Neutrino and antineutrino inclusive charged-current cross section measurements with the
MINOS near detector
P. Adamson,
8
C. Andreopoulos,
19
K. E. Arms,
14
R. Armstrong,
12
D. J. Auty,
23
D. S. Ayres,
1
C. Backhouse,
17
P. D. Barnes, Jr.,
29
G. Barr,
17
W. L. Barrett,
30
D. Bhattacharya,
18
M. Bishai,
4
A. Blake,
6
G. J. Bock,
8
D. J. Boehnlein,
8
D. Bogert,
8
C. Bower,
12
S. Cavanaugh,
9
J. D. Chapman,
6
D. Cherdack,
26
S. Childress,
8
B. C. Choudhary,
8,5,
*
J. A. B. Coelho,
7
S. J. Coleman,
28
D. Cronin-Hennessy,
14
A. J. Culling,
6
I. Z. Danko,
18
J. K. de Jong,
17,11
N. E. Devenish,
23
M. V. Diwan,
4
M. Dorman,
13,19
A. R. Erwin,
31
C. O. Escobar,
7
J. J. Evans,
13,17
E. Falk,
23
G. J. Feldman,
9
M. V. Frohne,
10,3
H. R. Gallagher,
26
A. Godley,
21
M. C. Goodman,
1
P. Gouffon,
20
R. Gran,
15
E. W. Grashorn,
14,15
K. Grzelak,
27,17
A. Habig,
15
D. Harris,
8
P. G. Harris,
23
J. Hartnell,
23,19
R. Hatcher,
8
K. Heller,
14
A. Himmel,
5
A. Holin,
13
J. Hylen,
8
G. M. Irwin,
22
Z. Isvan,
18
D. E. Jaffe,
4
C. James,
8
D. Jensen,
8
T. Kafka,
26
S. M. S. Kasahara,
14
J. J. Kim,
21
G. Koizumi,
8
S. Kopp,
25
M. Kordosky,
28,13
D. J. Koskinen,
13,15
Z. Krahn,
14
A. Kreymer,
8
K. Lang,
25
J. Ling,
21
P. J. Litchfield,
14
R. P. Litchfield,
17
L. Loiacono,
25
P. Lucas,
8
J. Ma,
25
W. A. Mann,
26
M. L. Marshak,
14
J. S. Marshall,
6
N. Mayer,
12
A. M. McGowan,
1,14,†
R. Mehdiyev,
25
J. R. Meier,
14
M. D. Messier,
12
C. J. Metelko,
19
D. G. Michael,
5,‡
W. H. Miller,
14
S. R. Mishra,
21
J. Mitchell,
6
C. D. Moore,
8
J. Morfı
´
n,
8
L. Mualem,
5
S. Mufson,
12
J. Musser,
12
D. Naples,
18
J. K. Nelson,
28
H. B. Newman,
5
R. J. Nichol,
13
T. C. Nicholls,
19
J. P. Ochoa-Ricoux,
5
W. P. Oliver,
26
T. Osiecki,
25
R. Ospanov,
25,x
J. Paley,
12
V. Paolone,
18
R. B. Patterson,
5
Z
ˇ
. Pavlovic
´
,
25
G. Pawloski,
22
G. F. Pearce,
19
D. A. Petyt,
14
R. Pittam,
17
R. K. Plunkett,
8
A. Rahaman,
21
R. A. Rameika,
8
T. M. Raufer,
19,17
B. Rebel,
8
P. A. Rodrigues,
17
C. Rosenfeld,
21
H. A. Rubin,
11
V. A. Ryabov,
32
M. C. Sanchez,
1,9
N. Saoulidou,
8
J. Schneps,
26
P. Schreiner,
3
V. K. Semenov,
33
P. Shanahan,
8
W. Smart,
8
C. Smith,
13
A. Sousa,
9,17
P. Stamoulis,
2
M. Strait,
14
N. Tagg,
16,26
R. L. Talaga,
1
J. Thomas,
13
M. A. Thomson,
6
G. Tinti,
17
R. Toner,
6
V. A. Tsarev,
32
G. Tzanakos,
2
J. Urheim,
12
P. Vahle,
28,13
B. Viren,
4
M. Watabe,
24
A. Weber,
17
R. C. Webb,
24
N. West,
17
C. White,
11
L. Whitehead,
4
S. G. Wojcicki,
22
D. M. Wright,
29
T. Yang,
22
M. Zois,
2
K. Zhang,
4
and R. Zwaska
8
(MINOS Collaboration)
1
Argonne National Laboratory, Argonne, Illinois 60439, USA
2
Department of Physics, University of Athens, GR-15771 Athens, Greece
3
Physics Department, Benedictine University, Lisle, Illinois 60532, USA
4
Brookhaven National Laboratory, Upton, New York 11973, USA
5
Lauritsen Laboratory, California Institute of Technology, Pasadena, California 91125, USA
6
Cavendish Laboratory, University of Cambridge, Madingley Road, Cambridge CB3 0HE, United Kingdom
7
Universidade Estadual de Campinas, IF-UNICAMP, CP 6165, 13083-970, Campinas, SP, Brazil
8
Fermi National Accelerator Laboratory, Batavia, Illinois 60510, USA
9
Department of Physics, Harvard University, Cambridge, Massachusetts 02138, USA
10
Holy Cross College, Notre Dame, Indiana 46556, USA
11
Physics Division, Illinois Institute of Technology, Chicago, Illinois 60616, USA
12
Indiana University, Bloomington, Indiana 47405, USA
13
Department of Physics and Astronomy, University College London, Gower Street, London WC1E 6BT, United Kingdom
14
University of Minnesota, Minneapolis, Minnesota 55455, USA
15
Department of Physics, University of Minnesota–Duluth, Duluth, Minnesota 55812, USA
16
Otterbein College, Westerville, Ohio 43081, USA
17
Subdepartment of Particle Physics, University of Oxford, Oxford OX1 3RH, United Kingdom
18
Department of Physics and Astronomy, University of Pittsburgh, Pittsburgh, Pennsylvania 15260, USA
19
Rutherford Appleton Laboratory, Science and Technology Facilities Council, OX11 0QX, United Kingdom
20
Instituto de Fı
´
sica, Universidade de Sa
˜
o Paulo, CP 66318, 05315-970, Sa
˜
o Paulo, SP, Brazil
21
Department of Physics and Astronomy, University of South Carolina, Columbia, South Carolina 29208, USA
22
Department of Physics, Stanford University, Stanford, California 94305, USA
23
Department of Physics and Astronomy, University of Sussex, Falmer, Brighton BN1 9QH, United Kingdom
24
Physics Department, Texas A&M University, College Station, Texas 77843, USA
25
Department of Physics, University of Texas at Austin, 1 University Station C1600, Austin, Texas 78712, USA
26
Physics Department, Tufts University, Medford, Massachusetts 02155, USA
27
Department of Physics, University of Warsaw, Hoz
˙
a 69, PL-00-681 Warsaw, Poland
28
Department of Physics, College of William & Mary, Williamsburg, Virginia 23187, USA
29
Lawrence Livermore National Laboratory, Livermore, California 94550, USA
30
Physics Department, Western Washington University, Bellingham, Washington 98225, USA
PHYSICAL REVIEW D 81, 072002 (2010)
1550-7998=2010=81(7)=072002(16) 072002-1 Ó 2010 The American Physical Society
31
Physics Department, University of Wisconsin, Madison, Wisconsin 53706, USA
32
Nuclear Physics Department, Lebedev Physical Institute, Leninsky Prospect 53, 119991 Moscow, Russia
33
Institute for High Energy Physics, Protvino, Moscow Region RU-140284, Russia
(Received 12 October 2009; published 8 April 2010)
The energy dependence of the neutrino-iron and antineutrino-iron inclusive charged-current cross
sections and their ratio have been measured using a high-statistics sample with the MINOS near detector
exposed to the NuMI beam from the main injector at Fermilab. Neutrino and antineutrino fluxes were
determined using a low hadronic energy subsample of charged-current events. We report measurements of
-Fe ( -Fe) cross section in the energy range 3–50 GeV (5–50 GeV) with precision of 2%–8% (3%–9%)
and their ratio which is measured with precision 2%–8%. The data set spans the region from low energy,
where accurate measurements are sparse, up to the high-energy scaling region where the cross section is
well understood.
DOI: 10.1103/PhysRevD.81.072002 PACS numbers: 13.15.+g
I. INTRODUCTION
Neutrino-nucleon and antineutrino-nucleon charged-
current (
N CC and
N CC) inclusive cross sections
above 30 GeV have been determined by several experi-
ments [1–3] with a combined precision of 2% [4]. The
measured cross sections at these energies have a linear
dependence on energy, which agrees well with the predic-
tion of the quark parton model (QPM) [5].
At lower energies, the cross section is both less well
measured and difficult to model due to overlapping con-
tributions from quasielastic processes (
þ n !
þ p), resonance excitation followed by subsequent
decay, and the onset of deeply inelastic scattering (DIS).
This energy range is of particular interest to ongoing and
future neutrino oscillation searches in MINOS, NOA [6],
and T2K [7]. Most cross section measurements in the E
<
30 GeV range [8–14] have uncertainties of the order of
10%. Recently, NOMAD [15] measured the cross section
down to 2.5 GeV with a precision of better than 4%.
However, this result relies on a particle production model
tuned to data [16] to predict the neutrino flux. In this paper
we present a measurement of the
N CC cross section
with a precision from 2%–8%, covering the 3–50 GeV
energy range using the MINOS near detector. Our analysis
uses a low hadronic energy subsample to determine the flux
shape [17,18].
Antineutrino-nucleon charged-current cross sections in
the E
< 30 GeV range suffer from the same complica-
tions listed above and tend to be even less well measured.
Several experiments reported results [11–13,19,20]; how-
ever data coverage in energy was sparse and these mea-
surements typically have larger than 10% uncertainty. Our
measurement has higher precision, with uncertainties
which range from 3%–9%.
The
N CC to
N CC cross section ratio, r ¼
=
, has been measured with a combined precision of
better than 1% at high energies [17,21] but only one
dedicated measurement [22] has been performed in the
E
< 30 GeV range. Gargamelle [22] reports measure-
ments of r from 1–10 GeV with precision of about 20%.
Our result substantially adds both coverage and precision
to the determination of r. The ratio is more precisely
determined than either cross section measured separately
due to a partial cancellation of most systematic effects and
a cancellation of the normalization uncertainty.
The results in this paper can be used to tune and improve
neutrino interaction generator models [23,24]. For ex-
ample, neutrino scattering data are required for the model-
ing of the axial vector contribution to the cross section
[25]. Also, the cross section ratio r is particularly sensitive
to the modeling of xF
3
, the parity violating structure
function, which enters into the numerator and denominator
with opposite sign, and to the antiquark content of the
nucleon, which contributes differently to neutrino and
antineutrino scattering. In addition, at 5 GeV, about 70%
of our event sample has negative 4-momentum transfer
squared, Q
2
, of less than 1:5 GeV
2
. This large, low-Q
2
sample provides model sensitivity to the low-Q
2
QCD
contributions (higher order QCD, higher-twist, and target
mass corrections) which are difficult to calculate.
Overview of the analysis
The
CC and
CC total cross sections as a function
of incoming neutrino energy E are determined from the
inclusive charged-current interaction rate and the incident
neutrino flux. A sample of CC events (‘‘cross section’’
sample) is selected and a subsample of these events with
low hadronic energy (‘‘flux’’ sample) is defined. A
Monte Carlo simulation which includes detailed detector
geometry and response is used to correct the flux and cross
section samples for detector acceptance and smearing
effects.
*
Present address: Department of Physics and Astrophysics,
University of Delhi, Delhi 110007, India.
†
Present address: Physics Department, St. John Fisher College,
Rochester, NY 14618 USA.
‡
Deceased.
x
Present address: Physics and Astronomy, University of
Pennsylvania, Philadelphia, PA 19104, USA.
P. ADAMSON et al. PHYSICAL REVIEW D 81, 072002 (2010)
072002-2
Neutrino and antineutrino differential cross sections,
d
;
=d, approach the same constant value, independent
of energy, in the limit of low-, where is the energy
transferred to the hadronic system. A method which ex-
ploits this feature is used to determine the energy depen-
dence of the flux from the flux sample, which is then
normalized using the world average cross section value
measured above 30 GeV. To accomplish this we make use
of the full range of our data sample, which overlaps with
the high-energy measurements in the 30–50 GeV region.
This ‘‘low-’’ method has been used previously at high
energies [17,18] and here it is adapted to the E<30 GeV
range.
The neutrino beam, detector and the Monte Carlo simu-
lation of the experiment are described in Sec. II. Section III
describes the event sample selection and the methods for
extracting the flux and the cross section. A discussion of
systematic uncertainties and results are given in Secs. IV
and V, respectively.
II. BEAM LINE AND DETECTOR
MINOS is a two-detector, long baseline neutrino oscil-
lation experiment using the NuMI (neutrinos at main in-
jector) neutrino beam at Fermilab. The oscillation
parameters are measured [26,27] by comparing the
energy spectra at the near detector located at Fermilab
and the far detector located 734 km away in the Soudan
Mine in northern Minnesota. In this section we describe the
neutrino beam, the near detector, and the Monte Carlo
simulation. More detailed descriptions of the beam line
and the MINOS detectors are given elsewhere [28].
A. Neutrino beam
The NuMI neutrino beam is produced from 120 GeV
protons extracted in a 10 s spill from the main injector
which impinge on a graphite target, with a typical intensity
for the data presented here of 2:2 10
13
protons on target
(PoT) per spill. Charged particles produced in the target,
mainly pions and kaons, are focused by a pair of toroidal
magnets called horns into a 675 m long decay volume
where the mesons decay to muons and neutrinos. The
decay region is followed by a hadron absorber where
remaining mesons and protons are stopped. The neutrino
beam then traverses 240 m of unexcavated rock before
reaching the near detector located 1.04 km from the target.
Data for this analysis were collected in ‘‘low-energy’’
beam mode in which the downstream end of the target is
placed 10 cm from the neck of the first focusing horn and
the current in the horns is 185 kA, with the polarity set to
focus positively charged mesons. The Monte Carlo simu-
lation predicts the composition of the event sample to be
92.9%
, 5.8%
, and 1.3%
e
þ
e
. Figure 1 shows the
simulated flux spectrum of the
and
in the beam. The
component of the beam, which results primarily from
focused
þ
and K
þ
, peaks between 3 and 4 GeV with a
long tail. The
component arises mainly from low trans-
verse momentum
and K
traveling through the neck of
both horns, where they undergo little defocusing. This
results in a spectrum with no focusing peak and greater
mean energy.
B. Near detector
The near detector is a tracking calorimeter composed of
planes of magnetized iron and plastic scintillator. A toroi-
dal magnetic field with an average strength of 1.3 T pro-
vides a measure of muon momentum from curvature and is
used to distinguish
and
CC interactions based on the
charge sign of the final state muon. In normal operational
mode the field is set to focus negative muons.
The near detector, illustrated in Fig. 2, consists of 282
steel plates, 2.54 cm thick, of which 152 are instrumented
with 1 cm thick scintillator planes. The scintillator planes
are made of 4.1 cm wide strips oriented 45
with respect
to the vertical and alternating 90
in successive planes.
The strips are read out with wavelength shifting fibers
connected to multianode photomultiplier tubes (PMT).
Every fifth plane throughout the detector is fully instru-
mented with a scintillator layer. In the upstream calorime-
ter region, comprising the first 120 planes, each of the four
intervening planes has partial scintillator coverage. The
calorimeter region is used to measure energy deposited
by neutrino-induced hadronic showers. Event vertices are
required to be within a fiducial volume contained in the
calorimeter. The downstream 162 planes of the detector
form the muon spectrometer.
In the low-energy NuMI beam configuration, the typical
interaction rate in the near detector is about 16 events in a
10 s spill. Events are separated using timing and spatial
information. The events accepted for this analysis were
from interactions occurring during a 13 s long gate syn-
chronized to the beam spill. The readout electronics con-
tinuously digitize the PMT signals in 19 ns samples
without dead time throughout the spill. In between beam
spills, cosmic ray muon data are recorded with less than
1% dead time.
Neutrino Energy (GeV)
0 1020304050
PoT
9
/10
2
Particles/GeV/m
1
10
2
10
3
10
4
10
5
10
NEUTRINO
ANTINEUTRINO
FIG. 1. The muon neutrino and antineutrino flux at the center
of the near detector as calculated by the NuMI beam simulation.
NEUTRINO AND ANTINEUTRINO INCLUSIVE CHARGED- ... PHYSICAL REVIEW D 81, 072002 (2010)
072002-3
The detector is calibrated in several steps that convert
the raw PMT signal to deposited energy [28]. The nonline-
arity of the electronics is measured with charge injection;
relative PMT gains are measured with an in situ light
injection system; variations in the light output between
scintillator strips and along the strips are corrected with
cosmic ray muons and a radioactive source scanner.
Cosmic ray muons which stop in the detector are used to
calibrate the measured signal to energy lost by muons
passing through the scintillator strips. The detector simu-
lation is tuned to emulate the actual detector response at all
stages in the calibration chain.
C. Beam and detector Monte Carlo simulation
A Monte Carlo simulation is used to model the produc-
tion of the neutrino beam, interaction of neutrinos in and
around the detector, and the detector response, which is
simulated using
GEANT3 [29]. The beam model includes a
simulation of secondary hadron production from proton
interactions [30] and the propagation of these hadrons.
Their reinteraction and decay products are also tracked
through the target, magnetic horns, and decay region.
This simulation produces an initial estimate of the flux,
which is later replaced by the flux extracted using the
method described below.
Neutrino interactions in the detector are simulated using
the
NEUGEN3 [23] event generator. The simulation of qua-
sielastic interactions, which dominate at low energies, is
based on the Llewellyn-Smith [31] model, while
intermediate-energy resonance interactions are simulated
according to the Rein-Sehgal model [32,33]. Both models
assume a dipole parametrization of the axial part of the
cross section that depends on the axial mass parameters
M
A
ðQELÞ and M
A
ðRESÞ, taken to be 0:99 0:15 and
1:12 0:17 GeV, respectively. A transition is made be-
tween resonance production and the DIS model by phasing
out the former and phasing in the latter over the hadronic
invariant mass range, 1:7 <W<2:0 GeV. The sum of the
resonance and DIS contributions are constrained to match
total cross section data.
DIS interactions, which dominate at high energy, are
based on an effective leading order model by Bodek et al.
[34]. The Bjorken scaling variable x is replaced by an
effective scaling variable that depends on two parameters
A
ht
and B
ht
, where A
ht
accounts for target mass effects and
higher-twist terms. B
ht
depends on the transverse momen-
tum of the initial state quark. The model is fit to charged
lepton scattering data [34] and gives the parameters A
ht
and
B
ht
and correction factors (C
v1u
, C
v2u
, C
v1d
, C
v2d
, C
s1d
,
and C
s1u
) for valence and sea up and down quark parton
distribution functions. The uncertainties on these parame-
ters were not readily available so a study was performed to
estimate them and their effect on this cross section mea-
surement (see Sec. IV).
The cross section in the transition region from resonance
to DIS is expressed as a sum of a pure-resonance cross
section and a nonresonance contribution from DIS. The
sum is tuned to describe low multiplicity final state data in
this region [23]. For DIS interactions, the final state had-
ronic system is modeled with KNO scaling [ 35], which
transitions to
PYTHIA/JETSET [36] at hadronic invariant
mass W ¼ 3 GeV. The total neutrino cross section is tuned
by a scale factor so that the cross section at 100 GeV
matches the world average of measurements.
The dynamics of hadron formation in the target nucleus
and reinteraction of hadrons after formation modify the
visible hadronic shower energy. These effects are simulated
using a cascade Monte Carlo anchored to N, pN and Fe
and pFe scattering data and validated against neutrino-
deuterium and neutrino-neon scattering data [37,38]. A
treatment of hadron formation time is included [39].
FIG. 2. Left panel: top view of the near detector, showing the calorimeter and muon spectrometer. The drawing is not to scale. Right
panel: transverse view of a near detector plane. The shaded area shows a partially instrumented active scintillator plane and the dashed
line within shows the boundary of the fiducial region. The dotted line shows the outline of a fully instrumented scintillator plane.
P. ADAMSON et al. PHYSICAL REVIEW D 81, 072002 (2010)
072002-4
III. ANALYSIS
The CC total cross sections are measured from the
inclusive CC scattering rate,
ð Þ
CC
ðEÞ, and the incident
neutrino flux,
ð Þ
ðEÞ. A sample of CC events, N
ð Þ
CC
ðEÞ,
is selected and then corrected for acceptance and back-
grounds to determine
ð Þ
CC
ðEÞ. A flux sample, F
ð Þ
ðEÞ,
consisting of the subset of N
ð Þ
CC
ðEÞ with low (in the lab
frame ¼ E
had
, the energy measured at the hadronic
vertex), is also defined and corrected for acceptance, back-
grounds, and for a small energy dependence using our
Monte Carlo model to yield
ð Þ
ðEÞ. The event recon-
struction and selection of these samples to form the cross
section are described in this section.
The data used in this analysis were collected between
June 2005 and April 2007 and correspond to an exposure of
2:45 10
20
PoT. The MC sample is almost double the
data, corresponding to 4:4 10
20
PoT.
A. Event reconstruction
Neutrino events are identified using the timing and
spatial pattern of energy deposited in the scintillator strips.
Muon tracks are recognized as a string of hit strips typi-
cally spanning more than 10 steel plates. For muons that
stop in the detector the energy is computed from range
according to the energy loss tables of Groom et al. [40]. A
systematic uncertainty of 2% is assigned to the energy
measured from range, arising from uncertainties in the
range tables, the variation in material composition, and
the accuracy of our track length reconstruction. The mo-
mentum of muons exiting the detector is measured using
the curvature of their trajectory in the detector’s magnetic
field. A 4% systematic uncertainty is assigned to our
knowledge of the absolute muon momentum measurement
from curvature. This is assessed by comparing the energy
measured with curvature to the independent measurement
from range using tracks that stop in the detector, and by
folding in underlying uncertainties in the detector’s mag-
netic field [41]. The resolution for muon momentum mea-
sured from range is 5% while that measured from curvature
has non-Gaussian tails and width of approximately 10%.
The vertex of a neutrino interaction is taken to be at the
start of a reconstructed track. Hit strips near the vertex
which are not included in the track are identified as coming
from hadrons produced in the interaction. Their summed
signal is converted to energy using a lookup table derived
from simulated showers to form the hadronic shower en-
ergy, E
had
. The response of our detector to single hadrons
was measured in an exposure of a smaller version of the
detector to a test beam [42]. The measured test beam
detector response was used to tune our simulations. The
absolute energy scale of the detector’s response to hadronic
particles is modeled to an accuracy of 5.6% [27,43], which
we take as the hadronic energy scale uncertainty in the
cross section measurement (see Sec. IV).
B. CC Event selection
The inclusive charged-current sample N
ð Þ
CC
ðEÞ is se-
lected using the following criteria:
(1) Fiducial volume: Selected events have a vertex po-
sition along the detector axis between 0.5 and 4.0 m,
measured from the upstream face of the detector. In
the plane transverse to the detector axis, the vertex is
required to be more than 0.5 m from the edge of an
active scintillator plane and outside of a 0.8 m radius
centered at the coil hole. The outline of the fiducial
region is shown in Fig. 2.
(2) Coil hole: The coil hole is uninstrumented and
variations in the material composition and magnetic
field are somewhat larger in the region around it. To
reduce the effect of these uncertainties, events with
tracks that spend a significant fraction of their path
length near the hole are removed from the event
sample. A minimum of 95% of hit strips in the event
is required to be farther than 0.3 m from the center at
closest approach (see Fig. 2).
(3) Track energy: The energy of the muon must be
greater than 1.5 GeV. This requirement rejects
neutral-current (NC) background events, which
populate the low-energy region, and short, poorly
reconstructed tracks.
(4) Track quality: The track fitting procedure yields a
measurement of the muon momentum with an asso-
ciated uncertainty. The track fit is required to be
convergent and have an uncertainty of less than
30%. In addition, we require the track’s longitudinal
start positions in each view to be less than six planes
apart.
(5) Neutrino energy: The reconstructed neutrino en-
ergy, E, which is the sum of the track and shower
energies, is required to be greater than 3 GeV
(5 GeV) for the neutrino (antineutrino) sample and
less than 50 GeV. The minimum energy require-
ments are imposed to minimize the overlap of the
inclusive CC sample and the flux sample, which is
substantial below these values. Above the maximum
energy cut, resolution of the track momentum mea-
surement from curvature degrades as the tracks
become straighter.
The event sample is divided into two categories depend-
ing on whether the track stops in or exits the detector. For
exiting events, the muon leaves the detector through the
back or side, or passes into the uninstrumented coil hole
region. The stopping and exiting samples are further differ-
entiated based on whether they end in the upstream or
downstream region (see Fig. 2) because of the difference
in sampling in the two regions.
The
CC sample is selected by requiring the sign of
the track curvature measurement to be positive. This sam-
ple has a higher fractional contamination from wrong-sign
events (misidentified
tracks) due to the much larger
NEUTRINO AND ANTINEUTRINO INCLUSIVE CHARGED- ... PHYSICAL REVIEW D 81, 072002 (2010)
072002-5
