Inclusive Radiative Corrections in COMPASSand Input Information for the Present and Future
RC Programs
Barbara BadelekUniversity of Warsaw
Precision Radiative Correctionsfor Next Generation Experiments
Jefferson Lab, May 16 – 19, 2016
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 1 / 49
Mo & Tsai and Dubna schemes
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 2 / 49
Mo & Tsai and Dubna schemes
Lowest order radiative processes
Mo and Tsai scheme: b) – d)
Dubna scheme b) – g) but replaces e) – g) by:
ccccccBadelek, Bardin, Kurek, Scholz, Z.Phys. C66 (1995) 591
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 3 / 49
Mo & Tsai and Dubna schemes
Mo and Tsai scheme: FERRAD, model indep., non-covariant
Measured cross section:
δR(∆) is a residue of cancellation of IR divergent termsσtails processes where real photons of Eγ > ∆ are emitted
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 4 / 49
Mo & Tsai and Dubna schemes
Mo and Tsai scheme: FERRAD...cont’d
Here: σmeas = vσ1γ + σtails = vσ1γ + σinel + σel + σqelv – virtual corrections + soft photon emission.
Range of kinematical variables from which the radiative tails contribute to the cross sectionmeasured at the point A(Q2, ν); parallel lines: W = const
Even if we measure at DIS, information onF1, F2 (or R, F2) needed down to Q2 =0!
Weak dependence of η(x, y) = σ1γ/σmeas on ∆(here E = 280 GeV)
Badelek, Bardin, Kurek, Scholz, Z. Phys. C66 (1995) 591
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 5 / 49
Mo & Tsai and Dubna schemes
Dubna scheme: TERAD (also: POLRAD), QPM, covariant
Measured cross section: Born × { resummed collinear γ+ remnant of exponentiation+ remnant of subtraction in σin(vertex) }
inelastic radiative tailand regularization=⇒ Dubna scheme ∆ indep.
QPM calculationsof RC for hadron current
elastic radiative tailsas in MT but covariant
“Vacuum polarisation” through running of α(Q2)
O(α2) in amplitude implemented
Weak loop correction also present
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 6 / 49
Mo & Tsai and Dubna schemes
FERRAD vs TERAD (µp, 280 GeV)
η(x, y) = σ1γ/σmeas ηF /ηTopen symbols = FERRAD open symbols = FERRAD without τ τ , qqclosed symbols = TERAD closed symbols = full FERRAD
Badelek, Bardin, Kurek, Scholz, Z. Phys. C66 (1995) 591
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 7 / 49
Mo & Tsai and Dubna schemes
Input information for polarised/unpolarisedand inclusive/semiinclusive RC calculations
The items below should be known forxmeas < x < 1 and 0 < Q2 < Q2
max
Spin independent structure function F2(x,Q2) (nucleon, nuclei)Spin independent structure function R(x,Q2)
Spin dependent structure function g1(x,Q2)
Quasielastic suppression factors(Q2) (nuclei)Elastic form factors(Q2) (nucleon, nuclei)
All the input infoirmation is collected in a COMPASS note 2015-6,for the moment not accessible for outsiders but this may be changed
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 8 / 49
Mo & Tsai and Dubna schemes
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 9 / 49
Extension of F2(x,Q2) down to Q2
= 0
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 10 / 49
Extension of F2(x,Q2) down to Q2
= 0
F2(x,Q2) and R(x,Q2) in the low Q2 region
(see e.g. Badełek, Kwiecinski, Rev. Mod. Phys. 68 (1996) 445)
F2 and R needed at: xmeas < x < 1 and 0 < Q2 < Q2max
They are either physics motivated fits or models of dynamic origin andhave to have a proper asymptotic behaviour:at Q2 →0 fulfilling the conditions for arbitrary ν
F2 = O(Q2),F1
M+F2
M
pq
q2= O(Q2).
and
R(x,Q2) =σL
σT=
(1 + 4M2x2/Q2)F2
2xF1− 1 =
FL
2xF1−→ 0 at Q2 → 0
and at Q2 →∞ joining the QCD improved parton model expressions.
Observe that:Growth of F2 with decreasing x is slower at low Q2
R(x,Q2) essentially independent of x in the low x, low Q2 region.
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 11 / 49
Extension of F2(x,Q2) down to Q2
= 0 Data at low Q2
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 12 / 49
Extension of F2(x,Q2) down to Q2
= 0 Data at low Q2
What do the F2 data show around Q2 = 1 GeV2?
PDG 2015
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 13 / 49
Extension of F2(x,Q2) down to Q2
= 0 Data at low Q2
What do the F2 data show around Q2 = 1 GeV2?... cont’d
10-1
1
10
10 2
10 3
10-2
10-1
1 10 Q2 (GeV2)
F
2 (x
= Q
2 /sy
, Q2 )
(× 4096)
(× 2048)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 8)
(× 4)
(× 2)
(× 1)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 2048)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 8)
(× 4)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 8)
(× 4)
(× 2)
(× 1)
y=0.8(× 4096)
y=0.7(× 2048)
y=0.6(× 1024)
y=0.5(× 512)
y=0.4(× 256)
y=0.33(× 128)
y=0.26(× 64)
y=0.2(× 32)
y=0.12(× 16)
y=0.05(× 8)
y=0.025(× 4)
y=0.015(× 2)
y=0.007(× 1)
H1 SVX 1995 H1 1994
ZEUS BPT 1997 ZEUS SVX 1995 ZEUS 1994
ZEUS QCD fitZEUS Regge fit
ZEUS 1997
hep-ex/0008069 ZEUS, Eur. Phys. J. C7 (1999) 609
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 14 / 49
Extension of F2(x,Q2) down to Q2
= 0 JKBB
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 15 / 49
Extension of F2(x,Q2) down to Q2
= 0 JKBB
Parametrizations of F2 in the low Q2, low x regionJKBB (Kwiecinski, Badełek, Z. Phys. C43 (1989) 43; Phys. Lett. B295 (1992) 263)
The starting point is the Generalised Vector Meson Dominance (GVMD) representation of thestructure function F2(x,Q2):
F2[x = Q2/(s+Q2 −M2), Q2] =Q2
4π
∑v
M4vσv(s)
γ2v(Q2 +M2v )2
+Q2
∫ ∞Q2
0
dQ′2Φ(Q′2, s)
(Q′2 +Q2)2
≡ F(v)2 (x,Q2) + F
(p)2 (x,Q2) (1)
The function Φ(Q2, s) is expressed as follows:
Φ(Q′2, s) = −1
πIm
∫ −Q′2 dQ′′2Q′′2
FAS2 (x′, Q′′2) (2)
Asymptotic structure function FAS2 (x,Q2) assumed to be given.By construction, F2(x,Q2)→ FAS2 (x,Q2) for large Q2.The first term in (1) corresponds to the low mass vector meson dominance.Contribution of vector mesons heavier then Q0 is included in the integral in (1).This integral can be looked upon as the extrapolation of the (QCD improved) parton modelfor arbitrary Q2 (including Q2 =0).The representation (1) is written for fixed s and is expected to be valid at s� Q2, i.e. atlow x but for arbitrary Q2 – and above the resonances.
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 16 / 49
Extension of F2(x,Q2) down to Q2
= 0 JKBB
F p2 in the low Q2, low x region...cont’d
JKBB...cont’d
Choosing the parameter Q20 > (M2
v )max where (Mv)max is the mass of the heaviestvector meson included in the sum one explicitly avoids double counting when adding twoseparate contributions to F2.
Q0 should be smaller than the mass of the lightest vector meson not included in the sum.
Representation (1) for the partonic part F (p)2 (x,Q2) may be simplified as follows:
F(p)2 (x,Q2) =
Q2
(Q2 +Q20)FAS2 (x, Q2 +Q2
0) (3)
where
x =Q2 +Q2
0
s+Q2 −M2 +Q20
≡Q2 +Q2
0
2Mν +Q20
(4)
Simplified parametrization (3) connecting F (p)2 (x,Q2) to FAS2 by an appropriate change of
the arguments posseses all the main properties of the second term in (1).
Apart from Q20, constrained by physical requirements, the representation (1) does not contain
any other free parameters except those which are implicitly present in FAS2 .
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 17 / 49
Extension of F2(x,Q2) down to Q2
= 0 JKBB
F p2 in the low Q2, low x region...cont’d
JKBB...cont’d
���
� �
���
�
����� ����� ��������������� �!� "$#&%�'(#*)+��,-��%�#�%�".�/#0���21�# ������'3��"+1�#546%�"87:9<;5=?>A@ 9�B 4C"D#0)+��E6%GF�@ 9 ����H�46%�"DIKJMLONP '3�/EE?,:%�4C"$#5�RQ�ST#�%�H$��#*)+� �.F�45#0)!,-����U/4V% �/�3EWJX,Y� I/E4� )+��Z[���3���/E5#5�\'5��%��]LON P %�,:� "�134^�K13EV�3�RQ�SY_a`<`�b P %c,(� "#0��46��"2HdEV�3�RQ�S�ef�hg P %c,(� "i��j$�2�����3�RQ�k�l()+��@ 9 Uc��E0�2�3�m�����:H�45U�� "n4C"Do\�<p 9 kmpq����46%r�/�:,-����Z<461�#546%�"d�s'*%���7(9�����a1�%��n,?������ZtF�45#0)�#*)+��Z ��#���us#0)+�s�t%�Z �3E&%�'?vn%�" "+�<13)d46�f��"+Z8w:��"+Z<��)+%�x P Z ��� )+��ZnEy4C"+��Q�S:#0)+�z�!%�Z �3E%�'�gs�c,(�3EE6����#a��E5k P Z %�#3#���ZtEy4C"+�3{<���t��EE�Q�S�#0)+�\�t%�Z �3Es%/'T|��2Z �3EV� }~��"+Z��iF�46��134C"d� }<4 P Z ��� )+��Z��*Z %d#3#���ZE4C"+��Q�Sf#0)+�8�t%dZ �3E�%�'\o�E/��213} Ss�m��J$����"+Z�pq%�Hq# P '3�/EETE4C"+��Q���"+Z�#0)+�O�t%dZ �3Em%�'s�.Z �3Em��#\�/E�k P Z %�#3#���ZE4C"+�3{<E6���*H$��Q�kmo�E6%rI2�/E("+%��3�t�/Ey4����#�4V%�"���"+1�� ��#��/4^"+#546�3�n�����f"+%�#�4C"+13E0�2Z ��Z$k
���
������������
� ���������
���������! #"�$&%('*),+�-�./'102'135476,874:95'#4!6�4!)<;>=<?�.*41-@),;BAC956�4!6D3�EFAC./6G4!6D3IH1./6,+J+C+K'*H,4L?M6D3ON7PMQ:PRTSLU ),+�)#8V-�35H,4L?M6D3W6,8>X)�4Y=*)D.Z?[6�-<+\=*),;:-@'V+\6<8^]`_ba�?�3dce'Tf�_ZgDhjik95'lH1./6,+V+Y+*'*H,4L?M6D3G+m)D./'n0d-<;L4L?oA^;�?M'*prq�?L4:924:95'�8:)TH,4Z6D.!+Y?s35pT?MH*)�4Z'*p?�3r4t95'vu@wx-�./'OaM3C-�0dyG'1.Z+#?�3zy�.{)THZ|>',4L+!gDhmik95'\'1.V./6�.!+v./'!AC.{'V+*'13^4}4:95'#+,4Z)�4L?�+,4L?MH*),;7)D35pI+!~>+<4!'102)�4L?MH\'1.J./6D.!+)Tp�p�'*p2?s3(�^-@)Tp�./)�4/-�./',h#�k-<;�;Y+!~^0�y>6,;�+O)�./'ep�)�4!)O8L./6D0�4:9G?�+O)D35)<;�~�+V?�+*��6*AY'13d+!~^0dyG6,;�+I)D./'lAC./',=<?M6>-<+V;o~An-�y<;�?�+195'*pdp�)�4!)j8�./6�0��`�WaLHV?�./HV;M'V+!g,�^�\�T�B��a:4:.!?M)D3@w�;M'V+!gj)D35p`�j%���a�+*�^-@)D./'V+ZgDh#cm;[6�y@),;C356D.J02),;�?���)�4L?M6D3-�35H*'1.*4!)<?s354L?M'V+e)D.{'�356�4Y?�35HV;:-@p�'*p^h�ik95'jHD-�.*=�'V+j./'!AC./'V+*'1354l4:95'����\�7��aLp�),+195'*p�;�?s35'*�Y6D3G;o~d+1956Tq#3�8t6D.] _l��� g,�����&��%�aLp�6�4V4Z'*p�; ?�35'!g�)D35p�¡l¢�aLp�)<+195'*p�Etp�6�4V4!'*p�gmA&)D./)D02',4Z'1.!?���)�4L?M6D3G+,h£T¤
H1 Collaboration, DESY 97-042
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 18 / 49
Extension of F2(x,Q2) down to Q2
= 0 JKBB
F p2 in the low Q2, low x region...cont’d
JKBB...cont’d
0
0.1
0.2
0.3
10-5
10-4
10-3
10-2
F 2
Q2 = 0.11 GeV2
0
0.1
0.2
0.3
10-5
10-4
10-3
10-2
Q2 = 0.15 GeV2
0
0.1
0.2
0.3
10-5
10-4
10-3
10-2
Q2 = 0.20 GeV2
0
0.1
0.2
0.3
0.4
10-5
10-4
10-3
10-2
F 2
Q2 = 0.25 GeV2
0
0.1
0.2
0.3
0.4
10-5
10-4
10-3
10-2
Q2 = 0.30 GeV2
0
0.1
0.2
0.3
0.4
10-5
10-4
10-3
10-2
Q2 = 0.40 GeV2
0
0.2
0.4
0.6
10-5
10-4
10-3
10-2
F 2
Q2 = 0.50 GeV2
0
0.2
0.4
0.6
10-5
10-4
10-3
10-2
Q2 = 0.65 GeV2
ZEUS BPC 95ZEUS 94
E665H1 94 H1 SVTX 95
DLCKMT
GRV(94)
BKABY
0
0.5
1
10-5
10-4
10-3
10-2
x
F 2
Q2 = 1.50 GeV2
0
0.5
1
10-5
10-4
10-3
10-2
x
Q2 = 3.00 GeV2
0
0.5
1
10-5
10-4
10-3
10-2
x
Q2 = 6.50 GeV2
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
10-6
10-5
10-4
10-3
10-2
10-1
F2 (model JK + BB)continuous - dF2 /dlogQ2
broken - dFVMD2 /DLOGQ2
dotted - dFpart2 /dlogQ2
points - ZEUS data (Vancouver 1998)
x
dF2/
dlog
Q2
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 19 / 49
Extension of F2(x,Q2) down to Q2
= 0 Martin-Ryskin-Stasto
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 20 / 49
Extension of F2(x,Q2) down to Q2
= 0 Martin-Ryskin-Stasto
F p2 in the low Q2, low x region
Martin-Ryskin-Stasto (Martin, Ryskin, Stasto, Eur. Phys. J. C7 (1999) 643)
Exploits further the idea of BBJK.
Perturbative and non-perturbative QCD contributions separated by the distanceconfigurations of the qq pair in the γ∗ → qq:
small distance configurations (k2T > k20) given by pQCD (unified equations, DGLAP +BFKL, unintegrated gluon distribution);
large distance configurations (k2T < k20) given by VMD (for low qq fluctuation masses,M2 < Q2
0), and additive quark model (for high qq masses, M2 > Q20).
Excellent description of the data throughout the whole Q2 region, including Q2 =0.
Fitted (at x <0.05) are 3 parameters of the gluon distribution; scales k20 and Q20 chosen as:
k20 =0.2 GeV2 (crucial) and Q20 =1.5 GeV2. Choice of k20 yields physically sensible g and
FL.
Interference between states of different qq masses is crucial for description of the data.
Importance of the perturbative contribution in the non-perturbative domain.
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 21 / 49
Extension of F2(x,Q2) down to Q2
= 0 Martin-Ryskin-Stasto
F p2 in the low Q2, low x region
Martin-Ryskin-Stasto...cont’d
γ*M′2
p
z,kT
Im Aqq+p
p
M2
Fig.1
Fig.5
10 2
10 3
10 4
10-2
10-1
1 10 102
W=245 GeV (x128)
W=210 GeV (x64)
W=170 GeV (x32)
W=140 GeV (x16)
W=115 GeV (x8)
W=95 GeV (x4)
W=75 GeV (x2)
W=60 GeV (x1)
ZEUS-BPCH1-95H1-94H1ZEUSMainusch-thesis
Fit A (k02=0.2 GeV2)
Fit B (k02=0.5 GeV2)
Q2(GeV2)
σ γ*p(
µb)
≈≈
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 22 / 49
Extension of F2(x,Q2) down to Q2
= 0 (Modified) saturation model
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 23 / 49
Extension of F2(x,Q2) down to Q2
= 0 (Modified) saturation model
F p2 in the low Q2, low x region
(Modified saturation model, Bartels, Golec-Biernat, Kowalski, Phys. Rev. D66 (2002) 014001)
• Original saturation modelof Golec-Biernat and Wüsthoffmodified by DGLAP
• 5 parameters fitted toE665, H1 and ZEUS data,x < 0.01, 0.1< Q2 <500 GeV2
(claimed to be valid down toQ2 = 10−5 GeV2).
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 24 / 49
Extension of F2(x,Q2) down to Q2
= 0 ALLM97
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 25 / 49
Extension of F2(x,Q2) down to Q2
= 0 ALLM97
F p2 in the low Q2, low x region
ALLM97 (Abramowicz, Levy, hep-ph/9712415)
Parametrization of the σtot(γ∗p) at W 2 >∼ 3 GeV2 (above resonances).Valid everywhere in x and Q2 (including photoproduction).Based on Regge–type approach; extention to large Q2 compatible with QCD.Observe that it is a fit of 23 parameters to all the dataFit contains contributions of the pomeron (P ) and reggeon (R):
F2(x,Q2) =Q2
Q2 +m20
[FP2 (x,Q2) + FR2 (x,Q2)
](5)
of the form
FP2 (x,Q2) = cP (t)xα(t)P (1− x)bP (t), FR2 (x,Q2) = cR(t)x
α(t)R (1− x)bR(t) (6)
where
t = ln
(lnQ2 +Q2
0
Λ2/ ln
Q20
Λ2
)(7)
and1
xP= 1 +
W 2 −M2
Q2 +m2P
,1
xR= 1 +
W 2 −M2
Q2 +m2R
(8)
Here M is the proton mass; m20,m
2P ,m
2R, Q
20 allow a smooth transition to
photoproduction. For Q2 � m2P , Q
2 � m2R, xP → x, xR → x;
cR, aR, bR, bP ↗ Q2 ↗; cP , aP ↘ Q2 ↗.
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 26 / 49
Extension of F2(x,Q2) down to Q2
= 0 ALLM97
F p2 in the low Q2, low x region
ALLM97...cont’d
(σeffγ∗p ≈ σtotγ∗p at HERA energies)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 27 / 49
Extension of F2(x,Q2) down to Q2
= 0 ALLM97
F p2 in the low Q2, low x region
ALLM97...cont’d
1
10
10 2
10 3
10 4
10 5
102
103
104
105
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.11 (x 128)
ZEUS BPC 95 ZEUS 94 E665ZEUS, H1 γpH1 94
H1 SVTX 95 low W γpDL
ALLM
CKMT
GRV(94)
BK
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.15 (x 64)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.20 (x 32)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.25
0.23 (x 16)
(x 16)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.30
0.31 (x 8)
(x 8)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.40
0.43 (x 4)
(x 4)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.50 (x 2)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
0.65
0.59 (x 1)
(x 1)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
1.50
1.50 (x 1/2)
(x 1/2)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
3.00
2.80 (x 1/4)
(x 1/4)
W2(GeV2)
σ totγ*
p (µb
) (sc
aled
)
6.50
7.16 (x 1/8)
(x 1/8)
Q2 (E665)
Q2 (ZEUS)
0.00 (x 256)
������������ ����������������������������� ��!�����"�#$ �������"&%'���(�()(*�%'�����"+��(-,.��"�%'�����"/�0,21435,.���)�6��798:3);��0�<��=�(%>�����;7�����?����@,$��%A������(��6"�;��B%>�0���;C��"?D��<��%�E�>��(�,.���F;��B(* ��B�>GH ����� -�I(*�('J����!K%'������(L�0��M����/>N� -O%A���������"�(��0,;��QP2>��>"��� R�0����SH>��>��6T��0������"R(AJ
UAV
1 10 100 1000 10000 10000010
-2
10-1
1
10
10 2
1 10 102
103
104
105
������������ �������������������! #"$��&%%'%(�"$������)*��%,+-��)."/�����)0��+2143$56+-���'7��98�:��:)���;,3,<��������%/=>�����"$���<?@%�A�'���2:BDC.�"$�E�A��F��).%���+G���IH2JKJ�LNMPO'C.�A�E�AQR:��:��FS��A������)NTU+-���F�K���F)��VG��).7W�����?%XI��+Y�����LNZ\[DZ2]\��)�T^7�����@7W���F)��VE=
]:_B. Badelek (Warsaw ) Input for RC RC JLab, 2016 28 / 49
Extension of F2(x,Q2) down to Q2
= 0 ZEUS Regge fit
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 29 / 49
Extension of F2(x,Q2) down to Q2
= 0 ZEUS Regge fit
F p2 in the low Q2, low x region
ZEUS Regge fit (ZEUS, Eur. Phys. J. C7 (1999) 609)
Combines the Q2 dependence of the VMD with the energydependence from the Regge model:
F2(x,Q2) =
(Q2
4π2α
)·(
M20
M2 +Q2
)·[AR · (W 2)αR−1 +AP · (W 2)αP−1
]where AR, AP , M0 are constants; αR, αP are reggeon and pomeronintercepts. Fixed: M2
0 = 0.53 GeV2, αR = 0.53.Remaining 3 parameters fitted to Q2 =0 data at W 2 >3 GeV2.
Result: αP = 1.097 ± 0.002.
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 30 / 49
Extension of F2(x,Q2) down to Q2
= 0 ZEUS Regge fit
F p2 in the low Q2, low x region
ZEUS Regge fit...cont’s
0
0.05
0.1
0.15
0.2
10-6
10-4
0
0.05
0.1
0.15
0.2
10-6
10-4
0
0.05
0.1
0.15
0.2
10-6
10-4
0
0.1
0.2
0.3
10-6
10-4
0
0.1
0.2
0.3
10-6
10-4
0
0.1
0.2
0.3
10-6
10-4
0
0.2
0.4
10-6
10-4
0
0.2
0.4
10-6
10-4
0
0.2
0.4
10-6
10-4
0
0.2
0.4
0.6
10-6
10-40
0.2
0.4
0.6
10-6
10-4
x
F
2 (x
,Q2 )
Q2=0.045 GeV2
x
F
2 (x
,Q2 )
Q2=0.065 GeV2
x
F
2 (x
,Q2 )
Q2=0.085 GeV2
x
F
2 (x
,Q2 )
Q2=0.11 GeV2
x
F
2 (x
,Q2 )
Q2=0.15 GeV2
x
F
2 (x
,Q2 )
Q2=0.2 GeV2
x
F
2 (x
,Q2 )
Q2=0.25 GeV2
x
F
2 (x
,Q2 )
Q2=0.3 GeV2
x
F
2 (x
,Q2 )
Q2=0.4 GeV2
x
F
2 (x
,Q2 )
Q2=0.5 GeV2
x
F
2 (x
,Q2 )
Q2=0.65 GeV2
H1 SVX 1995 ZEUS BPC 1995
ZEUS BPT 1997 ZEUS SVX 1995
E665ZEUS Regge fit
ZEUS 1997
10-1
1
10
10 2
10 3
10-2
10-1
1 10 Q2 (GeV2)
F
2 (x
= Q
2 /sy
, Q2 )
(× 4096)
(× 2048)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 8)
(× 4)
(× 2)
(× 1)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 2048)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 8)
(× 4)
(× 1024)
(× 512)
(× 256)
(× 128)
(× 64)
(× 32)
(× 16)
(× 8)
(× 4)
(× 2)
(× 1)
y=0.8(× 4096)
y=0.7(× 2048)
y=0.6(× 1024)
y=0.5(× 512)
y=0.4(× 256)
y=0.33(× 128)
y=0.26(× 64)
y=0.2(× 32)
y=0.12(× 16)
y=0.05(× 8)
y=0.025(× 4)
y=0.015(× 2)
y=0.007(× 1)
H1 SVX 1995 H1 1994
ZEUS BPT 1997 ZEUS SVX 1995 ZEUS 1994
ZEUS QCD fitZEUS Regge fit
ZEUS 1997
ZEUS, Eur. Phys. J. C7 (1999) 609
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 31 / 49
Extension of R(x,Q2) down to Q2
= 0
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 32 / 49
Extension of R(x,Q2) down to Q2
= 0
R and FL in the low Q2, low x regionBKS (Badelek, Kwiecinski, Stasto, Z. Phys. C74 (1997) 297)
A model for FL, valid at low x and low Q2;based on the photon–gluon fusion, essential at low xand extended to low Q2. Similar approach inNikolaev, Zakharov, Z. Phys. C49 (1991) 607, C53 (1992) 331
The model embodies the constraint FL ∼ Q4 at Q2 →0.
FL =
∫ 1
x
dx′
x′
∫dk0Tk0T
F 0L(x′, Q2, k0T )f(
x
x′i, k2T )
where F 0L comes from γ∗g fusion, is a longitudinal
structure function of the off-shell gluon of virtuality k2T and is calculated perturbatively; f isan unintegrated gluon distribution related to the “ordinary” g(y, µ2) by:
yg(y, µ2) =
∫ µ2dk2Tk2T
f(y, k2T )
Its evolution is controlled by (approximate) BFKL.
To extrapolate FL to low Q2 and to Q2 =0, evolution of g(y,Q2) and argument of αs(Q2)was frozen via Q2 → Q2 + 4m2
q .
HT contribution needed at moderate Q2, i.e. terms vanishing as 1/Q2 for Q2 →∞. Theywere assumed to originate from low values of the quark transverse momenta andinterpreted as coming from soft pomeron exchange (intercept = 1). Such HT has a properbehaviour both at Q2 →∞ and Q2 →0.
x / x’, kT
2
Q2
FL ^ 0 2k
f
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 33 / 49
Extension of R(x,Q2) down to Q2
= 0
R and FL in the low Q2, low x regionBKS (Badelek, Kwiecinski, Stasto, Z. Phys. C74 (1997) 297)...cont’d all with HT
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
0.09
0.1
10-4 10-3 10-2 10-1
R(x
,Q2 )
GRV off-shellGRV on-shellMRS(A) off-shellMRS(A) on-shell
Q2=0.1 GeV2
0.1
0.2
0.3
0.4
0.5
10-4 10-3 10-2 10-1
Q2=0.5 GeV2
0.1
0.2
0.3
0.4
0.5
10-4
10-3
10-2
10-1
Q2=1.5 GeV2
0.1
0.2
0.3
0.4
0.5
10-4
10-3
10-2
10-1
Q2=5.0 GeV2
x
0.1
0.2
0.3
0.4
0.5
10-1 1 10
R(x
,Q2 )
x=0.0001
0.1
0.2
0.3
0.4
0.5
10-1 1 10
x=0.001
0.1
0.2
0.3
0.4
0.5
10-1
1 10
x=0.01
0.1
0.2
0.3
0.4
0.5
10-1
1 10
GRV off-shell
GRV on-shell
MRS(A) off-shell
MRS(A) on-shell
x=0.1
Q2(GeV2)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 34 / 49
Extension of R(x,Q2) down to Q2
= 0
R and FL in the low Q2, low x regionBKS (Badelek, Kwiecinski, Stasto, Z. Phys. C74 (1997) 297)...cont’d all with no HT
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
10-4 10-3 10-2 10-1
FL(
x,Q
2 )
Q2=0.1 GeV2
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
10-4 10-3 10-2 10-1
Q2=0.5 GeV2
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
10-4
10-3
10-2
10-1
Q2=1.5 GeV2
GRV off-shellGRV on-shellMRS(A) off-shellMRS(A) on-shell
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
10-4
10-3
10-2
10-1
Q2=5.0 GeV2
x
0.1
0.2
0.3
0.4
0.5
0.6
0.7
10-1 1 10
FL(
x,Q
2 )
x=0.0001
0.1
0.2
0.3
0.4
0.5
0.6
0.7
10-1 1 10
x=0.001
GRV off-shellGRV on-shellMRS(A) off-shellMRS(A) on-shell
0.02
0.04
0.06
0.08
0.1
0.12
0.14
10-1
1 10
x=0.01
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
10-1
1 10
x=0.1
Q2(GeV2)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 35 / 49
Extension of R(x,Q2) down to Q2
= 0
R and FL in the low Q2, low x regionBKS (Badelek, Kwiecinski, Stasto, Z. Phys. C74 (1997) 297)...cont’d all with HT
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.009
0.01
10-4 10-3 10-2 10-1
FL(
x,Q
2 )
Q2=0.1 GeV2
0.005
0.01
0.015
0.02
0.025
0.03
0.035
0.04
0.045
0.05
10-4 10-3 10-2 10-1
Q2=0.5 GeV2
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
10-4
10-3
10-2
10-1
GRV off-shellGRV on-shellMRS(A) off-shellMRS(A) on-shell
Q2=1.5 GeV2
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
10-4
10-3
10-2
10-1
Q2=5.0 GeV2
x
0.1
0.2
0.3
0.4
0.5
0.6
0.7
10-1 1 10
FL(
x,Q
2 )
x=0.0001
0.1
0.2
0.3
0.4
0.5
0.6
0.7
10-1 1 10
x=0.001
GRV off-shellGRV on-shellMRS(A) off-shellMRS(A) on-shell
0.02
0.04
0.06
0.08
0.1
0.12
0.14
10-1
1 10
x=0.01
0.02
0.04
0.06
0.08
0.1
0.12
0.14
10-1
1 10
x=0.1
Q2(GeV2)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 36 / 49
Extension of R(x,Q2) down to Q2
= 0
R and FL in the low Q2, low x region
x range of points: ∼10−5 – 0.01 ICHEP04/Abstract 5–0161
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 37 / 49
Extension of g1(x,Q2) down to Q2
= 0
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 38 / 49
Extension of g1(x,Q2) down to Q2
= 0
Measurements of gp1(x) for proton
Q2 >1 (GeV/c)2 Q2 <1 (GeV/c)2
COMPASS, PLB753 (2016) 18 COMPASS DIS2016
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 39 / 49
Extension of g1(x,Q2) down to Q2
= 0
g1 at low Q2, method IBadelek, Kiryluk, Kwiecinski, Phys. Rev. D61 (2000) 014009
The following representation of g1 was assumed:
g1(x,Q2) = gVMD1 (x,Q2) + gpart1 (x,Q2) (9)
gpart1 at low x is controlled by the ln2(1/x) terms; it was parametrised as discussed in Kwiecinski,
Ziaja, Phys. Rev. D60 (1999) 054004. gVMD1 (x,Q2) was represented as:
gVMD1 (x,Q2) =
Mν
4π
∑V=ρ,ω,φ
M4V ∆σV (W 2)
γ2V (Q2 +M2V )2
(10)
The unknown cross sections ∆σV (W 2) are combinations of the total cross sections for thescattering of polarised vector mesons and nucleons. At high W 2: ∆σV = (σ1/2 − σ3/2)/2Assume:
Mν
4π
∑V=ρ,ω
M4V ∆σV
γ2V (Q2 +M2V )2
=
C
[4
9
(∆u0val(x) + 2∆u0(x)
)+
1
9
(∆d0val(x) + 2∆d0(x)
)] M4ρ
(Q2 +M2ρ )2
, (11)
Mν
4π
M4φ∆σφp
γ2φ(Q2 +M2φ)2
= C2
9∆s0(x)
M4φ
(Q2 +M2φ)2
, (12)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 40 / 49
Extension of g1(x,Q2) down to Q2
= 0
g1 at low Q2, method I...cont’d
Each ∆p0j (x)→ x0 for x→0. Thus ∆σV → 1/W 2 at large W 2, i.e. zero intercept of theappropriate Regge trajectories.
Results for the spin asymmetry, A1 = g1/F1, for the proton, and for different C:
C ?? C <0 ?
-0.02
0
0.02
0.04
0.06
0.08
10-4
10-3
x
Ap1 Q2 < 1. GeV2
SMC data
C=+4C=0C=-4
Q2 = 0.02 0.06 0.1 0.2 0.3 0.6 GeV2
0
0.05
0.10
0.15
0.20
0.02 0.04 0.06 0.08 0.1x
Ap1 Q2 < 1 GeV2
E143 data,
Ebeam = 16.2 GeV2
C=0
C=-2
C=-4
Q2 = 0.5 0.6 0.7 0.8 0.9 1. GeV2
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 41 / 49
Extension of g1(x,Q2) down to Q2
= 0
g1 at low Q2, method IIBadelek, Kwiecinski, Ziaja Eur. Phys. J. C26 (2002) 45
The following representation of g1 was assumed, valid for fixed W 2 � Q2, i.e. smallx = Q2/(Q2 +W 2 −M2):
g1(x,Q2) = gL1 (x,Q2) + gH1 (x,Q2) =Mν
4π
∑V
M4V ∆σV (W 2)
γ2V (Q2 +M2V )2
+ gAS1 (x, Q2 +Q20). (13)
The first term sums up contributions from light vector mesons, MV < Q0, Q20 ∼ 1 GeV2. The
unknown ∆σV are expressed through the combinations of nonperturbative parton distributions,evaluated at fixed Q2
0, similar to method I.
The second term, gH1 (x,Q2), represents the contribution of heavy (MV > Q0) vector mesons tog1(x,Q2) can also be treated as an extrapolation of the QCD improved parton model structurefunction, gAS1 (x,Q2), to arbitrary values of Q2: gH1 (x,Q2) = gAS1 (x, Q2 +Q2
0). The scalingvariable x is replaced by x = (Q2 +Q2
0)/(Q2 +Q20 +W 2 −M2). It follows that at large Q2,
gH1 (x,Q2)→ gAS1 (x,Q2). Thus:
g1(x,Q2) = C
[4
9(∆u0val(x) + 2∆u0(x)) +
1
9(∆d0val(x) + 2∆d0(x))
]M4ρ
(Q2 +M2ρ )2
+ C
[1
9(2∆s0(x))
]M4φ
(Q2 +M2φ)2
+ gAS1 (x, Q2 +Q20). (14)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 42 / 49
Extension of g1(x,Q2) down to Q2
= 0
g1 at low Q2, method II...cont’dNow fixing C in the photoproduction limit via the DHGHY sum rule.
The γ∗p scattering amplitude fulfills the dispersion relation:
S1(ν, q2) = 4
∫ ∞−q2/2M
ν′dν′G1(ν′, q2)
(ν′)2 − ν2(15)
whereG1(ν, q2) =
M
νg1(x,Q2) (16)
in the Q2, ν →∞ limit.
As a result of Low’s theorem: S1(0, 0) = −κ2p(n), G1 in the Q2 → 0 limit fulfills
the DHGHY sum rule: ∫ ∞0
dν
νG1(ν, 0) = −1
4κ2p(n). (17)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 43 / 49
Extension of g1(x,Q2) down to Q2
= 0
g1 at low Q2, method II...cont’d
At ν →0, eq.(15) is:
S1(0, q2) = 4M
∫ ∞Q2/2M
dν
ν2g1(x(ν), Q2). (18)
Now we define the DHGHY moment, I(Q2) as:
I(Q2) = S1(0, q2)/4 = M
∫ ∞Q2/2M
dν
ν2g1(x(ν), Q2). (19)
Before taking the Q2 → 0 limit of (18), observe that it is valid only down to some threshold valueof W , Wth
<∼ 2 GeV (above resonances). Requirement W > Wth gives the lower limit forintegration over ν in (18), where νt(Q2) = (W 2
t +Q2 −M2)/2M :
I(Q2) = Ires(Q2) +M
∫ ∞νt(Q2)
dν
ν2g1(x(ν), Q2
). (20)
Here Ires = contribution of resonances. The DHGHY sum rule now implies:
I(0) = Ires(0) +M
∫ ∞νt(0)
dν
ν2g1 (x(ν), 0) = −κ2p(n)/4. (21)
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 44 / 49
Extension of g1(x,Q2) down to Q2
= 0
g1 at low Q2, method II...cont’d
Thus action plan for extracting C in eq.(14):
take g1(x(ν), 0), eq. (14); C is the only free parameter,
put it into eq. (21),
take Ires(0) from measurements,
extract C from eq. (21).
Taking:
Ires(0) from photoproduction, Wt=1.8 GeV GDH, Nucl. Phys. 105 (2002) 113,
gAS1 prametrized by NLO GRSV2000 Phys.Rev. D63 (2001) 094005
nonperturbative ∆p(0)j (x) at Q2 = Q2
0 = 1.2 GeV2 from
1 GRSV2000 =⇒ C = –0.302 “flat” ∆p
(0)j (x) = Ni(1− x)ηi =⇒ C = –0.24.
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 45 / 49
Extension of g1(x,Q2) down to Q2
= 0
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 46 / 49
Extension of g1(x,Q2) down to Q2
= 0
g1 at low Q2, method II...cont’dByproducts: g1 from eq.(14) and the DHGHY moment, I(Q2), eq.(19). Results for the proton:
-1
0
1
Q2=0.01 GeV2 x = 0.0001
-1
0
1
Q2= 0.1 GeV2 x = 0.001
-1
0
1
Q2= 1 GeV2 x = 0.01
-1
0
1
10-4
10-3
10-2
10-1
Q2= 10 GeV2
10-1
1 10
x = 0.1
x Q2 [GeV2]
g 1(x,
Q2 )
broken lines – gAS1 , dotted – gL1 ,
continuous – total g1
-0.002-0.001
00.0010.0020.0030.0040.0050.0060.0070.0080.009
0.0001 0.001 0.01
x
xg1
TOT ASYM
VMD (GRSV) SMC DATA
-1
-0.8
-0.6
-0.4
-0.2
0
0.2
0.4
0.6
0 0.2 0.4 0.6 0.8 1 1.2
Q2
[GeV2]
I(Q2)
TOTASYM
VMD (GRSV)E91-023 DATA
data at Q2< 1 GeV2
points markresonancesat W < Wt(Q
2)
Figures from: Badelek, Kwiecinski, Ziaja, Eur. Phys. J. C26 (2002)45.
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 47 / 49
Outlook
Outline
1 Mo & Tsai and Dubna schemes
2 Extension of F2(x,Q2) down to Q2 = 0Data at low Q2
JKBBMartin-Ryskin-Stasto(Modified) saturation modelALLM97ZEUS Regge fit
3 Extension of R(x,Q2) down to Q2 = 0
4 Extension of g1(x,Q2) down to Q2 = 0
5 Outlook
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 48 / 49
Outlook
Outlook
In the precision RC calculations a part of systematic uncertaintiescome from a choice of the input information.
We have collection of models (expressions) for Q2 → 0 extrapolationsfor:
form factors, suppression factorsF2(x,Q2)
R(x,Q2)
g1(x,Q2)
These expressions are valid at low x, appropriate for the EIC
but they have to be updated! =⇒ TO BE DONE
B. Badelek (Warsaw ) Input for RC RC JLab, 2016 49 / 49