Near detectors and systematics IDS-NF plenary meeting at TIFR, Mumbai October 13, 2009 Walter Winter Universität Würzburg TexPoint fonts used in EMF: AAAAA.

Near detectors and systematics

IDS-NF plenary meetingat TIFR, MumbaiOctober 13, 2009

Walter WinterUniversität Würzburg

2

Contents

Initial IDS-NF questions Beam and detector geometry Systematics Results for high energy NuFact Results for low energy NuFact Near detectors for new physics (examples) Answers to initial questions Systematics requirements (for simulation) Summary of new physics requirements

3

Introduction: Initial questions

What is the potential of near detectors to cancel systematical errors?(implies: need to address what kind of systematics …)

When do we need a near detector for standard oscillation physics?

What (minimal) characteristics do we require? (technology, number, sites, etc.)

What properties do near detectors need for new physics searches?

4

Geometry of decay ring

Need two near detectors, because +/- circulate in different directions

For the same reason: if only std. oscillations, no CID required, only excellent flavor-ID; caveat: background extrapolation

(Tang, Winter, arXiv:0903.3039)

5

Geometry of the beam

Beam diameter ~ 2 x L x

We use two beam angles: Beam opening

angle:

Beam divergence: contains 90% of total flux (arXiv:0903.3039)

Beam divergence

Beam opening angle

6

Geometry of the detectors?

(ISS detector WG report)

What are the physics requirements forthe geometry of the detectors?

7

Geometry: Extreme cases

Far detector limit:The spectrum is the same as the on-axis spectrum, i.e., the detector diameter D < 2 x L x , where is the beam opening angle, for any point of the decay straight

NB: Point source approximation d >> s (size of source) not required for this limit. The extension of the source can be desribed by

Near detector limit:The detector catches almost the whole flux, i.e., the detector diameter D > 2 x L x , where is the beam divergence, for any point of the decay straight

8

Assumptions for NDs

Only muon neutrino+antineutrino inclusive CC event rates measured (other flavors not needed in far detectors for IDS-NF baseline)

No charge identification At least same characteristics/quality (energy

resolution etc.) as far detectors No explicit BG extrapolation Fiducial volume cylindrical No systematical errors considered, which are

potentially uncorrelated among ND and FD (they are present, but they cannot be improved on with the NDs)

9

Different ND versions?

Near detectors described in GLoBES by (E)=Aeff/Adet x on-axis flux and

Some ND versions:Near detector limit

Far detector limit SciBar-size Silicon-vertexsize?

OPERA-size

Hypothetical

Nearest point

Farthest point

Averaged

=1: FD limitDashed: ND limit

(Tang, Winter, arXiv:0903.3039)

10

Extreme cases: Spectra

Some spectra:

~ND limit ~FD limit

(Tang, Winter, arXiv:0903.3039)

11

Systematics treatment

Cross section errors: Fully correlated among all channels, detectors etc. measuring the same cross section, fully uncorrelated among bins and neutrinos-antineutrinos (30% cons. estimate)

Flux errors: Fully correlated among all detectors in the same straight and all bins, but uncorrelated among polarities, storage rings (2.5% for no flux monitoring to 0.1%)

Background normalization errors: as IDS-NF baseline (20%)

12

Systematics, qualitatively

Near detectors important forLeading atmospheric and CPV measurements

Flux monitoring (by NDs or other means) important for CPV measurement

Almost no impact for 13 and MH discovery (background limited)

(arXiv:0903.3039)

13

Relevance of statistics

Event rates (10 years) extremely large

Physics is limited bystatistics in FD, notspectrum in ND

Near detector locationand size not relevant(caveat: elastic scattering for flux monitoring)

However, for new physics searches, such as e ->

s, es, size

matters!

(arXiv:0903.3039)

14

Atmospheric parameters

Atmospheric parameters measured at L=4000km:

At L=4000km+7500km no impact of NDs!

Unfilled: 30% XSec-errors, no NDFilled: Near detectors

(Tang, Winter, arXiv:0903.3039)

sin2213 = 0.08, CP=0

15

CP violation measurement

(Tang, Winter, arXiv:0903.3039)

IDS-NF systematicstoo conservative?

3

16

Low-E NuFact

„High statistics“ setup from (Bross, Ellis, Geer, Mena, Pascoli, arXiv:0709.3889)

E=4.12 GeV, L=1290 km 5 1020 useful decays per

polarity and year, 10 years, 20 kt mass x efficiency

Reference: 2% system. Our ND3 with IDS-NF-like

storage ring PROBLEM: We need

decay ring geometry for some applications!

(Tang, Winter, arXiv:0903.3039)

17

Low-E versus high-E NuFact

(Tang, Winter, arXiv:0903.3039)

Low-E NuFact: Systematics estimate seems quite accurateNear detectors mandatory!

High-E NuFact: Qualitatively different, since two far detectorsNeed something like Double Chooz/Daya Bay systematics?

18

NDs for new physicsExample: SBL e disappearance

Two flavor short-baseline searches useful to constrain sterile neutrinos etc.

e disppearance:

Also some interest in CPT-invariance test (neutrino factory ideal!)

Averaging over straight important (dashed versus solid curves)

Pecularity: Baseline matters, depends on m31

2

Magnetic field if (Giunti, Laveder, Winter, arXiv:0907.5487)

90% CL, 2 d.o.f.,No systematics,

m=200 kg

19

SBL systematics

Systematics similar to reactor experiments:Use two detectors to cancel X-Sec errors

(Giunti, Laveder, Winter, arXiv:0907.5487)

10% shape

error

arXiv:0907.3145

20

Summary: Answers to initial questions

What is the potential of near detectors to cancel systematical errors?

Cancels X-section errors; possibly useful for flux monitoring etc. When do we need a near detector to cancel cross

section errors? If we only operate one baseline for sure! Mainly needed for leading

atmospheric and CP violation searches. What (minimal) characteristics do we require?

(technology, number, sites, etc.) Two near detectors; at least as good as far detectors for ; not

necessarily magnetic field, site and size hardly important (statistics high) What properties do near detectors need for new

physics searches? Also e, detection; as large as possible (statistics matters!); magnetic

field; site application-dependent; maybe more sites Near detector characteristics driven by new physics requirements?

21

Systematics requirements For a more accurate simulation, PPEG needs to know

systematics treatment The simulation results depend not only on the numbers

for some systematical errors, but also the implementation of systematics (cf., Double Chooz, Daya Bay!)

What systematical errors (and how large) are there correlated/uncorrelated among Bins Detectors Storage rings Channels at the same detector Channels measuring the same X-secs …

Possible alternative (discussed via mailing list some time ago): Show also curve with „no systematics“?

22

Summary of (new) physics requirements

Number of sitesAt least two (neutrinos and antineutrinos), for some applications four (systematics cancellation)

Exact baselinesNot relevant for source NSI, NU, important for oscillatory effects (sterile neutrinos etc.)

FlavorsAll flavors should be measured

Charge identificationIs needed for some applications (such as particular source NSI); the sensitivity is limited by the CID capabilities

Energy resolutionProbably of secondary importance (as long as as good as FD); one reason: extension of straight leads already to averaging

Detector sizeIn principle, as large as possible. In practice, limitations by beam geometry or systematics.

Detector geometryAs long (and cylindrical) as possible (active volume)

Aeff < Adet Aeff ~ Adet

23

What we need to understand(for new physics)

How long can the baseline be for geometric reasons (maybe: use „alternative locations“)?

What is the impact of systematics (such as X-Sec errors) on new physics parameters

What other kind of potentially interesting physics with oscillatory SBL behavior is there?

How complementary or competitive is a near detector to a superbeam version, see e.g.http://www-off-axis.fnal.gov/MINSIS/