Pages

Monday, January 13, 2020

Piece It Together

We’ve extensively discussed the problems with the Navy’s concept for surveillance and targeting and noted that it is unrealistic and will fail.  ComNavOps has offered ideas that would allow for effective surveillance and it’s time to pull some of that together and present a complete view of how that would work in the context of an amphibious assault, just to pick one relatable scenario.  What follows is presented as a fictional snippet with some analytical commentary mixed in.  The purpose is to illustrate the concept, not present a realistic battle simulation.



___________________________

Aboard the LHA, the Marine assault commander studied his composite screen which displayed the overall situation.  Initially, the screen had shown nothing other than the handful of known, fixed structures which were scattered around the assault site.  As the assault fleet had approached the target, UAV recon drones had been launched to begin filling in the picture.  This was an opposed landing and UAV attrition had been quite high but that was expected. 

The bigger problem was that no single UAV could survive long enough to provide anything remotely resembling the total awareness that misguided military planners of the early 2000’s and 2010’s had assumed and promised would be available.  The large scale Navy Fleet Exercises that had been re-introduced beginning in 2020 had shattered that delusion.  When the opposing fleet forces in the exercises had finally been allowed to utilize their full electronic warfare (EW) and defensive (anti-UAV) capabilities in an exercise, it had become instantly evident that the earlier concept of a few UAVs leisurely circling the battlefield and providing total situational awareness with all ships and aircraft seamlessly connected by a flawless network had been just a fantasy of wishful thinking.

Based on those shocking, eye-opening exercise results, the entire surveillance/recon/network concept had been reworked.  The revised concept had been exercised and refined and was now being put to its first actual combat test.  No one expected it to work perfectly and, in fact, that was one of the foundational assumptions of the system – failure was inevitable, anticipated, and built in!

The assault commander noted that the display was showing that the first assault wave was hitting the beach just about now.  Taking in the entire display picture, he noted the various areas (most of the assault area!) that were largely unknown.  Turning to his Recon Coordinator, he quickly designated and prioritized specific areas for additional, more intensive recon based on the assault plan.  The coordinator contacted the UAV carrier which shortly began launching swarms of small UAVs to the designated areas.

As the recon UAVs approached their target areas they were greeted by enemy laser, EW, missiles, and projectile weapons.  UAVs rained out of the sky or simply vanished in brilliant explosions of light and thunder.  Each UAV was lucky to survive long enough to transmit just a momentary, fragmentary sensor view.  That was perfectly normal in the new recon scheme and was the reason why recon UAVs were as small as possible, simple, dirt-cheap, and deployed in swarms.  As each UAV transmitted its fragmentary data before dying, the central recon data collection center on the UAV carrier began assembling the fragments of data into a cohesive picture.  It wasn’t necessary that any single UAV be able to surveil and transmit the entire battlefield.  It was only necessary that enough UAVs be able to transmit enough fragmentary scans to allow the central computer to assemble a pieced together picture.  Targets were updated and displayed in colors which indicated how old the data point was.  Brilliant red points were newly acquired targets and the color faded to duller red and, eventually, black as the target ‘aged’.  Thus, the operators viewing the data were able to instantly see and comprehend the newness or ‘value’ of the data. 

It was inevitable that many areas were not scanned at all when the designated UAVs were destroyed before being able to acquire and transmit usable data.  Those areas were displayed in patches of gray, letting operators know that the areas were un-scouted.  The Recon Coordinator noted the unknown patches and began designating additional swarms to the areas.

One of the problems was that some (many?) UAVs managed to collect data but the enemy’s EW efforts were disrupting the data transmissions.  Again, this failure was expected and built into the system.  The enemy’s EW disruption and jamming efforts were not, of course, 100% continuous and effective.  The UAVs, assault vehicles, and ships employed various techniques such as frequency hopping to try to shake off the effects of the enemy’s EW.  When momentary (often microsecond bursts) clearances occurred the communications/network computers prioritized which assets to communicate with instead of attempting to talk continuously to every asset.  Communications occurred in microsecond bursts rather than attempting to maintain continuous connections which required massively large broadband channels with unachievably high data rates and network stability.  As the exercises had shown, network stability in a combat zone was a laughable fantasy. 

Thus, the display of the Marine Colonel acting as the shore commander in his specialized Command and Control variant Amphibious Combat Vehicle (ACV), now sheltered behind a rock outcropping a short ways inland from the beach, was able to receive frequent, fragmentary bursts of data updates.  His situational display was, therefore, never complete and real time in quality but it was alive and functioning.  Noting a couple of areas that were especially lacking on his display but vital to his immediate operations, he designated the areas on the display for priority updates.  The controlling computers back on the UAV carrier adjusted their priorities and the Colonel’s display began filling in the requested areas more frequently at the expense of the lower priority areas.

This was certainly not the all-seeing, all-knowing surveillance system that the delusional Navy leadership of earlier times had promised but that had never been a realistic promise to begin with.  This was a fragmentary approach that allowed commanders to at least maintain a semblance of situation awareness in the face of intense enemy EW and defensive measures.  Yes, there were risks associated with prioritizing certain areas and allowing other areas to remain less monitored but that was the nature of war.

Back aboard the LHA, the assault Commander noted that the assault area display was slowly beginning to fill in.  While the surveillance picture was insufficient to give him a real-time display of the number of rounds remaining in every enemy soldier’s rifle magazine, it was adequate to provide an indication of where enemy units were and were not and in what approximate strength.  With this degree of information, he could begin adjusting his own positions and start calling in fire support where he needed it.

Looking for an opportunity to flank the enemy mechanized battalion that was anchoring the central defensive works, he ordered his Recon Coordinator to concentrate on a stretch of river far to the enemy’s right.  If he could slip a company of infantry in ACVs up the river, they could turn and cut into the enemy’s rear and isolate them from reinforcement.  Again, dozens of small catapults on the UAV carrier launched as one and a small UAV recon swarm formed up and headed for the designated river. 

It took about 30 minutes for the UAV recon swarm to reach the river whereupon they spread out and began their search.  The Commander, noting the time, turned his attention to the recon swarm.  As he watched, the UAV signals began disappearing.  In a matter of minutes, the swarm was wiped out without a single target being detected.  That, the Commander knew, could only have happened if the swarm had run into a large concentration of well equipped enemy forces.  It was safe to assume there was at least an enemy company and quite possible a battalion with attached anti-air units.  Sometimes, a flaming datum was as good as real data.  Having been saved from a possibly disastrous flanking attempt, the Commander marked the river as a likely company/battalion location.  The information would shortly be reflected on the display screens of his own units as communications processing time allowed.  He also passed the information on to the naval commander for air strike and naval gun attention.

 _____________________________ 




We see from this fictional snippet that effective battlefield surveillance is possible but only if we build some fundamental assumptions, based on reality, into our concept.

  • Failure must be built into the concept.  The system will be degraded but must operate effectively in a degraded form.
  • Attrition among UAV assets will be extreme and UAV assets must be dirt cheap and plentiful – almost unlimited in available quantity.
  • A dedicated UAV surveillance carrier will be required.
  • The goal must be broad, general, situational awareness, not total, real time omnipotence.
  • Communications will be severely degraded and must be designed to function as such.  A real time network is pure fantasy.  Communications and data transmissions will have to be on a ‘when possible’ basis.

What all this means is that we must drastically scale back our surveillance expectations.  Exquisitely detailed, real time information is not possible in an electronically contested environment.  We must learn to work with fragmentary information.  It is sufficient to know that a battalion size unit is in a given area – that we don’t know how many rivets are on each vehicle is not an impediment to effective operations and, more importantly, is not a requirement for effective operations.  Our surveillance goal should be to obtain the minimum information we need, not the maximum.

What all this means is that our current plans and equipment are, largely, wrong and useless.  The Navy’s touted distributed lethality concept is a prime example of an idea that will not work and urgently needs to be reworked (doubtful) or abandoned (preferred).  Our Triton and P-8 aircraft are unsuited for battlefield surveillance.  We lack a cheap, expendable surveillance UAV.  Our underwater unmanned vehicles are completely inappropriate for battle surveillance although they may play a role in the larger strategic surveillance requirements.  We lack a supporting vessel (UAV carrier) for mobile, cheap, short range surveillance UAVs.  We utterly lack any viable Concept of Operations (CONOPS) for naval battle surveillance.

In short, we currently have almost nothing of use for naval battle surveillance.  We desperately need to begin exercising our concepts in a realistic fashion so that we can start recognizing their inherent faults and start working towards a concept that is actually viable.

39 comments:

  1. Your points are good. There is no doubt that the current use of drones is more or less the product of the endless war on terror and Iraq where nobody was knocking drones down.

    Its interesting that a large ship like the WW2 New Orleans carried 4 long range float planes but a similar sized Burke carries none to just 2 shorter range helicopters. And most things I read see a ship carrying just one Fire Scout. Seems like a drop in distributed awareness.

    How many Fire Scouts (and or other drones) do suppose a Wasp could hold and launch and run if was not carrying marines but drone operators.

    ReplyDelete
    Replies
    1. The float planes were shot off using cats but landed on the water. The result was that the planes needed calmer seas to land than a helo does.

      As far as the Fire Scouts go the more important factor than the size of the helo is the number of support people/ facilities needed. With the maintenance and operators required it's probably similar to embarking a huey.

      Delete
    2. I guess my point was if you are going to distributed detection there were still four of them (and a sufficient amount of maintenance facilities to fix some things). Now we are down two at best two helicopters and or one drone per ship if that.

      Delete
    3. Which means retrofitting drone capabilities across the fleet (more than onsy twosy per ship. Or we can build dedicated drone carriers. Or probably better, both.

      Delete
    4. I agree a dedicated CV would be better but that seems not to be in the cards. If distributed we seem to be at less capability than WW2

      Delete
    5. "There is no doubt that the current use of drones is more or less the product of the endless war on terror and Iraq where nobody was knocking drones down."

      This is the classic, 'preparing to fight the last war'. We're jumping all the way in on drones, data, and networks because that's what worked in the 'war on terror'.

      Delete
    6. "I agree a dedicated CV would be better "

      Bear in mind that we don't need a true aircraft carrier to be a UAV carrier. A converted merchant ship with a couple dozen 'slingshot' catapults (Blackjack style catapults, for example) and a landing/snagging area is sufficient. We should be able to procure a used merchant ship and convert it for around $200M. There's our UAV carrier!

      Delete
  2. For me, important factor is the training and mindset the commander has. He knows his picture will be limited, needs to be updated and will be ever changing....compared to what we train of highly scripted exercises where we get a full continuance unlimited picture which is unrealistic, plus the enemy is an idiot that just does what we tell him to do.... Will that commander be ready when the system doesn't deliver? When the enemy gets a vote?

    ReplyDelete
    Replies
    1. "compared to what we train of highly scripted exercises where we get a full continuance unlimited picture"

      It's worse than that. We not only train with a full picture but our desire to ensure good results, combined with our zeal for safety, means our commanders have the complete picture BEFORE THE EXERCISE BEGINS.

      I'm desperately hoping that the reintroduced Fleet Exercises that will begin this year will bring a much needed element of realism. If not, then they'll just be a colossal waste of time and resources.

      Delete
  3. I like it.

    Only small nit is I think your data comm expectations are still too optimistic. There is no reason not to expect continuous broadband jamming from multiple redundant sources.

    Handshaking protocols are likely to be the biggest issue complicating comms in a jamming environment. It all takes time to establish a connection and hinders using microsecond windows. The other problem is that if you want to move video at all, you are leaving the robustness of FSK modulation and moving into 64 QAM or higher. Which very definitely doesn't like jamming at all. You see this in SCADA environments where new DHS regulations have forced traditional highly-robust low-bandwidth monitoring links into the new world of continuous video surveillance. And those reliable links have become problematic as a result.

    So in an ideal world, some kind of image recognition capability with onboard processing would be ideal. Its much easier to shout "Hey, I see X" than trying to move video or even a single static frame.

    But I really like the scenario and hope you might consider doing more along that line!!

    ReplyDelete
    Replies
    1. "I think your data comm expectations are still too optimistic."

      It's possible. I'm not a combat communications expert so I'm just engaging in low level speculation, though somewhat semi-informed. I'll also say that for every comment I get like yours, expressing doubt about combat comms, I get one claiming that I'm being overly pessimistic about combat comms and that we can securely, reliably communicate in combat. Who/what to believe? I don't think even the military knows because they seem to refuse to test combat comms under realistic conditions. So, I'm left to fend for myself. What I've described is my best guess/assessment but I freely admit I could be incorrect in either the optimistic or pessimistic direction. Does that make sense?

      The larger point is not whether my comm presentation is exactly correct but that the concept of lowered expectations and acknowledgement of failure is necessary to design a combat-functional system. I'll leave the intimate details to the combat comm techs.

      "if you want to move video at all"

      You don't! That's kind of the point of the post! We have to lower our expectations from full color, real time, 3D, holographic, multi-spectral video to bare bones detection. For example, it will have to be sufficient to detect a vehicle (radar return, for example) and not try to obtain full, hi-res video so that we can get the vehicle's serial number.

      Another approach is to allow the sensor UAV to determine the type of vehicle and simply transmit the tiniest bit of data that means, say, 'armored personnel carrier' rather than the full, raw video. Transmitting a single bit/byte has to be easier than transmitting full, hi-res video and both accomplish the same thing!

      Delete
    2. Consumer 5G and low earth orbit satellite internet technologies are more or less what the military is looking at for use on small UAV and missile, like the ~40mm MHTK,
      datalinks/seekers.

      My prediction is that small UAVs will act far more like the messenger pigeons of old than the UAVs of the GWOT. They'll scan an area and fly off to a virtual "coop" from which they can hand data off via highly directional 40-75-ish GHz datalinks before RTB or proceeding to another objective.

      Delete
    3. "They'll scan an area and fly off to a virtual "coop" from which they can hand data off"

      Interesting concept. Of course, that completely eliminates the real time surveillance that military has been working towards and it eliminates using the data for targeting on any but fixed targets.

      These are the kinds of things we should be testing, today, under realistic combat conditions.

      Delete
    4. The other side of EW is that to use it you have to transmit. That can be detected and localized. The use of EW also produces data points for the traffic analysis people on the other side. It's not a one way street.

      If you use EW against a competent foe it has to be intermittent and mobile similar to the doctrine on using MLRS and arty. Continuous or repeated use from the same location invites effective attack.

      Delete
    5. @Anonymous

      5G has about 500m range per tower how are you going to use that on the battlefield/sea? GO 6G and you are talking 100s of feet at best.

      Delete
    6. Better would some array of drones maintaining direct line of sight beam communications. But even that is degraded in bad weather.

      Delete
    7. "The other side of EW is that to use it you have to transmit. That can be detected and localized."

      Assuming you are talking jamming, sure. But the scenario had a couple of battalion-sized units in it.

      A practical jamming unit is likely about the size of a fridge. You would have to run power cables and fiber to it and be good to go.

      Everyone forgets just how good the Chinese are at mass producing electronics. I wonder how many units like this a couple of battalions could deploy?

      Obviously you would drop them off and hook them up well away from troop concentrations. Home on jam could ruin your day otherwise. But I see the concept of using this kind of jamming to be similar to drone swarms. Have so many units available that they are effectively disposable. Blow one up? Fire up another one...

      And you would employ the Software-defined radio (SDR) concept so you can upgrade and deploy new tactics on the fly.

      Delete
    8. " EW is that to use it you have to transmit. That can be detected and localized."

      "Blow one up? Fire up another one..."

      This illustrates one of my constant themes - that we all have a marked tendency to think in one-on-one terms. Yes, it is possible to spot a jamming transmitter. However, in the overall scheme of a battle with battalions of tanks, vehicles, men, aircraft, helos, missiles, artillery, etc., all on a mammoth scale (compared to the tiny efforts that we've come to think of as combat), a jamming transmitter is but one of many, many targets and it may or may not be prosecuted depending on what else is happening.

      So, yes, jamming can certainly be spotted but that doesn't mean that it will automatically and instantly be destroyed. Just a reminder to maintain the 'big picture' perspective as we discuss all this stuff!

      Delete
    9. I'm thinking in terms of having 50-100 running at any given time over an area a couple of battalions might occupy.

      At some point prosecuting them becomes an economic nightmare when you look at missiles and aircraft in harm's way.

      As you are saying, the trick is to make anything that emits cheap enough that its ubiquitous and prosecuting it becomes a waste of time.

      But then you, as the attacker, have to to be prepared to operate in a seriously jammed environment. Which is exactly what I expect to happen. I've been looking at data sheets from some of the combat radio manufacturers. Pretty depressing but I'll get back to it later.

      Delete
    10. "But then you, as the attacker, have to to be prepared to operate in a seriously jammed environment."

      Which was the premise of the post! Whether you believe my presentation is accurate or not, the rationale for it certainly seems correct and the military needs to come to grips with the requirement to operate in a degraded surveillance, communications, and network environment.

      A very good start would be to allow our own EW capabilities to be used to their fullest extent during exercises and, instead of just giving up, shutting them down, and continuing the exercise, we need to learn how to operate in spite of the problems. If the problems are really so severe that we can't operate then we clearly have the wrong equipment and wrong doctrine.

      Delete
    11. A useful jammer versus spread spectrum, frequency hopping and burst transmissions is not going to be cheap. Additionally if it is powerful, it and its power source are not going to be extremely small. EW equipment suitable for use against high level opponents is not going to be cheap. They will be rare enough that their losses over time will be felt.

      All of this means that they will be used, they can't be simply allowed to be destroyed at a rapid rate. Because otherwise you won't have enough. That means they will have to be selective in their use and take precautions not to keep them absolutely safe, but to lower the attrition rate to an acceptable level. That means intermittent use, mobility and figuring out novel ways to keep them alive.

      All of this is part of the arms race between EW and Counter EW. It's not just the equipment its the tactics too!!!! Just thinking you have a bunch of high tech gear isn't enough.

      Delete
    12. "A useful jammer versus spread spectrum, frequency hopping and burst transmissions is not going to be cheap. Additionally if it is powerful, it and its power source are not going to be extremely small. EW equipment suitable for use against high level opponents is not going to be cheap. They will be rare enough that their losses over time will be felt."

      Any references as I think you are way out in left field on this one?

      A jammer is a radio transmitter and very little else. The controller does not have to be part of the jammer and it doesn't emit.

      You can do narrow band that requires quite a bit of sophistication and lower output power, which is more applicable to airborne applications, or you can do wide band that requires quite a lot of power and works best on land. You can build wide band, fairly high power transmitters for Smarties compared to some of the other costs we are talking about.

      And a decent electronics factory could turn them out by the tens of thousands.

      Just because we do things a certain way, it would be unwise to assume the Chinese and Russians are doing the same.

      Delete
    13. "A useful jammer"

      The range of photos of Russian and Chinese jamming equipment is impressive. They run the gamut from man-portable backpack jammers to small, easily transportable land units to large truck mounted units. Each is aimed at specific signals but the variety is what's noteworthy.

      Delete
    14. "Each is aimed at specific signals but the variety is what's noteworthy."

      Entirely the point. The lower the frequency, the larger the unit needs to be. We are not just talking about data comms, we are also talking about jamming radar. The more sophisticated units are a nightmare to maintain and program. Go big, go ugly, go brutal as long as you can feed them enough power which you can't do to an airborne unit.

      And even jamming comms, really seriously wide band jamming requires larger units if you can't make them ubiquitous within the battle space. So make smaller units ubiquitous.

      If this was my decision, I would make EW such a core function it would be equivalent to battalion artillery. They may well have a similar value.

      Delete
  4. Disposable UAVs could be built, allowing the design to skip landing gear, and launched by catapult from smaller surface ships. Those one way UAVs could have greater range to give the destroyer (or even LCS) a wider surveillance range.

    Not a replacement for a UAV carrier, but given their high utility and higher predicted attrition the ability to launch as many as possible should include the whole fleet.

    ReplyDelete
  5. I think it would behoove the USN to experiment with a UAV carrier or something similar. Play around with the concept. Also, I know it's all hush hush but I dont like that it always seems like we are playing defense when its come to ECM. Can we even go on offense? Is it something we plan to do or is it just destroy the guys ECM kineticly?

    ReplyDelete
    Replies
    1. This would not strictly fit into the UAV category NICO, but it would count as one of your 'offensive weapons' I think:
      https://www.thedrive.com/the-war-zone/29789/spear-mini-cruise-missile-getting-an-electronic-warfare-variant-to-swarm-with-is-a-huge-deal

      Delete
    2. Dale, you've linked to the MBDA SPEAR-3 mini-cruise missile. There are a few problems/limitations with it.

      The link indicates the missile has a range of 85 miles. Presumably, that's the best case, maximum range and operational range may be less. Even so, 85 miles is not much! If a naval (we'll skip land combat uses for our purposes) force is within 85 miles of an enemy to use this missile, they've probably been heavily engaged for quite some time given that naval forces have weapons and sensors on the order of multiple hundreds of miles.

      The link also mentions the ability to loiter over the battlefield. Again, if 85 miles is the max range and you're talking about loitering then you had to have launched the missile at much, much closer range in order to have enough fuel to loiter.

      As noted in other comments, power is an issue for any EW weapon. A mini-cruise missile just can't provide huge amounts of power which leads to the question, 'what is this EW'ing against'?

      It's possible that this weapon might be more useful in land combat. I don't know. That's not my area. It seems to have limited usefulness in naval applications.

      Delete
    3. You're right of course - that range is basically piffle in naval operations isnt it. in the beach landing scenario above, maybe slightly more useful, but perhaps not. I would hope the technology is also available on other, longer ranged platforms, or at least ones that can "EW" against targets from a greater distance - wouldnt have a clue about how that actually works.
      I guess the other thing that could happen in this battlefield is an EMP weapon - although I'd assume most military grade electronics are at least partially shielded against such things.

      Delete
  6. Before the Champ cruise missile went totally black over 10 years ago it was reported that the system might turn off all electronics
    either permanently or short term. That was a long time ago. If that system is sent in first into a battle space the assumptions made by our military might look more reasonable.

    ReplyDelete
    Replies
    1. And if the Chinese version is sent in first … ?

      Delete
    2. Not going after Peter G but isnt that one of the problems today, how often do we hear about some super secret, in development, still testing weapon system that fixes all the stuff we are missing? It's ok for as a comment but I regularly hear professionals just casually mention some exotic or still being tested weapon as a sure thing that fixes everything and dont worry....just my cents, not sure what can be done about it.

      Delete
  7. (Don McCollor)..at the risk of bring foolish, your scenario suggests another approach to drones. If a drone swarm only survives for 30 minutes, they could be designed for single use -small, simple, (relatively) cheap, and with about the same operating lifetime. Delivery could be "seeding" them with a larger drone, missile, or even artillery shell over the area of interest (perhaps along with "decoys"). Perhaps no propulsion, instead parachute, aerodynamic shape, or glider airfoil.

    ReplyDelete
    Replies
    1. That's a fine idea but just be aware that the smaller we make the drone, the smaller and less powerful its sensor will be, the smaller the field of view, and poorer the resolution, and the less power we'll be able to supply it with for communications (very limited broadcast range) and sensor operation. So, there's a trade off and a balance point somewhere between small/cheap and effectiveness.

      Delete
  8. (Don McCollor)…(I was thinking of a drone about 6-12 x 3 x 3 inches undeployed) If it does not have to aviate to the area, a smaller power pack should do the same job (full power for a short time) for comm. I think that sensor resolution should not be an issue - weight of image sensors are about the same and video processing to stabilize images works wonder. The bottleneck is transmission. Speed is of the essence. As a heretic, transmit in "clear"... the enemy already knows about them, if they are shooting...

    ReplyDelete
    Replies
    1. Wow, that's small! The payload would be about the size and weight of a cell phone, if that. Sensor capability aside, would you be concerned about the very low power comms being easily overwhelmed by general broadband jamming? The Russians have demonstrated exactly that capability against Ukraine comms.

      Delete
  9. (Don McCollor)...now that you bring it up, the (COTS) capabilities of a cell phone (processing capacity, memory, GPS, cameras, communications, ruggedness, cheapness) would be a good starting point. Communications range, transmission power, and (especially) jamming is a problem. Relatively few higher power transmissions might solve this. Being quickly expendable and not actively flying, the battery can use up all its power mostly for communications in a few minutes. With sensors, some could be dedicated to visual, some thermal. The question is what is the minimum information that would be acceptable?

    ReplyDelete
  10. I think eventually the Navy and the Army , are going to come to see that both drones and missiles will be highly degraded by ECM and ECCM jamming.They will be blinded and /or lose their control links.

    what I see as a solution is for drones and missiles to be fiber optically connected with their launching source. This will provided an unjammable control/ info/intel transmission link, ,much like the old TOW missile is and the FOG AT missile would have been.

    My guess is they can now make such control wire almost as light as spider silk enabling a trailing wire of 50 -100+ miles for such "wire-guided " drones/missiles.

    ReplyDelete

Comments will be moderated for posts older than 7 days in order to reduce spam.