The Atlas
The table here is a key to the Current Atlas: Strait of Georgia to Juan de Fuca Strait published by the Canadian Hydrographic Service, and referred to below as the Atlas¹. It identifies the appropriate Chart in the Atlas to use at approximately hourly intervals through a year. The table was created by an implementation of the algorithm described inside the front cover of the Atlas, and on page 4 of that book.

Using the Tables
The tables are useable if you simply open the page (salishtracks.com/tables) and accept the data generated by default parameters. However, you can change those parameters to specify exactly how you want to construct the tables. How you do that is described immediately below. Once you have specified new paramters, click the "Go" button just above the tables (and beneath the parameter table). Also described below are the toggles you can use to collapse or expand the data for individual months, or the entire year.

To set parameters, look under "Options" at the top of the table (if you don't see it, click the "Show Options" button). You will see a number of text- and check-boxes.

• Round to hour: check this if you want all times rounded to the nearest hour (as is done in most "Atlas Tables"). SalishTracks caclulates all times to the minute and then rounds to the nearest hour. This can easily result in two consecutive periods (say, 07:35 and 08:25) rounding to the same hour. In this case, the item with the smallest error wins. (In case of a tie, the later item wins.)
  
  If you do not round (i.e, accept the default), the Tables will step through each Chart in every tidal cycle. You will not step from, say, the penultimate Chart in a rising tide to the second Chart in a falling tide.
• In the Cutoff columns:
  • Large is the tidal scope at or above which a tide is "large".
  • ? corresponds to the 1.8m breakpoint shown in the Atlas pictograph (inside front cover and p. 4). I interpret the 1.8m as saying tides whose scope is greater than, or equal to, 1.8m (but less than the Large cutoff, 3.0m) may be either medium or large. Tides with smaller scope (but larger than the 0.6 scope of a Small tide) may be either medium or small. In this interpretation, no tide is unambiguously "medium".
  • Ignore: check this if you want to ignore the ambiguity in medium tides mentioned immediately above. If you check "ignore", every entry in the table will display just one Chart number. Any tide that is neither 'small' nor 'large' will be medium.
  • Small is the tidal scope (in meters) at or below which a tide is "Small".
• Clicking the button Salish Setup will populate the options with "standard" values.

To understand exactly what these parameters mean, you might read the section below, "The Algorithm".

If you "round to hour", there will be a line in the table for every hour of the year; that is about 8800 lines. There may be more if you do not round (and that is how SalishTracks recommends you use the table).

With so many table entries, navigating to the day you're interested in can be awkward. To facilitate navigation, you can collapse or expand monthly data. Hover the cursor over the minus-sign (or plus) just right of the word "Day" in the column headers. If you leave the cursor there a moment, you should see text describing what will happen if you click on that symbol: "collapse all months" or "expand all months". If you click on the minus-sign, the approximately 720 lines of data for each month will be replaced by the first line of data for that month.

Similarly, you can expand or collapse data for individual months, clicking on the plus or minus sign you'll find in the Day field of the first record of each month. If you're interested in just one month, you can collapse all records, then expand the month of interest.

Finally, you can navigate from the Tables to other pages using the menu at the top of the page.

The Algorithm
The algorithm for identifying the appropriate page is described in the Atlas, both inside the front cover and again, with examples, on pages 4 and 5. It is, to put it generously, succinct. The annual "keys" available from several vendors are testimony to the difficulty the average navigator has in deciphering the algorithm. What follows is another attempt, though this too may be too "succinct".

The Atlas divides all tides into six categories: rising small, rising medium, rising large, and three similar falling tides. A particular sequence of charts depicts current patterns that evolve over the course of any one of these tides. For example, charts one through eight show the current patterns over the course of a rising large tide. Not all tides require eight pages; the "small rising" tide, for example, requires only 6.

Tides, of course, vary according to where they are measured. The Atlas is keyed to tides at Pt. Atkinson, a point on the north side of the entrance to Burrard Inlet, British Columbia, in which Vancouver is sited. The tidal data are published on-line by Fisheries and Oceans Canada. The URL may change; it is best to google for Pt Atkinson Tides. Use the official Canadian Government web site. Other sites publish data for Pt. Atkinson, but the data do not agree precisely. (You can tell Google to search only Canadian government websites using the Site specification, e.g., Pt Atkinson Tides site:.ca). By using the filter to "Show Extremes only", the salishtracks tables should show the same times of high and low tide, and in addition will provide the range (change to the following extreme) and category (e.g., Large Rising). Of course for navigation purposes, rely only on the official Canadian website.

Looking at Pt. Atkinson tides, we find a high of 4.1m at 3:46am on Jan 1, 2015, followed by a low of 3.4m at 8:25. The scope of the tide is -0.7m. The tide is falling, i.e., going from high to low. The pictograph on page 4 of the atlas tells us that any falling change up to 1.2m is "small" (and any rising tide above 3.0m is "large"). Thus, this is a small falling tide. Currents in the period beginning at the 3:46am high are described on Chart 37. As the tide falls, the currents that develop are described by charts 38, 39...43.

A point of particular importance is the time-step as you go from one Chart to the next. The text suggests each step is one hour: "...select the hourly increment on the curve" (step 2, "How to Use Part A", p. 4). This works if the time from low to high tide is eight hours. Each of pages one through eight would show the (approximate) current pattern for one hour of the cycle. Page one would show the current pattern for the first hour, and so on.

Of course, the time from high to low tide is rarely eight hours. Furthermore, the number of pages for the various tides varies from six to eight. The text admits the difficulty in the Note on page 5: "The duration of time between high and low water...varies considerably...". It then goes on to suggest that you measure the first time intervals as hours from the start of a tidal cycle, and measure later intervals from the end of the cycle. Suppose, for example, a small rising tide lasts 279 minutes (4 hours 39 minutes), as did a tide on the morning of Jan 1, 2015. The six charts would span 6 hours, but if you measure from the first hour, you would be looking at Chart 20 at slack, rather than the last Chart, 21. Taking the text's suggestion, the proper sequence would be charts 16, 17, 18, 19 and 21 (or possibly 16, 17, 18, 20, 21). This suggests the pattern shown by one Chart (19 or 20) would never develop.

It should be clear that this is not the way current patterns evolve; they change smoothly. Nature will not allow them to jump suddenly from one pattern to the next. It seems reasonable to assume that every tidal cycle displays all the current patterns shown for the particular cycle. A medium falling tide shows all seven current patterns for such a tide (charts 30 through 36). Further, it seems reasonable that the Charts for a particular tide apply for equal time intervals, rather than one hour intervals. The first Chart in a sequence applies from the first moment of a high (or low), and the last Chart applies right up to the moment of the following low tide.

The Atlas was written long before we all had computers on our desks and in our pockets. The algorithm on pages 4-5 was written for computational convenience. That this resulted in skipping a page, or using a particular Chart a half-hour early, may not matter a whole lot in light of all the other approximations involved in the tables. When a computer is available to do the work, though, it is just as easy - in fact, easier - to calculate time intervals precisely. Doing so may also improve, if only slightly, the accuracy of the Tables. It is the way the tables here operate, by default.

The decision about rounding-to-the-hour is up to the user of these Tables. The user can check the box "Round to Hour". With this option, times are calculated to the nearest minute, then rounded to the nearest hour. In the case of a short tidal cycle, this could result in the same hour being repeated, once with one page, then with the next. Rather than display this ambiguity, the tables here report the page with the smallest rounding error. This is perhaps most easily understood by looking at a particular short-duration tide with the "round to hour" option unchecked (the default), and then with it checked.

There is one more difficulty to address with regard to the algorithm. The pictograph shows three sizes of tide, so we need only two "breakpoints" to determine a size: the maximum change for a "small tide", and the maximum change for a "medium" tide. The pictograph, though, shows three, the "extra" value being 1.8m for both a rising and falling tide. What is this for?

The default algorithm used to produce tables on this site takes the middle breakpoint (1.8m) to mean the level below which some tides may develop as either a small or medium tide (but never a large tide), and above which some tides may develop as either medium or large (but certanly not small). You can choose to ignore the problem by clicking the "ignore" box at the top of the Tables. In this case, the middle breakpoint is ignored. This is the default. Alternatively, you can un-check the ignore, in which case the table will suggest two possible charts for each phase of every medium tide. In every case, the first Chart corresponds to a medium tide. The second, or alternative Chart, corresponds to a large tide if the scope is larger than the breakpoint, or a small tide if the scope is smaller than the breakpoint.

There is some additional ambiguity in the breakpoints noted in the pictograph: 3.0m (for large tides) and either 0.6m or 1.2m (for small tides): do these numbers represent a floor, or a ceiling? Is 0.6 the scope of the largest small rising tide, or of the smallest medium rising tide? It is unclear, and publishers seem to disagree on the answer. Therefore this table offers a default, and allows the user to pick an alternative. By default, parameters are set so a "large" rising tide is any tide with a scope of 3.0m or more. A small rising tide is a tide whose scope is 0.6m or less, and a small falling tide is a tide whose scope is 1.2m or less. The user is free to choose different breakpoints. If you click the "Salish Setup", you will restore these defaults.

The time indicates the beginning of the period to which the given Chart applies. The line for each tidal extreme (i.e., high or low) indicates the level of the tide (in meters), the change in level (scope) over the ensuing period until the next extreme, the duration of the tide in hours:minutes, the year and date. Any time the date changes, the date is included in the line. To avoid a cluttered appearance, only the time and Chart are noted on all lines.

Finally, note that these tables are designed to be easy to copy into a spreadsheet. You can select the entire page, copy it, and past it into excel. There should be no gaps - every row will contain at least an hour and Chart number.