"A NEO Search Program Using the Pan-STARRS PS1 Telescope"

ranchocucamongabrrrΤεχνίτη Νοημοσύνη και Ρομποτική

6 Νοε 2013 (πριν από 3 χρόνια και 11 μήνες)

143 εμφανίσεις

3
rd

Annual Progress Report


Submitted to the


National Aeronautics and Space Administration



for



Grant No.
NNX08AR22G

"
A NEO Search Program

Using the Pan
-
STARRS

PS1 Telescope"

Program Manager: Lindley Johnson


For the period

29 March 2010

to
XX March
2011



Submitted by


Robert Jedicke

& Richard Wainscoat

Principal Investigator
s

Institute for Astronomy

University of Hawaii

1.

Executive
Summary

For most of the time period covered by this report the PS1 telescope failed to
perform at the expected level. While the PS1 survey officially began on 13 May
2010 the system did not discover its first NEO until 16 September 2010. The
discovery rate was l
ow till the end of January 2011 such that by that time the
system had only recorded about a couple dozen
NEO
discoveries.

But on the night of 30 January 2011 (UT) the MOPS team was given control of
the surveying to showcase the system’s NEO discovery cap
ability and it
established a record by discover
ing 19 NEOs
. The night’s success catalyzed
changes in the general PS1 survey strategy that were implemented in the last half
of February 2011. Bad weather and an unfortunate mechanical failure of the
camera’
s helium cooling lines
then played a factor in reducing the NEO discovery
rate from the end of February thru early March 2011.

However, in the
one
-
week

period from 9
-
16 March 2011 PS1 reported 29 NEO
candidates to the MPC. About 10 of them made it on th
e NEO confirmation page
but only 2 are currently listed as PS1 discoveries. We are confident that a large
fraction of the NEO candidates reported to the MPC are in fact NEOs


the
reason that so few of them have become discoveries is because during that w
eek
the PS1 Image Processing Pipeline was particularly bad at processing the data in
a timely manner. Most of the candidate NEOs were reported to the MPC 1
-
2 days
after they
were
detected such that the tracklets provided insufficient leverage for
followup
.

Despite the problems described above we are optimistic about PS1 NEO
performance in the coming year. We believe that the PS1 survey strategy is now
about as good as it can be for NEO discovery given the constraints of surveying
for other science goals a
s well. We have seen major improvements in the image
processing pipeline performance and expect data to be available to MOPS in a
timely manner for the remainder of the year. In short, we believe that PS1 will be
either the first or second NEO discovery
system this year.

The portion of this grant used to support PS1 Operations is mostly used to cover a
small number of budget items such as the PS1 observers, power, etc. Much of the
funding to PS1 Operations from this grant during this time period was not
received due to PS1 not performing well. Furthermore, PS1 Operations was
supposed to provide software support for MOPS that has not materialized so we
have used some of the original PS1 Ops funding to support a MOPS software
engineer
.

MOPS continues to fu
nction well and rise to the challenge of processing the ever
changing PS1 data stream.

2.

Grant Organization

This grant is divided into two components:

2.1.

PS1 Operations Support (PS1 OPS)

This component of the funding is devoted to supporting the PS1 telescop
e
operations. The PS1 telescope

operations

is currently being handled by the PS1
Science Consortium (PS1SC), a group of about a dozen institutions that
contribute about $3M/year to support PS1 operations. Since this grant provides
about $0.5M/year for PS
1 operations (with numerous caveats) it implies that the
NEOO program is supporting about 14% of the total operations cost. PS1 OPS
covers all operational aspects of the PS1 facility including but not limited to
observatory maintenance, training and arran
ging of observers, camera
maintenance, surveying strategy, telescope support,
the
image processing

pipeline (IPP)

and delivering the image data products to
the consortium through
an image and detections database.

The PS1 OPS funding from this grant are del
ivered directly from the co
-
PIs to a
co
-
I, Ken Chambers, who is the PS1 Director in charge of all PS1 OPS. The NEOO
funding is used to support a small and specific sub
-
set of components of the PS1
OPS.

PS1 operations does not cover the opera
tions of any ‘science clients’ i.e. scientific
analysis and interpretation of the processed PS1 data.

2.2.

Moving Object Processing System (MOPS) Support

The MOPS
is a PS1 science client that handles

all aspect
s of identifying moving
objects in the PS1 data.
This includes but is not limited to linking transient
detections provided by the IPP on a single night into ‘tracklets’
,

that are then
lin
ked across nights into ‘tracks’,
which are then fit to orbits to determine if the
detections are consistent with being

a solar system object. If the detections are
consistent with being the same solar system object the detections are added to
the MOPS relational database. MOPS takes care of attributing tracklets to
previously known objects and precovering orphaned track
lets when they can be
associated with a newly derived object.

Roughly half of this grant’s funding is directed towards MOPS support to enable
continued improvements of the NEO detection capability, monitoring of the
system performance, modifications to wor
k with the PS1 data as it evolves into a
fully functional sky survey, NEO visual confirmation

by ‘NEO Czars’, data quality
checks, scientific analysis, feedback to the PS1SC on survey strategy efficacy, etc.
The vast majority of the grant support is devot
ed to paying salaries for key MOPS
personnel as described below.

3.

NEOO Funding Profile


Figure
1



Delivered and expected funding profile from the NEOO program on this grant
.

The
original grant funding was for a total of $2,525,00
0 over 3 years. Delivery of the funds to PS1
was delayed for a number of reasons. The NEOO program manager expects to deliver the
remaining funding during the first year of extension.

For the
first six months of the
one
-
year

period covered by this report

(
roughly
April 2010 thru March 2011 inclusive)

this grant was underfunded relative to the
expected profile. The NEOO program manager delayed delivery of the funding
until PS1 demonstrated an ability to discover NEOs. The delivered funding to the
MOPS co
mponent of the grant was higher than the OPS component because most
of the problems with NEO discovery were due to PS1 operations upstream of
MOPS.

Late in 2010 we received an injection of funding because PS1 began finding NEOs
and it appeared that many of

the operational problems were being addressed (see
below for details). Much of the new funding was directed to PS1 OPS since that
component of the grant was underfunded relative to MOPS as described above.
More funding was delivered later in 2010 and ea
rly in 2011

and distributed
between MOPS and PS1 OPS. The new funding was distributed between the
components in favor of MOPS even though the original grant proposal stated that
funding would be equally distributed between the two. This is because the
pr
oposal assumed that the PS1SC would provide a software engineer to MOPS to
continue MOPS integration into the PS1 system but that engineer never
materialized. Instead, MOPS began the process of hiring a new dedicated junior
software engineer.

4.

PS1
OPS

A de
tailed status report on

PS1 Operations as of 6 December 2010 is provided in
the attachment
PS1SC_Status_Report_December_2010
.Chambers
.pdf

(Ken
Chambers, PS1 Director, co
-
I on this grant, and
de facto

manager of the PS1 OPS
component of this grant).
It pro
vides a comprehensive description of the state of
the system at the end of 2010 and

mentions

how
the Modifed Design Reference
Mission
(MDRM)
will improve NEO discovery efficiency. The MDRM was
implement
ed

late in 2010

but by late January 2011 it was clear that it was not
providing the expected jump in NEO discoveries. In February 2011 MDRM
-



implemented

and it does seem to be successful at increasing the NEO
discovery rate to a respectable level (see section
5.5
). Due to a string of bad
weather and an unfortunate break in the camera’s helium lines in early March we
cannot yet provide an accurate assessment of the expected i
mprovement PS1 will
see in terms of NEO discovery.

In general, we are finding that on a clear night
with an operational telescope we are managing to get 2
-
3 NEOs on the MPC
Confirmation Page.

5.

MOPS

5.1.


MOPS
Personnel
supported by this grant

The following

individuals
’s

salaries were supported wholly or in part by this grant

during the period of this report

(
in alphabetical orde
r
)
:

5.1.1.

Larry Denneau (Manager & Software Engineer)

During the time period of this report
25
%
of
Denneau
’s salary was supported on
th
is grant.


Most of his time was spent

commissioning MOPS for operations

with
PS1 data
.

E.g.

ensuring that the data reported to MOPS by PS1 was in the correct
format for MOPS ingestion and trying to handle the very large false source
detection rates.

5.1.2.

Mikael Granvik (post
-
doc)

Granvik wa
s
supported

100% on this grant

throughout this report period until

his
employment
completed

on

4 March 2011
.

He has accepted a position at the
University of Helsinki.

In the absence of
significant

PS1 data he 1)
devel
oped a
technique to measure PS1 asteroid detection efficiency

2) provided orbit
determination support for PS1 asteroid identifications 3
) work
ed

on determining
the number and orbital properties of tran
sient natural Earth satellites 4
)

supervised graduate s
tudent Eva Schunova on a) identifying NEO families and b)
simulating the production and detection of tidally disrupted NEO families and 5)
assisted in the analysis of genetically related NEO pairs with colleague Dr.
Shinsuke Abe
and
6) was critical to impl
ementing human visual confirmation of
PS1 NEO candidates
.

5.1.3.

Eva Schunova (non
-
IfA graduate student)

Schunova is a Ph.D. graduate student
in the Faculty of Mathematics, Physics and
Informatics at Comenius University, Bratislava, Slovakia
.
She received a
Slov
akian national scholarship to perform research anywhere in the world at a
location of her choice and she approached us to work on NEOs and Pan
-
STARRS.
Since she would cost no money we were happy to accept her help after confirming
her abilities with her r
eferences.
She arrived at the IfA on 22 April 2010 and was
supported by her
scholarship

through mid
-
September.
She has been working on
determining whether any NEO families exist in the known NEO population using
rigorous statistical techniques. We expec
t that it will lead to a submitted Icarus
paper within a couple months. A follow
-
on paper where she will set limits on the
rate of tidal disruption by close Earth approaches should be submitted well before
the end of the year. She then decided that she w
ould like to complete her Ph.D.
work on the research she had begun with us
and has
obtained permission from
her home institution to do so. Thus,
Schunova has been supported since
22 Sep
2010

by this grant

at a graduate student level

to work on NEO familie
s
.

5.1.4.

Denver Green (software engineer)

Due to the PS1SC not providing PS1 software engineer for MOPS we advertised
for a junior software engineer in Feb 2011. At the same time Will Burgett, the
Pan
-
STARRS Program Manager, offered us the option of buying 50%

of their new
MOPS software engineer

to obtain 100% of his time
. (Note that PS1 and Pan
-
STARRS are different entities.
)

We decided that this was a better idea, and
provided a more experienced engineer in a
timelier

manner, than us hiring an
engineer ours
elves. Thus, Green began working on MOPS

on 1 Mar 2011

under
the direct supervision of
the MOPS Manager (Denneau). We have developed a
prioritized work plan thru the end of 2011.

5.2.

Other MOPS Personnel

Other individuals who were critical to MOPS developmen
t, support and NEO
research during this report year
:

5.2.1.

Tommy Grav (
Johns Hopkins
)

Grav was an IfA

post
-
doc working on MOPS for three years at its inception. He
has remained active in following and contributing to MOPS development and was
particularly helpful in applying his knowledge from analyzing WISE spacecraft
data to PS1 this year.

5.2.2.

Henry Hsieh (
IfA Hubble Fellow)

Hsieh is working with Kleyna (see below) to identify comets and especially main
belt comets in the PS1 MOPS data. This work relies only on morphological
parameters of the detection

so they use postage stamp images centered on each of
hu
ndreds of thousands of detections to identify cometary activity.

5.2.3.

Robert Jedicke (co
-
PI)

Jedicke and Wainscoat continue to perform PI duties.

5.2.4.

Jan Kleyna (IfA
-
NAI postdoc)

Kleyna is working with Hsieh on main belt comet detection. In particular, Kleyna

developed the MOPS postage stamp server that provides postage stamps for all
detections that appear in tracklets that are 1) attributed to a known object, 2) in a
derived object or 3) in a tracklet containing at least 4 detections.

5.2.5.

Andrea Milani & others
(University of Pisa)

The MOPS team has relied heavily on many people through the years but Andrea
Milani stands out above the rest as having made critical contributions to the
system without financial compensation
.

He and Fabrizio Bernardi, Davida
Farnocc
hia and Givoanni Gronchi spent 4 weeks at the IfA in Aug 2010 to refine
their orbit determination software for use with MOPS and also to study
opportunities for removing false source detections based on their morphological
properties.

5.2.6.

Richard Wainscoat (co
-
PI)

Wainscoat and Jedicke continue to perform PI duties.

5.3.

MOPS Development
s

MOPS
has been

a mature software system
for many years that was simply
waiting
for
PS1
data
. Thus,

r
elatively little MOPS d
evelopment
has taken

place

in many
years
. Instead, most

of Denneau
and Granvik’s time
during this report period
was spent
trying to
get PS1 data
into a format that could be
processed through
MOPS.

The PS1 data flow and quality changed dramatically every month or so
and often from night
-
to
-
night due to image q
uality or other processing issues
.


The fundamental problem was/is the false source detection rate. For most of this
report period the false detection rate was 10
-
100x larger than specified in the
MOPS input requirements. There was little that MOPS coul
d do to handle data
with this false detection rate


instead,

the PS1 surveying

strategy needed to be
modified

to acquire 4 images at each boresite location to reduce the false tracklet
rate to a manageable level. Even then, it was extremely rare to obtai
n 3 or more
nights in a lunation with 4 images at each boresite so it was impossible to link
tracklets across nights as MOPS was designed.


Once 4 images were regularly acquired during surveying MOPS was quickly able
to form tracklets

containing 3 or more detections. Much to our disappointment,
the false source detection density was still so high that
on many nights there
could be hundreds of false 3 or 4 detection tracklets


a set of false detections
that appear to be real objects m
oving at a consistent rate between the image
frames. The only way to ensure that we could extract real objects and fast moving
NEOs from these tracklets was to initiate visual examination of each tracklet
candidate. This was a disappointment to the MOPS
team because it was contrary
to the intended MOPS
operations that were

not supposed to require human
intervention.

Denneau quickly put together a ‘NEO Czar’ page to facilitate the examination
process. It presents the user with postage stamps of the dete
ctions in each
tracklet along with important information such as

e.g.

the MPC NEO digest score,
rates of motion,
and
whether the object is linked to a known object. We also set
up a gmail NEO Czar calendar to coordinate who on the MOPS team would
examine
the tracklets. In the beginning, Denneau, Grav and Granvik did much
of the czaring until the process was refined enough to train Wainscoat, Jedicke
and Schunova. We are also in the process of training volunteers from the
‘Friends of the IfA’, a group of

educated and eager astronomy enthusiasts who
support the IfA financially and organizationally. The NEO Czaring page is also
being improved and we expect that the entire process should take no more than
15
-
30 minutes per clear night.

One of the problems w
ith NEO Czaring is that the data from the PS1 Image
Processing Pipeline does not arrive on a regular schedule. In theory the
processed data should be available to MOPS before local noon but in practice it
rarely arrives till late in the afternoon. This m
akes it difficult for the NEO Czar to
process and report MPC candidates to the MPC in a timely manner.

Despite the problems
,

PS1 began reporting NEO candidates to the MPC on 28
August 2010 and our first confirmed NEO discovery was 2010 ST
3

on 16
Septembe
r 2010
.

As of the time of writing PS1 has
XX

confirmed NEO
discoveries including
XX

PHAs.

We expect the discovery rate to increase even
more in the first half of 2011 and then level out as the PS1 surveying finally settles
into a consistent pattern.

T
he quantum leap in PS1 and MOPS performance was catalyzed by the Pan
-
STARRS PI (Nick Kaiser) choosing to give the MOPS team one good dark night
for a ‘NEO Demo Night’ to showcase what PS1 can do. NEO Demo Night is
described in detail below (section
5.4
) but its main effect was to show that PS1
could

be extremely effective at identifyi
ng NEOs if the survey strategy was
des
igned to find them. Based on the stunning success of NEO demo night

PS1
adopted a more asteroid and NEO friendl
y survey strategy
beginning in mid
-
Feb
2011
suggested by the MOPS team

about six months earlier
. While there is only
limited data to support the efficacy of the strategy the first results look promising
as

shown in

Figure
2
.

Figure
2



Left: eccentricity vs. semi
-
major axis for 2911 derived objects from early in 2010.
Right: same distribution for 1266 derived objects from early in 2011.

Figure
2

shows
the
dramatic improvement in

PS1+
MOPS

performance over
the past year. The left side of the figure shows the results for derived
objects, multi
-
night tracklet linkages
,

for ‘demo month’ in early 2010 while
the right side shows the same distribution for derived
objects using the
latest PS1 su
rveying strategy implemented
part
way through

Feb 2011
.

A
year ago we were plagued by false detections that generated false NEO
signatures for false tracklets and linkages. Now there are zero false NEO
candidates in a sample

of 1266 derived objects.

The semi
-
major axis
distribution in the main belt clearly shows the large scale structure of the
belt including the

6 resonance that cuts off the distribution of objects on
the inner edge and the mean
-
motion resonances that caus
e the lack of
objects at about 2.5 and 2.9 AU. We identify Hilda’s at about 4 AU and
several Jupiter Trojans near 5.1 AU.

5.4.

NEO Demonstration Night
1

In
early
January

2011

the
Pan
-
STARRS PI
,

Nick Kaiser
,

invited the
MOPS
team and
PS1SC Inner Solar

System key

project (KP1) to devise an
observing plan for an entire

night of observing using his PI discretiona
ry
time. The ‘
NEO Demo

Night’ would be

dedicated to
demonstrating

PS1's
NEO and PHO discovery capability
.

The night of
30

January

2011 UT
was
selected du
e to it being near new moon and the
good weather forecast
.

This grant’s co
-
PI Wainscoat devised an observing plan consisting of the
evening and morning

sweetspots


between

60
-
90 degree
s

solar

elongation
(
the ESS and MSS respectively)

and opposition solar

system fields (OSS)



1

Thanks to Larry Denneau for providing much of the text in this section.

near the ecliptic

and opposition (see
Figure
3
)
.

The PS1 s
urveying
was
composed of

4
-
visit ‘quads’

to each footprint using 45
-
second w
-
band
exposures
. As a test at low surveying altitudes we decided to perform a

small

chunk


of r
-
band exposures at low

solar elongation in the
ESS
.

It
was thought that the r
-
band might provide better image quality and depth
at low altitude because of the decreased atmospheric dispersion relative to
the wide w
-
band.


Figure
3

Realized surveying fields for NEO Demo Night on the night of 30 January 2011 UT.
The solid light gray fields were observed while the empty gray polygons were optional fields.
The chunk of fields on the right is the ESS, the large chu
nk of fields to the right of center are
near opposition and the ecliptic, and the chunk of fields on the left close to the Sun are the
MSS. The ecliptic is the red solid line and the galactic equator is represented by the blue solid
line.

Overall, 596

exp
osures (432
near opposition and

164

in the

sweetspot
s
)
were acquired at 149

different boresights covering about

1000
deg
2
.

Everything operated very smoothly on NEO demo night with IPP
processing the images nearly as they were acquired.

By 9:16

PM
HST 29
January 2011
we

had submitted our first object to the
MPC and the object was posted on the
NEOCP

and b
y late
afternoon on 30
January 2011
HST
about

20 objects had been accepted to the NEOCP.

A

M
OPS processing glitch held up processing
of
a couple

OSS chun
ks until

Monday yet

by mid
-
Monday at least 30 objects were on the NEOCP.

While the original design of the Pan
-
STARRS surveys was to provide self
-
follow
-
up

of all asteroids this could not be accomplished during NEO demo
night. Instead, we informed all our followup collaborators (in particular:
Tholen & Michelli on the CFHT and UH 2.2m, Ryan at the Magdelena
Ridge Observatory, McMillan at Spacewatch, Lister

and Armstrong

at the
Faulkes Telescope and Las Cumbres Observatory,
and Robert Holmes at
Astronomical Research Observatory
)

about the upcoming demo night and

Figure
4



Maximum number of NEOs discovered at a single

site in a single night for the 5
most prominent NEO survey programs. No other program discovered more than those
shown in this figure. The month and year of the accomplishment for each program is given at
the top of the program’s column.

forewarned the
MPC.


Unfortunately, the w
eather was poor on

the
mainland after the 29th so our follow
-
up partners were less

effective than
usual

and a significant portion of the
follow
-
up

confirmation was left to
Tholen and Michelli on

the CFHT and UH
2.2m.



Figure
4

shows that a

total of

19 NEOs
were confirmed as
discover
ies
by
PS1 on Demo Night, establishing a record for objects discovered on

a single
night

at a single site
.

There were never so many candidates reported by a
single site to the MP
C NEOCP and we estimate that at least a few of the 11
objects that made it on the NEOCP but were not confirmed were actually
NEOs. Furthermore, we know that about an equal number of NEOs were
‘hiding in plain
sight’ in

the data (Jedicke AJ 111 #2, 1996)


they were
detected but their rates of motion did not flag them as being NEOs. When
PS1 implements the MOPS surveying plan we expect to extract those NEOs
from the data as well. Finally, due to the remaining false source detections
and the unavailability

of trail fitting software, the identified NEOs were
constrained to have rates of motion between about 0.1 deg/day and 2
deg/day so that we could not identify very close NEOs with long trails or
NEOs moving at very small rates. Thus, despite establishing
a single night
NEO discovery record we suspect our NEO detection efficiency in the fields
we observed was less 50%.



Figure
5



Left: Main belt object detection efficiency as a function of apparent V magnitud
e in
the ESS. Right: The same but for an
OSS

chunk.

The solid line in both figures is a fit to the
efficiency.

On the other hand, the main belt object detection efficiency was ~90% for
objects that were well above the detection threshold
(see

Figure
5
)
.

Note
the difference in the fitted efficiency curve in the ESS and OSS. The
magnitude at which the ESS detection efficiency is 50%
of
the maximum is
about V=22 but in the OSS the curve fades gently instead of fast and the
sa
me point occurs at about V=21.5.
Ken Chambers,

the PS1 Director and
co
-
I on this grant, suggests that this effect is due to

a known IPP bug
related to

the
variance.

(This problem has been fixed and the latest PS1
data is a vast improvement over what MOPS

has processed in the past.)

Andrea Milani's
known_server

software attributed over 18,000
numbered
MBOs
in the NEO Demo Night data or

close to 200 asteroids per
PS1
image
footprint.

There are likely another equal number of known
unnumbered objects in the
data as well as another equally large number of
unknown objects. Unfortunately, for various reasons these data have not
yet been reported to the MPC.

5.5.

MDRM
-


The second version of the Modified Design Reference Mission (MDRM
-

)
was implemented midway thru o
bserving cycle (lunation) 136


essentially
the month of February. MDRM
-



suggested by the MOPS team

in
September 2010

as a new surveying strategy that would have minimal
impact on other PS1SC science but a dramatic improvement on the
detection and
fol
low
-
up

of asteroids and especially NEOs.

The first
implementation of the MDRM (

) late in 2010 neglected to enable many of
the key aspects of MDRM
-


that we did not realize improved
asteroid discovery capabilities. Furthermore, continuing bugs in th
e IPP
processing exacerbated the surveying problems. However, with the
catalyst of NEO Demo Night (Section
5.4
) MDRM
-

was finally
implemented. The
strategy

is illustrated in
Figure
6
.



Figure
6



Schematic representation of the MDRM
-

su牶敹eng⁳瑲t瑥g礮
The basic idea of MDRM
-

is to obtain many more tracklets for each
object in a single lunation such that at least one or two of the tracklets
contains 3 or 4 det
ections. With the recent IPP bug fixes the probability
that a 3 or 4 detection tracklet is real is nearly 100% and these tracklets can
be used to ‘anchor’ the identification of derived objects


tracklets linked
across nights within a single lunation. Th
e right side of
Figure
2

shows the
very clean result of implementing this technique and the IPP bug fixes.
The derived objects show zero contamination from false NEO candidates
and all the derived orbits look real and give a snap sho
t of the solar system
in the observation direction out to Jupiter.

5.6.

MOPS Summary

The MOPS continues to perform well and is adapting to the ever improving
PS1 data stream. The system has demonstrated the ability to identify
objects in tracklets on a nightly

basis and to cleanly link tracklets across
nights into
derived

objects.

Figure
7



Total number of objects reported to the MPC in 2010. For various reasons PS1 has
only reported known multi
-
opposition objects t
o the MPC and only for a subset of the
available 2010 data.

Despite the numerous processing problems in 2010
Figure
7

shows that
PS1 still managed to report observations to the MPC for a competitive
nu
mber of minor plants.

In

fact, PS1 ranked 3
rd

in terms of the number of
numbered+unnumbered objects reported to the MPC behind the WISE
spacecraft and the CSS G96 site.
It is important to note that

PS1 has not
reported all the available observations due to the processing problem
s
described above


we expect to report 2
-
3x more detections for 2010 after
all the year’s survey data is reprocessed through the IPP and MOPS. It is
likely that once all the available data is reported to the MPC that PS1 will
take the lead in number of r
eported objects for 2010.

PS1’s NEO discovery rate was not good until the end of 2010
but

it
continues to improve. We believe that the 2011 NEO discoveries reported
to the MPC provide a better snapshot of expectations for the year (see
Figure
8
). But even the first few months of 2011 do not provide a good
indication of what we expect from PS1 since it incorporates only a tiny
portion of data acquired with the new MDRM
-

surveying strategy.



Figure
8



PHAs and NEOs reported to the MPC in 2011

6.

Outlook

6.1.

Funding

We have filed
and been granted
a first year NCE request with NASA thru
the
NSSC No
-
Cost Extension Web Form
. We have had discussions with the
NEOO Program Manager who has indicated hi
s intention to respect the
funding of the original grant and provide us the remaining $1050K in
funds this year. We expect to receive $300K early in the year and the
remainder late in 2011.

6.2.

PS1 Operations

PS1SC Operations is approaching survey maturity
from the standpoint of
the NEO discovery program. There are a few issues to be resolved with e.g
.
fine

tuning the 3pi surveying mode, resolving image processing delays that
often provide data to MOPS 2
-
3 days late, and the remaining false source
detection

problems. But the issues are no longer shackling MOPS
performance and we expect to see continued improvements that ensure
MOPS will find even more NEOs thru the 1
st

NCE year.

6.3.

MOPS

The outlook for MOPS during the upcoming report period is excellent. The
numerous improvements in PS1 operations will allow MOPS to operate
close to the manner in which it was designed and we expect to report
record numbers of minor planets, NEOs and PHAs

to the MPC this year.

With the promise of the remaining grant funding we are currently seeking
a n
ew postdoc

and have 3 strong candidates. We expect to make an offer to
one of them within days. We will also continue funding

Slovakian
grad
uate

student

Ev
a Schunova for her assistance with NEO czaring, data analysis
and research. The shared funding of the Pan
-
STARRS software

engineer
,
Denver Green, should allow MOPS to achieve its design goals by ensuring
that the PS1 input data meets the MOPS requirements
.

We will hire a new
junior software engineer late in 2011 to ensure a smooth transition to a new
MOPS software engineer after Green reduces his time on MOPS effective 1
Jan 2012.