AmateurRadio.com

We often read that POTA is the “fastest-growing activity in U.S. amateur radio.” I’ve said it, too. Several times. I think satellite operations are a distant second growth activity. But do we actually know much about POTA activity growth through empirical data? My impressions are based on guesstimates from data on Canada where I analyzed a national survey of hams conducted originally by RAC. In 2001, some 37% of those Canadian hams said they engaged in portable operations, including POTA, SOTA and the like. It ranked 9th among 37 common ham activities in the survey. They reported to the tune of 15.7% participating in satellite operations. Alas, we do not have parallel data on U.S. hams so these are inferences made from comparative data. But data nonetheless.

It is not unusual at all for most of us to gain our impressions of ham radio activity through social media. Heck, there are Youtubers who say they make their living from their channel(s) and associated activities! Those who do not work at it full time or don’t even monetize their outlets add significantly to the mix of media impressions. So if POTA activity gets a very large share of social media output, our best guesses are shaped by those impressions in the absence of empirical data themselves. This applies to things other than amateur radio, like politics. You get the point. In a vacuum of data, we get captured by social media.

To illustrate, here’s a Google Trends result from using both “Parks on the Air” and “POTA” as search terms with a five-year window ending circa January 25, 2025. In the dual trend lines, POTA wins as a search term but both show a marked, continuing, and slightly growing interest through search terms on Google. The brand “POTA” seems to have currency in Google Searches as a frequently-used proxy for all social media output. Thus, social media impressions are strong in this one!

We need data to complement what we see through social media. In this article, I present a brief snapshot on POTA activity among U.S. hams. I use data actually reported to the POTA website to describe where all of the current U.S. POTA sites are located. I also examine 2025 annual data for all activators as well as all activations reported to their website. I have not seen any research like this reported by the POTA administration or other hams. If readers do see what I’ve missed, I’d appreciate receiving the link. I’m good on QRZed.

Here’s a snapshot of key POTA elements using these data. I use maps and basic statistical tools to help describe patterns in POTA activity. The results are organized into POTA sites, activations and activators in 2025. I think the reader will gain much greater insight into the POTA program activity since inception in 2017 through these data.

POTA Sites Across the U.S.

Owing to the origination of the POTA program as a continuing expansion of the National POTA program by the ARRL, these official sites include national parks. These favor undeveloped areas (and cannot be developed without permission) outside of urban centers. POTA adds many, though not all, state-controlled parks and similar preserved entities. They add or remove POTA entities on an ongoing basis. As of December 2025, the map below illustrates the locations of all POTA sites in the U.S.

The map shows definite visual clusters of sites in the Northeast. Although not shown, this region also has a high density of licensed hams. Florida, Missouri and Utah also have clear and distinct concentrations of POTA sites. These are followed by Michigan, Washington, Arkansas and California. These states have internal clusters within the state characterizing POTA site distributions. Note that these are both Federal and state-controlled parks and related sites (wildlife management areas, etc.).

The data show that POTA sites are far from uniformly distributed across the U.S. Given the basis for founding the POTA program on the sunset of NPOTA by the ARRL, we would not expect them to be for U.S. Parks alone were founded to be undeveloped protected areas away from urban centers. This map puts a finer point to where official POTA sites are located. In future articles, I’ll show more details at finer spatial scales on the characteristics of POTA site locations.

POTA Activations

While POTA sites are all over the U.S. and not uniformly distributed, how frequently each one is activated by hams is far, far less uniform in nature! I took each POTA site’s lifetime activation numbers and put them on the map below. After examining a histogram of the distribution (not shown here), these intervals were placed on the number of times each park was activated: 0-1, 1-1000, 1,500-2,500, and 2,500-5,381 (highest). They seemed to reflect the observed cut-points in the activation data.

While most parks have been activated at least once—although difficult to see on this scale—I emphasized parks by size of the circle, colorizing each category. Those with the largest circle shown in red are the most highly activated POTA sites in the U.S. Where are they? Mostly along the Appalachian Trail, near the popular vacation spot of Orlando FL, in the Midwest, and one in the Northwest in Washington. Those one step down in size shaded in blue tend to follow this same pattern with the exception of one in Colorado.

These results show that there is some linkage of POTA park activation frequency to distance from the market of hams for activation. In consumer economics, the “friction” that distance to any consumer choice will generally favor shorter choices over longer ones, especially over time. This tends to affect the frequency of consumer choice such as which POTA park to activate. The map of most popular activations in the POTA program is consistent with this long-standing consumer behavior paradigm.

Friction of distance in consumer behavior represents the time, effort, and financial costs required to access goods or services, causing a reduction in interaction as distance increases. This concept drives consumers to minimize travel, favoring closer options, which impacts store choice, purchase frequency, and online engagement by reducing cognitive and physical load. 

The reader might look at this map and think that this is not clearly reflecting this phenomenon. While it does depend on the market of hams and that subset who participated in POTA since the 2017 inception, it is a fair question. I’ve put another visualization of how unequal the distribution of activations across parks in this program is from 2017-2025. The curve in the graphic below reflects the following. The number of activations for each park was sorted from highest-to-lowest with the cumulative percentage computed for each. The ranks from 1 (most activated) to 11,996 (least activated) came from the original sort. The cumulative percentage of the total activations that each park represents is plotted against the rank. (In statistics and economics, this is a type of Lorenz Curve.) This curve tells us just how much of the total lifetime activations are reflected by each park. If each one had an equal share, the line would be flat and straight from zero percent, rank 1 to 100 percent, rank 11,966 (number of POTA parks).

I have placed three lines noting the share that the top 100 parks (green), 500 parks (blue) and 1,000 parks (red) are of all activations ever reported to the POTA program. The green line intersects the 20 percent line of all activations. In other words, 100 parks account for one-fifth of all activations. The top 500 parks account for some 40 percent. Finally, the 1,000 top parks out of the almost 12,000 POTA sites account for one-half (52%) of all POTA activations since the program began in 2017. This is a highly unequal distribution demonstrating the dominance of a small portion of POTA sites in total activations.

POTA Activators in 2025

Who are POTA activators? This is in intriguing question as social media profiles suggest they are significantly comprised of hams who face housing restrictions on home antennas outdoors. Not all of them, but enough to make a clear inclination toward portable operations in the activity space of Parks on the Air^(tm). I’ve shown two pie charts below to illustrate the composition of POTA activators in terms of license class. On the left are all amateur licensees in States (no territories) as of December 2025. On the right are POTA activators in 2025 using U.S. call signs, individual licenses (no clubs).

Note: I always remove those licenses that have reached their expiration date. The FCC ULS data system does not. This is for their convenience since an expired license is dormant but can be recalled for activation by the operator within two years. The FCC database management team simply keeps this license record in the database in case of that recall. This is for their operational convenience. This often confuses hams who just download the ARS data from the FCC ULS system and compute totals without fully understanding the curation protocols in the data themselves!

As has been made widely known, Technicians are the largest operator class in the U.S., here over 49 percent. Generals are next at about one-quarter with Extras comprising about one-fifth. The dormant Advanced Class (about 4 percent) and Novice Class (less than one percent) round out the distribution of operator classes at the end of 2025.

POTA activation is dominated by Extra and General Class licensees. Over half (57%) hold Extra Class licenses with almost forty percent being Generals (39%). Only two percent (or 257 hams) reported activations in the POTA program. Another two percent (rounded to the percent) of Advanced tickets participated in POTA park activation. These data show that it’s an HF-privilege game, largely, as only 257 Technicians used the meager HF privileges they have to activate a POTA park, unless they were using other modes facilitated by their band privileges. It’s almost an all-General/Extra affair in POTA at this point.

Some POTA activators are far more active than others. To illustrate this, consider the box plot shown here. The average number of activations in 2025 was just over 23 (mean = 23.4) but ranged from 1, the minimum, to 3,396! The median number is 5 with the mode just being a single park activation (the mode being the single most occurring number). The distribution of individual ham POTA activations, not unlike what we saw for parks, is highly skewed. This suggests that the social media haze of the fastest growing activity might be based on the extreme number of activations by a small number of hams.

This box plot represents the percentile distribution of the number of total activations by each POTA activator reporting in 2025. As the bottom text details, out of the activations last year, one-quarter reported 2 activations (Q1, or lowest quartile). One half reported 5 activations (median). Some three-quarters activated 19 or fewer POTA parks (Q3 or third quartile). Those activators reporting 20 to 3,396 are classified as extreme values. I’ve put a black rectangle around the hams whose number of activations reflect statistically extreme values in this distribution. They are shaded in blue. These extreme activators contribute a lot to the POTA program while the vast majority of portable operators reporting to the POTA program reported less than 20 last year. Do the extreme activators garner the bulk of the social media presence regarding POTA activity?

To flesh out where these extreme POTA activators are located, I’ve created the map below. There are light green and dark green points. Together, they represent all 2025 POTA activators reporting to the program. The dark green points are the license locations of these extreme POTA activators. They are concentrated along the Appalachian Mountain trail and environs in the subregion. They are scattered throughout the rest of the U.S. in small clusters within states we have mentioned before. I’ll demonstrate more below. But I want to emphasize that the dark green points reflect hams who repeatedly activate POTA sites in extremely higher numbers than the vast majority of POTA participants.

To help the reader see the connection, I have put all POTA activators on a map with the most activated POTA sites as shown above. This map also has urban and metropolitan areas designed on the base map. This will become more important in future articles on the POTA program. But for this snapshot, it confirms the role that distance to POTA sites plays in the frequency of activating any POTA sites.

In an analysis not shown here, I estimated a spatial regression model (spatial lag specification) predicting the number of activations by the distance in miles to the nearest POTA site for the ham operator. These are for those hams who participated in POTA during 2025. The results show that for every mile further the nearest POTA site is, the ham activated one-half fewer times during the year (i.e., a reduction of 0.5 activations per mile). I’ll study this more in future articles but suffice it to say here, where hams are located does affect the choice of parks and the number of times they activate them.

Thoughts

This snapshot provides the first look at the POTA program from a national scale using data rather than social media impressions or conversational anecdotes. What do we see?

There are a small number of POTA sites that account for at least half of all activations since the program began. Not surprisingly on the heels of this finding, there are a small number of extreme activators who account for a significant share of last year’s POTA activations. These extreme activators are scattered throughout the same regions as the most activated parks. Perhaps I will examine those patterns in a future article but the empirical fact remains: POTA is dominated by a small share of extremely active operators, almost all of whom hold General or Extra Class licenses. The largest license class, Technicians, are just not part of the game in this activity space in 2025.

As in most consumer behaviors, distance to the “product,” here a POTA site, shapes the frequency of consumption. My initial regression model showed a one-half activation reduction over the year for every mile that the nearest POTA site it to hams who reported activity during 2025. This result begs for further analysis. Has it been this way since the beginning of the POTA program (2017)? Does the ham’s location in the rural-to-urban residential continuum affect this “friction” of distance? Perhaps we will find out in future articles.

This descriptive snapshot gives the reader a clearer picture of where POTA sites are and which ones are most popular for activators. But the majority of activators pale in comparison to the extreme activators. The median for 2025 was just 5 activations, illustrating that one-half reported less than five with the other half more than five. The mode was a single activation. Are the social media reports indicative of these “small timers” or of the extreme activators? Well, watch social media portraying POTA for yourself. There could be a lot of experimentation by hams in 2025 who just did a POTA activation to see if it was for them, then decided it was not. Or they just did it as a social activity with ham friends and the schedule precluded more activations than just one or a few last year. We don’t know. But it appears that longer trip distance to activate a POTA site might largely eat away up to a half day, something that many amateurs would not have very frequently available for this activity. These data are consistent with that pattern.

Clearly the initial finding that 2025 POTA activators activated less as the nearest official site was further away begs for additional analysis. Distance requires an increased time commitment. Those on social media who both frequently activate POTA sites, especially serially in a “rove,” devote a significant amount of time, money and preparation in doing so. Nothing wrong with this. But how common is it within the market of amateurs interested in POTA operations? Not very, these data suggest. It begs the question of where are most ham operators located relative to existing POTA sites? Are they too far away to be practical for their sphere of obligations to activate official POTA sites very much?

I’ll explore this avenue in future articles because it is important to gain our impressions from data rather than the pantheon of what we see on social media, particularly where the outlet is monetized on the content creation surrounding their POTA activity. We reported results suggesting that it may just be a small group of all activators who live in some favorable propinquity to POTA sites themselves or who monetize their time expenditure for POTA content creation (i.e., they are “working”). This leaves out many hams because of the choice of sites by the POTA program and the legacy of National Parks as the foundational base for the program itself relative to where they live.

I’ll close with the observation that Technicians are only two percent of the reported activators in 2025. It seems a shame that these licensees don’t have more of a place in this activity space. Getting 2 meter simplex working, for instance, would be very challenging for many POTA sites. Are there other viable options to engage the largest group of hams in the U.S. license classes into more portable activations in parks? Yes, SOTA is one. It takes SOTA sites with significant prominence to facilitate such contacts that could count as a two-fer (SOTA and POTA). SOTA sites are not uniformly distributed to offer universal access either. It seems a wasted opportunity to have POTA sites that do not overtly engage Technicians to participate at a higher rate than merely two percent of all activations in the most recent year. I’ll look into this in future articles.

Listeners to the ICQ Podcast where I appear monthly as a Presenter probably heard me say that I was successfully treated for a highly aggressive prostate cancer (adenocarcinoma, Gleason 8) almost two years ago at the Mayo Clinic. While in Rochester for two weeks, I had a lot of time in bed recovering from the surgery before being released for home back in Mississippi. For me, I tried to keep my mind on things besides the cancer treatment as I had the top prostate cancer hospital in the U.S. treating me and the top robotic surgeon, Dr. Igor Frank, at the helm. So what does a ham think about in this circumstance? Let me tell you…

Antennas! I worked out plans for a half dozen portable antenna designs that had been smoldering in my brain before the unexpected diagnosis and biopsy. Don Field, Editor of Practical Wireless, expressed an interest in first dibs on each manuscript resulting from my experiments with portable HF antennas. Here are a few that are published or in-press as of this writing. And more to come.

Eiffeltenna

Based on the almost whimsical Youtube video by Jim Heath W6LG (now sk), I considered the photography lighting tripod, ubiquitous in the camera industry. Jim put together a quick-and-dirty (unusual for him) vertical antenna based on a type of tripod construction that electrically isolates the three legs from the telescoping vertical part. Brilliant! They are inexpensive so I bought a few from Amazon or eBay to experiment with. I paired the tripod with a 17′ telescoping whip and an inductor coil, finishing off the “fine French dining” concept with a tablecloth underneath of Faraday Cloth. Following my focus on making each section having a low resistance electrical connection when extended, it works very well. Very quick to set up, take down, and pack. It appeared in the October 2025 issue of PW. Bon appetit!

Delta Vee AutoLoop

Loop antennas of all designs have fascinated me since I was a teen building classic ones for AM BCB DXing. I have a sort-of horizontal loop around the edges of my roof due to HOA restrictions. )I have also written about that in PW-see March 2023 issue.) When Chameleon released their Tactical Delta Loop, I took a look at my friend, Steve KM9G’s take on it (see Temporarily Offline on Youtube). He found it was very flat across most HF bands. Hmm. Lawrence Cebik’s earlier models of loops, including the Delta geometry, showed that it has harmonics based on the design band and that height above ground places a significant part in the feedpoint impedance. What antenna magic have the Chameleon engineers come up with? Whatever it was, it priced out at over $400 USD.

From watching TO’s channel and the brief review of Michael KB9VBR, it appears that they use a fixed 5:1 balun to get the impedance down to the 50 ohm ballpark. I don’t understand TO’s flat SWR but antennas nearly on the ground can do funny things. So my take on the problem here was we can choose to optimize the balun wind for a certain band with some higher band harmonics working against a tuner for a reasonable match or just have a single band antenna. But wait. Why not use an ATU at the feedpoint to optimize the match across most HF bands? This only becomes practical with a light-weight battery-powered, RF-sensing ATU mounted directly at the “bare wires” from the loop’s geometry. That’s what I did here to create what I call the Delta Vee AutoLoop. I use a $40 used surveyor’s tripod without the head made by Manfrotto. It was purchased at a local electronic recycling center but I bought another on eBay for the same price. The head is what surveyors want so one without it is far cheaper. It has a standard bolt that I connect to an adaptor for the mount point as explained in the PW article.

Randy K7AGE says he’s building one to use on his parked truck when out doing POTA or other portable operations. This antenna appeared in the January 2026 issue of Practical Wireless.

Random Copper Stick

I had been puzzling over these carbon fiber masts since they came out. My friend George KJ6VU cautioned to me that they wouldn’t load up and they played havoc with his Packtenna wire antennas if they come in contact with them. So, bah humbug! Then I ran across Ben VE6SFX’s Youtube video on an experiment with on using Faraday tape on the outside of the carbon fiber mast. He said it worked as a random wire type of antenna! So, after working my brain for months on an angle for this, Ben’s experiment gave me a direction.

The Random Copper Stick was built by using a carbon fiber telescoping mast and copper tape—both 1 inch and 1/8″ widths—applied from the bottom to the top for each section of the disassembled mast. The very top section was removed as it was way too small to hold the tape. This gave me a reasonable length for a random wire antenna. My experimental measurements with a 17′ wire (20 meter measurement) showed a fairly clear phenomenon that Ben didn’t get into in his Youtube video. As shown below in a slide from a talk that I give to clubs on these experimental antennas, the carbon fibers disturb the relationship between the physical and electrical lengths of the wire as they are closely connected to the mast. (Score one for George KJ6VU’s observations!) Thus, using a 9:1 Unun with the mast works very well with an ATU at the transceiver. To quote my lawyer friend, Mike N5DU, I was shocked and amazed at how an antenna that I was convinced would not work turned out so well.

If you fancy taking a walking stick to operate portably, consider the RCS. It’s very easy to build, back and put up. I built one for my friend Scott K0MD to take with him on his trips to pair with his Icom 705. My article on the Random Copper Stick appears in the February 2026 issue of Practical Wireless.

Wave Caster Vertical

It is scheduled to appear in the April issue of PW but that should coincide with Spring Break weather at the beach. You can build the Wave Caster to Hang Ten while working some HF bands, lol. This one is also very easy to setup if you have anything in the portable site environment to clamp the mount. in the Wave Caster, I use an idea that Chuck KK6USY published on his Youtube channel in a coujple of videos. Particularly, he solved the problem of winding an antenna wire around a spool without it getting too much stress and eventually breaking. Chuck used a small resin reel with a ring terminal to solve this problem. Well, that solved a problem I wrestled with on ways to multiply antenna designs using these inexpensive carbon fiber masts with wire that wouldn’t take much time for the portable operators. (Not all have a half day to go to a POTA site that can be far, far away with a lengthy setup.)

With a Super-C photography clamp, almost any sturdy surface can be used to “brace” the vertical so it stays up while the temporary operation plays out. Just don’t forget it when you leave because I can tell you it is easy to do! This one is scheduled as of this writing to appear in the April issue of Practical Wireless magazine.

I have a couple of other designs that I am finalizing as the wild weather in the South permits. They may appear in PW but I’ll post a notice here if they do.

My time at Mayo Clinic was well spent. I cannot overstate how positive the medical treatment there was. Mayo treats 26,000 prostate cancer patients per year and have been rated #1 for many years. There was never a moment where I felt like I was a number on a lengthy list. I was fortunate to get connected with Dr. Igor Frank there as the “top gun” on robotic prostate cancer removal. I gave him a small momento as an expression of my appreciation as I was discharged. I understand that it may have made an appearance at the Department Christmas Party.

Change has pulls and tugs toward the past and forward into the future

We hear the word culture bandied about all the time. It means different things to different people. Social scientists define it as follows:

As a sociologist and statistician, one thing I’ve observed about amateur radio in the U.S. is that culture has a strong pull towards the past. We often hear this coming out when hams begin some comment with their “tenure” in the hobby: “Well, I’ve been licensed X years and I know…” Substitute your own number of years for X in this sentence. The listener is supposed to genuflect toward this tenure in the hobby as containing superior knowledge and wisdom. Social linguists call this “indexicality” to indicate what reference is being used in the argument. (For bench testers, think “reference plane” in VNA calibration.) So many amateurs “index” their understanding of the hobby relative to when they were first licensed, especially if it was during their teen years. That understanding “indexes” everything that comes afterwards and results in much of the verbal conflict on the air and on social media. Or, in person, to the astute observer at hamfests, lol.

The power that such indexicality has on the hobby is related to the demographic composition of amateur radio at this time. The demographer Ron Lesethage documented how the age composition of a population is related to the values for having children, a predicate for population replacement. In most all developed countries, child-bearing age women no longer see giving birth as an important part of their future. This is a clear historical change with respect to how they define what is important in their lives as women, unlike their mothers and women in most previous age cohorts. This “index” by women in developed countries is an example of the power that these belief benchmarks have on society. With the dominance of Baby Boomers in the hobby, is it any wonder why so the view of many hams in terms of technology is much closer to the Spark Gap than it is, say, using digital modes like FT8?

What does this have to do with a Spark Gap transmitter?

It is a metaphorical reference point to the technological origins of the wireless which, by definition, was amateur in nature. If we consider an automobile as the ham collectivity passing through time, at what point should the Spark Gap transmitter leave our rearview mirror as a guidepost? Is that vision a stifling tug against our speed toward what is visible through our front windshield? Many historians and innovators say yes it is. Here’s an AI-assisted image to illustrate what I mean.

So many new innovations are before us. But so many index our progress against the earlier periods more adjacent to the ancient Spark Gap transmitter. This is a continuum, of course, but listen to the naysayer commentaries on recent innovation. Why are they “bad” for the hobby? Are they “killing” amateur radio? I’ve noted previously that change is often labeled as pending death to those whose indexed standards are in decline. Right now, it’s those of the Baby Boomers as they hold positions of power and influence in the hobby while they progress toward Silent Key status.

Remembering the cultural origins of an activity like ham radio should become history at some point. That is, every new initiate shouldn’t be held to norms of beginning at the beginning. It is thought that an overemphasis on history can stifle innovation. This doesn’t mean that history should be forgotten. This is not a binary argument. Quite the contrary, there is a balance and a place for history. New initiates into the hobby shouldn’t be pushed toward the historical beginning but acquire it as part of the present and immediate future’s innovation. The latter is often what draws newcomers to the hobby space.

What does an emphasis on the past do for progress?

There are several elements to the detriment of an over-reliance of early history as a main part of the culture of a group like amateur radio. An article on this by the Thought Lab says the following about it.

How an overemphasis on history reduces innovation

• Risk aversion: When experienced professionals rely too heavily on “how we’ve always done it,” they become hesitant to venture into the unknown. The fear of failure can stop the pursuit of unconventional, and potentially groundbreaking, ideas.
• Limiting frames of reference: Extensive experience and historical precedent can create a mental model of what is possible, making it difficult to conceive of entirely new possibilities. In this environment, alternatives and fresh perspectives are often overlooked.
• False confidence in expertise: A deep knowledge of the past can create an “illusion of expertise” that leads to overconfidence. This mindset can close people off to new information and different approaches, stunting creative growth.
• Subconscious bias against novelty: Research has shown that many leaders have an unconscious bias toward familiar, established solutions, especially when motivated to reduce uncertainty. This bias can cause them to reject new ideas, even if they outwardly claim to want creative thinking.
• Misguided strategic choices: As seen in Soviet technology policy, an incorrect assumption about the historical trajectory of innovation can lead to big, irreversible bets on the wrong path. Instead of building on existing strengths, leaders may shift their focus toward an ineffective strategy, weakening their own sector. 

How a balanced understanding of history promotes innovation

• Learning from past successes and failures: By studying the history of an industry, innovators can see what has worked and what has not. This prevents the repetition of past mistakes and allows for the identification of successful strategies that can be applied in new contexts.
• Understanding complex origins: History reveals that modern innovations are often built upon a long lineage of prior technologies and discoveries. This understanding gives innovators context for where to focus their efforts and avoids a simplistic or misleading view of progress.
• Questioning assumptions: Historical perspective allows innovators to question entrenched narratives and conventional wisdom. It helps them re-evaluate their beliefs by comparing them to a wide range of past scenarios, which can lead to new insights.
• Gaining resilience: Studying how past innovators and companies overcame challenges can inspire a more resilient approach to obstacles. Instead of seeing setbacks as reasons to give up, they are viewed as a necessary part of the journey.
• Revealing long-term impacts: History helps put the ethical ambiguities of new technologies into perspective. By observing the unintended consequences of past innovations, creators can better consider the potential long-term risks and societal impacts of their work. 

Readers can identify these issues within the hobby by just reading and listening for a bit. National and local organizations are legion for this “we have always done it this way.” Witness the ARRL Sections which arose when one of the Founders (Maxim) was organizing regional bodies to pass messages. Is this not a Spark Gap in the rearview mirror today? This geography to serve amateur radio in the United States is almost ludicrous. See also Onno’s article on changing the current culture in amateur radio. I could go on but this is a family-rated blog, lol.

What can we do about this demographic transition in our culture?

How can amateur radio more effectively deal with the shackles of a fossilized culture where tradition rules innovation? Do we need the Spark Gap in our rearview mirror as a guidepost? Let’s just acknowledge that no amateur could sit down at a workbench and design and build a modern transceiver! Take a Kenwood TS-590SG. It’s far from the leading edge. But could you design one? Then build that design? I couldn’t. Why would we? As Rob Sherwood has written (and I have analyzed), we have the best receivers that we have ever had in the current market. Even though purchasing semi-homebrew radios, like the BitX variety (and I have), push hams toward tinkering, they are not “production” quality for many ham radio activities—like contesting, DXing, and so forth. Yes, some do spend most of their time in tinker-mode rather than production-mode. What we know from Canada is that we have strong segments in “production mode” activities as well as segments in experimentation where homebrew radios have a better use-case for the population. Should we pressure all newbies toward the past when getting them interested in the present and future? My friend, Dan KB6NU, just wrote about this same topic.

Yes, I built a crystal radio as a young teen, using the Fox Hole model with a pencil lead, razor, toilet paper roll for the coil, and so forth. Led the building of an FM and AM station as well. Learned a lot. I still build a lot of things. These activities are highly useful as educational tools. But we must face that we are appliance operators today due to the sophistication of the technology. At best, we are appliance enhancers by homebrewing accessories, modifying “appliance” radios, and so forth. Yes, some do build and operate fully homebrew rigs. Bill Meara N2CQR of Solder Smoke comes quickly to mind. Nothing detrimental with that per se but should we have a norm that every ham should follow suit? When we get the Spark Gap out of our rearview mirror as a guidepost for the future, we will make much more progress with post-Baby Boomers for reasons I’ve outlined above.

My bespoke portable HF antenna, called the EiffelTenna, is featured in the October issue of Practical Wireless. I was inspired by a video of Jim W6LG on his Youtube Channel as well as the further work of Jason VE5REV via Twitter (X). It’s a fun build, inexpensive, is very portable, and works 40, 20, 15 and 10 meters. I use Faraday Cloth for the counterpoise and place the tripod directly on it.

The 40 meter operation works as a center-loaded vertical, something I posted on regarding the inductor coils recently. If you intend to build the EiffelTenna, check out that article too. The EiffelTenna base alone would be good for Technicians since it works on 10 meters without a whip or coil. A stainless steel whip on top of the tripod makes for a solid vertical with its own mounting base. For windy conditions, I use some 1lb ankle weights attached to each tripod leg using the built-in velcro straps.

Shown below is the EiffelTenna deployed for testing on my driveway. (Click for full image) It is setup for 40 meters using a JPC-12 inductor coil. Others work as well or better so this was just the option used here because it’s adjustable. The RF sweep has the coil bypassed using KB9VBR’s trick for use on 20 meters. Nearly 50 ohms with SWR of 1.06 at 14.154 MHz. Note how relatively small the counterpoise cloth is in this picture.

This full antenna system packs down into an inexpensive camera tripod bag ($16 via Amazon). A RigExpert antenna analyzer is underneath the Faraday Cloth for matching in the field. Coax from RG-316 with a ferrite bead choke is wound on a wire winder printed by my public library for the cost of resin ($4). Blue ankle weights were purchased at Academy Sports while on sale.

The EiffelTenna uses traditional vertical antenna concepts with unexpected objects serving as both a ground mount and a radiating element. Thanks Jim W6LG and Jason VE5REV for the inspiration!

Blog Author Note: This is a paper written with Dr. Scott McIntosh of Lynker Space. A PDF of this article is available by clicking here. Reproduction with attribution is permitted.

Sunspots to amateur radio operators are central to daily operations, especially on the HF bands. Ever since Schwabe began counting and plotting daily observed sunspots in 1826, the leading perspective has been that sunspots follow a given, time-ordered, sinusoidal pattern of rise and fall, largely around an eleven-year cycle. The pattern of differences among cycles has been the topic of much speculation, generally without any actual empirical observation of those factors. Until recently, the sunspot cycle is virtually just a given construct based largely on daily sunspot counts, summarized to each month.

There is almost no theoretical discussion of sunspot cycle antecedents, only their effects on propagation. For instance, Nichols (2015) exclaimed, the “top experts” are unable to identify the peak or trough of the solar cycle or the timing of the transition from one cycle to another. Previous work, especially in amateur radio, has focused almost wholly on atheoretical “curve-fitting” style models of the sunspot cycle with observed sunspots gathered for almost two centuries being just a given phenomenon (see Howell and McIntosh 2022a).

With the open acknowledgement that previous predictions have not been very accurate, it is puzzling as to why better theoretical explanations have not been sought by amateur radio experts on propagation patterns (e.g. Luetzelschwab n.d.). This devotion to the theoretically unexplained eleven-year sine wave function as a sterile paradigm in the face of the empirical anomalies reduces scientific progress in understanding the sunspot cycle. Here is why.

Philosophers of science have long debated the role of prediction and explanation in scientific progress. Douglass (2009) summarizes the debate as follows. “Prediction is important because we can be surer that the scientist generating the theory has not fudged or somehow subtly made his theory inconsistent or less clearly applicable to certain contexts by virtue of some torturous, ad hoc accommodation. Prediction also allows for the generation of new (hopefully supporting) evidence. Explanation is important because it helps us think our way through to new predictions.” To make progress in the scientific understanding of the sunspot cycle, we need both theoretical understanding coupled with better predictive capability.

This was the thrust of our 2022 article series in RadCom. We published papers in the July and August issues outlining the long prevailing scientific paradigm on the sunspot cycle, noting that it was largely devoid of a formal theory predicting its rise and fall. We outlined a major challenging theoretical paradigm, led in its creation by the second author, on not only predicting Cycle 25 but offering the beginnings of why the amplitude and modulation of such cycles behave the way they do. In essence, this marked a change from mere prediction toward explanation, a cardinal sign of growth in any area of science. As we have reached the midpoint of Cycle 25, it is time to see how this argument is faring.

Competing Sunspot Cycle Paradigms

The expert panels convened by the NASA/NOAA/ISED organizations (hereafter, NNI) over the past several sunspot cycles have published their own forecasts of the next one. They have done this without any disclosure of the specific model used or the specific substantive theoretical perspective driving them. They do not disclose their methodology but state that it is the consensus opinion of an expert panel reviewing more than fifty various models submitted to them for consideration. Unlike most all peer-reviewed scientific work, the official sunspot cycle forecasts are a theoretically unexplained given resulting from an expert opinion panel whose deliberations are not open to public inspection. Their forecasts have largely failed to be very accurate when later compared to the observed sunspot numbers in the predicted cycle (see Howell and McIntosh 2022a,b for a full discussion).

The second author’s team, hereafter called the McIntosh team, developed both a theoretical foundation and empirical forecast of Cycle 25, publishing the methods they utilized and what substantive concepts shaped them. Unlike the official NNI forecasts, the McIntosh team’s work is public for all to read. We strongly encourage the readers of this paper to review our 2022 articles for details as they are indeed nuanced arguments.

Suffice it to say that the competing McIntosh paradigm emphasizes not the mere curve-fitting exercise that so many amateur radio prognosticators subscribe to in their forecasts (e.g., Cohen 2020) but two new key conceptual elements of the Sun’s dynamo. This was new ground. As we illustrate below, much of the scientific community resisted an open consideration of these ideas at the beginning.

One concept is the Terminator, a landmark event in the sunspot cycle delineating the start, end, and overlap of sunspot and magnetic activity cycles. This event does not correspond to the statistical minimum or maximum in the number of sunspots but to an underlying shift in part of the sun’s dynamo that shapes the entire cycle’s behavior. It arises from the famous Hale Magneto Cycle (Howell and McIntosh 2022a).

A second concept is the timing of the Terminator within the approximate eleven-year period. Taken from our earlier paper:

“This variability, when viewed through the lens of an insular sunspot cycle, lends itself to the anomalies noted by prominent amateur radio propagation enthusiasts. The delay in Termination frequently leads to the forecast of a poor cycle approaching, even another Maunder Minimum, by hams. More critically, the longer the time between terminators, the weaker the next cycle would be. Conversely, the shorter the time between terminator events, the stronger the next Solar Cycle would be. This is the cornerstone thesis in the new competing paradigm which successfully addresses several anomalies observed by Nichols (2015), Nichols (2016) and Luetzelschwab et al. (2021).” (Howell and McIntosh 2022a: 40).

We suggested in 2022 that the key question is whether we are indeed in a crisis stage of a paradigm shift, using the perspective of the well-established Kuhnian model of scientific revolutions (Kuhn 1962). The evidence of such a crisis state would include the following two elements. Firstly, if the competing McIntosh team’s model produces a better empirical forecast than the official NNI’s forecast does, then the theoretically-based paradigm that is a better forecast pushes toward a crisis state. Secondly, what further shapes a crisis state is if other scientists flock to the empirically-superior, theoretically-explicated one. This pattern of behavior, measured largely through citations of the competing paradigm’s exemplars, propagates the new paradigm to the field. If scientists use the competing paradigm’s exemplary papers to shape their own work, then the revolution is taking shape through the collective behavior of other scientists (Kuhn 1962).

We offer a narrated illustration in Figure 1 of the stages and processes of Kuhn’s classic explanatory model applied to these two competing paradigms. We begin on the right side of the wheel of paradigm change. Effectively, the initial “boundary maintenance,” or resistance by adherents to shift from the traditional paradigm embedded in the NASA/NOAA/ISED predictions, eventually gave way to peer reviewers’ objectivity. This occurred through reviewers and editors evaluating the increasingly massive empirical evidence based upon all sunspot cycles for which there were data constructed by the McIntosh team as they revised their initial 2012 work. This was not accomplished very quickly or very easily. As Kuhn (1962) stipulated, this is not at all unusual for a competing set of ideas which threaten the “normal science” embedded in a reigning paradigm.

Nevertheless, the existing normal science “puzzle solving” produced many anomalies in the prediction of both the amplitude and the timing of adjacent sunspot cycles. This acknowledgement that we just do not have a sufficient understanding to produce very accurate forecasts created increasing doubt in adherents to the current paradigm after facing the massive amount of evidence from the original McIntosh team’s paper.

The “boundary maintenance” from 2012 when Science Magazine rejected the initial paper on the new theory began to give way some years later. This occurred through the McIntosh team’s surprising Cycle 25 prediction of a far higher peak in sunspots than the NASA/NOAA/ISED (NNI) official predictions and why they made this forecast. Remember, the official sunspot forecast for Cycle 25 contained no explanations of how they were derived, only that a panel of experts came up with them. This “exemplar” article was published in 2020, some eight years after the initial “boundary maintenance” rejection in 2012. As these results are compared to the errors in previous expert panel forecasts by more and more scientists, this set in motion increasing collective doubt being attached to NNI’s undisclosed methods. This behavior is shown as “model drift” in Figure 1.

Once this model drift occurred after the McIntosh team’s Cycle 25 forecasts were published (2020), the empirical race was on to see which forecast would be more accurate. Modern website technology made this a monthly comparison with the release of each new count of sunspots (shown in Figure 2 below). When the second “exemplar” paper on the timing of the Terminator event appeared in 2023, this undoubtedly significantly enhanced the motivation of other scientists to read and consider the competing paradigm’s exemplars to use as a basis of their own work.

Modern technology speeds up scientific awareness of new works as compared to periodic print journal publication. So, this social network technology makes the process identified by Kuhn back in the 1960s as a “revolution” in paradigm-change an even more valid metaphor today. Should a growing number of other scientists base their published work on the exemplars of the McIntosh team, then direct “model competition” sets in. These collective acts by others in the scientific community are behaviorally manifested through increasing citations of the exemplars in the competing paradigm. If the competing paradigm’s empirical superiority continues, it is only a matter of time before the full model revolution occurs, quickly resulting in a rapid change to a new accepted paradigm. It is our assessment, as illustrated in Figure 1, that we are clearly in the model competition stage as we stand today.

Where Is Paradigm Competition at the Peak of Cycle 25?

In this paper, we evaluate the status of this potential revolution in our shared understanding of the important sunspot cycle. This is based on the two elements described above:

1. Empirical superiority of the McIntosh team exemplars which introduced their paradigm to this field of science. Is the McIntosh team forecast for Cycle 25 demonstrably more accurate than those offered by the NASA/NOAA/ISED Panel of experts?
2. Does the pattern of citations of the two exemplar articles published by the McIntosh team show that other scientists are adopting them? If this adoption is considerable, then the evidence compounds in favor of their new paradigm.

We now provide evidence on both elements of the issue at the approximate middle of Cycle 25. It will show that the results underscore our assessment of where things are in Figure 1.

Statistical Comparisons of the Two Cycle 25 Forecasts

Using the Austrian Space Weather Office website, we produce in Figure 2 the smoothed monthly sunspot numbers for Cycle 25 and for the two competing forecasts. We use the approximate peak time in Cycle 25 to delineate our comparisons. In other words, if this were an athletic competition, what is the score at the end of the first half of play?

In Figure 2, the vertical line is this demarcation as of August 2024 in the time series. Note that the NASA/NOAA/ISED (hereafter NNI) forecast had somewhat of a “false start,” to borrow a track-and-field metaphor, in that after their first set of numbers went public (light blue line), they released another revised forecast. This one shifted their forecast start back some six months (dark blue line). No public explanation was given by the NNI group. We use this revised NNI forecast in our analysis. The McIntosh team forecast is in the red line.

For comparative illustration, there are four other data series. The average monthly sunspot cycle number since 1750 is in green. The three observed sunspot numbers include the daily sunspots (light green line) and the key smoothed monthly sunspots in black. (There is a short series of estimated daily numbers in orange, shown after the final monthly figure.) This represents a visualization of the forecast and observed monthly sunspot numbers. The series includes 32 months of data, our approximation of the first half of Cycle 25.

In this graph, the NNI forecast does appear consistently lower than the observed monthly sunspot data after the summer months of 2022 while the McIntosh team numbers appear generally higher. The exception is near mid-cycle where the observed sunspots spike above both projections. Neither set anticipated this sharp rise in monthly sunspots. But are these two forecasts just a random walk around the observed monthly sunspots? Statisticians have addressed questions like this for some time because time series graphs are somewhat subject to various interpretations. We make statistical comparisons using standard methods for this in Figures 2 and 3.

In our 2022 RadCom paper series, we presented the McIntosh team forecast for a complementary index of solar propagation influence, the Solar Flux Index (SFI, abbreviated as f10.7). We use the NNI forecast for SFI to further compare the statistical accuracy of an atheoretical expert opinion forecast versus the theoretically-driven McIntosh team model.

We use the standard text by Theil (1966) for the analysis of forecast comparisons. One measure of the statistical accuracy of two time series is the mean absolute error (MAE) represented by the formula of

where yⁱ and xⁱ are the respective data values for each time series compared at the i^th time interval. That sum is divided by the number of points in the time series (or n) to yield this average absolute error in numbers of monthly sunspots.

Another test that is metric-free is the mean absolute percentage error (MAPE) which is the percent version of MAE. It is the sum of the actual minus forecast divided by the actual which is averaged over the number of temporal observations:

The third consideration we make is to test for the equivalence of the two forecasts (i.e., are the different forecast series just random walks?). In this part, we use the Diebold-Mariano test or D-M (Diebold and Mariano 1995) which compares the mean difference in the squared-error or absolute error of each forecast to the observed data. This S₁ test is applied to both the MAE and the MAPE (Theil 1966). The D-M S₁ test can use alternate kernel densities in this computation. To safeguard our comparisons, we compute tests using both a uniform and a Bartlett kernel for the standard error estimation. Each produced similar results so the uniform kernel is presented in our results. See Diebold and Mariano (1995) for details. We used Stata 17 software for our computations using the dmariano script (StataCorp 2017).

As shown in Figure 3, the McIntosh team series results in an average monthly forecast error of 26.8 sunspots. For the NNI forecast, the average monthly error is 45.3 sunspots. The McIntosh team forecast is 18.73 sunspots more accurate on average each month. The D-M test shows that this is statistically significant: the McIntosh forecast is significantly more accurate than the NNI prediction for the first half of Cycle 25. The most recent surge in monthly sunspots during 2024 was predicted by both forecasts but the McIntosh predictions were more on track in the graph with those observations.

The second panel of Figure 3 contains the percent form (MAPE) of the forecast errors for each group’s projections. The McIntosh forecast averaged a 25.4% monthly error, lower than the NNI expert panel’s 38.3% error each month. This is a 12.9% difference between the two, reflecting a statistically more accurate forecast (p=.0000).

The Solar Flux Index (SFI) is also a critical index for propagation. It rivals the SSN in importance for daily HF operations. We use this forecast to complement the ones for monthly sunspots. The SFI graph is in Figure 4. For NNI, all monthly errors are on the high side whereas the McIntosh team’s hover on the low side of zero (i.e., matching the observed SFI). Both anticipated the rise upward during the summer months of 2024 but were off in their respective predictions. Over the first-half of Cycle 25, the McIntosh forecast averaged 25.65 Index points closer to the observed SFI (17.2 vs. 42.9). The D-M test suggests that this is a significant edge in favor of the McIntosh prediction (p= .0000).

Putting the SFI forecast errors in percent form, the lower panel illustrates a consistent over-prediction of the monthly Solar Flux Index by both. The NNI’s numbers are visibly off-base by 20 percent or greater in the graph. Some segments of the McIntosh predictions are also off by 10 to 20 percent. Overall, however, the average percent error is 27.8% for the NNI forecast and some 16.6% less for McIntosh at 11.2%. As with the MAE metric, this difference in percent form is statistically in favor of the McIntosh series (p = .0000).

In short, the McIntosh team has empirically superior forecasts for both monthly sunspots and the Solar Flux Index, two leading indices for propagation used by amateur radio and many other spheres of radio transmission practice. They are uniformly statistically significant in favor of the McIntosh theory-driven approach as compared to the expert panel forecasts from NNI.

Evidence of Paradigm Change Through Bibliometric Analysis

To examine evidence on other scientists adopting the new McIntosh team paradigm, we used methods of bibliometric citation analysis (De Bellis 2009: Chapter 8; Prabhakaran et al. 2018). This is a set of methods used to measure the impact and influence of scholarly works through the patterns and frequency of citations in various contexts (Andres 2009; Alphasoft.com n.d.). This set of metrics measures the behavior of the scientific community toward the competing paradigm which Kuhn (1962) shows is the key to paradigm change.

Traditional citation counts from the print medium tend to be much slower than scientific discovery is actually produced because of the circulation of print media (De Bellis 2009: Chapter 8). Because of this, alternative metrics were developed to measure how Internet-based tools enhance the sharing of scholarship. These tools include paper pre-print servers, online exchange of papers, and other discussion networks that are in daily use to stay abreast of the latest emerging knowledge. These “alternative metrics” including social media and online publishing are used in this part of our analysis through the Altmetric system (see Astrophysics Data System).

Following Kuhn’s approach to paradigm-change, we studied the two papers that the second author identified as the exemplars (Kuhn 1962) introducing and illustrating his team’s competing paradigm. The Astrophysics Data System (or ADS) maintained by Harvard University on behalf of the Smithsonian Astrophysical Observatory (SAO) under a NASA grant was the source of our citation analysis. The ADS system (available at (https://ui.adsabs.harvard.edu/) facilitates meta-analysis of papers in astrophysics with both bibliometric computation and visualization of results. We utilize this system to analyze citation and discussion metrics for both exemplar papers as well as all publications for the leader of this scientific team (the second author here). While we do not report a full bibliometric analysis (for example, see Prabhakaran et al. 2018), the compilation of citation metrics does suffice to gauge the initial attention and influence that this competing paradigm is having on the field of solar physics and amateur radio itself.

Figure 5 summarizes these metrics for the two exemplars. The first paper, introducing the overlapping Hale magnetic activity cycles and the relationship they have to sunspot amplitude, has 73 total citations. The bar chart on the upper right shows the trends in citations for this paper (note that 2025 is not yet fully realized). The scientific output analysis by Altmetric gives it an “attention score” of 858. This ranks number one of almost two thousand articles in the journal. Among over one-half million articles published during the same period, it ranks 813. It is in the top 5% of all research ever tracked by Altmetric.

The second exemplar introduced the Terminator timing construct, illustrating it using all the data on sunspot cycles in existence, by associating it to patterns within the 22-year Hale Cycle. This paper extended the first exemplar’s idea of potential causes of the amplitude by adding a conceptual basis for what “kick starts” the next cycle. This paper has 18 citations, moving up quickly in the year after publication as shown in the bar chart. It has an attention score thus far of 654. This article is ranked number one of almost 1,600 articles in this leading journal, Solar Physics. By comparison to about one-half million articles published in the field at the same time, it ranks 999. More importantly, it too ranks in the top 5% of all research articles scored by Altmetric.

Turning to all papers published by this scientific team’s leader, Figure 6 summarizes the same type of citation analysis. Emphasizing the period of 2020-2025 for when the competing paradigm was introduced, there are 457 papers considered in this figure. There are 5,901 citations of these papers, only a thousand of which are self-citations, necessary to build and continue a line of scholarship. The citations by other scholars are the key element for exemplar adoption. There are about two thousand citations, “normalized” to the volume of other articles published around the same time. This puts the citation patterns into the context of the scientific problem as a comparison to numbers of citations per se (see the ADS website for details). The bar chart extends the scope prior to the year 2020 to check the scholarly output by this team’s intellectual leader. The result is a steady increase in citations by other scientists in peer-reviewed papers, an indication of a very productive scholar growing in a career that is being recognized by other scientists in their own work.

The H-index is the most popular one in use for scientific comparisons. A value for H means that the author has that number of papers that are each cited by a minimum of the same number. The H-index number in Figure 6 increases prominently in 2020 to about 45, continuing through 2025. Note that an H-index value of 40 is outstanding and over 60 is exceptional (Hirsch 2005).

The read10-index reflects a decade swath of readership citations of the author’s publications. It shows the works published in 2010 (pre-paradigm introduction) and 2020 (paradigm introduction) as having the highest values, well over 100. This trend line shows the immediate interest in the author’s works over a long period of time (a decade), a sign of prominence in science.

The i100-index, however, might be the most illustrative for our purpose to ascertain how the paradigm is being adopted by others in this field of science. It illustrates the number of publications with at least 100 citations, a challenging hill to climb in science. The growth in the (purple) i100 line shows that a minimum of 100 citations for papers by McIntosh steadily increases after 2015 but especially after 2020, the year of publication for the first exemplar paper. This is also indicative of movement toward paradigm-adoption by others. The tori-index corroborates this trend as his papers become central citations by other scholars. The i10-index began to also spike when the 2020 paper came out and continued after the second exemplar appeared. This index surpasses 130, suggesting that many papers by the author have been each cited by a minimum of 10 other authors.

The bibliometric portion of our analysis shows strong evidence that the peer scientific community is heavily engaged in the competing paradigm. The two exemplars have been substantially growing in peer citations to a level of prominence. They have garnered the top attention in the respective scientific journals where they appeared, no small feat for any scientist. The H-index score shows that the lead scholar producing this new paradigm has reached an outstanding region, further evidence of movement toward direct model competition in Kuhn’s model of paradigm change.

Is There Demonstrable Progress in the Revolution?

Our goal has been to determine if there is Kuhnian movement (Kuhn 1962) toward a revolution in the long-standing paradigm for the sunspot cycle at the midpoint of Cycle 25. We identified two elements of evidence: the empirical superiority of the NNI versus the McIntosh team forecasts and the degree of scientific adoption of the competing paradigm’s exemplary papers.

Using long-established methods in forecast comparisons, our results leave little objective doubt that the theory-driven forecast by the McIntosh team is superior. For smoothed monthly sunspot counts covering the first 32 months of the Cycle, the McIntosh team forecast is 19 spots more accurate, a 13 percent and statistically significant improvement over the NNI numbers. (Note that we used the NNI’s revised forecast after they adjusted to some six months behind their original Panel’s predictions.) We included forecasts for the Solar Flux Index (f10.7) over the same time horizon. The McIntosh team’s SFI forecast is 26 index points or 17 percent more accurate. The empirical superiority, at least at mid-cycle, clearly favors the McIntosh paradigm.

The bibliometric analysis we presented on how the two exemplary papers have been received by the scientific field showed strong evidence of engagement and adoption with the competing paradigm. The overall standing of scholarship by the lead scientist was a second element surrounding this new paradigm. It too demonstrated a clear upturn in citation metrics after the publication of the two exemplar papers.

The citation numbers have been continually increasing since the first (2020) and second papers (2023) appeared in peer-reviewed journals. We noted in Figure 1 the boundary maintenance by keepers of the long-standing paradigm who rejected the original paper in 2012. It took nearly a decade (from 2012 to 2020) of continually increasing the amount of scientific evidence involving the linkages between the Hale Cycle to the sunspot cycle’s behavior to reach a successful peer-reviewed publication. With the observed rapid increase in citations of the two exemplar papers, Kuhn’s concept of a non-linear, revolutionary adoption of a competing paradigm appears indeed to fit the bibliometric results.

To underscore Kuhn’s notion, the more contemporary “attention” metrics for the two exemplars show that each is the number one ranked article in the respective publishing journal. Each is also in the top five percent of all research articles ever tracked by Altmetric. The two exemplars have clearly captured the attention of the field and the associated reporting on it challenging the status quo paradigm.

We find the bibliometric citation results to also be strong evidence that the competing paradigm is indeed now within direct model competition as illustrated in Figure 1. It may well take until the end of Cycle 25 to determine the extent that a paradigm revolution has occurred. It will depend on the continued reception of the McIntosh team’s published results as they continue their research program. This adoption would be spurred along by a continuing forecast superiority during the second half of Cycle 25. Those monthly comparative results are available on the Austrian Space Weather website for all to see.

We likened this study to that of checking the score at half-time of an athletic event. However, we must wait until this cycle is compete to render a full assessment of who wins the scientific competition. We plan to revisit this analysis at the appropriate time. The available evidence at half-time, nonetheless, clearly favors progress in the revolution involving our understanding of the critically important sunspot cycle.

References

Astrophysics Data System (ADS). Online resource [https://ui.adsabs.harvard.edu/].

Alphasoft.com n.d. “Exploring Bibliometric Methods: Citation Analysis in Research.” Online resource:  https://alfasoft.com/blog/products/scientific-writing-and-publishing/exploring-bibliometric-methods-citation-analysis-in-research/.

Andres, Ana. 2009. Measuring Academic Research: How to Undertake a Bibliometric Study. Oxford: Chandos Publication.

Cohen, Nathan. February 14, 2020. “Are You Ready for the Next Solar Cycle?” https://forums.qrz.com/index.php?threads/are-you-ready-for-the-next-solar-cycle.692443/.

De Bellis, Nicola. Bibliometrics and Citation Analysis: From the Science Citation Index to Cybermetrics. 2009. Lanham MD: Scarecrow Press.

Diebold, Francis and Roberto Mariano, “Comparing Predictive Accuracy,” Journal of Business and Economic Statistics, 13:3, 253-263, 1995.

Douglas HE. “Reintroducing Prediction to Explanation.” Philosophy of Science. 2009;76(4):444-463. doi:10.1086/648111).

Hirsch JE. (2005) An index to quantify an individual’s scientific research output. PNAS 102(46):16569–72.

Howell, Frank M. and Scott W. McIntosh. 2022a. “On the Cusp of a Scientific Revolution: Part I.” RadCom July: 36-43.

Howell, Frank M. and Scott W. McIntosh. 2022b. “On the Cusp of a Scientific Revolution: Part II.” RadCom August: 76-77.

Kuhn, Thomas. 1962. The Structure of Scientific Revolutions. University of Chicago Press.

Luetzelschwab, Carl. n.d. “A Look at All Twenty Three Solar Cycles.” Retrieved from http://k9la.us/A_Look_at_All_Twenty_Three_Solar_Cycles.pdf

McIntosh, Scott W., Sandra Chapman, Robert J. Leamon, Ricky Egeland & Nicholas W. Watkins. “Overlapping Magnetic Activity Cycles and the Sunspot Number: Forecasting Sunspot Cycle 25 Amplitude.” Frontiers in Astronomy and Space Sciences. 295 (12), December 2020.

Nichols, Eric P. 2015. Propagation and Radio Science: Exploring the Magic of Wireless Communication. Newington, CT: American Radio Relay League.

Prabhakaran, T., Lathabai, H.H., George, S. et al. Towards prediction of paradigm shifts from scientific literature. Scientometrics 117, 1611–1644 (2018). https://doi.org/10.1007/s11192-018-2931-3.

StataCorp. 2021. Stata: Release 17. College Station, TX.

Theil, Henri. 1966. Applied Economic Forecasting. Chicago: Rand McNally.

With the copyright terms used by The Spectrum Monitor, my cover article in last month’s August issue is now freely available. I’ve put a PDF of the article with cover page and table of the issue’s contents over at FoxMikeHotel.com. I hope you will take a look at it if you do not already subscribe to TSM. Ken Reitz publishes a highly informative magazine in a reasonably priced PDF-only format. It’s more than singularly focused on amateur radio but the hobby sort of drives the car. At $24 a year, it is very, very reasonable.

An Op-Ed piece that I’ve written by invitation from Editor Paul McLane is scheduled to appear in Radio World soon. That is the industry magazine for radio broadcasting and associated technologies and activities. I’m sure that not everyone will agree with my assessment of where both the AM and FM radio industries are on the “death and dying” spectrum. My argument is based on publicly available data from the FCC, Edison Research and Nielsen, plus some thoughts about the industry from various outlets on the Internet. But all are publicly available.

As Paul McLane of RW and I have discussed privately, we do not have a key leading indicator of local media market prosperity available publicly. That is ad-revenue to AM stations in that market. One can license the data at some expense but not publish it. These data are indeed a bread-and-butter product of the collector. Thus, that is that caveat to my findings that this leading indicator is simply not available…unless the vendor releases it.

My argument is that even with this omission, the outcomes of AM ad-revenue within the local media market are not (yet) manifested in levels of annual shutting of AM stations or audience reach to warrant a prognosis of the death of the industry. It may be worthy to state that this might be a valid prognosis in some local media markets as my analysis shows. The famous author Stephen King’s closure of several AM stations appeared in the New York Times as a sign of the death of AM radio. As Allan Wiener, owner of WBCQ shortwave and several AM stations in Maine wrote me, this is a very struggling local media market. It is not a national issue.

The annual percent change of both AM and FM stations from my article is reproduced here. Yes, there has been about a 1% annual decline in station licenses since 2010. One percent per year. It has also been apparent that FM station licenses have been in a similar pattern of decline. Thus, whatever the local media market ad-revenue to AM stations is, FM radio is also suffering some small annual decay as well. Things are not unique to AM radio, regardless of the myriad of statements that AM is “dying.”

The alternative thesis that I show evidence supporting is consistent with Wiener’s observations. It is not a national trend but a “shake-out” in some local media markets. As I show, even after market size and audience reach is controlled, the absolute number of AM stations is a better predictor of the number of station licenses relinquished over this period of time. This also holds true for FM station closures, something that adds stronger support for the shake-out interpretation since it not unique to AM broadcasting in local media markets (called DMAs).

One question that arose to Editor Ken Reitz of TSM after my article appeared came as an e-mail letter from John Schneider W9FGH. He questioned whether the numbers were skewed by “licensed and silent” AM stations. I produced the map of L&S AM stations below using the FCC LMS for licenses tagged as such. My narrative response appearing in the September 2025 TSM issue is reproduced verbatim below in italics. I greatly appreciate thoughtful questions like this from John W9FGH, a long-time contributor to TSM and its predecessor, Monitoring Times.

My response to the Editor, published in the September 2025 issue, with a note that Ken sent me some National Radio Club information on listener reports on “licensed but silent” AM stations:

As you know, it’s important to have some basis for comparison when focusing on one narrow phenomenon so as to avoid siloed thinking. If ‘licensed and silent’ status AM stations are an indicator of ‘dying’ markets, then the FM broadcast industry is in as much or more trouble than is AM! According to the FCC’s Licensing and Management System (LMS) as this is written, there are 130 Full Power AM stations in the LMS status. But there are more (148) Full Power FM stations in this license category. These include 101 Full Power, 43 Low Power and 9 Booster stations. The upshot to me is that ‘silent’ stations have an unknown basis. Temporary financial issues, death of owner while license is active, storms putting them off the air, and so forth.

My market ‘shake-out’ thesis, which Allan Wiener seems to also embrace from his experience in a challenging media market [author’s note: see Allan’s letter to me reproduced by TSM in the September 2025 issue], would actually be buttressed if the LMS AMers are in DMAs where more outright cancellations occurred (they may be on the skids but have not lost or turned in their license yet). I produced the map which has a base of DMA-level number of AM cancellations for the years 2010-2025. The 130 AM stations that are tagged ‘Licensed and Silent’ in the FCC LMS as this is written are overlaid as points, symbolized by blue stars.

I’ve not done a tabular summary but here’s what I see here. There are few ‘silent’ AMers in DMAs with the lowest number of cancellations over the 15-year period. Most are in markets with the higher numbers of AM cancellations.

On the NRC reports, these may well be very prescient. But note that a listener report that s/he hears nothing on a given date gives preference to close-by listeners as these stations aren’t likely the Clear Channel occupants. It may be an indicant of full non-operation but it could also be temporary until it reaches the FCC list as shown above. We do not know.

Without doing the extended analysis, my take on these data is that, should I add them into the officially cancelled AM licenses, it would only make the results stronger with regard to the ‘market shake-out’ interpretation as well as point to the FM sector as having as much of a problem as does AM. The totals of another 130 AM stations would change the absolute numbers by about 3 percent, but not the conclusions.

So I hope you take the time to read the TSM article in full. We need less heat and more light on issues like this. While no study is complete enough to reach closure on the issue, it is enough to state with care and reason that there is no public evidence that AM broadcasting is dying. Rather, it is changing with several local media markets facing serious challenges to remain profitable. The same goes for FM broadcasting as my analysis illustrates.

The Departments of Communications and Continuing Education at Georgia College & State University are hosting me for a public talk on this study during their Student Media Day on October 17, 2025. We are celebrating the 50th anniversary of the college FM station I founded back in 1975 as WXGC, now WGUR. If you are in the area of Milledgeville GA, you are invited to attend.

CHOTA 2025 is nigh!

Here in the U.S., Churches and Chapels on the Air is a new concept. I was fortunate to play a leading role a couple of years ago in herding a few cats on this side of the pond to get on the air in this very special event. It’s led, for over a half century now, by WACRAL, The World Association of Christian Amateurs and Listeners.

Even though most US operators will grouse about the propagation over the past few months with all of the solar storms, we are at the high point in Cycle 25. And this makes the opportunities to jump the Pond from the US to England possible! Let’s give it a go…

This year, I’m activating Covenant Presbyterian Church on Ridgewood Road in Jackson, MS. This is in partnership with Jim Armstrong AK5J, President of the Jackson ARC. He and I will work with Chris AF5OQ and Wanda (KC5IBO) Dunn as a team using my portable contest station (see below).

I am not sure of which antenna we will use. It’s likely that a 31′ vertical will get us 40M and up (see below). We shall see as Jim and I scope out the church grounds for a place in the shade (lol). In 2023, my church got front-page coverage on the state’s largest newspaper as well as a human interest feature on Jackson MS TV. It’s worth a shot to get the word out about ham radio on the church grounds this year. It could pay off big time for you, your club or other group.

We will likely hang out on 20 meters, although most of the Brits settle on 40M and some 80M. Our team will monitor 40M and QSY when we hear some CHOTA calls. I’ll suggest to John G3XYF who coordinates the event to encourage those operators in England to check the 20M calling frequency at the top of each hour and call the US. John usually posts some suggested frequencies near the date for each band. It’s important to recall that the English hams are about 5-6 hours ahead of us in the day. Our team plans an early start to try to jump the Pond if conditions permit!

Here’s a screenshot of the CHOTA Rules…and don’t forget to “register” by sending John [email protected] that you’ll be operating in the event. I’ve just done that myself.