Read on Twitter
Thanks! I haven’t studied specific landers and I wouldn’t comment on specific companies’ landers without their permission. I am (so far) studying a simple, generic case with one engine mounted under the centerline. I can compare predictions based on lander mass for that case. 1/n https://twitter.com/LuisBandeo/status/1241700159530156032
2/ As I said in this poster back in July, the 5-ton (landing mass) Apollo Lunar Module blew about 2.6 tons of soil per landing. Using the equation we derived from Apollo landings (a 2.5 power index) this predicts a 40-ton lander will blow 2.6*(40/5)^2.5 = 470 tons of soil! 
3/ But it is also possible that other effects would occur under such a large lander. In Apollo a deep crater did not form. In vacuum the engine exhaust spread out enough so the soil did not mechanically fail and flow, forming a crater. It just scoured away the surface material.
4/ For a much larger lander, it is possible the gradient of pressure on the surface will overcome the strength of the highly frictional lunar soil, causing it to shear and flow, forming a cup under the engine. That would change the gas flow patterns and greatly enhance ejection.
5/ Here’s an example of soil failing under pressure of a rocket plume. Notice how the darker sand layers are shoved down as the crater approaches. Then the layers are pulled up in the final image due to enhanced upward erosion inside the crater wall. This didn’t happen in Apollo.
6/ We call this mechanism “bearing capacity failure” or BCF. It is possible this mechanism could “turn on” for a much larger lander when its engine is very close to the lunar soil. We don’t know yet what the limits are for this and several other cratering mechanisms to “turn on”.
7/ In some of the projects we are studying these deep cratering mechanisms. In other projects, like in this picture I tweeted last night, we are studying the simple scouring process, where rocket exhaust doesn’t form a deep crater but just erodes soil off the flat surface.
8/ So to answer the question, what if the lander is 100 tons, I had to clarify that this current new result is only talking about surface scouting rates. It isn’t talking about the potential for deep cratering. A lot more work is needed to answer the deep cratering question.
9/ So as I said above, the Apollo-derived equation tells us that the surface scouring *alone* would eject 470 tons of soil for a 40 ton lander, *if* the equation can be extrapolated to larger landers. I suspected it could not be extrapolate that far. The new result confirms it.
10/ Just for grins, what if we plug in a 100-ton lander into that equation? It predicts 2.6*(100/5)^2.5 = 4,650 tons of blown soil! 
We know this is wrong on multiple levels. It predicts a hole so deep it disrupts the gas flow so the “flat surface assumption” would be wrong.
We know this is wrong on multiple levels. It predicts a hole so deep it disrupts the gas flow so the “flat surface assumption” would be wrong.
11/ With this background, I can now explain what this new result really means. It has only limited benefit, but still it is a step of progress to solve the physics. All it says is that the Apollo-derived equation slightly over-predicts the flat-surface erosion rate. How much?
12/ I will be able to answer in a couple of days. I have to set up the computer simulations and run the new cases, calibrate the results against the existing data sets, and produce the new estimates. Even then, there will remain major gaps in our understanding of the physics.
13/ @mastenspace is working on the deep cratering physics with @astroaddie and me. Will a large lander blow a hole in the soil? This is a totally un-solved branch of physics but we have made progress. https://twitter.com/mastenspace/status/1230932878944559105?s=21 https://twitter.com/mastenspace/status/1230932878944559105
14/ Doctoral student @wachambers has been studying cratering effects in both low gravity and vacuum at the @UCF Center for Microgravity Research with @astroaddie. https://twitter.com/drphiltill/status/1185287766994763776?s=21 https://twitter.com/DrPhiltill/status/1185287766994763776
15/ Here, Wesley is working with an undergraduate student researcher to perform a drop-test to measure cratering in asteroid conditions — vacuum and ultra-low gravity. This helps us solve the physics for the Moon, too. https://twitter.com/drphiltill/status/1185288986958487557?s=21 https://twitter.com/DrPhiltill/status/1185288986958487557
16/ And here is the video of the drop-test. It is a very brief test, but it provides enough time in microgravity to measure the eruption of asteroid regolith under a firing rocket thruster. https://twitter.com/drphiltill/status/1185289539444772864?s=21 https://twitter.com/DrPhiltill/status/1185289539444772864
17/ You can see the erupting gravel (asteroid regolith) on Wesley’s computer screen. This is from the drop-test you saw in the previous tweet. This eruption is not erosion; it is a *another* cratering mechanism, which we call “diffused gas eruption”. https://twitter.com/drphiltill/status/1185291831376011264?s=21 https://twitter.com/drphiltill/status/1185291831376011264
18/ In this thread I mentioned 3 cratering mechanisms: (1) ordinary erosion or surface scour, which was dominant in the Apollo landings; (2) bearing capacity failure shoving a hole under the engine; (3) diffused gas eruption. There are 2 more known mechanisms.
19/ They are: (4) Diffusion-driven flow or DDF, and (5) Shock-induced fluidization. DDF is when gas flows *through* the soil inducing a drag force throughout the bulk of soil that exceeds the soil’s strength, causing it to begin flowing in bulk. We saw that in some experiments.
20/ Shock-induced fluidization is when a pulsed rocket engine repeatedly slaps the soil inducing kinetic energy (rearrangements of grains, analogous to vibrating molecules) increasing its “granular temperature” T, so it “unjams” and becomes a granular liquid. (Image: Liu & Nagel)
21/ There are at least 2 special cases of shock-induced fluidization. One is where the shockwave simply slaps the surface of the soil. The other is when the shockwave diffuses between the soil grains, carrying repeated waves of kinetic energy deep into the soil. This occurred...
22/ ...when NASA’s Phoenix Lander landed on Mars. Manish Mehta and @Doctor_Astro discovered this and named this cratering mechanism “Diffused Gas Explosive Erosion” (DGEE). Here’s a video of their experiments showing DGEE:
23/ Here: I sped up their high-speed video (which had slowed things way down) and looped it a few times. You can now see pulsing in the soil, which is caused by the pulsed plume repeatedly slapping the soil, and how this effectively fluidizes the soil making a deep crater.
24/ The Phoenix lander had pulsed engines. Because of that, it was able to fluidize and sweep away the soil that had covered the ice at the landing site. So it was DGEE that led to the first view of subsurface ice on Mars — the white patches in this image under the lander.
25/ Cool side note: see how the ice patches are hexagons? I don’t think that is an accident. The lander used 12 engines in six pairs. Each pair was close enough to act like a single plume on the soil. So there were 6 evenly-spaced plumes hitting soil. (Art by NASA/ @MissionArtist)
26/ The rocket exhaust hits the soil then spreads out until it hits the gas from one of the neighboring pairs of engines. Thus, each pair of engines dominates an area defined by the hexagonal interaction of all 6 pairs. I think that explains the hexagon-shaped ice. Could b wrong!
27/ My point: there is a VAST amount of unsolved physics in how sand is moved by gas. When people think of unsolved physics they think Dark Matter, Quantum Collapse, and Consciousness. But the motion of a pile of grains is equally unsolved, & yet we see it everywhere! (Wikimedia)
28/ This is why we have no equation that can predict how much lunar soil is blown by rocket exhaust, even in the case where we ignore the deep cratering mechanisms, pretend the soil stays flat, and assume just 1 engine scouring its surface. Even then the uncertainty is giant.
29/ NASA is developing several instruments to go on lunar landers, and some will measure the plume/soil interactions. @Ryan_N_Watkins is on the Heimdall camera team, which will see these effects and do more lunar science. https://phys.org/news/2019-07-camera-high-resolution-images-video-lunar.html
30/ And even though we don’t have the physics figure out we know enough to begin planning methods to control the blast. @CLASS_UCF is now starting a series of robotics competitions in Puerto Rico to develop landing pad robotics technology. Pics here from NASA’s Desert RATS.
31/31 Another area of research is making a concrete-like material from lunar soil to build landing pads. Image: lunar soil construction at @SwampWorks_KSC. There is more R&D I can’t tell due to proprietary info. This thread barely scratched the surface. More to come!
