .cm" 2 =6.973*10"2 watts.cm" 2 (6.937*102 watts.m" 2 ' -2 =693.7 watts.m -103 -Energy to Quanta -2 -? -1 1 watt.m =4.57 uE.m .s 1 ly.min _ 1=693.7 watts.nf 2 * 4.57 uE.m~ 2.s =3187 u E . m ~ 2 . s _ 1 -2 -1 -2 -1 1 meal.cm min =3.187 uE.m .s Dahlgren's data to Quanta for Solar Radiation models 3.187 uE.m~ 2 .s" 1 /mcal .cm" 2 .min" 1 * 60 sec/min" 1 * 1 mE/1000 uE *0.5 (PAR)= 0.0956 mE.m" 2 .min~ 1 /mcal.cm" 2 .min - 1 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0062952/manifest

Comment

Related Items