If you are fundamentally opposed to (or trying to avoid paying taxes on) an aboveground structure, you could bury a vault to put at least a small pressure tank in. With use of a "constant pressure valve" or a variable-speed "constant pressure pump" you can get away with quite a small pressure tank (2-5 gallons) but you really can't run a pump without any pressure tank, at least not remotely efficiently (you could, I suppose, put a pressure-relief valve in the line at the top of the well, and run without a pressure tank - the excess pressure would recycle water down the well, but that's a terrible waste of power...)
If you really need water in the winter, bury the whole thing below frost line - one common method is to use 4 or 6 foot diameter concrete well tile (as used for shallow dug wells) placed over/around the drilled well casing to provide the frost-free vault - you can insulate the top and sides for better effectiveness. While you have a backhoe in to dig for that and the power line, dig some trenches below frost line to where you want to have hose spigots, and put "frost-free hydrants" at each one (they have a valve at the bottom, and leak away the water in the upright pipe between uses.) Might as well have the backhoe dig 100 holes for trees while you have it, too.
If not fundamentally opposed to a structure, an aboveground structure (small toolshed for the orchard) above the wellhead might be a less expensive approach, unless the local taxing authority makes it an expensive approach after all (that can happen, and is best to be aware of when deciding.) The area with the pressure tank can be heavily insulated and electrically heated to 45F or so if you really need water to hoses in freezing weather, or you can drain the whole system in the fall and start it up again in the spring if you don't really need to use water from hoses in the winter.
All sort of issues here - while there's a short list of things that one normally looks for with extreme short cycling (such as a blown bladder in the pressure tank) that doesn't touch the 90 PSI issue...
If, in fact, water is going back down the well, the check valve in the pump (there is pretty much always one built into deep well pumps) is bad. Depending how your system is set up, there may or may not be another one inside the house, which would also have to be bad. My own system, after considerable reading and opinion-gathering, has only the check valve on the pump 300 feet down my well. When it goes, it's pump-pulling time. That would be the case even if there were others, and others might actually cause that to happen sooner according to a number of experienced well folks, which is why the other one I bought is still sitting in a box, not installed. In another few years I'll put a new pump on the shelf for when this one dies, so it's not a panic buy when (not if) I need it.
80-90 PSI is VERY high and almost certainly indicates a problem with the pressure switch. Certainly on my well the emergency relief valve lets go at 100PSI. On 40 off 60 is usually the upper range of domestic well pump "common settings" (30/50 and 20/40 being the other two commonly seen.) Ideally you'd know what your setting usually is or it may be recorded on a sticker somewhere on the system. You mention it going to the very high pressure when first switched on after being off for a while - what was it cycling between when you found it?
You could TRY hooking up a garden hose and opening that valve, then turning the pump on to flush out the crud that's been stirred up. It's possible-but-not-likely that this MIGHT clear a bit of crud that is jamming the check valve open. If nothing else it will get the stirred-up crud out of the well before the pump is repaired or replaced.
If your well is prone to throwing dirt, you may want to add a "spin-down filter" (I put mine after the relief, but before the rest of the system - this was contrary to the manufacturers instructions - they suggested after the pressure tank - but in line with the whole point of having it.)
Best Answer
In general, moderate silt won't damage a typical electric well pump, as there are no close clearances (virtually all are not a piston with seals, it's a set of impellers which have space between them and the silt just flows through with the water.)
Heavy sand over long periods of time can cause significant wear, but that does not look like what you show, and in any case you either have to live with it, (and replace your pump more often) or possibly reposition the pump to try and minimize it. You can use various filters to keep it out of the plumbing if it persists, but it may not persist once you have flushed things out a bit.
Are you pumping into a 5 gallon bucket inside the building? Put a hose on, run it outside, and run it for an hour or so, keeping an eye/ear on it so you can shut it off just as soon as water stops coming out, if it does.
Pay attention to when that happens, and make the occasional check of how long it takes to fill a bucket for an estimate of flow rate (which will vary over time as the water level in the well drops) and you can get an estimate of your well's recharge rate. You'll also get an idea of how the silt is progressing. If your well recharge rate is high enough, it may just keep pumping the whole hour, that's fine.
Leave it several hours or a day to fully recharge, and repeat for perhaps a week, perhaps less depending what you see for collected silt.
Your previous post has some data regarding the "volume of water in the well" which implies a 6 inch casing and counting the water below the pump (10 feet off the bottom of the well) which you cannot pump. In reality (if those figures are still accurate - water-tables not being fixed) you have about 110 gallons "stored in the well" you can access, and anything more than that you get out over a period of time informs the recharge rate of the well (typically stated in gallons per minute.) So if you pumped out ~200 gallons over the course of 30 minutes and then the water stopped, you could infer that about 90 gallons flowed in over the 30 minutes, implying 3 gallons per minute inflow. But if your pump pumps out slower than water comes in, the water would never stop - but you'd still have a number for the minimum rate of inflow you could infer. That's merely information, but it's often good information to have.
In "flushing out the well" you certainly want to pump out MORE than the "stored volume" each time, since water WILL come into the well (and if the well has not been used for a long time, the initial inflow may indeed carry some silt with it.)