Yes.

Yeah.

That's right.

Yeah.

OK.
It really is a pleasure to be back here.

It's it's fun to see familiar faces and
buildings and

some unfamiliar ones as well.

And both categories and

I feel a bit like an archaeologist
because the students were.

Were really excited to learn that I was
here when the shuttle mural went up

on the S.P.S. wall and so I feel like
I'm shedding light on ancient history.

For the students but
it's been a great visit so far so

right I'll talk to you today about
a project that I have been working on for

some time to test gravity by
the technique of literally as

arranging and so just as a motivation.

If you do a google search for

quantum gravity you get a picture like
this and if that's not a cry for help.

I don't know what is so

you know there's a problem in physics
a core problem that two of the pillars

of physics gravity as expressed by
generality and quantum mechanics

are fundamentally incompatible and efforts
to bring those two things together.

Tend to introduce scalar fields and
make for

measurable effects like violations of the
equivalence principle of time variation of

gravitational constants and
it's also true that the dark sector and

physics dark energy dark matter
are based on a framework

of generality of a T. as our fabric for
understanding cosmology and so

cosmology is sort of built on
the general T.V. framework.

If we don't have that part right

things like dark matter dark energy
especially could be misinterpretations of.

Of gravity at large scales.

So it's something we
certainly ought to test.

Gravity is the weakest of the fundamental
forces despite its apparent strength

in our daily lives and so
we really should subjected to the.

Toughest test that we can so

testing gravity and looking for
General Timothy is not a new pursuit.

In fact when Einstein.

Developed his theory of
general to Vittie he

already understood that there were
three experimental domains in which

this might be tested this idea might
be tested The first is the Perry

hellion advance of mercury in simple
terms the elliptical orbit of Mercury.

Well process around and point to
different parts of the sky over time and

it's a very small effect forty
three arcseconds percent tree and

in fact dwarfed by the Newtonian facts
there are other things that put torques on

Mercury's orbit but it had been understood
as long as since I think the eight hundred

fifty S. that Mercury's orbit
had this anomalous procession.

And it was an outstanding
mystery in physics in fact

it was one of the first motivations for
what I would call searches of dark matter.

It was thought that interior planet
to mercury which they called Vulcan

was perturbing mercury in
moving the orbit around and so

there were a lot of concentrated surface.

In the last part of the seventeenth
century looking for one thousand sentry

I was get that backwards looking for this
planet Vulcan nobody ever found it but

it was the first time when an anomaly
in gravity was pinned on dark matter.

So that's kind of a lesson we should
pay attention to but Einstein

calculated what his general to be theories
should produce on Mercury's orbit as early

as nine hundred fifteen and right off
the page fell forty three arc seconds for

Sentry the exact anomaly
that had been sought for

many decades and it's said that he
had heart palpitations he was so

excited to see this result come off
the page and now it's been tested to very

high precision the deflection
of light near the sun.

This was a famous test of relativity that
Sir Arthur Eddington a mountain expedition

to look to take photographs of
the sky during an eclipse and

compare to photographs of the same
region without the sun present to see

the stars that are revealed when
the sun is blocked moving and this is

an exaggerated scale in the sense that
the sun's disk is half a degree across and

here is a scale in degrees and where
the stars were that they were able to to

photograph and then on a different
scale an expanded scale in arcseconds

it shows the deflection measured for
each of these stars and so

you'll notice that the ones that are
closer to the sun are deflected more but

they tend to be deflected
away from the sun.

This is something that even Newton would
have predicted to happen but Newton's.

Prediction would have been exactly
half that of General to a T.

but I'm Stein imposed a sort
of curvature on space

that Newton did not have half
of this effect you might say is

the equivalence principle photons are just
particles in their following trajectories.

But the other half is from
the curvature of space and so.

Edington showed that you know this
effect did happen at the scale that I

predicted and that was a big news event.

It took till nine hundred sixty to realize
the third classic test of general a T.V.

that clocks run slower deep
in a gravitational field.

It took technology to catch up but
it's interesting to note that without

understanding and
appreciating this the G.P.S.

constellation would under really fail to
give you predicted positions on earth

after an hour or so
important is this gravitational redshift.

So it's actually every time you drive
somewhere in your car you have G.P.S.

It's using general to
Vittie at some level.

So that's pretty interesting.

Now in the lunar orbit.

There are a number of ways that
we can look at gravity and

you know in general terms
you can just appreciate that

the moon's orbit is a very
clean dynamical system.

If you go for billions of years without
any friction or drag or things like this.

Having serious effects on on its dynamics.

So right now.

Lynn arranging provides the best test
of the various aspects of gravity

it test the equivalence principle that two
objects fall the same in a gravitational

field these two objects of the earth and
the moon falling toward the sun in a solar

orbit to an acceleration difference
of a part in ten of thirteen.

So that's a position.

Experiment.

And I'll talk more about that in a bit it
test the time rate of the gravitational

constant to better than a part
in ten of the twelve per year so

that means and think about this.

Kuzma logically the universe is ten to
the ten years old ten billion years old.

Roughly.
So if gravity has changed.

If the gravitational constant
has changed by less important.

Ten of the twelve in the last year

then it's changed by less than
a percent over the age of the universe.

If today's rate is
representative of past history.

So that's interesting because of you know
magnetism is the origin of this thing

called frame dragging is something that
also the lunar orbit is sensitive to

this is the thing that Gravity Probe B..

Attempted to measure and
got a twenty percent result Well our long

literally is arranging has
had a point one percent.

Constraint on this physics.

Geodetic procession one of our squared
force law also things that lunar ranging

has the best limits on.

And the project that I'll describe
in a little bit is aiming for

a factor of ten improvement on
all of these physical effects.

So let me tell you a little bit more about
the weak equivalence principle and it

comes in to flip side of the equivalence
principle which comes into flavors.

What is the weak flavor which is
really a composition dependence.

Of acceleration due to gravity based
on the sort of nuclear constituents of.

The material and you could say
that it probes all energies but

gravity itself so if you think about
what makes up the mass of brick.

It's got you know electromagnetic
forces that are contributing it's got

kinetic energy of all
the particles in the nucleus.

So it's got strong nuclear forces it's got
weak nuclear forces those are all playing

a role but it's not really bound
gravitationally gravitational energy is

a negligible fraction of laboratory
scale objects something like a part

in ten of the twenty seven mass is
due to gravitational binding energy.

I'm not held together gravitationally.

But once you get to
astronomical size bodies.

You get to test what's called the strong

equivalence principle where you're
measuring gravitational energy at so.

Off and so the energy of assembly of
the earth the gravitational energy sort of

embodied in the formation of the Earth
is about a half of a part per billion.

Of the Earth's total mass energy.

So that's still a small number but it's
a lot better than the ten to the minus

twenty seven the laboratory
scale objects and so

we can ask the question does this
type of mass this gravitational

self energy have gravitational
properties does it itself gravitate

if you have a lump of self energy
does it feel gravitational force.

Does it have inertial forces that
hard to push it around and so

the earth has part of its mass energy
budget in this form of self energy and

we can ask how gravity is pulling and
pushing on on gravity and

that gets out on a very non-linear
aspect of gravity itself interaction.

So it turns out that literally laser
ranging provides the very best way to test

this piece of physics and
just to sort of flesh it out a little bit.

If you'd like to see it in this form this
half of a part per billion comes from

integrating the mutual
interaction gravitational

potential between any two
mass elements in the earth.

So if you do a sort of double
integration over over the volume.

You're obviously going to end
up with something that has

you know two powers of the mass
one power of radius and

some numerical factor which
is pretty close to one and

you can express the gravitational
to inertial ratio for the earth.

Which is if the equivalence principle is
satisfied is identity it's one always but

you can express any deviation from
that as some one minus parameter.

Times this self energy fraction.

OK so this would be.

Sort of describing the.

Degree to which equivalence
principle is is broken as a function

of the fractional self energy in
the object and for the lunar orbit

that makes a thirteen metre difference or
polarization of the moon's orbit.

And that means if A does one in
the interpretation of being one either.

Plus or minus one would mean
that either the self energy

has no inertial properties so it doesn't
take anything to push it around or

it has no gravitational properties and
it doesn't gravitational attract So

those are the extremes but
anywhere in between.

You'll see some potential range signal and

this is what let's learn arranging to test
the strong qualms principle to reasonably

high precision and
just to put that in picture if

the equivalence principle is violated so
let's imagine that we do have a violation.

So that gravitational inertial
mass are not equal and let's say

that all of that will cause earth and moon
to fall toward the sun at different rates.

It's important to realize that
the moon is not orbiting the Earth.

Contrary to what you might think it's
orbiting the sun and there's a real reason

I say that which is that the moon
is beyond the realm where Earth

gravity dominates once you're forty Earth
radio from the center of the earth solar

gravity dominates in the moon is sixty and
you know well beyond forty Earth radio.

So it's really in a solar
orbit with a strong earth.

You know.

Perturbation and so
what would happen is Earth and Moon would

fall at different rates and it would look
like a polarization of the lunar orbit.

So think of it this way.

Let's pretend the earth has greater
inertial mass than gravitational mass and

let's pretend Meanwhile that the moon
is normal equal amounts of inertia

gravitational So that means the Earth.

You might think of it.

It's sluggish to move a target to push it.

Harder to push it around than it should
be or you could say that it's got weak

gravitational coupling Either way or
the same it's going to take an orbit.

That's exteriors.

To the average orbit of the moon as
the moon sort of weaves around the Earth's

orbit and so from our perspective on
earth that would look like the moon is

pushed the moon's orbit
is pushed toward the sun.

So that's where you get
this thirteen meter.

Polarization signal and
there's a maybe better

pictorial representation of this that
showed up in the newspaper at one point.

So Earth's orbit is this thick blue line
the Moon's orbit we and actually one

of the consequences of the moon being in
a solar orbit is if you drew this to scale

this part of the orbit would never be
convex pointing toward the sun it's always

concave But of course then you wouldn't be
able to see it very easily on the page.

But the moon's average orbit
is this dotted line and

then the an equivalence
principle violation.

Those two separate and
there's a signal that we can measure.

So literally is arranging
has been around for

a long time since the Apollo astronauts
first put reflectors on the moon

in one thousand nine hundred eighty nine.

We've been able to send
laser signals to it and

get a round trip travel time and
early on we had something like twenty

centimeter range precision
which is pretty good.

It started before that it
was two hundred meters.

We knew how far the moon was
two hundred meters suddenly

it collapsed by three
orders of magnitude and

that was using a large telescope
a two point seven meter telescope.

But older technology.

And one nine hundred seventy S.
laser technology question.

Yes I do and
I'll show you some pictures of it.

In the mid eighty's a transition
actually away from the big

telescope back to small telescopes.

But.

With much better technology detector and
laser technology and

the effort I'll describe I should
put some more points on this.

This was an early slide is down
sort of another order of magnitude

really by going back to the big telescope.

That's all we really did.

So the project is called Apollo and

we're looking for order of magnitude
improvements in this range perception by

using a large telescope
back to the telescope.

Shooting twenty pulses per second which
is a higher rate than other efforts.

The best latest detector technology
we can get our hands on we

are able to gather multiple photons per
shot which sounds pretty pathetic but

it's actually a radically new regime for

this experiment and were intent on
integrating the experiment in the US

is that sort of diverged over the years
of the people collecting the data weren't

themselves doing the analysis and so
were that's not a great way to do science.

And so we also came up with I think
the best acronym we're using the Apache

point Observatory the technique is
Lunar Laser ranging all you have to do is

demote the R. and come up with a no you
have Apollo and NASA can't help but

fund that and then N.S.F. fortunately
has come in as a roughly equal partner.

So it's a collaboration that started
at the University of Washington with

Eric out of Berger Christopher Stubbs and
myself and then we've sort of scattered to

different institutions and
is also involved he's the I would

say the grandfather of literally is
arranging he's the first one back in one

nine hundred sixty S.
maybe sixty eight who realized that you

could test general to Vittie and the
equivalence principle using the moon and

he communicated this to Bob Dickey at
Princeton and actually by by figuring out

which airplane barbecue was going to be
on and he purchased a ticket on the same.

Flight and

he said it wasn't hard to recognize
the physicists who got on the plane and

he sat next to them and told him about
this and Bob Dickey was very impressed and

sort of accelerated the installation
of reflectors on the moon.

So in fact here we are Apollo eleven
the very first lunar landing and there's

already a reflector there that they built
they designed and built in six months.

We can't do anything like that these days.

Doesn't seem six months isn't enough
time to even submit a grant proposal.

So here's the reflector That's Buzz Aldrin
and he's installing a seismometer and

Neil Armstrong is taking the picture and
what these things are our

collections little arrays of quarter
keep prisms solid pieces of glass.

With mutually perpendicular back surfaces
have the property that light entering will

rattle off all three surfaces in urgent
exactly opposite the direction it came in.

So it's like a crystal Ector something
that seems like exactly back the way

from which it came and
so Apollo eleven and

fourteen each have one
hundred ten by ten array.

These are thirty eight millimeter one and
a half inch.

Cubes corner cubes the Apollo fifteen
array has three three hundred of those.

Devices So it's a bit larger.

Where are they on the moon the Apollo
reflectors are here Apollo eleven fourteen

and fifteen Apollo twelve didn't land
one Apollo thirteen had trouble and

didn't land on the moon but
Apollo fourteen or fifteen completed

the The installation of these reflectors
and you need at least three to

make this project work if you measure the
distance to one reflector from the earth.

You could theoretically pivot the moon
about that fixed point measurement and

you don't know where
the center of mass is and

that's what you care about as soon as
you measure two points you can rotate

the moon about those two fixed points but
you still have ambiguity.

About where the center mass is the third
one locks the orientation boom you've got

the center.

You've eliminated the degrees of freedom.

All the same the Russians played to and
landed a.

An unmanned lander rover.

Here on Luna seventeen mission and another
one here on the Luna twenty one mission

and they're comparable these two
are comparable in performance to the three

These two are comparable in performance
to the smaller of the two Apollo arrays.

OK so how does it work.

I.

At one point developed a nice animation.

To show what goes on.

So aside from being slowed
down in the scale being wrong.

Well OK there's one thing
that's right about this.

There are fifty pulses in
route at any given time for

our experiment we shoot
twenty pulses per second.

It takes two and a half seconds
to get to the moon and come back.

So it's about fifty en route.

So we launch from the telescope and

immediately the atmosphere
starts verging the beam.

At one arcsecond let's say
that's about the atmospheric.

Turbulence level and so at the time
by the time it gets to the moon one

arcsecond is two kilometers across most of
the light is lost on the lunar soil goes.

Of course there's no sound in space.

And then the light that's lucky enough

about one out of thirty million of our
photons manages to find the reflector.

And then that spreads by diffraction
from the corner cubes at

maybe characteristic seven arc seconds or
so and so

it's maybe fifteen kilometers
across of the earth and so

most of it doesn't hit our telescope and
it's roughly comparable factor of

a part in thirty million makes it so
it's like if you went.

It's like winning the lottery
to actually hit a photon.

Here one in thirty million chance.

But what if you won the lottery and they
called you up and said there's a one in

thirty million chance the money will
make it to your bank account but

we get to play the lottery a lot with ten
to the seven hundred photons per shot.

And so that even the odds for us and
by the time it's all said and

done we get something like one
photon per pulse on a good night.

So our detector and our laser.

It's at five thirty two so
same as this laser pointer.

Here's the telescope three and a half
meter telescope our laser is actually in

this box and
there are some people for scale.

If you've heard of the Sloan Digital Sky
Survey that was taken from an adjacent

telescope at the same peak
in southern New Mexico.

So this is a very high elevation
site with good atmospheric seeing

good image quality flexibly scheduled.

So we usually get one hour blocks of
time every few nights maybe eight or

ten chunks out of the month and
the whole point is for

us to map the shape of the Moon's orbit so

as long as we're sampling along that
curve and get several and weather knocks

us out of maybe forty percent of
our nights thirty forty percent.

As long as we get several
points around we're OK.

At the in the laser boxes a two foot
by four foot section of laser bench.

We have this neodymium AG laser
that's point by flashlamps

it's cavity down to frequency doubled
to five thirty two nanometers

I like to say don't trust
the font trust the laser pointer.

That's the real color.

It's about one hundred pico second pulse
with which is physically about this long

in space about an inch or so.

Post it twenty times a second.

It's got Therefore one hundred milligrams
per post gives it about a to what average

power and if you divide hundred Milledge
rolls by ninety one hundred seconds.

You get one gigawatt So
that's it's a nuclear power plant for

one hundred because seconds.

We expand the beam to three and
a half meters to fill the aperture so

it's less of an eye hazard to
people in the Dome or Tele or

sorry pilots planes are attracted
like Mas to this thing.

It's also less damaging to the optics
it doesn't thermal heat them it doesn't

ruin the telescope for
subsequent observers

we use a an avalanche photodiode
array is the detector.

It's an array in our case four by four.

Sixteen elements each
thirty microns in diameter.

We've upgraded to a forty Micron damager.

Detector and
these are one hundred microns apart.

Now the advantage of that is we have lots
of buckets to collect any photons if we

have multiple photon multiple
photon hit when we're lucky.

We have we spread it out.

We make our point spread function or
are focused image kind of.

Spread out across these
tech the tech tears.

So we're not susceptible
to first photon bias.

If you think about an avalanche
photodiode detector the first

photon that text makes an avalanche and
once you've got an avalanche going you

can throw a snowball into
the avalanche nobody cares.

It's just there's avalanches in progress.

So you could bias your measurement to
early the earliest photon you detect and

that's a danger we need to be able
to detect multiple photons and

know when they came in so
that helps here the other thing it

helps is that we have spatial
information because we make or

putting this is the focal plane if
we had a single element detector and

we lost the signal we don't know which way
to move to hunt it but in this case as

if we start losing the signal it's going
to pile up perfectly in one corner and

we know which way to move the telescope so
it's much more efficient.

We might not.

A lot of dead space here
between the elements but

we put a lens a little ray in front and
so we recover the fill factor.

Every photon that would be headed for

a square around this element gets
channeled onto that element.

Just some pictures there's the laser box
mounted permanently on the telescope.

We call the Utah box for reasons I'll
let you discover this is our phone

booth which one of our more nerdy
associates is started calling the Tardis.

Then we have a corner cube prism that's
mounted to the secondary mirror which you

can't see here.

Very well but it's another one of these
corner cube prisms that returns light back

from the direction came and that's very
important because that gives us a what we

call our for do show measurement of
when the light left the telescope and

we detected by the same exact
optical path detector electronics

etc at the single photon level.

So we have a start time basically for
our travel to the moon.

This is lighting up the telescope
mirror with their laser beam.

Again let's see how close the that's
pretty close color match.

Here's a basically giant laser
pointer shooting at the moon.

And so we get really strong returns
compared especially compared to what

came before this is time on the wall
clock that's four minutes of time.

This is how far off.

We are from our predicted round trip time.

So we have a prediction a model
that tells us how far we expect it.

How long it.

We expect it to go take to go to the moon
and come back these points are background.

Photons and then we see a very bright.

Thick line.

Those are the returns from the reflector
we see here's the Apollo fifteen reflector

Here's Apollo eleven.

You can see that one is fat and
one is then and

it's because these
reflectors are physically.

Tilting as a function of
what's called libration

how the moon is oriented exactly it's plus
or minus seven degrees in both axes and so

in general there is in your
corner in a far corner.

So even if you know precisely its
orientation a laser pulse hits it some

light comes back early some light comes
back later that spreads the pulse out in

time and you can see that
the physically larger arrays spread to

return more than the physically
smaller a Ray I like to write down

this is the measurement representative
measurement from this series of photons.

That's how many millimeters of the moon
at that moment with an error bar.

So there are.

You know I'm generally an advocate of
low perception math and calculation in

order of magnitude estimates but
there is a time and place for digit.

Yeah.

Up here.

Yeah good I that is because we have

entered detector photons that
penetrate beyond the depletion region.

Create a photo electron that doesn't
have an electric field to tear

tell it where to go and there's a random
walk for a while diffusion until it

happens to hit the electric field that
drives it into the multiplication region.

So we get late avalanches and
you can see the tail here.

I should also mention this is
from our local corner cube.

So this is what our system does.

And then this everything else is
what the reflector does to it but

this asymmetric tail is just
a feature of our detectors.

Yeah.

Yeah.
This one Apollo eleven is closer

to the physical center so
it is going to have a smaller number

more on that particular night these
were both taken on the same night.

Now this number will change from

three hundred fifty six to four
hundred seven depending on where.

Where the moon is and
it's a let the whore bit.

And even over the course of
an hour as the Earth rotates

at four hundred meters per second.

That number can change
quite rapidly as well.

So the geometry has a large
role on the actual number.

So this only has meaning when it's
associate with a particular timestamp

OK sort of a closer look
at the corner cube and

its mathematical model and then if we
can Vall that with the trap a zoid

which is the mathematical function you'd
expect from a rectangular array that's

tilted the pole shape turns into
a generally a trap as OID and

you can evolve it with a theoretical trap
as odd that we expect on that night and

you get this blue curve that
fits very nicely the broad.

Run poles from the reflector and

just sort of pictorially as as you
tilt the moon different orientations

you get different trap a Zoids
in we see a good match.

We basically I can tell you.

Even if nobody had told me that
dimensions of this reflector over time

I could have said it's about a metre
wide by point six meters this way and

it's oriented within about
a degree of the nominal mean

Earth positions of the astronauts
did a good job orienting it and

nobody can convince me that
the lunar landings were faked.

It doesn't work on me.

OK So this is just raw fast
we broke all the previous

records by roughly a factor of seventy in
terms of our photon return rate and I've

been bragging about you know photon for
shot level that's compared to you know.

Sort of one every five hundred pulses.

For.

Sorry that's when every fifty.

That's the best that Texas has done.

And typically it's more like
one out of every five hundred.

So we can also operate at full moon which
is important for several reasons one

of which is that the equivalence principle
signal is maximized at full moon.

If you think about this polarization
of the Moon's orbit toward or

away from the sun that signal
is maximum at full a new moon.

So it's really nice to be of operate
a full moon we can't operate at new moon

the sun is a major scourge for us.

So this is sort of our data
accumulation rate sort of steadily

ramping up on the different are flatters
Apollo thirteen fourteen eleven in

the two Soviet reflectors more
on that later uncertainties

per night perf Lector
are typically a few millimeters.

Now bear in mind that previous
experiments were typically

one really two or three centimeters.

So even off this chart so all of our
measurements are quite high precision.

And it's really just because
we're getting tons of photons.

It's just beating down the statistics.

Our average median nightly range error
is about one and a half millimeters.

Skip that slice not terribly important.

We are working very hard on the model of
development right now you have to model

everything that can influence the Earthman
range to make sense of the measurements

and that includes not only obviously
some relativistic gravity model for

how the solar system works and
in body kind of mechanical.

Dynamical system but these are not
point masses that are they are finite

masses that have shape and the Earth and
Moon torque each other in the sun torques

the bulge of the earth in the bulge of the
moon and that leads to rotational dynamics

and if we could measure to the centers
of these objects that would be great but

we're on the surfaces so
we have to understand.

The.

Tilt and orientation.

And there are lots of things that can
displace the site a high pressure

atmospheric system comes in and
it loads the crust down it's a lot of mass

locally and it pushes the crust down a few
millimeters in groundwater after heavy

rains the area can be depressed by
a few millimeters and even ocean

loading in New Mexico where this telescope
is water piling up on the California

coast pushes down the region and off
that New Mexico sinks a little bit too.

So there are all kinds of things
like that that we have to model.

And we we know a lot of
the facts that we haven't yet

include in the model so
we basically have our work cut out for us.

I like to put it in Rumsfeld terms
these are the known knowns we also

have a list of known unknowns and then
we are scared of the unknown unknowns.

But anyway that that's just a point
that we got a lot of things to work

on in the model and
if you're a little miffed by that.

Keep in mind that the science signals
were after a very well defined discrete

frequencies and most of these things
are broadband influences and so

it helps to do a good job modeling
them and good sampling but

it's not likely to mimic
the physics that we're after.

Now one thing we are doing that.

I think is kind of.

Nice is.

We've got a superconducting of
emitter installed at the site and

it's a levitating niobium sphere at
liquid helium temperatures that is so

sensitive to gravity that if you raise
the thing one millimeter it senses that

gravity is weaker because you're
farther from the center of the earth.

And we can use that to
characterize the tides measure

the vertical displacements of the crust.

And I really like it because we're
using gravity as a tool in a.

Experiment to test gravity so it's not
circular don't worry about that but

here's For example a month.

Of Time series of raw data.

These are not massaged this is not a model
these blue points make up the raw data.

You see the lunisolar tides here
that are roughly correspond

with a half meter amplitude of vertical
motion of the site to the tides.

Obviously we need to take care of that.

When you have something sense of like
this you see some fun things like

one day I saw the grimmer
signal have three steps

of unequal size in a span
of a few minutes and.

I called the observatory to say
has anything been happening there

are moving any big masses around for
instance to do anybody to big Brito for

lunch and it turns out they say yeah
we have these three big concrete

blocks that we store basically right
under where your groove emitter

sits on the other side of this wall and
we use it once a year to do low test

the crane to make sure the crane is
capable of carrying heavy loads without

busting before we pick up an expensive
telescope mirror for instance.

And so

I can tell which one they move first they
move the one that's farthest from our.

Remember first and then this one in the
one that's closest made the biggest step

and then the next day I
saw them come back in.

But we can also see that asymmetric you
know gravitationally asymmetric telescope

dome as it moves around a different
as Miss We see a gravitational

influence from that and things like
after some of the big earthquakes.

On the Earth we see the earth in
this breathing mode the whole earth

rings in this sort of spherically
symmetric breathing mode

at a twenty minute interval.

So I've learned a lot about geo
physics by having this thing around.

Let's see this is part of
a time series of the grad.

Signal The blue is after you
subtract a tidal signal and

most of that is loading most
of it's from Ocean loading but

you can see that we now have data to
tell us what winning is looking like and

after we model the loading
we're left with a red trace.

And some of the jumps in the red
tracer from that telescope down moving

the discreet discontinuous pieces.

We also have near the site
precision G.P.S. installation.

So we can see how the site is
moving over time we can see a North

drift in East drift in a vertical I don't
know what you call that but it's big.

It's many millimeters we should know
about this and so this is something that

you know we might rely somewhat on models
but we also feel as experimental is much

better having measurements
on what the site is doing OK

let me tell you about one little story and
let me ask what's the usual stop time for.

OK so one story about a surprise that we
had which is we always expect one hundred

percent from anything we expected the
reflectors to operate as if they were new.

But what we got instead was something
like a ten percent performance that did

dept drop down to one
percent during full moon.

And we somewhat jokingly started calling
it the full moon curse and we thought it

was just low number sampling we you know
some werewolf is causing problems or

whatever but it stopped being funny
anymore after a while it was always there

and so this is a plot of a lunar phase so
full moon is right here and

the return string from our reflector
on a logarithmic scale we see

why variety of conditions were very
sensitive to the atmospheric conditions.

But there's a big hole around
full moon where we never get.

One photon for shot.

And that you know has held up over
time and it's not just that we're have

a high background level yes
the background does go up a full moon but

the signal is coming down and signal to
noise suffers but it's really a drop in

signal strength not just signal
to noise and you can see it.

I'll skip the top and

you can see it in the early literally as
arranging record in the early seventy's.

There's not much of a dip here full moon
but by the late seventies to the eighty's

we started to see something
cut in sort of a factor three.

CUT HERE IN OUR stuff that's
maybe a factor of ten or fifteen.

So it's appeared to get worse over time.

And so.

What we believe to be
wrong with it is that

there was a thermal problem likely
from dust and the idea is that

normally light comes into the corner
Cuban reflects out with barely any loss.

OK there's a little bit of
surface reflection here but

this is total internal reflection.

It's perfect should be perfect but
if you put dust on the front surface.

Now you have to traverse that front
surface twice and you get a pretty

significant dimming But what's worse is
that for Moon these arrays are placed so

they point at the Earth at full moon the
sun is basically right behind the earth.

So the sun is shining right down
these sort of recessed cavities and

when you put light on to this
dark dust you heat it up and

it turns not red hot but
Power Point is only so

good but you heat it up and you create
thermal gradients within the corner.

Q And when you have a thermal
gradient the refractive index now

depends on temperature and so
you have a refractive index radiant so

it's hot near the surface
cool near the tip and for.

Light paths that go deep into the corner.

Q. they experience a cool route and

travel through more quickly because
the refractive index is as low or

near the surface for exterior rays
you stay near the surface and

warmer material and you get retarded or
delayed because it's a slower path

in the what what that means is you send
a plane wave in but what comes out has

a spherical wave front and stay verging
And so you're losing light and so

the signal intensity drops down when these
corner cubes are exposed to to light.

We did a lot of modeling of
corner cubes these are models and

these are laboratory
experiments in my lab for

just checking that we can model
the diffraction patterns.

There's the total diffraction pattern from
a corner Q When these are horizontal and

vertical polarization come ponens So
we try to understand that well enough so

we can understand what happened to that
diffraction pattern as you tilted it and

what happened is you crank up
a thermal gradient within the corner.

Q.
And

we found that it only
takes about a four degree.

Kelvin difference between the surface and
the vertex to

actually destroy the central
radiance of the return.

So that's not big the moon goes up and
down between you know two

fifty one hundred fifty to three
hundred fifty Kelvin about.

And so
that's a huge range a little residual of

four degrees of thermal imbalance
not hard to comprehend.

And we had this great opportunity to
test this idea when the moon crossed.

Through the Earth's shadow.

It's a celestial light switch turn
off the Sun see what happens.

And here's a nice pictures that resulted.

If nothing else I would have
been happy with the pictures but

this actually spot is from our
laser beam hitting a high band

of clouds the actual target
we're aiming for is about here.

But it hit some clouds and

reflected in the real laser
went all the way to its target.

I wish it were that easy to see the thing
hitting the moon but it's just not so

here's what I expect as the illumination
goes down and comes back up.

We expect that the hot front face will
cool off as it radiates to space becomes

a cold front face and then it goes back
to hot when the sun comes back and

so hot face cold face and as zero
crossing as it transitions from being

thermally heated to the main source of
cooling and when it crosses through zero.

We expect the signal
to get really good and

then we expect that to happen again on
the tail end so that that's a prediction.

And here's the actual data from the
eclipse these lines mark when the eclipse

started and ended we get this
nice peak and then a law so

bad that we lost the signal and then
after the light came back we saw another

peak and this Grey Range is the historical
range that we've always seen and

we've never seen anything
outside of this range and

look how you know order of
magnitude stronger signal.

So we were pretty convinced that this has
a thermal cause and we likely know why

we had the last little fun bet is
that we after forty years of not

seeing this Lunokhod one
the first were Soviet.

Rover.

We were able to find it.

And you know its construction
is a little bit different but

it has a twin in Lunokhod too so there
are we've been using Lunokhod two for

decades successfully even though it's
a weak reflector it's gotten weaker

even weaker than the Apollo reflectors.

We're still I would use it but
nobody had found Lunokhod one and

I'm going to give you
the opportunity to find it.

So somewhere on the slide which is.

Less than or about a kilometer
across lurks Lunokhod one and

I'll give you a hint it looks like that.

And I haven't covered it up.

Anybody think this it.

It's right here.

It's the little thing that has a little
piece of light in the shadows going

after the last words craters have
shadows going off to the right.

So the light is clearly coming from
this lower corner shining this way and

this thing is sticking up and
making a shadow going the other way.

So the Lunar Reconnaissance Orbiter
still going around the moon took

some nice high resolution images of
the surface and found this little point.

And also got some good altimetry introduce
surface so we had a good definition

of where this thing was and
within a month we had

an opportunity to try to find this thing
and we opened up our gate normally

one hundred seconds is all we open our
detectors eyes for we opened it for

a little bit as wide as we could and
we saw this thing right at the edge of our

gate and that should always make you
suspicious when you see something right on

the edge of some instrumental
transition but we changed the timing so

the gate was a little bit later and later
and later and this thing stood right out

and was very strong we saw two thousand
photons in this first eight minute run.

And it was so
strong I said to the press that.

This thing has been silent for forty years
it's got a lot to say after all that time

which they like pretty well but really
the best quote came from a colleague

your discovery gives hope to all of us who
lost something during the seventies and

it works on so many levels.

It definitely winds.

Just a little cap is that as
a side consequence working

on this laser range and we're shooting
lasers through the atmosphere and we do.

I want to hit any pilots because that
would not only likely shut us down it

would shut down laser Guide Star adaptive
optics programs at places like tech and

Gemini and
it's just bad news to hit an airplane.

So the F.A.A.
has been happy to have spotters human

spotters watching the sky with a kill
switch standing out in the cold blistering

wind with bleary eyes in inattention
they're still happy with

it to have a kill switch to shut off the
laser if they see an airplane come by and

it's a hassle at a remote
location to schedule spotters.

I have found.

So I developed with a colleague
intellectual engineering at U.C.S.D.

a phased array antenna that we could
mount on the front of a telescope and

using the ratio of
a directional beam pattern

to an more broadening pattern we can tell
when an airplane is out in front and

we're listening just to the transponder
signals that it is issuing

forth as ground stations and
other aircraft interrogate and

it just responds on the directionally
we listen to the chatter and.

It's a very nice or

robust way to detect it through
clouds light conditions don't matter.

Never falls asleep never get believe.

And we've installed it at Apache
point to telescope and I'm right now

in the process of building them for tech
one Subaru in the two Gemini telescopes.

And connect to has already
started operating and has F.A.A.

approval to not use spotters anymore so
they're delighted.

And this thing doesn't cost more than
half a year's worth of spotter pay.

So it will pay itself off
it's my entire jobs program.

I'm very proud of it.

OK So

there's a last lied so
you know Apollo is a fairly new.

Operation that is able to get millimeter
range precision between Earth and

moon and we expect that to give order
magnitude gains in all those physics

parameters that Lynn arranging is provided
over the years and I didn't mention it but

you know Lynn arranging is out in
front of all the other techniques in

these tests of relativity pulse
our timing is a close second.

Maybe a factor of ten behind.

And it's possible that pulse or
timing could leap in front of littering.

But for now at least in arranging
holds the front seat if

we can get more out of the model.

We will be able to.

Put all of these tests
of fundamental gravity.

To good use.

And so now we're doing a lot of grappling
with the model we thought we could

rely on other people to do that and
it turns out you know the saying If you

want something done right
you do it yourself.

So we're getting into that mode and we
have had some nice surprises along the way

to keep things interesting as we await the
good science and I think I'll stop there.

Thank you.

Of the.

I've heard those rumors.

Right right.

In fact you know there are definitely
good opportunities to improve.

The measurement because
the dominant source of error

in our measurement is that
finite sized Ray and if we could

have individual corner cubes that are
sparsely position and resolve each one.

So that we get more of a comb
instead of a big mushy trap as oid.

We would do far better and we would
have an incentive to improve the ground.

System right now if I make a laser
post it's fifty pico seconds instead

of one hundred I don't
get anything from it.

I would have to go down to two pico
seconds before I got much benefit

with the present arrays and
that's kind of too far to push it.

Would not be easy.

So my ideas just have a bucket
of almost like golf balls but.

Corn cube prisms and just scatter and

you don't care where they go
you'll figure it out later.

Some will point to Earth and some want so
you could do it very simply and crudely of

course if you're going there with a person
you would do a proper job and but

yeah there's a great opportunity and there
are some new reflectors being designed.

That would also help mitigate dust
because that's something we know over

the long term you have to contend with.

I don't know if the Chinese
are are amenable but

I'm not on talking terms but I should ask.

I know the guy who is the sort of lead
developer of the next technology next

generation reflectors I don't know if
he's been talking to the Chinese or

not I know he's got his thumb out and he.

He wants to talk to anybody who's going so
he's probably tried and

maybe has been successful.

I don't know.

Well that's a good question.

A longer wavelength would certainly
be less sensitive to the dust

Very well I think you

know one thing we could do is just yank
the frequency doubler out of our laser.

And we'd have a double
The wavelength sixty four.

But the main problems with that for

us would be a our silicon based
detector is very crummy at sixty four.

It's almost transparent to
photons be it would become more

of an eye hazard because it's invisible
it would be hard to work with that.

We would have twice as many photons.

So there is one benefit but

I've never been audacious
enough to try it and infrared.

That.

It does

it doesn't appear to be
the data appear to show that

it took a few decades to settle in and

early on the Apollo fifteen array
was worked fine through full

moon with no dip and
then that appeared to develop but

it looks also like well I guess another
data point the Texas Station early on

ranged I think the Apollo fourteen while
the astronauts were still on the moon and

then did it again after they left and
didn't see any difference.

But there are mechanisms for
moving dust around the surface

of the moon electrostatic charging from
photo line is ation by X. rays and U.V.

from the sun can positively charge
the Greens in then if they get

dislodged from some other disturbance like
micrometeorites they can loft around and

move my best guess is that there's
about a fifty percent fill

factor of dust right now so
to the eye it wouldn't look that bad.

But that's enough to
cause these are facts.

So what we do it is tough.

That's the toughest thing we have to do in
a sense because we have a one arcsecond

sized beam we need to point within one
arcsecond before we see anything in our

detector I didn't really say it's only
one point four arcseconds Square.

So it's also very tiny
postage stamp on the sky.

So what we do in practice is
I've got a model of the moon

that has it's sort of orientation at
the time in our top a center correction

depending on where we are and so it's also
that crater so we point first at a crater.

That's nearby or
some physical feature a hill or

something that stands out and
we know where that is relative.

The other so we move to that and
get the telescope offsets and

then we do a small move over.

Now sometimes that
the reflectors in the dark and

that's a more challenging thing
because you have to pick a crater.

That's on the illuminated Crescent and

then move over blind and
you don't have any references.

Once you're there but
typically we're within one arcsecond so

very small hunt pattern is
enough to pick it up and

most of the time we find
it within the first minute.

Once we found one.

And we're locked into the sort of
offsets we skip to the other one and

we pick it right up.

So that doesn't tend to be
we hunt once per night and

once we find one we can run the table.

You know.

Well there is a coherent back
scatter phenomenon in the moon.

So the full moon is actually forty
percent brighter than one would predict.

In fact I do have a slide that shows.

Our own measurements of
the lunar background.

So there's a spike and

you might not believe it because there
are some other specks too but the moon.

Does get forty percent brighter at
full moon than it would be within

about five degrees of full moon part
of that is called Shadow hiding.

You know you're not the sun is coming in.

It's a male you're looking so

you don't see the dark parts of
shadows of of fluffy objects.

But the other aspect is you do
get this coherent back scatter in

the crystals of you know it's
basically class the moon

is kind of obsidian glass and
so you do get some of that but.

Not to the not to the extent that we
would expect any confusing signals.

Partly because they're spread.

We're hitting terrain that you
know typically we're shooting

at a section of the moon.

That's Telt at relative to us.

Only if we hit right
in the middle is flat.

So we're hitting something that's tilted
and so our two kilometer beam comes

back with a lot of you know spread in time
and so it just looks like background.

Yeah

I think so.

First of all both rovers moved around for

maybe ten months on the moon during
the lunar day when it had solar power and

then they were parked at night and
oriented toward the earth so

that the reflector was facing Earth and
they could get a signal back and

so they did arrange to it several times
while it was still moving around but

there were no reported sightings
of it after it had stopped.

So one of the theories was it
parked in the wrong orientation.

I did run across one old Russian
reference that said yeah in one

thousand nine hundred five.

We did see it.

But the coordinates were never properly
transmitted to the international

scientific community in the West and so
it sort of remained a secret little too

was found just because we got lucky and

found it just in a search but
Lunokhod one wasn't found and

it wasn't long after that that other
reflector started appearing on the moon we

knew where those were and
people just stopped looking.

And it's a hard it's
hard enough to find one.

Even when you know where it is
especially the earlier laser

ranging operations where you get maybe
ten photons back in ten minutes.

So it was just a hard game and
only even with our much

higher photon rate it was almost a lost
cause to search a five kilometer region.

And it's not but the beam with this two
kilometers that's not hard to search

spatially it's the temporal
dimension that is the most limiting.

Thank you.