UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The mid-depth temperature minimum in B. C. inlets MacNeill, Margaret Rose 1974

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

Item Metadata


831-UBC_1974_A6_7 M33.pdf [ 4.31MB ]
JSON: 831-1.0053207.json
JSON-LD: 831-1.0053207-ld.json
RDF/XML (Pretty): 831-1.0053207-rdf.xml
RDF/JSON: 831-1.0053207-rdf.json
Turtle: 831-1.0053207-turtle.txt
N-Triples: 831-1.0053207-rdf-ntriples.txt
Original Record: 831-1.0053207-source.json
Full Text

Full Text

THE MID-DEPTH TEMPERATURE M I N I M U M IN B . C . INLETS by i MARGARET ROSE MACNEILL B . S c . , University of British Columbia, 1970 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in the Department of Physics and the1 Institute of Oceanography We accept this thesis as conforming to the required standard. THE UNIVERSITY OF BRITISH COLUMBIA April 1974 In p r e s e n t i n g t h i s t h e s i s in p a r t i a l f u l f i l m e n t o f the r e q u i r e m e n t s f o r an advanced degree at the U n i v e r s i t y o f B r i t i s h Co lumb ia , I ag ree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s tudy . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y purposes may be g r a n t e d by the Head o f my Department o r by h i s r e p r e s e n t a t i v e s . It i s u n d e r s t o o d tha t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be a l l o w e d w i thout my w r i t t e n p e r m i s s i o n . Depa rtment The U n i v e r s i t y o f B r i t i s h Co lumbia Vancouver 8, Canada i ABSTRACT A springtime mid-depth temperature minimum has often been observed in many B . C . inlets. The size and extent of the minimum varies markedly from year to year. This paper examines the temperature minimum more closely,in Bute, Knight and Jervis Inlets. Pickard (1961) suggested that a major factor affecting the size of the temperature minimum layer might be the outflow winds which blow down most B . C . mainland fjords during winter months when the Arctic air mass moves south to cover tne interior of the province. Using Abbotsford Airport as a station representative of outflow (no wind recording devices available in Bute, Knight or Jervis) for Bute, the size of the springtime temperature minimum was compared to the outflow of the previous winter for the period 1954-1973. There seems to be a rough linear relationship between the two. During 1972, 1973 and 1974 monthly cruises were made to Jervis, Bute and Knight - (making it possible to follow winter cooling on a month to month basis. This analysis seems to indicate that in Bute, at least, most of the cooling in the winter occurs during outflow situations. The actual formation of the temperature minimum layer (as shown in the cruises of February and March) appears to be partly i i caused by down-inlet advection of cold water from the head. It is possible that outflow winds may cause the disturbance which is the origin of the cold advection. i i i TABLE OF CONTENTS PAGE ABSTRACT i LIST. OF TABLES v LIST OF FIGURES vi ACKNOWLEDGEMENTS viii Chapter I. INTRODUCTION 1 1.1 A General Introduction 1 1. 2 Region of Study 2 1.3 Source of Data 4 II. YEARLY CHANGES IN WEATHER CONDITIONS AND PROPERTIES OF THE INLETS 5 11.1 Variation in Heat Content from Spring to Spring.. 5 11.2 Weather Conditions in Bute Inlet 8 II.2.A Temperature 8 II. 2.B Wind 10 II. 3 Heat Budget Considerations 14 II.4 Discussion of Yearly Data 22 iv III. RESULTS OF MONTHLY CRUISES PAGE III. 1 Description of the Changes in Bute Inlet 28 I I I . l . A 1972-1973 Winter '. 29 III. 1 .B 1973-1974 Winter 35 I I I . l . C 1973 Spring 38 I I I . l . D Discussion 40 III. 2 Some Quantatative Comparisons of Outflow and Cooling at Bute 4 40 III. 3 Cooling in Knight Inlet 41 III.3.A 1972-1973 Winter 42 III.3.B 1973-1974 Winter 43 IV. ^SUMMARY AND CONCLUSIONS 45 DATA SOURCES - BIBLIOGRAPHY 49 V LIST OF TABLES TABLE PAGE I. Dimensions of Bute, Jervis and Knight 86 II. Wind Summary from Abbotsford Airport for December, 1972, showing a period of out-flow from December 2nd to 4th and 5th to 7th 87 III. The average wind speed, average air temper-ature (minus 3C"for Bute), duration, and heat loss due to sensible heat transfer to each outflow recorded at Abbotsford from 1954-1973 88 LIST OF FIGURES FIGURE PAGE 1. Temperature, salinity, oxygen, and density depth profiles for Bute 4, June 5, 19 69 showing the temperature minimum layer 50 2. British Columbia southern mainland coast showing the location of Jervis, Bute and Knight Inlets and their longitudinal sections 51 3. Springtime temperature-depth profiles for Bute 4 from 1964-1971 showing the temperature minimum each year 52 4. Yearly changes in Z, AT and for Bute 4 from 1951-1973 53 5. a) - d) comparison of yearly changes in AQ for Bute 2, 4,6, and 8 from 1956-1973. e) yearly changes in AQ for Jervis 3 from 1956-1973. f) , g) changes in AQ for Knight 5 and g from 1956-1973 54 6. Average south coast winter air temperature (minus 2.5 °C) for the period 1951-1973 56 7. Comparison of the strength and duration of outflow between Abbotsford and Squamish 57 8. Annual cycle of the monthly mean for components of heat transfer for Port Hardy Region, 1963. (J. Elliott, 1965). . 58 9. The changes in average winter air temperature, heat loss due to conduction during outflow, and water cooling ( AQ) for the period 1954-1973 59 10. A comparison of Q e and Q^ for 7 outflows recorded at Abbotsford between 1971 and 1974; and one outflow recorded in Bute Inlet 6 0 vii 11. Comparison of average winter air temperature (south coast minus 2.5 C°) with total yearly heat loss due to sensible heat transfer during outflow (1954-1973) 61 12. Comparison of water cooling, A Q, with total yearly heat loss due to sensible heat transfer during out-flow (1954-1973) 61 13. Temperature-depth profiles for Bute 4, May 1958 and June, 1969 showing unequal surface warming.. . 62 14. A comparison of T of the temperature minimum with total yearly heat loss due to sensible heat transfer during outflow (1954-1973) 63 15. The average air temperature (south coast minus 2.5 C°) and periods of outflow (at Abbotsford). (between cruises; for the winters of 1972-1973 and 1973-1974 64 16. Temperature-depth (a) and Salinity-depth (b) profiles for Bute 2 for the cruises between October 16, 1972 and June 26, 1973, and October 1973 and March. 1974 65 17. Temperature-depth (a) and Salinity-depth (b) profiles for Bute 4 for the cruises between October 16, 1972 and June 26, 1973, and October 1973 and March 1974 65 18. Temperature-depth (a) and Salinity-depth (b) profiles for Bute 6 for the cruises between October 16, 1972 and June 26, 1973, and October.. 1973 and March 1974 65 19. Longitudinal sections of (i),.temperature and (ii) sal-inity for Bute Inlet for selected cruises for 1972, 1973, and 1974 72 20. Comparison of cooling between winter cruises with heat loss due to sensible heat transfer during outflow between the cruises for 1957-1958, 1967-1968, 1972-1973, 1973-1974 80 21. Selected longitudinal sections of (i) Temperature and (ii) Salinity for Knight Inlet for 1972, 1973 and 1974 81 v i i i ACKNOWLEDGEMENTS The author wishes to express her sincere gratitude to Dr. G.L. Pickard who suggested the problem and gave valuable advice and guidance throughout the course of this study, to Dr. Pond who suggested improvements to the manuscript and to the Meteorological Division, Department of the Environment for pro-viding unpublished material. Special thanks are given to the crew and captain of the vessel C.S.S. Vector who were always helpful and agreeable on inlet cruises;..and to Murray Storm and Dan Hsiao who both provided invaluable assistance. Finally, thanks to all the students who assisted in collecting and analyzing the data aboard ship on these cruises. CHAPTER I INTRODUCTION 1.1 A General Introduction Since 1951, the Institute of Oceanography, U.B.C. has con-ducted cruises in B.C. mainland, Vancouver Island and Alaskan inlets for the purpose of studying oceanographic conditions. One of the features which is immediately evident from a qualitative study of the accumulated data is the annual formation of a mid-depth temperature minimum (visible in the spring) in many of the inlets. In his 19 61 paper on B.C. inlets, Pickard documents this feature, commenting briefly on the nature of the cold layer, and speculating on possible factors affecting its formation. Figure 1 shows the cold layer in Bute (Station 4) as it typically appears in a temperature-depth profile (accompanied by the corresponding salinity-depth, oxygen-depth, and 0"t (density)-depth profiles). The suggested mechanism of formation is that the upper layers are cooled in the winter to a temperature below that of the deep water. Then the net input of heat in the spring warms the surface layer, leaving a temperature minimum layer usually at a depth of from 10 to 100 metres, positioned between the warmer surface water and the nearly isothermal deep water. During the summer, autumn and early winter the minimum erodes and gradually disappears. 2 The coastal climate which provides the winter cooling to produce the cold layer mentioned above is characterized by periods of "outflow winds" - cold winds which rush down coastal valleys when a mass of continental arctic air moves south to cover the interior of the province. It was suggested by Pickard (1961) that these outflow winds (locally called "Squamishes") may be a signifi-cant factor in causing turbulent mixing and cooling of inlet waters. The purpose of this paper is to establish more quantitatively the significance of outflow winds in their contribution to the overall winter cooling and formation of the temperature minimum in the inlets by studying: (1) year to year changes in the size and extent of the cold layer - i.e. using it as an indicator of cooling that has occurred during the previous winter (2) monthly changes in the heat content of the water column during winter and spring months for years that such data exist, and comparing these changes to the frequency and strength of periods of outflow. I. 2 Region of Study The three inlets where data have been collected on a regular basis for the past twenty years are Jervis, Knight and Bute. Of these inlets, Bute has been studied most frequently and systematically, with at least one spring cruise yearly since 1951, as well as several 3 years of monthly or bi-monthly cruises. Consequently, data from Bute have been used as a basis for this study, supplemented by data from Knight and Jervis. Figure 2 shows a map of the south coast of B.C. with longitudinal sections of the three inlets, illustrating their relative size, geographical position and bottom topography; Table I gives the dimensions, sill depth (Pickard, 1961) and mean annual river discharge into each inlet (Trites, 1955). The numbers appearing on each inlet in figure 2 indicate the positions of standard stations where oceanographic data were collected. The runoff in Bute and Knight is controlled chiefly by snow-melt, maximum runoff occurring between May and July, minimum runoff occurring between January and March (Trites, 1955). The main sources for both inlets are rivers which rise in the Chilcotin (fed in part by the extensive snow and ice fields of the Coast Range) - the Homathko and Southgate Rivers flow into the head of Bute, the Klinaklini River flows into the head of Knight. A secondary (fresh water) source for Bute is the Orford River which discharges into the inlet on the eastern shore twenty miles from the head (at station 4). Jervis differs from Bute and Knight in that its fresh water source is controlled by precipitation and snowmelt - it experiences runoff maxima in the spring and fall (Pickard, 1961). There are no rivers from glaciers running into Jervis. 4 1.3 Source of Data All oceanographic data used in this thesis were collected by the Institute of Oceanography, U.B.C. Extensive use has been made of the southern inlets spring cruise series begun in 1951. To provide a study of month to month changes, a series of 20 cruises were made in Bute, Knight, and Jervis between June, 1972 and March 1974 - nine of these occurring during the winter months. Observations of temperature, oxygen and salinity were taken at standard stations (figure 2). Temperatures were measured with Richter and Wiese reversing thermometers mounted on Atlas or N.I.O. water bottles, with a bathythermograph, and with a Plessey Model 9060 S.T.D. (Salinity, Temperature, Depth recorder). Dissolved oxygen was determined by Winkler titration on board ship. Salinity was determined used an Autolab (inductively coupled) Salinometer in the shore laboratory as well as from S.T.D. records. At each station, a secchi-disc reading was taken together with observations of wind, air temperature, cloud cover and sea state. The Atmospheric Environment Service, Department of the Environment (previously the Meteorological Branch, Department of Transport) provided all the meteorological data. Average south coast temperatures were found in the publication "Monthly Meteorological Observations in Canada"; detailed wind data was made available through the Vancouver office of the Atmospheric Environment Service. CHAPTER II YEARLY CHANGES IN WEATHER CONDITIONS AND PHYSICAL PROPERTIES OF THE INLETS II. 1 Variation in Heat Content From Spring to Spring Figure 3 shows temperature-depth profiles for an eight year period 1964-1971 in Bute Inlet, station 4. The change in shape and extent of the cold layer from year to year is considerable. For example, o in June 1969, the temperature of the minimum is 5.2 C at a depth of 80 metres - the profile is thin and elongated; but in May, 1964, the temperature of the minimum is 8. 6°C at a depth of only 40 metres. Because there were few cruises conducted during the winter months, and even fewer at precisely the time when the upper waters were coldest (before the surface warming process began), a comparison of these spring temperature profiles provides the one available method of comparing total winter cooling from year to year; a rough qualitative study of long term weather conditions in B.C. shows, for example, that the pronounced temperature minimum of 1969 mentioned above followed the extremely cold winter of 1968-1969 whereas the slight minimum observed in 1964 followed a very mild winter. To establish a quantitative comparison, the cold layer temper-ature profile has been characterized by three parameters - AT, AQ, and 6 Z (figure 3). Z is the depth of the minimum, AT is the difference in temperature between the minimum and the temperature of the iso-thermal deep water (this is taken as the temperature of the water at 200 m) and A Q - the "heat deficit" - is the area of the temperature-depth curve bounded by 200 metres, the temperature of the 200 m water and the curve itself - giving a measure of the heat loss the inlet suffers in the winter. For convenience, AQ will henceforth be expressed in units of C°« metres and often be referred to as the "heat deficit". In actual fact, AQ is the heat content of the upper 200 metres of the water column per unit area (divided by the density of water, £u> , and the specific heat of water, c^> ) subtracted from the heat content of the upper 200 metres of the water column per unit area (divided by0*> and Cp ) if it were at a uniform temperature equal to the temperature at 200 m. Using this definition, the colder the water is, the larger is AQ. In figure 4, these parameters are plotted for Bute 4 from 1953-1973. 200 metres is an arbitrary boundary chosen as the deepest extent of penetration of surface effects; but it should be noted that although winter cooling doesn't usually penetrate this far, there are several winters since 1953 when the isothermal deep waters begin as deep as 300 metres, above which is the negative temperature gradient indicative of the penetration of cooling from above. 7 Variations in the parameter Z seem to be more a result of spring warming and advective effects than winter cooling - the more upper layer warming that has occurred in the spring the larger should be Z. (Later on it will be shown how advective effects can quite drastically change Z - Section III.l.C.) The yearly variation in spring warming will vary the calculated values of A Q independent of winter cooling. A warm early spring will more quickly warm the upper waters than a late cold spring; also, the colder the waters after the winter, the faster they will gain heat due to a greater air-water temperature differential, (although this will be compensated for some-what by increased stability with colder water which inhibits heat transfer),. Even the time when the spring measurements were taken ranges over a two month period from early May to late June. The warming during this period, depending on weather conditions and river runoff (entrainment of cold water), can be considerable . For example , in 1968 a time lapse of one month in measurements taken at Bute 4 (May 14-June 14) made a 25% difference in AQ. The previous year, a three month lapse made only a 10% difference. A plot of yearly changes in AQ for Bute stations 2, 4, 6, and 8 is shown in figures 5.a-d. Although AQ for each station generally fluctuates in a similar manner, there are specific differences between stations. Looking in progression from station 2 to station 8, a general 8 increment inAQ is apparent - i.e. the top 200 metres at station 8 are colder than the top 200 metres at station 2. For the two years 1965 and 1966, Bute 2 and 4 show an increase in AQ, but Bute 6 and 8 show a decrease. Another discrepancy seems to be the decrease in A Q at Bute 2 from 1956 to 1957 when all other stations appear to be colder in 1957- than 1956. In the analysis in Chapter II.4, comparing cooling and outflow in Bute, station 4 has been used as a representative station for the inlet, being less influenced by advection than station 2 and more likely to have a water column of uniform temperature before cooling than stations 6 and 8. Figures 5.e, f & g present changes for some stations in Knight and Jervis Inlets for the same period. In contrast to Bute, Jervis is warmer and Knight is colder, although the scales of the fluctuations are of the same order. If one looks at each station separately for these inlets, Knight, like Bute, exhibits an increase inAQ from mouth to head; Jervis, on the contrary, does not. II. 2 Weather Conditions in Bute Inlet A. Temperature. Because no meteorological stations have existed in south coast B.C. inlets for any extended length of time (although temperature records were kept sporadically at the head of Knight Inlet from 1963 to 1966, and a meteorological station at Squamish 9 has been recording wind since 1970), there is no possibility of exactly correlating local climatic conditions with the size of the temperature minimum; but it is possible to use an indirect method to piece together a weather picture for Bute and the other inlets. First, a winter air temperature was derived for Bute. There are no temperature recording stations in close proximity to Bute so rather than choosing a specific station and approximating the Bute air temperature by this station, the mean air temperature for the "south coast" was used - this temperature (found in "Monthly Meteorological Observations in Canada") is derived by averaging the mean air temperatures from all south coast stations - this region extending from north of Bute Inlet to Howe Sound. A comparison of the overall south coast monthly averages with the specific monthly averages in Knight Inlet reveals a fairly consistent difference of approximately 2.5 C° (Knight being colder) - although winter temperature differentials are slightly more extreme than summer ones .^Assuming that Bute and Knight are closely correlated as far as air temperature is concerned (the reason behind this assumption is explained in section II. 2 .B), it seems reasonable, then, to use the south coast variations corrected by 2.5 C° to describe Bute. Since, in this yearly comparison, it is only necessary to know the average temperature over the whole winter, the monthly mean air temperatures for the south coast were averaged 10 for the months of November, December, January, and February -these being the months when most of the cooling probably occurs. O This mean value minus 2.5 C is plotted for the years from 1951-1973 (figure 6). The average air temperature representative for Bute varies from a low of -1.7 °C (1956) to a high of 2°C (1967). B. Wind. To determine even a rough wind picture for Bute Inlet when no continuous wind measurements have been made inside it involves some conjectire. Although there are wind recording meteorological stations situated on the coast within a hundred miles of Bute (e.g. Powell River and Port Hardy), all of these stations are open to the northwest and southeast winds that blow along the coast in the winter (Elliott, 1965) and none of them is in an advantageous location for exposure to outflow winds from the interior. Because the outflow winds are the winds of interest in this study it is definitely not representative to use any of these meteorological stations as a model for Bute. Besides, the special topographical nature of the inlets protects them from the coastal winds mentioned above (north-west and southeast). The following discussion explains how a station (representative of outflow) was chosen. Winter outflow winds blow mainly in November, December, January and February. They occur when a large region of high pressure 11 (cold air) moves down to cover the whole interior of B.C. and a strong inland to coastal pressure gradient causes this continental Arctic air to rush seaward down coastal valleys. Consequently, strong, cold winds blow down the inlets all along the coast wherever a valley penetrates to the interior. Therefore, when Bute Inlet receives strong outflow winds down the Homathko Valley from the Chilcotin, so does Knight Inlet down the Klinaklini River valley; at the same time Howe Sound receives winds from the Cariboo down the Squamish and Lillooet River Valleys as does the Lower Fraser Valley down the Fraser Canyon and Harrison Lake. So, if a station can be found that registers outflow winds, it probably represents, approximately, weather conditions in the fjords all up the coast (with corridors to the interior in the form of river valleys) in general outflow situations; and it follows that it is feasible to extrapolate its record of northeast (outflow) winter winds to describe Bute Inlet (and Knight). Such a station is Abbotsford. (Vancouver never seems to experience outflow winds, apparently being too far east to be in the path of winds blowing out of Howe Sound, and too far west to be in the path of winds blowing down Harrison Lake and other valleys on the north side of the Fraser Valley.) It is in an ideal location to record the outflow winds funnelled into the Fraser Valley. • Wind records from Abbotsford, then, have been used as a record of outflow winds along the coast in general, and more specifically as a rough record of outflow in Bute Inlet. 12 It is not possible to generalize an exact correlation between Bute and Abbotsford but where corresponding outflow data exists between Abbotsford and other B.C. inlets; for example, Squamish (at the head of Howe Sound); a reasonable correlation is found. There are corresponding records for the two station for November, 1970; December, 1970; January 1971; December 1971; January 1972; December 1972; January 1973,and February 1973. For example, in January 1971, both stations recorded outflow conditions from the 10th to the 14th: Abbotsford for a duration of 78 hours and an average speed of 9.4 m/sec; Squamish for a duration of 106 hours and an average speed of 8.9 m/sec. In December 1971, Abbotsford recorded an outflow of 71 hours with an average speed of 8.6 m/sec; during the same period, Squamish recorded an outflow of 100 hours long with a speed of 7.6 m/sec. A simple quantitative comparison is demonstrated in the graph in figure 7. For each outflow, average speed was multiplied by duration of the outflow for.•.•both Squamish and Abbots-ford and the results plotted in figure 7. For the few points available, a direct relationship can be construed if a straight line is drawn through zero to best fit the points. The slope of the line is 1. Although out-flows at Squamish are of longer duration than Abbotsford, outflows at Abbotsford are stronger (i.e. higher speed). It may well (and probably 13 does) follow that as one looks still further north (to Bute and Knight) the duration of the outflows increase even more; hopefully the relation-ship as direct, if not one to one. The wind information from Abbotsford is hourly, the wind speed recorded in miles per hour (converted to m/sec). An outflow situation is defined for the present purpose as having an average speed of more than 4. 5 m/sec, a duration of greater than eighteen hours, o o and a prevailing direction from 0 - .90 . The first two critera are fairly arbitrary; anything less intense has a fiarly large probability of being merely local. In scanning the record, outflows are easily identified by direction alone (see Table II). Each outflow can be characterized by year, month, date, average speed, duration in hours and average temperature throughout the outflow. See Table III. Temperature records from the head of Knight Inlet during six outflow o situations show the average temperature to be 3 C colder than Abbots-ford. Because Bute and Knight receive cold air from approximately the same interior region, their temperatures during these periods are taken to be the same, and the "average outflow temperatures" recorded in Table III are really the average temperature at Abbotsford during the outflow corrected for Bute by subtracting 3 C. 14 II.3 Heat Budget Considerations In order to study the heat loss in an inlet in any particular winter, and the processes involved in this loss, it is necessary to look at the heat transfers (by currents, absorption of solar energy, evaporation etc) which make up the heat budget of the inlet. The simplified equation (Pickard, 1963) for the balance of heat energy in a column of unit plan area, extending from the sea surface to the sea floor, can be written as QsU-r) + Q b + Q e + Q h + Q v = Q T (1) where Q s is the incident short wave (solar) radiation received on a horizontal surface at sea-level, both direct and diffuse, r is the fraction of short wave radiation reflected by the sea surface. Qk is the net heat transfer by long wave radiation from the sea and the atmosphere. Q e is the heat transfer by evaporation and conden-sation of water vapour. Qh is the net sensible heat transfer across the air-sea interface. Q v is the net gain of heat in the water column by advection. 15 Ql> is the resultant rate of gain or loss of heat by that column of water. The unit used for Q will be langleys/day (1 gram calorie/sq. cm. day). As was mentioned in the preceeding section, the complete lack of meteorological stations in Bute Inlet (or Jervis or Knight) makes it difficult to estimate with accuracy the variables needed to calculate all the heat budget components above. Consequently, it has been necessary to neglect the radiation terms Q s and Qb in the year to year comparison of the cold layer with the heat loss occurring due to outflow winds. The justification for dropping these terms becomes more apparent on examination of their daily and yearly variability. Figure 8 shows the seasonal fluctuations of Q s, Qb, Qe, and Qh for the Howe Sound region of the B.C. coast during the year 1963 (Elliott, 1965). In the winter months, Q s drops to less than 100 langleys/day; data from "Monthly Meteorological Observations in Canada" shows that the variation in the January monthly mean value of Q s for Port Hardy over a period of 5 years (from 1969-1972) is + 15 langleys/day, which is small compared to a net surface heat flux of approximately -200 langleys/day (calculated value for the Port Hardy region for 1963 which was a fairly mild year; J. Elliott, 1965). During 16 an outflow the sky is usually clear, so the daily variation in Q s for this period would be small (In the order of 10-20 langleys/day) com-pared to the possible heat loss due to conduction of as much as 400 langleys/day (see equation 3) during an outflow (in the inlets). It is presumed that Port Hardy and Bute Inlet have comparable fluctuations in Q§ . From figure 8, it can be seen that fluctuates very l ittle seasonally. During the winter, the water temperature, which governs the amount of long-wave radiation emitted to a clear sky, (essentially black body radiation proportional to the fourth power of the absolute temperature of the surface) is fairly uniform, so the value of remains nearly constant. The sky is clear during an outflow, so it is reasonable to predict that in this situation wi l l contribute l itt le to the variability of the value of surface heat exchange. Because of variation in cloud cover, cloud height, and cloud type, there wi l l be some variation in the winter mean value of from year to year but probably not any more than the variation in Q s . It should be added that in the winter when Q s and are of comparable value (approximately 100 langleys/day), their short term variations tend to cancel each other out. If the sky i s clear, Q s increases but Qj-, also increases, and (being of opposite sign) the result is no change in the net radiation; l ikewise, on a cloudy day, Q s and Qb 17 are simultaneously smaller (although they don't decrease in the same proportion with cloud cover). This cancellation will limit the variation these terms introduce into the average total heat loss over the winter. The remaining terms to deal with in the heat budget equation are 0^, Qe' a n d Q v The formula (Sverdrup, Johnson, and Fleming - The Oceans, 1942) for calculating the net heat transfer by conduction across the air-sea interface is Q h = -C p • A h • dT/dZ (2) where Q n is the heat transfer away from the sea surface by sensible heat conduction, langleys/day. Cp is the specific heat of air. Aft is the eddy conductivity for heat, and dT/dZ is the gradient of temperature in the air. With Aft proportional to the mean wind speed times height and dT/dZ proportional to the difference in temperature between the air and the surface water divided by height, the formula becomes Qe = Qh (total) = -k • Cp • U * AT • D (3) where Q e is the total amount of heat transferred to the air (by conduction^during an outflow, in langleys. U is the mean wind speed during the outflow in metre s/sec. 18 AT is the difference in temperature between the air and the surface water in C . D is the duration of the outflow in days. Cp is the specific heat of air. and k is the constant of proportionality. The method of determining U and D has been explained in section II. 2. B: A single A T was found for each outflow by computing the difference between the average air temperature for this period (sec-tion II.2.B) and the average winter water temperature in Bute Inlet (there are no continuous surface temperature records for Bute so an average o winter temperature, 6. C, was calculated from all existing cruise data). The value of k, the constant, was taken to be 3.32, using the method of J. Elliott (1965). Although the absolute value of the constant in the final analysis of the data is unimportant because the comparisons are only relative, its inclusion allows to be calculated in common units of langleys/day. The results of the calculations are recorded in Table III; the last column shows the total heat loss due to sensible heat transfer for each outflow, which in turn, is plotted against the year (from 1954-1973) on the graph in figure 9 (dashed line). The heat loss due to evaporation (Qe) in the winter is as great or greater (greatest evaporation occurring when cold, dry air flows over warm water as in an outflow situation) than heat loss due to conduction 19 (e.g. figure 8, 1963), although it follows the same pattern of fluctuation - being likewise proportional to the mean wind and also dependent on temperature in that the vapour pressure of the air is a function of temperature. The formula (Sverdrup, Johnson, and Fleming - The Oceans, 1942) for the calculation of Q e is Q e = -L • A h • de/dZ . 6.2 X 10 - 5 (4) where Q e is the amount of heat transferred to the air by-evaporation in langleys/day. L is the latent heat of vaporization. Ah is the eddy conductivity, and de/dZ is the gradient of vapour pressure in the air. Although the data isn't available to calculate heat loss due to evaporation in Bute Inlet for each outflow in Table III, a ratio of to Q e using equations 2 & 4 (assuming the eddy conductivities have the same value) gives Qh dT/dZ Q7 = ° - 6 6 diTdz ( 5 ) if C =0.24 and L = 585. The gradients above can be replaced approximately by the difference in temperature and the difference in vapour pressure at the sea surface so that (5) becomes % = .66 T o ~ T a (6) Qe eo " e a 20 where T 0 = temperature at water, " C. T a = temperature at air, 0 C. e Q = vapour pressure of water, mb. e a = vapour pressure of air, mb. Because T 0 - T a and e Q - e a are of comparable value, Q e will in general be larger than Qh, but for air colder than water (e.g. during an outflow), as the temperature of the air becomes very cold, the difference T 0 - Ta will increase faster than the difference e Q - ea (the value of ea is restricted because it can not go below zero) and Qh and Qe will become closer in value. Figure 10 plots Qe against Qh for 7 outflows in Abbotsford between 1971 and 1974 - the only years dew point records can be readily obtained for Abbotsford. The point marked by a star represents an outflow recorded in Bute Inlet on February 8, 1973. Note the decrease in the slope of the line drawn as larger outflows are approached (in general larger temperature), although in the region of the graph where the points lie (over 100 langleys/day), the curvature of line is less than for outflow below 100 langleys/day. If it were possible to calculate the heat losses due to evaporation for every outflow (the data being available), and then add the results to the values of Qh already calculated, a graph of the data points would look similar to the graph of Qh alone in figure 9, only the relative size 21 of the fluctuations would probably be damped somewhat, so that the low values of heat loss would appear slightly higher and the high values, slightly lower. Qv is the only term which cannot be satisfactorily estimated using either direct or indirect methods. The inlets are by no means closed systems; yearly records suggest that Bute, Knight and Jervis are subjected to regular intrusions from outside the sill at various times of the year. In Bute, these intrusions can affect the upper waters directly (mid-depth intrusion) or indirectly (deeper intrusions which come in over the sill at 350 metres and spill into the basin) by the displacement of deeper water toward the surface or by the diffusion and upward mixing of new deep water properties. Generally, they provide a subsurface warming effect which counteracts winter cooling. They can also influence the inlet in the summer and fall so that it does not always settle to the same temperature just prior to the onset of winter cooling - the range is usually 8. 5°C to 9.0° C. An exact quantitative assessment of the intrusions on the basis of once-yearly cruises is highly speculative; thus it is possible to be aware of the limitations they impose on the heat budget calculations, but not possible to make numerical evaluations of these limitations. The discussion of monthly cruises will consider advection more closely. 22 II.4 Discussion of the Yearly Data The graph in figure 9 shows winter outflow, average winter air temperature (as explained in section II. 2.A), and water cooling (for Bute 4), all plotted against the year on the same graph in order to explore their interrelationships. The similar trend in the yearly fluctuations of all these variables is immediately apparent. High outflow years such as 1957 and 1969 are also characterized by low average winter temperatures and large amounts of winter water cooling (AQ); low outflow years such as 1958, 1961, and 1964 have high average temperatures and little water cooling. The relationships between the three variables are more clearly seen if the variables are plotted against each other - see figures 11 and 12. For the twenty years plotted, a linear relationship provides a reasonable fit (an estimated, not a least squares fit) between outflow and the average winter air temper-ature; the graph in figure 11 shows an intercept of 1.4°C for zero — 4 ° / outflow and a slope of 5 X 10 C /langley. In view of the nature of the origin of the outflow winds (i.e. the movement of the Arctic front southward to cover the southern Interior), this direct relationship seems reasonable; the advancing cold front brings both outflow winds, and the cooler temperatures which govern the average monthly air temperature on the coast. The relationship between outflow and water cooling is also of 23 a linear nature (figure 12). Although a straight line can be drawn through the points, the scatter is large. (Including Q e with in the heat loss due to outflow would probably produce a similar result.) The crude approximations used to calculate the above variables have introduced considerable error into the positions of the points on the graph in figure 12. First, the values for outflow are, as previously mentioned, only an extrapolation from Abbotsford data. It is noted, for example, that for three years - 1965, 1966, and 1972 - when the graph in figure 9 shows a very large temperature minimum layer resulting from proportionately low amounts of outflow, there is a much bigger difference in the average December and January temperatures between Bella Coola and the South Coast. This difference is probably an indication that the north coast is experiencing much more outflow than the south coast, and this outflow may extend as far south as Bute Inlet, but not to Abbotsford. Consequently, outflow for these years is underestimated. Also, several effects have made the yearly comparison of total winter water cooling subject to error. One of these effects, already discussed, is advection. Another effect to be aware of, although it occurs only rarely, is "cold" remaining stored in subsurface waters (around 100 metres) from one winter to the next, the result of a very severe winter. 24 The phenomenon can be seen quite clearly in 1957-1958, when monthly cruises were maintained over the winter. Notice in figure 5 how AQ at stations 6 and 8 (near the head) decreases much less than at stations 2 and 4 (near the mouth) from 1957 to 1958. This is due in part to a warm intrusion which has only extended as far as station 4 by this time, but is mainly due to the cooler temperature of the 200 m water at Bute 6. The sequence of bi-monthly temperature profiles over 1957-1958 shows clearly that this water was cooled in the severe winter of 1956-1957 and because of the high stability at the head of the inlet (due to more fresh water at the surface), this cold water never completely mixed into the surface waters during the summer of 1957. This has nothing to do with the cooling of the 1957-1958 winter but nevertheless affects the calculated value of cooling, A Q, calculated from the spring profile in 1958. The large values of and AT for 1951 (see figure 5.b) are also almost certainly due, not to the 1951-1952 winter which was relatively mild, but to the cold remaining after the extreme 1949-1950 winter. The third effect, also already discussed, is the unequal amount of spring surface warming that occurs before the measurements are taken each year. An extreme example of the inequality appears in figure 13 which compares two spring temperature profiles for Bute 4 from May 14, 1958 and June 5, 1969. The top 50-100 metres of Bute 4 25 in 1969 seem to have warmed up 3 C° - a significant amount. If the depth of the minimum hasn't increased, the increase in due to this warmed up layer of water would be 185 C° - metres, a difference of 75% from the original calculation; whereas in 1958, no significant surface warming seems to have taken place since winter (i.e. the minimum is close to the surface) and the value of A Q computed for the cold layer should represent quite closely the maximum winter value before warming. Therefore, the degree of spring warming very significantly affects the value that A Q has attained since its maximum winter value; it was first thought that a crude correction could be made for spring warming when calculating A Q / by merely increasing A Q the required amount to compensate for the increase in temperature of the water above the minimum. Upon study of the data from the cruises which took place between March, 1973 and June, 1973 it became evident, however, that the formation and erosion of the minimum in the spring was by no means a gradual diffusion of heat from the surface and entrainment of cold, salty water upwards so that erosion of the cold layer occurs uniformly - but that the minimum migrated up and down several times over at least a period of several months regard-less of surface warming (Section III.l.C); and that consequently the type of compensation performed above (for 1969) was meaningless and could only lead to further error. It is nevertheless probable that if correction 26 for the above discrepancy were possible, the point representing 1969 (and also 1957, another very cold year) on the graph in figure 12 would be located closer to the straight line drawn. Stability of the water column is a determining factor in the degree and depth of cooling; the less stable the water column, the deeper the cooling goes, but the more stable the water column, the colder the upper waters should become. Pickard (1961) shows that in the 1957-1958 winter seasonal variations in stability of the water column in Bute occur chiefly in the top 20 metres and that the head of the inlet experiences less change in stability than the mouth - i.e. the water column at Bute 8 is almost as stable as in the summer whereas the rest of the inlet becomes considerably less stable in the upper 20 metres. This head to mouth variation in stability should cause variations in the degree and depth of cooling at different stations along the inlet. Again, because of insufficient data (i.e. monthly mean discharge in the Homathko River or winter measurements of stability), the exact influence of year to year stability variations is difficult to gauge. With these limitations in mind, the relationship between outflow and cooling as illustrated by the graph in figure 12 can be construed as one of approximate direct proportionality. A plot of outflow v.s AT of the minimum for Bute 4 (figure 14) is also linear 27 with somewhat less scatter in the large outflow region than a plot of outflow vs AQ. Probably the parameter AT is more nearly conserved over the spring months (spanning the time of measurements - for example, section III.3.C, station 4) whereas AQ is more subject to change because of the penetration of solar effects from above and advective effects from below. The minimum is always located at a depth of strong salinity and density gradient (e.g. figure 1 is an extreme case) which prevents mixing of water properties across the interface; also, as seen in section III.l.C, the cold layer seems to be advecting mouth-ward and this feeds more cold water to down-inlet stations. Chapter III will serve to support the theory of direct relationship between out-flow and cooling and provide more insight into formation of the minimum with monthly data. CHAPTER III RESULTS OF MONTHLY CRUISES III. 1 Description of the Changes in Bute Inlet A series of cruises was conducted to Bute Inlet (as well as to Knight and Jervis) during the years 1972, 1973, and 1974 which provided data spanning two full winters and made it possible to observe the cooling process in more detail. In the 1972-1973 winter, the cruises took place in mid-October, mid-December, early February, late February, and late March. In the 1973-1974 winter, the observations were made in late October, mid-December, mid-January, mid-February, and mid-March. Three unfortunate gaps when there were no observations made for over six weeks occurred in January, 1973 and November of both winters; these gaps tend to diminish the probability of successfully correlating water cooling with natural phenomena (more specifically, outflow winds) and advective effects. In regard to outflow winds, the ideal approach to this type of study (i.e. physical oceanographic cruises) would be, of course, to observe conditions within the inlet immediately preceeding and immediately following an outflow and to analyze the ensuing change. Because the shiptime was booked months in advance, an outflow (predictable only a few days ahead) could not be monitored in this way; but more frequent cruises would have provided a more complete record. 29 Figure 15 shows the dates of the cruises scheduled over the winter months, with respect to conditions of outflow and average weekly temperature (South coast). Note that a major outflow situation developed during each winter. For the purpose of examining changes in physical properties throughout the water column between winter cruises, S.T.D. plots of salinity and temperature taken at the time of each cruises at stations 2, 4, and 6 are reproduced in figures 16, 17, and 18 (a & b). Because this is chiefly a qualitative comparison with emphasis on the out-standing changes detectable, the accuracy of the S.T D. traces should be sufficient (making sure, of course, that relative changes in temper-ature and salinity are verified by bottle casts). To elucidate the dynamic processes involved in heat exchange and formation of the minimum, longitudinal sections of temperature and salinity for selected cruises appear in figures 19 (a-g); these sections integrate the separate profiles into an overall picture of the inlet at the time of each cruise. A. 1972-1973 Winter. Between the October cruise and the o December cruise an outflow of average temperature -4 C and average y speed 8.7 m/sec occurred from the 2nd to the 7th of December. During the interval between the cruises there was a gradual decrease in the average weekly temperature until the last week in November when there was a sudden decrease to below freezing. The low temper-ature was maintained throughout the outflow and continued afterward 30 for another week; the cold spell ended just before the December cruise. In October (figure 16), the temperature of the inlet from the mouth to Bute 3 is quite uniform throughout the water column («8°C); toward the head of the inlet (at Bute 4, 6; and 8 - not shown), a remnant of the cold layer still remains at 100 metres. A week following the December outflow described above, the upper waters of the whole inlet have cooled considerably. Bute 2 is well mixed with respect to temperature to a depth of 40 metres (temper-ature of 6.2'C), the upper waters of Bute 4 are colder (6.0°C) than Bute 2 but well mixed to only 30 metres (the salinity profiles reflect the well mixed layer - at Bute 4, the salinity is approximately 29.4 %. ). Bute 6 has no well mixed surface layer (a fairly gradual vertical temper-ature gradient exists from 0 to 80 metres) but the temperature of the surface waters at the mouth of the inlet - e.g. 6.3°C at Bute 2. This strong head to mouth surface temperature gradient (usually of the order of only 1 0° difference) seems characteristic of the inlet following cold outflow and is probably partly due to stability differences and partly due to the gain of heat the outflow wind presumably undergoes as it flows down inlet. Notice that below 150 metres (not shown but extending to about 400 m), the water column has become warmer and more saline ( at all stations) since October - bottle cast data show a 31 temperature increase of as much as 0.3 C and a salinity increase of 0.1 to 0.2 as well as a decrease in oxygen content at Bute 2 in this region; Bute 6 experiences an increase of 0.1 to 0.2 C ° in temper-ature and a slight (less than 0.1 °/BO ) increase in salinity from 150 to 400 metres. This phenomenon is presumed to be the result of a mid-depth intrusion from outside the s i l l . The large increase in salinity (0.4 % e) in the top 75 metres of the water column at Bute 6 (figure 18) (also seen at Bute 7 and 8) which has formed a quite homogeneous salinity layer may be the direct iresult of outflow winds pushing cold , low salinity water mouth ward and causing greater upward mixing of mid-depth water properties. In contrast, at Bute 2, although the upper 20 metres have increased in salinity, from 20 to 50 metres (to the bottom of the well mixed temperature layer) there is a decrease in salinity. At the beginning of February, the top 20 metres of the inlet is considerably colder than in December along the whole length, but beneath the colder water there seem to be definite anomalies from station to station. The subsurface waters at Bute 2 are markedly warmer and more saline (figure 16) from 50 to 100 metres (0.3Cand 0.4 % 0 ); on the other hand, at Bute 4 a temperature minimum with a lower salinity than December has appeared at 70 metres although below this at 150 metres the water is warmer by 0.2 C* ; and at 32 Bute 6 and Bute 8 there has been extensive cooling from 50 to 100 metres. The cold layer observed in the top 100 to 20 metres from Bute 2 to Bute 8 is probably a direct result of the outflow that was occurring at the time of the cruise. During the entire fifteen hours spent in Bute Inlet in February 8, a north wind of 10 m/sec was blowing; the average (daytime) temperature observed was 4°C. From the fifth to the seventh of February, Abbotsford was also recording an outflow of 8.8 m/sec and 2.5°C (average of day and night), so this north wind in Bute was probably the tail end of the same outflow (the next day, in Knight Inlet, no outflow was recorded). With an average temperature of 0°C (the temperature would drop below freezing at night), an average wind speed of 10 m/sec (assuming little change over the entire outflow), an average water temperature of 5.5°C and an average vapour pressure of 4.5 mb (both from measurements taken at the time), the calculated heat loss during the outflow wind would be 165 langleys/day due to sensible heat transfer and 300 langleys/day ' due to evaporation (equations 3 and 4 ). This result is similar to the result of 160 langleys/day due to sensible heat transfer and 290 langleys/day due to evaporation found using the method of extrapo-lating from Abbotsford (Table III). 33 In light of the cold temperatures and outflow in early-January (figure 15), the net cooling at Bute 6 and 8 seems predictable, but what of the net warming at Bute 2 (also Bute 1). The anomalies observed are best explained by looking at the longitudinal sections of temperature and salinity for Bute for February 8 in figures 19.a.i & ii which show the warming below 20 metres at Bute 2 to be the apparent result of a disturbance along the inlet at about 75 metres which appears as a warm, up-inlet advection from stations 3 to 6 along the 29.2 -29.4 % 0 isohalines (there maybe as a result some mouthward move-ment of colder water beneath this, causing the temperature minimum at Bute 4). It is unclear whether this disturbance is in any related to an intrusion over the sill into the inlet at about 200 metres (8.3°C and 30.6 %») or whether it is perhaps a direct result of the outflow wind which is blowing (or they may all be interconnected). It is noted that there is virtually no surface salinity gradient along the length of the inlet which is usually the case even in winter (the head being usually much less saline than the mouth) - in fact, the least saline surface waters are found at Sutil 3 (outside the sill) which has a surface salinity of 27.9 % B compared to a surface salinity at station 8 of 28.9 % o . Actually the upper 50 metres at Sutil 3 are considerably less saline than the rest of the inlet. Two weeks later, on February 28, longitudinal section 19.b 34 shows the warming from the advection along the 29.4 %«, isohaline to have penetrated close to the surface as far up-inlet as Bute 4, for example the change in position of 6.8 C isotherm. The isohalines in figure 19".b indicate that the inlet is more well mixed from 50 to 100 metres (29.4 %„ and 29.8 % 0 isotherms are spread further apart). By March 2 7, a temperature minimum has formed all along the inlet at about 80 metres depth. From figure 18 it can be seen that the cold water at station 6 (as well as 7 and 8 not shown) has mixed down deeper to 100 metres since February although the water above 50 metres has warmed substantially. From 30 to 90 metres, there is an oxygen maximum of 5.5 ml/1. Bute 8 has a very uniform temper-ature of about 5.8°C and a uniform oxygen content of 5.7 ml/1 from 25 to 90 metres (salinity of about 29.2 %„) with a shallow warmed surface layer. The large temperature minimum of 6.2 C which has appeared at Bute 4 is attributed to advection from up-inlet (it seems to be too much cooling to come only from the surface at this time of year) and can be seen on the S.T.D. traces for stations 2 (figure 16) and 3 (not shown) as well, only to a lesser degree. At the depth of the temperature minimum at stations 2 (figure 16) and 3 (more prominent at station 3), the S.T.D. traces show several salinity inversions over a 20 metre range which may indicate the active leading edge of the advection. See figure 19.c i & i i for an overall view of the cold layer in Bute. 35 B. 1973-1974, Winter. The changes between October and December of 1973 (in contrast to 1974) are not very pronounced (figures 16, 17, 18). The only outflow to occur during this period is in the first week of November, with an average speed of 5.6 m/sec, a dura-tion of three days and an average temperature of 2°C. The average temperature for the entire interval (i.e. weekly averages) never goes below 0.5°C. There is no mixed surface layer along the inlet as in December, 1972 . At Bute 2, cooling has penetrated to only 100 metres (vs 150 metres in December, 1972). Between the December and the January cruise, a strong out-flow occurred (figure 15) with an average speed of 8.6 m/sec, average temperature of -0.5°C, and duration 5 days. The changes are seen again in figures 16, 17, and 18. The inlet seems to be in a similar state to that following the December outflow of the previous winter. A fairly well mixed layer is apparent at Bute 2 and 4 from the surface to 50 metres 6.2°C at Bute 2 and 5.8°C at Bute 4, where there has also been a decrease in the vertical salinity gradient. At Bute 6 there is a very cold (5.4°C to 5.6°C) mixed surface layer 35 metres deep beneath which is a sharp thermocline and halocline (the previous winter at Bute 6 there was no mixed upper layer but the cooling apparently penetrated deeper). The decrease between December and January in the temperature and salinity at about 125 metres at Bute 6 (figure 18) may 36 (as was suggested for December , 1972) be the result of increased upward mixing of deeper water to replace surface water pushed mouth-ward by the outflow winds. At Bute 2 the water below 150 metres is warmer and more saline (than December) indicating again, the begin-ning of an intrusion from outside the s i l l . There were no outflows registered at Abbotsford for the remainder of the winter; although the cold front of the Arctic air mass was in constant north-south flux over the B.C. interior during February and March, so outflows may have occurred in Bute, being further north. A short, cold spell penetrated to the south coast on March 7th and 8th. The changes that occurred in February and March are reminiscent of the previous year (see the profiles in figures 16, 17, and 18). In February, the water column to 400 metres at Bute 2 is warmer than January by 0.7 C° from 10 to 130 metres and by 0.2 C° © to 0.3 C from 130 to 400 metres. The salinity in the top 150 metres (below 25 metres) has increased by 0.2 % 0 and .the oxygen content has decreased. Bute 4 is warmer from 10 to 70 metres; bottle cast data show a decrease in oxygen and no significant change in salinity although the water column is more mixed from 25 to 75 metres. The surface at Bute 4 through Bute 8 was observed to be very cold, from 1-2*C, ice having been observed on the inlet on this cruise. Overall, the water column at Bute 6 has cooled since January to 125 metres (even though from 10 to 37 35 metres is warmer). This region is also less saline and more oxygenated. As in February 1973 then, there seems to be a sub-surface warming effect influencing the inlet most prominent at Bute 2 and disappearing by Bute 6, which again, may be an up-inlet advec-tion resulting from outflow winds or an intrusion (8.5°C and 30.7 %* ) originating outside the sill and entering the inlet between 200 and 300 metres. See figure 19. f. The profiles (figures 16, 17, 18) for March show the inlet from station 2 to 6 to have cooled and become more saline from 50 to 150 metres, especially from stations 6 to 8 and to have become warmer above 50 metres (approximately). Overall, the inlet has experience net cooling. The longitudinal sections 19.g. i & ii give a clearer picture of the upper 100 metres and how they are more mixed than February - note a spreading of the 28.5 %» and 29.4 isohalines (especially at the head). Again, as in 1973, there is some indication of an up-inlet advection between 25 and 50 metres which may be helping to mix the water and causing a down-inlet advection of cold water along the 29.3 % 0 isotherm which may become the temperature minimum (note temperature minimum at Bute 4). It may well be that outflow winds not recorded at Abbotsford have caused the cooling and down mixing especially evident up-inlet of Bute 4, but such additional outflow still does not in itself explain the temperature minimum formed 38 at Bute 4; also, it is worthy of note that Knight Inlet experiences no such cooling during this interval. C. 1973 Spring. Because the springtime cold layer in Bute Inlet has been used as a measure of winter cooling (Chapter II) in the twenty year analysis of the effects of outflow winds; it is of interest to follow its changes between its formation and early summer (all measurements used in Chapter II were taken during this time period). The spring of 1973 has been selected for this purpose. Figures 16, 17, and 18 for Bute 2, 4, and 6 respectively show how the temperature profile changes markedly from March to April. At all three stations, the first sign of the cold layer appears in March. The surface has warmed and a temperature minimum can be seen at 80 metres. At Bute 2, the stability has been low enough to allow down-mixing of surface warming to the minimum - a fairly uniform temperature of 7.8°C exists from 0 to 80 metres; at Bute 4 and 6 there is a gradual decrease in temperature from 8°C at 0 metres to 6.2*C at 80 metres. As one moves from station 6 to 4 to 2 the temperature of the minimum increases from 6 C toi6.2*C to 7.2°C. A section giving a picture of the whole inlet is shown in figure 19.c - the striped region represents water of less than 7.2*C. In April, the pronounced temperature minimum has disappeared from Bute 2 as more warming and mixing has occurred. But at stations 39 4 and 6 the cold layer still persists and has moved 50 metres toward the surface as has the corresponding salinity. The gradient of temper-ature at the surface is very extreme. The area of the region of less than 7.2°C (figure 19.d) has decreased. The temperature of the minimum at Bute 4 has also decreased from 6. 2"C to 6. 5°C. It appears that this upward movement may be caused by an intrusion coming into the inlet from outside the sill at 150 to 200 metres and pushing the cold layer up. Note the deepening of the 8.2°C isotherm in figure which spreads apart the 8.0°C and 8.2*C isotherms. By June, (figure 19.c) the cold layer has advanced to Bute 1 although much warming has occurred from the surface and the " 7.2°C" region has diminished in width from 50-75 metres to 2 5 metres. At Bute 4, the minimum has become colder (back to 6.2°C). There has been a sinking of the intrusion deeper into the inlet and the cold layer has responded by settling down to approximately 75 metres (this may also be a result of a surge in river discharge in late May). It is evident from the above discussion that the dynamic pro-cesses within the inlet can quite dramatically alter the orientation of the cold layer and that the cold layer itself advances (at least in this instance) mouthward, draining cold water from the head and intensifying the temperature minimum at down-inlet stations. 40 D. Discussion . For the two years studied, it appears that the formation of the (springtime) minimum is a combination of 1) cooling and turbulent mixing from the surface during outflow situations 2) ad-vective processes which seem to help mix the upper 100 metres of the inlet and also cause down-inlet movement of cold water, and finally 3) spring surface warming. It is possible that these outflow winds may also be involved in the advective process - by pushing surface waters out of the inlet, thus triggering the events observed which eventually seem to lead to the subsurface movement of cold water toward the mouth. The only other year it is possible to follow the formation of the minimum in any detail is 1958. There was little or no outflow in the immediately preceeding winter. The extremely small minimum that is observed was formed only by solar warming of the sur-face waters - no subsurface advection occurs to make a deeper minimum. Ill. 2 Some Quantitative Comparisons of Outflow and Cooling at Bute 4 For a limited number of years sufficient winter measurements have been taken to correlate a change in the heat content of the water column in the inlet over a particular time interval (usually one to two months). This calculation effectively reduces the problem of spring surface warming encountered in section II. 4 (when trying to compare 41 yearly totals) and at least reveals or clarifies subsurface and advec-tive effects (a problem also discussed in II.4).because the time lapse is smaller. The years of available data are 1957-1958, 1967-1968, and the two successive winters described in section III. 1. Bute 4 has been chosen as a representative station to base the comparison, being far enough up inlet to avoid direct effects of subsurface intrusions over the sill and yet not being so far up inlet that stability inhibits down-mixing to a large degree (as at Bute 6 and Bute 8). Figure 20 plots the correlated outflow - cooling data. The correlation appears linear for the six points on the graph. If the data are plotted on the twenty-year graph of total winter outflow vs total winter cooling (figure 12 - the data under discussion are marked by A ) the points lie along the line already drawn to demonstrate the possible linearity of this relationship. The points extend along the line to an outflow of only 2400 langleys. Ill. 3 Cooling in Knight Inlet For the purpose of study, Knight can be divided into two parts - an inner basin (stations 4 to 11) that is up to 540 metres deep, and an outer basin (stations 1-3) that is only 200 metres deep. The two basins are separated by a shallow sill 65 metres deep. Due to the 42 topography of the inlet (figure 2), stations 1 to 4 are exposed to the northwest winds which blow along the coast all winter (they may also be less subject to outflow winds); consequently they experience cooling throughout the whole winter and temperature changes reach the bottom. Strong tidal action probably helps mixing here also. The inner basin of Knight is protected from the coastal north-west winds which affect stations 1 to 4 but is probably more influenced by outflow winds so it cools like Bute Inlet with greatest surface cooling taking place during outflow. A study of the temperature minimum in Knight in past years reveals that in contrast to Bute, the minimum (A T from the deep water) is always less pronounced but the cold layer extends much deeper than Bute. In both the 1972^1973 and the 1973-1974 winters the salinity gradient in the upper 20 to 150 metres was observed to be much less pronounced in Knight than in Bute which should make the upper waters less stable in Knight and account for the mixing of surface effects (i.e. cooling) down deeper. A. 1972-1973 Winter. Advective processes seem more large scale than Bute. Looking from longitudinal section 21.a to 21.b for example, it appears that an intrusion over the shallow sill deep into the inner basin in October has resulted in the mixing and warming of the subsurface waters of the inner basin by December. The upper waters have cooled (to 100 metres at Knight 5), presumably by the 43 December outflow, but there seems to be some inhibition of the penetrations of cooling from Knight 7 to 9. There is a very strong negative vertical temperature gradient and positive oxygen content gradient from 0 to 50 metres at stations 7 to 11 (surface temperature of less than 4°C). By February 9, the cooling has gone quite a bit deeper but there is still no mixed surface layer. As in Bute on February 8, the surface salinity gradient from head to mouth is especially small. Between the 8th and the 27th of February, an intrusion (30.8 %„and 6.4°C) has entered the inner basin (there is some indication it has already begun on February 9) from outside the sill (figure 21. c). It seems to mix and cool the water below 25 metres in the inlet so that by March (not shown) a cold layer of fairly uniform physical properties (6.4°C and 30.7 %. to 30.8 %. ) 150 metres thick has formed. It is unlike the temperature minimum in Bute in that it has more uniform physical properties with no distinct temperature minimum, and is much thicker. B. 1973-1974 Winter. Like Bute, Knight experiences little cooling before the December cruise in 1973. There is some indication of an advection along the 31.0 %« isohaline at 50 metres depth at Knight 7 and 9. In mid-January, after an outflow, the inlet is cooler 44 and less saline to 200 metres from stations 5 to 7 (see figure 21.d) but like December, 1972 there seems to be inhibition of the down-mixing of cooling from stations 8 to 10. This region appears as a temperature maximum at 75 metres on the temperature-depth profiles of stations from 8 to the head which may indicate it is an advection along the 30.8 °/00 isohalines. A large spreading between the 30.6 and the 31.0 %» isohalines since December from stations 7 to 11 shows mixing has occurred in this region. Below the temperature maximum layer, a temperature minimum layer from stations 8 to 10 could be mouthward advection along the 31.2 %«, isohaline. Sometime between January 14 and February 14 (see figures 21.e. i & ii) an intrusion from the outer basin has entered the inner basin as in February of 1973 of temperature 6.4"C and salinity 30.6%* to form a well mixed cold layer as far as Knight 7 (by March this well mixed region extended to station 9). CHAPTER IV SUMMARY AND CONCLUSIONS The presence of a springtime temperature minimum has long been observed in B.C. mainland inlets. In this study, an attempt has been made to show that outflow winds play a significant part in the formation of the minimum in inlets with valleys penetrating to the interior. First, the temperature minimum was examined in Bute Inlet in the spring months (May and June) from 1951 to 1973. Within the crude approximations used 1) to estimate the outflow winds and temperature in Bute, lacking a meteorological station in the inlet, and 2) to evaluate the total winter cooling from the temperature minimum, a rough description of the correlation between outflow and cooling ("heat deficit") is one of direct proportionality. Although the points on the graph (figure 12) of outflow vs A Q have a lot of scatter in the region of greater outflow, this is probably because the years that the water is coldest are the years when most spring warming occurs in the shortest time - which leads to a relative underestimation of AQ. In the light of the nature of outflow winds, the above correlation seems highly reasonable; outflow winds and cold temperatures on the 46 coast spring from the same source - the movement of the Arctic air mass south - so they almost always accompany one another. For the rest of the winter, the inlets are usually mild - 0-5 C •- and experience little wind (being sheltered from coastal winds); so once the water has reached a uniform temperature little cooling, other than surface, occurs. For example, in 1958, a rare year of almost no out-flow winds, only the top 20 metres of the inlet became colder than 7*C, and below 50 metres there was no cooling at all. The monthly cruises which visited Bute, Knight, and Jervis every two to seven weeks over the 1972-1973 and the 1973-1974 winters provided a more detailed look at the cooling process and more insight into the formation of the minimum. To carry the quantitative comparison one step further, the change in the "heat deficit" , Q between cruises was calculated for Bute (also for several other winters for which the data was available) and plotted against the heat loss due to outflow during that interval (figures 20 and 12). These data points are free from the error of 1) surface warming and 2) unknown subsurface effects (in the monthly data these effects can be followed) but regardless, they still lie on the line drawn through the points of the yearly study, tending to support the theory of a linear relationship between cooling and outflow, at least for lower values of outflow (less than 1000 langleys total). 47 A qualitative examination of the data from the two winters seems to indicate that as suspected, most of the cooling in Bute occurs during an outflow situation. The 1973-1974 winter shows slight net cooling in February and March when Abbotsford registers no outflow but it is felt that this is because the front of the Arctic air mass was in frequent north-south flux over the central interior at this time and Bute, being further north probably felt the effects of this to some extent in the form of outflow winds and cooler temperatures. (A year of much cooling in February was 1951 when records show Bute to have cooled a significant amount in the upper 50 metres between the end of January and mid-February; not surprisingly, an unusually (for this late in the winter) long, cold outflow occurred in early February). Knight, also, experiences most of its cooling during outflow although advective effects in the inlet appear to be stronger than in Bute and can influence the depth of cooling considerably in the short term (months). After the cooling process, it seems that advection plays a large part (for these two years at least) in both Bute and Knight in actual mixing and formation of the minimum. Whether these advections are the eventual result of a chain reaction begun by the disturbance of the inlet by outflow winds or merely a part of some larger system of events that originates outside the inlet is difficult to ascertain without 48 further, more detailed study. The only other year of such complete data is 1957-1958 - there is no outflow this year and the minimum seems to form only by surface cooling (and spring surface warming) and not advection. As has been reiterated several times, a further study would be needed to make any definite statement about the exact link between outflow winds, cooling, and formation of the minimum. The first step in a future project would be the installment of anemometer and air temperature and humidity recording devices to establish out-flow winds and general weather patterns more accurately. The second step would be the installation of three thermister chains (situated at Bute 2, 4 and 7) to record changes in water properties throughout the winter months. Monthly cruises are obviously not sufficient to produce a detailed picture of winter cooling and spring formation of the minimum. 49 DATA SOURCES, BIBLIOGRAPHY Data Reports of the Institute of Oceanography, University of British Columbia, Vancouver, British Columbia, British Columbia Inlet Cruises 1951-1972. Meteorological Branch, 1951-1973. Monthly Record, Meteorological Observations in Canada. Canada Department of the Environ-ment. Elliott, J.A. , 1965. A Study of the Heat Budget Components for the British Columbia and S.E. Alaska Coast. Manuscript report #18, Institute of Oceanography, U.B.C. Pickard, G.L. , 1961. Oceanographic Features of Inlets in the British Columbia Mainland Coast. J.Fish. Res. Bd. Canada 18 (6): 907-999. Pickard, G.L. Descriptive Physical Oceanography. Pergamon Press Ltd. , Headington Hill Hall Oxford, 1963. Pickard, G.L. and R. W .•; Trites. 1957. Fresh Water Transport Deter-mination from the Heat Budget With Application to British Columbia Inlets. Jour. Fish. Res. Bd. Canada. 14 (4) 605-616. Sverdrup, H.U., M.W. Johnson and R.H. Fleming, 1942. The Oceans. Prentice-Hall, Inc., N.Y. Tabata, S. , and G.L. Pickard. 1957. The Physical Oceanography of Bute Inlet, British Columbia. Jour. Fish. Res. Bd. Canada 14 (4) 487-520. Trites, R.W. 1955. A Study of the Oceanographic Stricture in British Columbia Inlets and Some of the Determining Factors, PhD thesis, University of British Columbia, Vancouver, B.C. 22 27 4 50 100 200 C * 3 0 0 4 0 0 23 28 5 24 Density (Ot) .0 29 30 6 7 2 31 8 T 4 Oxygen (ml/I) 32 Salinity (%o) 9 Temperature (°C) <—Temperature Cn O Figure 1. Temperature, salinity, oxygen, and density depth profiles for Bute 4, June 5, 1969 showing the temperature minimum layer. 52 Temperature (°C) Temperoture (°C) I 8 9 7 8 9 Temperature (°C) Temperature (°C) Figure 3. Springtime temperature-depth profiles for Bute 4 from 1964-1971 showing the temperature minimum each year. 1951 2 3 6- 7 8 ' ' ' ^ | 9 i960 I 2 . 3 4 5 Year 6 7 8 9 1970 I 2 3 Figure 4. Yearly changes in Z, \ T and 6 Q for Bute 4 from 1951-1973. On CO 54 d) Bute 8 o < 100 6 7 8 9 I960 1 2 3 4 5 6 7 8 9 1970 1 2 3 Year Figure 5. a) - d) comparison of yearly changes in A Q for Bute 2, 4, 6, and 8 from 1956-1973. .• 55 100 h 0" 6 7 8 9 I960 I 2 3 4 5 6 7 8 9 1970 I 2 3 Year Figure 5. e) yearly changes in A Q for Jervis 3 from 1956-1973. f), g) changes in AQ for Knight 5 and 9 from 1956-1973, .•I-en Figure 7. Comparison of the strength and duration of outflow between Abbotsford and Squamish. 58 Total Outflow AQ (Longleys) (C°-m) 9000 300 r Year Figure 9 . The changes in average winter air temperature, heat loss due to conduction during outflow, and water cooling ( A Q ) for the period 1 9 5 4 - 1 9 7 3 . Cn CO 400 0 100 200 300 400 .Qh-heat loss due to conduction v" (langleys/day) Figure 10. A comparison of Q e and Qj^ for 7 outflows recorded at Abbotsford between 1971 and 1974; and one outflow recorded in Bute Inlet. 61 0 2000 4000 6000 8000 Total Yearly Outflow (Langleys) Figure 11. Comparison of average winter air temperature (south coast minus 2.5 C") with total yearly heat loss due to sensible heat transfer during outflow (1954-1973). Figure 12. Comparison of water cooling, A Q, with total yearly heat loss due to sensible heat transfer during outflow (1954-1973). • 0 2000 4000 6000 8000 Toiol Yearly Outflow (Langleys) a) Bute 4 - June 5,1969 Temperature (°C) 7 8 10 1 ' 1 1 J 1 warming i i \ i 7 8 Temperature (°C) b) Bute 4 " May 14, 1958 Figure 13. Temperature-depth profiles for Bute 4 , May 1958. and June, 1969 showing unequal surface warming. 63 Figure 14. A comparison of A T of the temperature minimum with total yearly heat loss due to sensible heat transfer during outflow (1954-1973). o Mar. S972 1973 1974 Figure 15. The average air temperature (south coast minus 2.5 C°) and periods of outflow (at Abbotsford) between cruises for the winters of 1972-1973 and 1973-1974. . CD Figure 16. Temperature-depth (a) and Salinity-depth (b) profiles for Bute 2 for the cruises between October 16, 1973 and June 26, 1973, and October 1973 and March 1974. Figure 17. Temperature-depth (a) and Salinity-depth (b) profiles for Bute 4 for the cruises between October 16, 1972 and June 26, 1973, and October 1973 and March 1974. Figure 18. Temperature-depth (a) and Salinity-depth (b) profiles for Bute 6 for the cruises between October 16, 1972 and June 26, 1973, and October 1973 and March 1974. 66 16.BUTE 2, 1972-1973 o) Temperature (°C) 7 8 6 7 8 50 100 150 200 250 I i i lr i) October, 1972 28 29 30 o r—i 1 1— 50 100 150 200 -i 1 — 7 T 250 ii) December, 1972 iii) February 8,1973 iv)February 28,1973 v) March 27, 1973 vi) April 30, 1973 28 29 30 28 29 30 28 29 30 28 29 30 28 29 30 I b) Salinity (%o) 16. 'BUTE 2, 1973-1974 a) Temperature - Depth Profiles (°C) 6 7 8 6 7 8 6 7 8 6 7 8 6 7 8 2 0 0 h 2 5 0 L _ l 1 d 1 I L L J I i i i 1 | I i • i I I I i » i I viflOctober 26, 1973 viii)December 2b, 1973 ix) January 14, 1974 x) February 14, 1974 xi) March 14,1974. 28 29 3 0 28 29 3 0 28 29 3 0 28 29 3 0 28 29 3 0 2 0 0 h 2 5 0 J 1 I I 1 I i • i b) Sal in i ty -Depth Profiles (%0) 68 17. BUTE 4, 1973-1974 18. BUTE 6,1973-1974 o) Temperature -Depth Profiles (°C) 6 7 8 6 7 8 6 7 8 6 7 8 6 7 200 h 250 I i | d 11 i | 1 i i il I I • i i l l I i i i vii)October 26, l973vB)December II, 1973 ix) January 14,1974 x) February 14, 1974 xi) March 14, 28 29 30 28 29 30 28 29 30 28 29 30 28 29 30 200 r-250 I 1 J I LI I I I i 11 I i I I b) Salinity-Depth Profiles (%o) 72 Figure 19. Selected longitudinal sections of (i) Temperature and (ii) Salinity for Bute Inlet for 1972, 1973 and 1974. 74 75 Su3 Station O l 2 3 4 5 6 7 8 I 1 1 : 1 1 1 1 ' ' Su 3 Station 0 1 2 3 4 5 6 7 8 7 6 7 79 80 0 1000 2000 3000 • Outflow (Langleys) Figure 20. Comparison of cooling between winter cruises with heat loss due to sensible heat transfer during outflow between the cruises for 1957-1958, 1967-1968, 1972-1973, 1973-1974. 8 1 Figure 21. Selected longitudinal sections of (i) Temperature and (ii) Salinity for Knight Inlet for 1972, 1973 and 1974. 82 83 84 I Station 3 5 7 9 I Station 3 5 7 9 85 86 87 Table I. Dimensions of Bute,, Jervis, and Knight Name Length Mean Width Mean mid-inlet depth Outer Mean annual sill fresh water depth discharge nautical miles m m m^/sec Knight Bute Jervis 70 41 48 1.6 2.0 1.7 295 510 495 64 355 185 410 410 180 Table II. Wind Summary from Abbotsford Airport :'for December, 1972, shGwing a period of outflow from December 2nd to 4th and 5th to 7th. DATE 0 .. 1 2 3 4 5 6 7 8 9 10 11 H 12 0 U 13 R 14 15 16 17 18 19 20 21 22 23 PREV OIR. MEAN SPEED MAX VEL °1 SSW 14 ssw 17 ssw 20 s 14 ssw 18 S 23 S 25 s 20 s 22 S 23 S 24 S 22 SSW 22 SW 14 SSW 14 SW 13 SW . 11 SW 9 SSW 7 SSW 5 SSW 8 ssw 10 ssw 5 ssw 10 SSW 1 5 . 4 25 02 sw" 7 ssw" 6 ssw 7 SW 7 SW 9 SW 8 sw 4 ssw 5 " " C ESE 16 E 10 ENE 15 ENE 16 ENE 11 ENE 17 ENE 15 ENE 13 ENE 16 ENE 18 ENE 14 ENE 16 ENE 15 ENE 14 ENE 20 ENE 11 . 6 20 03 ENE 19 ENE 19 ENE 21 ENE 20 ENE 20 ENE 19 ENE 17 ENE 18 NE 21 ENE 15 ENE 16 NE 24 ENE 23 ENE 22 ENE 24 NE 26 NE 27 NE 25 NE 26' NE 28 NE 28 NE 27 NE 24 NE ENE 2 2 . 1 28 04 " NE 27 " NE 26 ENE 25 NE 23 NE 24 ENE 25 ENE 21 ENE" 18 " NE 7 ENE 10 ENE 11 E 12 E 13 E 8 ENE 7 ESE 4 ESE 4 ESE 5 E 13 ENE 9 ENE 10 E 6 E NNE 10» 3 ENE 13.4~" ' 27 05 E 3 NE 8 NE 7 E 6 E 15 ENE 17 ENE 18 ENE 18 ENE 17 ENE 23 ENE 21 ENE 23 NE 24 ENE 20 NE 19 NE 23 NE 15 NE 23 .NE 25 NE 28 NE 26 NE 22 NE 25 NE 23 NE 1 8 . 7 28 06 '. NE 27 ENE 29 ENE 27 ENE 28 NE 26 ENE 28 ENE 31 ENE 30 NE 33 NNE 25 NE 31 NE 27 NE 27 ENE 25 ENE 25 ENE 27 NE 26 ENE 20 NE 25 ENE 21 ENE 20 ENE ' 18 ENE 23 ENE 20 ENE 2 5 . 8 33 07 ENE 20 NE 20 ENE 23 ENE 24 ENE 18 NE 24 NE 23 ENE 17 ENE 14 ENE 23 ENE 24 NE 22 NE 19 ENE 22 ENE 21 ENE 21 ENE 17 ENE 15 ENE 15 ENE 12 ENE 6 ESE 5 SE 4 ESE 11 ENE 1 7 . 5 24 08 • E 10 E 10 E 10 ENE 9 NNE 5 WNW 3 C E 8 ESE 6 SE 3 NNW 2 ESE 6 E 9 ESE 2 C SE 3 ESE 2 C C C C C C NNW 3 E 3 . 8 10 09 C N 4 NE 4 NNE 3 NNE 2 NNE 3 N 2 C C NNW 2 ENE 4 C NE 6 WNW 5 C WNW 5 NNW 5 C NNW 3 C C C NW 4 N 6 SVL 2 . 4 6 10 - .., w 3 "C ENE 2 wsw 7 C C . NW 2 C " C C E 3 ENE 4 C SW 4 SW 1 W 2 NNE 6 C C C C NE 3 NNW 3 C SVL 1 . 7 7 11 NNE 3 C ENE 2 N 3 NE 2 N 3 NNE 2 NE 4 NE 3 NE 3 ENE 3 NNE 2 ENE 5 NNE 7 NNE 4 ENE 3 NE 2 C ENE 5 E 3 NNE 4 NE 3 NE 4 ENE 15 NE 3 . 5 15 12 ENE 25 NE 24 ENE 17 NE 22 NE 22 NE 22 NE 18 ENE 18 ENE 16 NE 19 ENE 16 ENE 14 ENE 16 ENE 13 E 8 E 8 ESE 5 E 7 C C N 3 N 3 c C ENE 1 2 . 3 25 13 NNE 3 C C ENE 2 N 4 NNE 4 NNW 2 N 4 NE 2 ENE \ 2 C NE 3 NNE 4 N 6 NE 3 NNE 5 SE 3 C i N ,7 C ESE 6 NNE 5 NE. 5 C NNE 2 . 9 7 1* C C C C W 3 NNE 2 NNW 2 N 5 ' N 3 N 3 NNE 4 NE 5 C C NNE 2 NNE 5 NNE 3 NNE 5 NE 7 E 6 NE 6 NE 9 NE 8 ENE 9 NNE 3 . 6 9 15 NE 7 NNE 7 N 5 ENE 7 E 3 NE 2 NE 4 ENE 4 ENE 4 NE 5 NE 3 E 8 ESE 6 N 6 E 5 NNE 7 ENE 7 NE 10 ENE 10 ENE 7 NE 6 E 10 ' E 8 ENE 7 SVL 6 . 2 10 16 ENE 7 ' "E 5 ENE 2 NNE 3 N 4 ENE 3 NE 7 ENE 2 E 3 NE 10 NNE 5 NNE 6 NE 7 NNW 7 NNE 7 NE 4 NNE 7 NNE 8 ESE 6 NNE 7 NE 7 C NNE 6 C NNE 5 . 1 10 17 NNE 6 ENE 5 NE " ' 3 ESE 3 NNE 4 NE 3 NNE 7 NE 4 NNE 4 NNE 4 NNE 6 WNW 3 NNE 5 NNW 5 N 4 C C WSW 5 'SSW 4 SW 8 SW 6 ssw 6 WSW 9 WSW 8 NNE 4 . 7 9 18 ' ~WSW'~NNW" 7 * "' C NNW 4 'NNE 4 ENE 5 ENE 5 NNE" 6 NN'E"' 7 NNE ' 7 ENE 11 NNE 8 NE 5 ENE 7 NNE 5 ENE 4 N 4 NNE / 7 C NNE 6 S 12 S 24 s 18 S 21 NNE 7 . 5 24 89 Table III. The Average Wind Speed, Average Air Temperature (minus 3 C° for Bute), Duration, and Heat Loss Due to Sensible Heat T ransfer to Each Outflow Recorded at Abbotsford from 1954-1973 Year Month Date Average Average Duration in lang- Q h (total) Speed in Temper- in days leys/day in lang-m/sec ature in leys C 1954 Jan 10-12 8.7 -0.5 1.4 167 234 15-18 8.0 -111. 0 4.1 451 1851 19,20 5.9 -12.5 1.4 358 501 21,22 5.8 -9.0 1.5 284 426 23,24 4.8 -11.5 1.7 272 462 1955 Jan 1,2 6.8 -0.7 0.8 149 119 • Feb 25-27 7.2 -1.3 2.0 175 349 Nov 11-14 8.5 -11.0 3.4 462 . 1571 Dec 12-14 8.4 -3.0 1.8 257 462 17-18 8.8 -7.0 1.7 388 660 1956 Jan 26,27 6.5 -4.0 1.4 210 294 Dec 4-6 12.6 -9.7 2.5 654 1634 1957 Jan 8,9 5.6 -5.0 1.8 202 363 11 5.9 -5.7 1.0 236 236 13 7.5 -6.3 0.6 r Z 9 2 175 14,15 13.5 -7.3 1.4 653 914 20 6.8 -1.3 1.2 158 190 23-25 8.5 -11.4 3.3 494 1631 Feb 1,2 11.7 -2.5 3.7 326 1208 18-22 9.6 -2.5 1.4 272 381 Nov 14-16 4.5 4.0 1.1 46 51. 1958 Nov 14-16 7.7 -0.8 1.8 174 314 24-26 7.7 1.0 2.3 127 292 Dec 5-8 6.0 -5.0 3.2 221 706 90 Year Month Date Average Average Duration in lang- Qh (total) Speed in Temper- in days leys/day in lang-m/sec ature in leys C 1959 Jan 1-3 9.8 -12.5 2.1 564 1184 20,21 6.0 -3.2 1.1 171 188 Feb 6,7 7.7 -2.0 0.8 201 161 15-19 7.9 3.0 3.9 79 307 Nov 11-13 10.9 -2.0 1.6 281 449 15,16 10.8 -8.5 1.6 516 825 Dec 16,17 7.0 2.0 0.8 96 77 30,31 6.9 -5.0 1.4 254 ;356 1960 Jan 19-20 5.7 -2.0 3.4 149 508 20,21 7.9 -0.8 1.2 171 205 Feb 24-26 9.6 - 2.5 1.8 111 199 29 9.0 -1.2 0.7 204 143 Dec 14,15 6.9 2.5 0.9 78 ~70 1961 Jan 25 7.4 2.0 0.6 95 57 Dec 9,10 8.0 -4.5 1.3 266 346 1962 Jan 17-20 9.6 -8.5 4.1 458 1876 Feb 4,5 7.1 3.0 0.9 72 65 19,20 9.1 2.0 0.8 195 156 22-24 10.9 0.0 1.5 217 325 Dec 23,24 7.7 -1.2 0.8 181 145 1963 Jan 9-11 10.9 -7.0 2.1 466 978 18 6.9 -2.5 0.8 189 151 28-30 11.9 -4.5 2.8 406 1137 Dec 8-10 10.5 0.2 1.6 199 318 16,17 7.1 -1.3 0.9 388 349 1964 Dec 15,16 13.1 -12.5 1.7 783 1331 18,19 6.5 -6.0 L.3 256 333 1965 Jan 23 6.3 -0.8 0.9 137 123 Feb 2 5.2 -0.8 1.0 117 117 Nov 22,24 7.0 2.0 2.0 38 176 Dec 26-28 6... I -6.2 2.5 246 616 91 Year Month Date Average Average Duration Qh in lang- Q^ (total) Speed in Temper- in days leys/day in lang-m/sec ature in leys C 1966 Jan 3,4 5.5 -7.0 1.3 235 305 23,24 8.0 -3.5 1.1 239 263 Nov 7,8 8.0 2.0 0.6 110 66 12,13 6.0 4.0 1.0 40 40 28-30 6.4 -6.2 2.0 •74 148 1967 Jan 24,25 7.5 0.0 0.7 370 259 Nov 1-4 7.1 4.0 3.2 47 151 Dec 6 6.4 2.5 0.8 69 58 17-21 10.6 -6.0 2.9 45 121 1968 Jan 25-29 10.6 -6.5 3.9 434 1691 Dec 13-15 7.1 0.0 1.1 140 154 19 7.4 0.3 1.0 137 137 26-31 9.1 -14 4.8 595 2 858 1969 Jan 19-22 8.2 -9.0 4.0 403 1 614 24-28 10.9 -12.0 4.7 648 3 047 1970 Jan 15,16 9.0 -4.0 1.5 288 432 Feb 27,28 11.9 3.5 1.9 98 187 Nov 21-23 11.1 -4.0 2.3 372 855 26-28 7.9 -0.8 1.5 172 258 Dec 20,21 9.1 -5.0 0.8 309 247 1971 Jan 10-14 7.4 -12.0 4.1 415 1700 Dec 23-26 8.7 -10., 4 1.6 600 1000 1972 Jan 24-27 1.0..1 -13.0 4.0 500 2 000 Feb 1 5.8 -7.0 1.2 300 400 Dec 2-7 8.8 -9.0 5.4 350 1 860 1973 Jan 3 9.8 -7.0 1.1 350 370 5 7.9 -11.0 1.2 500 600 7-10 7.4 -7.0 2.7 200 700 Feb 5-7 8.8 -0.5 2.6 1@0 450 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items