Centretown Observatory:

As you may recall, I had an idiot moment in the wee hours of October 7, and accidentally deleted the entire Ottawa folder from my weather system. Twelve years' worth of accumulated hourly readings, and over eighty years of daily readings, were gone, along with all of the templates from which the Ottawa pages were generated, all of the GnuPlot specifications from which the graphics tables are generated—all gone.

I have now begun to rebuild the Ottawa database. To get the system going again after the deletion, I copied over Cornwall's records, so that I would at least have some data to work with.

With that out of the way, thank goodness that I had the processed Ottawa pages to go from. At present I am labouriously restoring the Ottawa pages:

• Copy the processed page into the Ottawa Templates folder, and add the ".wxtp" template suffix.
• Add the page specifications into templates.ini, which lists the pages to process, when to process them, and what to do with them.
• With the page, and the corresponding page for CTO, open in a text editor, copy the data-injection codes ("MetScript tags") from page to page, replacing the injected values.
• At some point, re-create the tables' specifications (again, should be able to borrow CTO's), so that generated graphics appear again.

Once that is done, I will wipe the Ottawa database; then I will have to capture and feed into the system daily readings going back to 1938, and hourly readings going back to 2012—which is when this station began recording readings from Ottawa. Fortunately, this process can be largely automated. It will, however, take some time.

Please—wherever you are in these volatile and dangerous times—be well.

-Bill

As many of you know, one of my passions in life is DXing—trying to receive distant radio stations; in my case, primarily in the AM Broadcast Band.

I began DXing on the first night of my arrival in the greater-Ottawa area. That was back in 1986. Fast-forward to 2024, where I have finally logged my one-thousandth distinct station. The pace of new loggings has remarkably slowed in recent years, due to three primary factors:

1. Over the years, the amount of radio noise has been increasing almost exponentially, in conjunction with our use of so many gadgets which use radio, or generate radio signals as a by-product of their primary function.
2. As we approach Solar Maximum, its negative effects on the AM band have been pronounced. This limits the 'catchment area' to semi-local stations.
3. I've logged so many stations in the Northeastern US and Southern Canada that the area is becoming 'fished out,' with few remaining stations. I estimate that I have logged about a fifth of all stations in North America.

-Bill

2024-10-26

This afternoon, we replaced the Primary Data-Capture server. It was becoming prone to random shutdowns. That particular computer had been donated to the CTO many years ago, and has served its role very well.

The replacement computer is a low-end laptop with a smallish Solid-State Drive. Running the latest Debian, performance is most impressive.

The old computer will now be scrapped, as it has always had a broken screen. Good-bye, gypsy3.

I'm pleased to say that the changeover was quick, as it had been planned out in great detail. In the end, we lost our live weather feed for no more than 15 minutes.

I've got a few things to do before I try to reconstitute the YOW (Ottawa) weather database. In the meantime, the YOW website will remain static.

2024-10-20

I'll bring you up-to-date on developments here at CTO.

Recently, I came to a realization that I had many Gigs of data that weren't properly backed up. So, I did a survey of all the places that data were stored, and then I wrote up a whole set of bash scripts to zip them up.

First, I did a baseline backup. And it was during this time that an unforunate coincidence of a frudulently-misrepresented thumb drive (causing a bad ZIP file to be created), coupled with insufficient fore-checking on my part, led to my deleting the YOW location. That means I've lost the airport website page templates, and the 80-some years' worth of data that I had painstakingly fed into the system to start with. It will take some time to rebuild the data, and some time to reverse-enginneer the page templates. In the meantime, I'm deep-scanning all of my prior backups, to see if I can recover any of that lost data; but, until then, the Airport Website will be stuck on Oct. 7.

The up-side to all of this is that I now have a backup system that does a monthly backup of all data associated with that month; monthly culls of stale and transient data; and a yearly backup of all data. Here are some stats:
- Baseline Backup was 92 Gigabytes
- Monthly Backup is 2 Gigabytes
- Yearly Backup is 14 Gigabytes.
I have a 64 GB thumb drive for monthly backups, and a true 1-TB thumb drive for Yearly backups. At this rate, I'll be good for a few years; and with storage costs always trending downwards, it'll be cheaper and cheaper in future to archive all of my data--and it will all fit into my pocket.

Some other notes:

• I recently refurbished my server, restoring several peripheral connections that had 'gone down' over time.
• I saw something of the massive recent geomagnetic storm (i.e. the Northern Lights). Best one I ever saw was in the late Eighties or early Nineties.
• I'm brushing up on my electronics. I was big into it during my teens, have always had a general understanding of it, and am itching to learn something new, because I have a few ideas.
• I intend to add a magnetometer to the roster of CTO sensors at some point in the coming year.
• I am having to replace the primary capture laptop, donated to the CTO some years back by The Powell Group. Something is failing, and it's increasingly prone to random power-loss events.
• Plans for the further evolution of MCH are being formalized and finalized. Among them: direct access to previous years, months, and days; inclusion of Sky-Brightness and UVI information into the dataset; facilities to make it easier for the curator to maintain the dataset; and a few minor bugfixes. Needless to say, the MetScript tagset will be expanding considerably.
• All telescopic observations ceased from the CTO location last autumn; the trees in back have overgrown almost all available sky.
• As the sun approaches maximum activity in its 11-year cycle, there are (on average) daily X-ray flares, bimonthly radiation storms, and fairly regular aurorae, all of which strongly affect mediumwave radio signal propagation. Additionally, our non-optimal urban location has seen an exponential increase in 'QRM'—human-caused interference—in recent years, which just makes it that much more difficult to DX. One look at the list of routers in the neighbourhood makes it clear that some of that interference will never go away. As a result, there have been times when even semi-local DX has been rare. I am stuck, for the time being, at 998 stations received in the AM Broadcast Band. I am, however, noticing an odd thing: while even stations from just over and across the border in New York State are absent, some of the clear-channel stations still come in. Examples from recent evenings include 700-WLW, Cincinnati OH; 750-WSM, Nashville TN; 840-WHAS, Louisville KY; 890-WLS, Chicago IL; 990-CBW, Winnipeg MB; 1040-WHO, Des Moines IA; 1120-WHO, St. Louis MO; 1170-WWVA, Wheeling WV; and to some extent 870-WWL, New Orleans LA. On any given evening, you may only hear one or two; but over the course of a week, you will probably hear all of them.
• I have refigured the antenna of the CYCTO beacon, to improve the radiation pattern and hopefully improve the chances of reception. One evening soon, as a fun learning project, I will traverse the neighbourhood, radio in tow, and map out the coverage area. Hopefully I can deduce the radiation pattern from the results. I will also try triangulation, to see how closely I can pinpoint the station from multiple distant baselines.
• The CTO Technology Museum is sad to announce that two of its vintage radios have had be to retired due to critical subsystem failure; the circa-1981 Realistic Pro-2008 Eight-Channel Programmable Scanner's flourescent display has failed; and our beloved 1988-era DX-440 direct-entry communcations receiver now refuses to turn on. We will hold onto both, in the hope that my burgeoning electronics skills may ultimately come to the rescue.

Enough for now.

2024-09-22

Our Internet modem broke down in early July, necessitating its replacement. At the same time, CTO underwent some fairly significant upgrades. I brought the site back online a couple of days ago. In addition to upgrading our weather sensors, and re-engineering our data capture software to make use of them, I've been working lately to restore functionality in the query-type pages. At this point, most of them are working again.

I have plans for MCH; I intend to expand the number of daily data the program will track; I intend to introduce a few new features that will make it easier to access data from past years, months and days.

I am slowly working on PacDB, v5; the general-purpose database engine. At present, I'm working on indexing.

In personal DXing news, I am up to 998 stations. I've been that close for months; this solar season is really cutting down on DXing.

-Bill

2021-12-07

First off, I have established the Demmery Software website, to facilitate the distribution of StationBase and other products.

Second, I am turning my attention back to MetCommand-Home (MCH), our first product. I'm going to tidy up how it stores forecast terms.

Third, I am beginning to work on a database-and-indexing module, informed by my experience creating StationBase. No timelines at present.

Fourth, I am going to do a little work on the site, to ensure that everything works as advertised.

-Bill

2021-02-01

I have news regarding StationBase. I have been entering my database of stations, and coding as I go. I now have all 841 of my stations entered. Now I can test the other functions properly, including the newest addition: radio bands, to help you filter through your various DX catches. Dates and times can now be entered in several formats, and all Combination Boxes (where you can type something, or select an item from a list) now process typed entries correctly. The log function is working, and the whole thing is stable. I still have a few functions to implement, and test; and a major code cleanup to do; but I am beginning to think that I can get this thing released by the end of March.

StationBase allows you to track your DX loggings. It allows you to record station details, and then enter loggings for that station. For each logging, it records both the location from which it was logged, and the set of equipment used to make the logging. StationBase will accept just about any frequency for a logging (from 1 Hz, to 9,999 THz). Frequency bands allow you to view stations from a single band at a time. There is some set-up required, but it's all automated, and geared to entering data quickly.

-Bill

2020-12-25

StationBase continues to come along nicely. I have managed to alter the databases (tables). StationBase now tracks the frequency as a two-part value: an explicit frequency value, and a wavelength type (Hz, kHz, MHz, GHz, THz). This is new and allows frequencies to be expressed in alternate ways. I've done the same with daytime and nighttime power, and eliminated the PSAPower field.

Testing to this point indicates that the changeover was nigh-perfect. The routines all just-worked, the first time. Let me tell you: that is a new experience for me!

My intent at this point is to continue to enter my station data. As mentioned, while the program is still in development, the core routines are stable enough at this point to allow me to use it operationally.

One thing I'm hot to add to the program is support for radio bands. In the release version, you will be able to specify radio bands, by frequency range, and then filter your station list accordingly.

I've given up on predicting an availability date for the program; it will be ready when it's ready.

To my Christian readers, Merry Xmas.

-Bill

2020-12-16

I had noticed for some time that wxCondition was not recalculating the Wind Chill Index when borrowing wind values from Environment Canada. Early this morning, I fixed it. Wind Chill values are now restored for CTO.

Make that version 2.22 for the utility in question.

-Bill

2020-12-15

The system suddenly stopped reporting local values at around 20:54 last night. I did some poking around:

• The Airport values were still being captured, and their webpages being regenerated.
• Values were backlogged in the staging folder, which meant something funny with wxCondition—the program that allows borrowing values from other sources
• Tried running the script that handles wxCondition. and had a crash where it's trying to borrow values. I had a hunch.
• Looking at the Airport data for last night, I can see that the 20:00 readings set had two usually-reliable values set to 'nil'. At that point, I realized that the program was trying to do something with nil donor values.
• Short-term solution: simply fill in the values for the airport data. The system went *snork*, and all the values were in the system two minutes later.
• Long-term solution: check for nil donor values, and halt their processing at that point. Done. We are now at Version 2.21.

In StationBase news, I recently discovered that I could make the program so much more useful by allowing users to specify a frequency and Power magnitude (ie: mW, kHZ, GW, THz) for each station. For that, I had do convert the station data--not an easy task. Now I realizethat I must also do that with log entries; I'm in the middle of that now. It will push back the release date, but I will have a much more useful program, capable of tracking all of my DX, and not just the AM band.

-Bill

2020-12-08

Ongoing work on StationBase is coming along nicely. I am using it operationally, now. I've fixed any number of bugs, and rewritten much of the code in response to the bugfixes. Ugh.

The big delay now is in getting the data input from my loggings spreadsheet. I have 800 stations to input, and about an average of four loggings per station. I'm almost at 300 stations now; it'll be at least a couple more weeks before I finish that task.

A pleasant surprise has been the homebrew database and indexing routines. They are working like champs! In the end, for anything other application that I may want to write in future, if that application uses databases, then I've got a package of routines ready to go.

Again, StationBase is for tracking DX catches (stations received over a long distance; hence 'DX'). Loggings are attached to station records, and both a location and an equipment set (list of equipment used) can be recorded. Stations can then be filtered according to location or equipment set.

I originally designed it just for AM DXing; though if you don't mind keeping your frequencies in kilohertz, it can track almost any station.

I am now anticipating an intial release in 2021, Quarter One.

-Bill

2020-11-24

This morning I noticed a bug in my setup and corrected it. As you know, the primary sensors give a false low reading for the relative humidity. I had been going with the backup reading if higher, and a value calculated from Environment Canada when not available. Unfortunately, this morning the backup sensors went down, and I was stuck with a reading of 1% for the RH. I've corrected that now, by substituing the backup value unconditionally; and I borrow from EC only when the backups are down.

-Bill

2020-11-22

I have replaced the batteries in the primary capture receiver, and rebooted. We have primary temperature figures again. As usual, we have no relative humidity; figures usually come from backup sensors, or, when they are down, are borrowed from Environment Canada.

So, at this point we have live readings for everything but wind velocity and bearing, and precipitation. Wind values are borrowed from EC, and precip values are determined manually.

The long-awaited StationBase is coming along nicely. I've had to fix piles of small bugs; but it is beginning to look good, and behave well; and I'm about to start using it as my primary station tracker (I currently use a spreadsheet). To that end, I am busily determining the latitude and longitude for the 817 stations I've logged to date. Release may well happen around the New Year.

In case you're new, StationBase is designed to track loggings of radio stations. A list of stations logged is the main feature. When recording a logging, the exact location, and even the equipment used, can be included in the details. The station list can be filtered according to location, equipment used, or through a custom search. There are some analysis tools, which suite will only grow in the future. I couldn't find anything that did what I wanted it to; hopefully someone else will also find it useful.

-Bill

2020-10-17

After a short period of rest, my backup sensors are finally working again. I'm calibrating backup data capture (having just dismantled, and then rebuilt, the backup data-capture stage), and I anticipate local readings again by this evening.

Just goes to show you...

-Bill

2020-10-12

I have given up on the backup sensor. Recently, the receiver ran out of battery; it did not revive with a replacement. Therefore, I have taken it out of the system.

Also, for now, the main temperature sensor is out, so I'm borrowing the value from Environment Canada. I suspect the problem is just a low battery in the transmitter; it's not so easy for me just to wander out and grab a couple of replacement cells. So there may be some delay in returning the sensor to service.

-Bill

2020-09-25

We recently lost the wind sensors. I speculate that the battery lost its capacity to recharge. Anyway, for now I'm borrowing the wind readings from Environment Canada. Also, I've been fighting with my backup sensors. I almost have them adjusted again.

Finally, I'm actively engaged with StationBase, my DX'ers station-tracking tool. I'm trying to use it operationally, fixing things as I find them. I'm guessing it'll be ready for a beta test around November, contingent upon my day job's not getting crazy again.

-Bill

2020-07-23

Oh, man, it's been hot.

A quick peek at This Month's records show that this July is on track to break the record for the warmest July. So far, the average high temperature has been 34.3°C. The warmest month on record is faulty, being ascribed to June, 2018—a time when the CTO wasn't even measuring outside temperature.

Anyway, it's been hot, and mostly dry.

Be well,

-Bill

2020-07-12

We had a brief outage yesterday, as one of the cats used the data-capture stage as a jumping-off point to greater heights.

Luckily, it was early in the morning; and it took just minutes to fix up.

Be well,

-Bill

2020-06-05

A couple of days ago, we had an outage of the wind vane and anemometer. This meant that all wind data are missing for about two days. The instruments came back by themselves; they may repeat the performance.

I am updating the documentation for mcOCR, and will include it in the download packages, which are now available through the Portal Page. Current version is 3.29, beta 8.

That's about it, for now.

Be well,

-Bill

2020-05-30

I have officially released mcOCR, version 3.29 beta, in binary packages for Windows and Linux environments (64-bit). Each download includes the binary program, configuration files and sample data for two different instances. Have fun.

The production release is a few weeks off, at which time I will also release the source download.

You can reach mcOCR through the portal page, MCH Suite of...

Be well,

-Bill

2020-05-29

The other day, I missed a call asking for information on viewing Comet SWAN. That just never happens. I did try to return the call.

In case the caller visits again, here goes:

First off, it's best to use some visual aid. If you have a pair of binoculars, use them. Second, if you have an unobstructed view of the south, that's better. If it's not already ruined with light pollution, then that's even better.

The best time to go out is around 21:15, local time. It's getting dark by then—and to get a good view of a comet, you need a dark sky.

Okay, so you're standing around outside, wondering what to do next. That involves getting yourself oriented, First, stand facing the point where the sun set a little while ago. That's Northwest. Southwest, where you want to be looking, is to your left. Face it.

What you'll be looking for, specifically, is the constellation Leo. There are plenty of resources out there on what Leo looks like, etc. Use them. Specifically, you're looking for the bright star Regulus; and, lucky for you, Leo actually looks like a crouching lion. It should be about a third of the way up the sky. Leo will look like it's ready to leap into the ground (ie: it's leaning forward).

Now that you're oriented, you can begin looking for the comet. Scan to the left of Regulus, out to a couple of handwidths (at arm's length). You're looking for a little hint of a fuzzball—a star that looks bigger and fuzzy. In the binoculars, a hint of a tail might be seen.

This is going to sound odd, but a trick that astronomers use to view really faint objects is "averted vision," where you look away from what you're trying to view, and catch it in your peripheral vision. Sounds nutty, but it works. Usually. Sometimes. Just try it.

That's about all I can say on the subject. There are some good viewing resources out there; just google "comet swan".

Be well,

-Bill

2020-05-20

We had a brief server outage on Monday. Too much current on a domestic circuit. Luckily, I was present at the time, and had everything back up within five minutes. The only thing we've lost is logging statistics; everything else looks fine.

I'm into final "nitty-gritty" testing with mcOCR, and will release it as a beta on Friday. I'll try to release both the source and a binary package.

In weather news, the cool start to this month seems to have wrapped up, and we'll mostly make up for it as the month of May closes on a very warm note.

In unreleated news, eight or nine years of painfully limping around, due to a bad left knee, have come to a head, as I can hardly walk at the moment. I believe I now have a much better handle on the problem—not just "arthritis", and I'm going to seek medical attention for it. I believe my problems can largely be fixed with a simple bit of arthroscopic surgery.

Be well,

-Bill

2020-05-15

mcOCR, version 3.29 beta—the application which captures weather data from an LCD display panel—is about to be released. It's already deployed on the backup-capture server, and running well. I do have to tighten up change detection in the settings pages; that won't take long.

I began working on this version last fall. I just got lazy for a while (I had a big commercial project that required all of my effort), but I'm back on-track now.

I've stlll some further plans for the app, including making it easier to adjust character and segment contrast values, and adding a few more control statements to the .ini file. But that will be some months away; my attention is now turning back to StationBase. I'm also planing some minor tweaks for the weather server; nothing major.

Be well,

-Bill

2020-05-12

I've updated the Home Page a little further. I've jettisoned the national radar image, and replaced it with the latest VHF/UHF activity forecast map, which links back to its source at William Hepburn's DX Info Centre. Used properly, it can help in predicting Very (VHF) and Ultra (UHF) High Frequency signal propation, as well as the potential for ducting events.

Okay, I've been talking it up about mcOCR over the past few months. Version 3.30 is now just a day or two away from going into beta. My bug list has a single entry left, just a combination-box item that disappears when selected. I've fixed a number of bugs with this release, including that the extra-sky timer had stopped working. That is now fixed; and I've really tightened up the performance of the Character Properties screen, particularly in the way it handles control-scan-area settings. New features include a 1500-entry scan-history buffer, for each of up to four locations. Each entry is editable, and you have the option, when saving, of regenerating the output files for the edited entries.

I've updated my backup and retention policies, and have been implementing the changes on the three computers involved in the CTO's operation.

In actual topical news, the early spring weather here in Ottawa has cooled off. Here at the CTO, you can clearly see where this month started out above the recent average, then cooled off from there. Patience, please. I see that today's forecast high temperature is still on the cool side, and it's possible that we'll set a new minimum record in that category, the existing coolest-high for this date being 11.9 degrees, in 2013. It then begins to warm up, and we should have a return to normal temperatures by the weekend. Beyond that, the chance of El Nino occurring this year is low, and indications still are that we will have a decent summer.

Oh, and I can't resist taking another jab at the climate doubters, who took glee in saying, "Look—it's snowing on Mother's Day! So much for Global Warming!" Well, I have a shoulder-tap for you. On average, here in Ottawa, it snows one in every three years, in early May. It used to be even more often. So, chances are, you've been saying the same thing every three years—you might as well be saying "January" instead of "May"—so take a look at the averages and shut up, already.

Be well,

-Bill

2020-04-12

I've updated the Home Page. I've dropped the satellite image of the Atlantic, and replaced it with an image of the Planetary K-Index. This is an index of geomagnetic activity ("space weather"); when the line start climbing into the orange and red zones, the possibility of enhanced VHF DX is increased (i.e. FM bands, sometimes TV).

In other news, mcocr, version 3.30, is near to being ready. It includes a very useful history feature, where you can go back, edit previous readings, and have the output files re-generated. I'm also working on some bugfixes. More news soon. Capturing weather readings from an LCD display just gets easier and easier!

Apart from that, not much is going on at the moment. We're locked down for COVID-19, and I'm going a bit stir-crazy. Have to make the most of my daily walks.

Be well,

-Bill

2020-03-12

We had a bit of a disaster yesterday. The cats got up into the data-capture stage again, and in the process of readjusting, there was a problem, and some of the .ini file was lost.

Thankfully, there was an older backup of the .ini file. It's been updated, most of the data have been captured, and the system is settling down into its routine again.

The big takeaway on this one: backup your .ini file. I now have.

Be well,

-Bill

2020-03-01

Let the record show that this March came in like a lamb.

The forecast for the coming week speaks of flurries and showers, and daytime highs at the airport will all be above zero, starting tomorrow.

Blasting was conducted yesterday on the Rideau River, just upstream of Rideau Falls, to clear the way for the annual river ice breakup. Another sign of impending spring.

The Rideau Canal skating season has ended. This year, for the first time, some segments of the skating surface could not be opened. A trend.

Sunset has come back about an hour and a half, and sunrise has come back an hour, from their limits.

The spring time change is in one week (Sunday, March 8). Don't complain to me—I'm a huge fan.

Lastly, Spring begins on March 19—less than three weeks away.

Be well,

-Bill

2020-02-16

We are just at the cusp of everything changing. We are about to enter the two-month period of greatest change, during which everything—temperatures, length of day, solar elevation at Noon—changes the fastest. At present, the sunlit portion of the day is lengthening by about three minutes daily. Sunrise has come back nearly 45 minutes, and sunset more than an hour. To top it off, I am expecting an early spring and a warm, though wet, summer.

In about another week, the south-facing sides of snowbanks will begin to acquire that rotten look, whenever the sun shines. The spring time change is less than a month away (and don't you dare whine about it to me; I am completely unsympathetic with people who are so hidebound by routine that they have become intolerant of any variance; how can anyone live like that?). Even now, on sidewalks which are lower than the adjacent yards, ice is forming overnight from meltwater.

The sun is getting higher, and soon will begin to cast shadows that are more vertical than horizontal; and spring will be here is just over a month.

The average temperture is now rising daily, though from the way we calculate it, it's difficult to see. Take a look at the temperature graph for this month.

In other news, I have mcocr (the program that captures the readings from an LCD display) again adjusted just-so, and data capture again is near perfect. I am working on mcocr in the meantime; I'm implementing a history feature, which allows editing and re-issuing of older data, with a buffer for about five days' worth of readings, at five-minute resolution. I'm taking my time. Once that is complete, I will add user-configurable tolerance settings, to reject large changes in any given measurement. And, by that point, mcocr should be a nearly perfect, research-grade, data-capture tool, suitable for full public release.

One concept that I have been toying with is an auto-calibration mode. Perhaps colour dots could be affixed to four corners of the display. On the next image read, the dots could be found again and, if they have moved, the positions of the characters could be recomputed, relative to the dots. Nothing too tricky; it's just basic arithmetic. I might try to implement it in a future release.

-Bill

2020-01-31

We had a data outage for about eight hours, overnight. The cats got up onto the data-capture stage and nudged the camera.

Fortunately, once it was discovered, it was easy to adjust the camera, and image, and recalibrate.

-Bill

2020-01-03

Perihelion is in two days. If you're unfamiliar with the term, perihelion is the point in its orbit where the earth is closest to the sun. This is a direct consequence of the earth's orbit's being elliptical.

There are some ramifications, though if they're not pointed out, they tend not to be obvious. For example: our 24-hour clock is based on an average day, and the assumption that all days are of equal length. But they're not. When Earth is farthest from the sun, our orbital speed drops a little. This causes the sun to 'lag behind' where it should be in the sky. At other times, Earth is moving slightly faster than the 'average', causing the sun's apparent motion through the sky to 'run ahead' of where it should be.

At perihelion, Earth is moving fastest in its orbit, and so the rate of change is the greatest. Incidentally, it is precisely this 'speeding up' and 'slowing down' that are responsible for the fact that the earliest and latest sunrises and sunsets of the year do not occur on the solstices (for example, here in mid-Northerly latitudes, we've just passed the latest sunrise of the year—well after the winter solstice).

Another consequence of our elliptical orbit is that it is symmetrically 'shaped.' The sun is at one focus, but it could be at the other, and the orbit would be the same shape.

One more fun perihelion fact: that perihelion occurs in January means that the southern hemisphere receives slightly more solar radiation than the northern, in summer, and slightly less, in winter.

In other news, it's been, all in all, a mild winter season thus far. There's been a bit of freezing rain—unavoidable these days—but generally it has been mild. I would expect that general trend to continue, and we could be in for an early spring this year.

Wishing you a successful 2020!

-Bill