CPeven scalar boson production via gluon fusion at the LHC
Abstract:
In view of the searches at the LHC for scalar particle resonances in addition to the 125 GeV Higgs boson, we present the cross section for a CPeven scalar produced via gluon fusion at NLO in perturbative QCD assuming that it couples directly to gluons in an effective theory approach. We refine our prediction by taking into account the possibility that the scalar couples to the topquark and computing the corresponding contributions through NLO in perturbative QCD. We assess the theoretical uncertainties of the cross section due to missing higherorder QCD effects and we provide the necessary information for obtaining the cross section value and uncertainty from our results in specific scenarios beyond the Standard Model. We also give detailed results for the case of a 750 GeV scalar, which will be the subject of intense experimental studies.
1 Introduction
The Higgs boson is the only elementary scalar particle of the Standard Model (SM) and the first such particle whose existence has been confirmed experimentally [1, 2]. Many wellmotivated extensions of the SM predict additional electrically neutral and colourless scalar particles (elementary or not). Experimental searches to hunt for these hypothetical new particles are thus of extreme importance. The LHC experiments have the potential to discover new scalar resonances by using similar techniques and signatures as in the search for the SM Higgs boson. In particular, in this context both the ATLAS an CMS experiments have recently reported an excess in the diphoton mass spectrum at an invariant mass of GeV [3, 4]. Even if it is still premature to speculate if the source of this excess is indeed a new resonance or merely a statistical fluctuation, this excess is exemplary for the kind of signals that may arise from the production of heavy singlet scalars in the current run of the LHC.
While specific models with colourneutral scalars may differ drastically in the principles which motivate them as well as in their spectra and interactions, the computation of QCD radiative corrections to their production cross section could share many common features that are independent of the underlying model. In particular, the production of hypothetical Higgslike scalars at the LHC may be favourable in the gluonfusion process due to the large value of the gluongluon luminosity. In this case, the modeldependence of the production cross section may only enter through the coupling describing the (effective) interaction of the Higgslike scalar with the gluons and quarks.
The purpose of this paper is to provide benchmark cross sections for a colorless CPeven scalar in an effective theory which is similar to the one obtained for the Higgs boson in the Standard Model after integrating out the topquark. The cross section for the scalar can be obtained from the cross section of the Higgs boson for a wide range of models, by appropriately taking into account the corresponding Wilson coefficient. Our computation is valid through NLO in perturbative QCD and is based on recent results for the SM Higgs boson gluonfusion cross section to that order [5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20]. We consider mass values for the new scalar particle ranging from 10 GeV to 3 TeV, which is the same range as the one targeted by the searches of ATLAS and CMS in the Run 2 of the LHC [21].
In some extensions of the Standard Model (see, for example, ref. [22] and references therein) the new Higgslike scalar has a nonnegligible coupling to the Standard Model topquark. For scalar masses below the toppair threshold the Wilson coefficient in our effective field theory (EFT) can be made to account for a coupling of the scalar to the topquark. However, for scalar masses around and above threshold, the effective theory approach is not accurate. For this purpose, we extend the effective theory by adding a Yukawatype interaction of the topquark with the scalar and compute the additional contributions generated by this coupling.
While we can correct our NLO crosssection in the effective theory for contributions of light Standard Model particles through NLO in QCD, it is not possible to do so in a modelindependent way for contributions from new relatively light particles with a mass smaller than about half the mass of the scalar. The presence of such particles will invalidate our effectivetheory computation. To estimate this effect, we consider the difference between the effective theory NLO crosssection and the exact NLO crosssection for a 750 GeV scalar which couples to a new toplike quark, as a function of the mass of the quark. We find that the two crosssections can differ by more than the QCD scale uncertainty for toplike quark masses as heavy as twice the mass of the scalar. Nevertheless, we observe that one can still make use of the effective theory computation for such a model also for lower values of the quark mass in order to estimate the relative size of the QCD corrections ( factor).
This article is organised as follows. In section 2 we present the effective theory setup of our calculation and discuss how the results can be adapted to suite a large class of models through the Wilson coefficient. We also discuss the theoretical uncertainty that is associated with the NLO QCD corrections in the effective theory. In section 3, we discuss the inclusion of finite width effects in the calculation and provide a fit of the zerowidth cross section to enable the adaptation of our results to models where the scalar resonance is expected to have a nonvanishing but narrow width. In section 4, we investigate the validity of our effective theory approach under the assumption that the coupling between the scalar and the gluons is mediated by a heavy top partner of varying mass. In section 5 we go beyond a purely effective description of the scalar interaction and allow for a nonvanishing coupling between the Standard Model top and the scalar, for which we present the NLO QCD corrections. We conclude in section 6 and provide appendices with tables of cross section values for a wide range of masses of the scalar.
2 NLO QCD corrections in the effective theory approach
Let us consider a model where the SM is extended by a colourless singlet scalar of mass and width , which only couples to the SM in a minimal way. For now, we assume that the only SM fields that the new scalar couples to are the gluons, through an effective operator of dimension five. We will discuss Yukawa couplings to heavy quarks in Section 5. Such a model can be described by a Lagrangian of the form
(1) 
where collects the kinetic term and the potential of the scalar . The vacuum expectation value is [23],
and is a Wilson coefficient parametrizing the strength of the coupling of the particle with the gluons. Except for the value of the Wilson coefficient, this is the same dimensionfive operator as the one coupling the SM Higgs boson to the gluons, after the topquark has been integrated out.
Since the scalar couples to gluons, it can be produced at hadron colliders. Its hadronic production cross section can be cast in the form
(2) 
where is the mass scale introduced by dimensional regularisation and is the scale of new physics, representing collectively the masses of the heavy particles in some ultraviolet (UV) completion of the theory. At all orders in perturbation theory, the cross section is independent of the arbitrary scale , with the scale variation of the square of the Wilson coefficient cancelling the one of the matrixelements . Since does not depend on , the scale dependence of the Wilson coefficient itself is universal, and it does not depend on the (renormalizable) UV completion of the effective theory. In particular, the renormalization group evolution equation that relates the Wilson coefficient at two different scales reads [24]:
(3) 
with the QCD function. We see that, as expected, the evolution equation only depends on the lowenergy effective theory, and it is independent of the details of the UV completion. As a consequence, if we divide the production cross section by the value of the Wilson coefficient at some reference scale ,
(4) 
we obtain a universal ratio that does not depend on the UV completion (up to corrections due to differences that the various highenergy theories may induce on the width of the scalar particle ). In particular, since the dimensionfive operator in eq. (1) is the same as the dimensionfive operator mediating Higgs production in gluon fusion in the SM, we can extract the righthand side of eq. (4) from Higgs production in gluonfusion,
(5) 
Here is the gluonfusion production cross section of a Higgs boson of mass and width , in a variant of the SM with massless quarks and the topquark integrated out, is the topquark mass and is the Wilson coefficient multiplying the SM gluonfusion operator evaluated at our reference scale . In the scheme it reads [25, 26]
with , being the strong coupling constant in QCD with flavors. We note at this point that eq. (5) is valid to all orders in perturbation theory, and so neither nor depend on the arbitrary scales and . They do, however, acquire a dependence on the scale once the perturbative series is truncated at some finite order. It is of course possible to further separate into a factorization and a renormalization scale, and , as it is customary, and we will assume in the following that this is done implicitly. In addition, also acquires a dependence on our reference scale order by order in perturbation theory, resulting from the truncation of the perturbative expansion of the ratio of Wilson coefficients.
Equation (5) is the basic formula which allows us to predict the hadronic production cross section of a generic CPeven singlet scalar through gluonfusion to high order in perturbative QCD. Indeed, the gluonfusion cross section is known through NLO in QCD in the large limit [19] (for ) as an expansion around the Higgs threshold, and eq. (5) easily allows us to convert the result for the Higgs boson into the corresponding NLO cross section for a generic CPeven scalar.
At this point we need to make a small technical comment: both the ratio of Wilson coefficient and the Higgs production cross section on the righthand side of eq. (5) admit a perturbative expansion up to NLO. Multiplying these two perturbative expansions, as suggested by eq. (5), introduces additional terms into the perturbative expansion which are of NLO accuracy or higher, thus beyond the reach and the control of our NLO computation. In ref. [19] it was shown that the numerical impact of these terms is captured by the scale variation at NLO if the central scale is chosen as , and the effects of such terms are therefore accounted for by the scale variation uncertainty assigned to our prediction.
A second comment regards the dependence of the Standard Model Wilson coefficient on the various scales. The SM Wilson coefficient depends on the reference scale and the topquark mass . While the value of is arbitrary and can be chosen freely, in practise we choose the value of to be the mass of the scalar, , and we set the value of the mass of the top quark to its physical value. In doing so, we may introduce large logarithms into the computation, order by order in perturbation theory. For example, we see that at NNLO and NLO the Wilson coefficient contains a logarithm of the form . Similarly, the Wilson coefficient expression for the scalar will contain logarithms of the type . In order to avoid large logarithms due to widely disparate scales in the ratio of the two Wilson coefficients, it may be preferable to evaluate the SM Wilson coefficient with a topquark mass value rather than its physical value. The value of the top mass entering the SM Higgs crosssection can easily be changed with a simple rescaling,
(7) 
Let us now discuss our predictions for the production cross section . In Figs. 1  4 we show the effective theory cross section as a function of through NLO, at a protonproton collider with a centerofmass energy of 13 TeV. The theory uncertainty bands correspond to a variation of the scale in the range . The value of the top mass is set to GeV and we use the PDF4LHC15 [27] parton distribution function (PDF) set. We observe that for the range of scalar masses between 10 GeV and 3 TeV the NLO scale variation band is always contained inside the NNLO band. This is also true for the lowest values of this mass range, (Fig. 1), where one observes especially large corrections at NLO and NNLO. For higher masses, the scale variation is reduced. As an illustration, we show in Fig. 5 the scale dependence for the cross section of a CPeven scalar with a mass of 750 GeV.
Sources of theoretical uncertainty affecting the gluonfusion cross section at NLO other than scale variation have been considered in detail in ref. [19]. In particular, these sources of uncertainty are due to the lack of NLO parton densities and to the truncation of the threshold expansion for the NLO correction. In order to estimate the uncertainty on our computation, we follow faithfully the uncertainty estimation prescription of ref. [19] with one modification: the uncertainty due to the lack of NLO parton densities is estimated here as the envelope of the corresponding uncertainties formed by using the CT14 [28], NNPDF30 [29] and PDF4LHC sets, which are themselves computed according to the prescription of ref. [19]. The reason for modifying the prescription in this case is that the PDF4LHC set leads to very small uncertainties for scalars in the region GeV and, in particular, to a vanishing uncertainty estimate for a scalar with mass around 770 GeV. This feature is not shared with individual PDF sets. We therefore use, conservatively, the envelope of CT14, NNPDF30 and PDF4LHC, which leads to an uncertainty due to the lack of NLO parton densities at the level of for scalars in the range GeV TeV. This uncertainty remains of the order of a few percent also at lower masses, but it increases rapidly to for GeV.
3 Finite width effects and the lineshape
The results of the previous section hold formally only when the width of the scalar is set to zero.
In many beyond the Standard Model (BSM) scenarios, however, finitewidth effects
cannot be neglected. In this section
we present a way to include leading finitewidth effects into our results,
in the case where the width is not too large compared to the mass.
The total cross section for the production of a scalar boson
of total width can be obtained from the cross section in the
zerowidth approximation via a convolution
(8) 
where is the virtuality of the scalar particle. This expression is accurate at leading order in . For large values of the width relative to the mass, subleading corrections and signalbackground interference effects are important and are not captured by eq. (8). Let us also note that in order to use eq. (8) faithfully, one needs the width as a function of the virtuality of the scalar, which may bear a substantial model dependence. However, it is often the case that the width can be approximated as
(9) 
The invariant mass distribution in this approximation for the production of a CPeven scalar with a mass of GeV and total width from 2 to of the scalar mass is shown in Fig. 6. This result has been obtained from an interpolation of the zerowidth cross section values of Tab. 6 in Appendix A. We caution that if the results of Appendix A are used to derive cross section values with nonzero width effects following the strategy outlined in this section, an additional uncertainty of the order should be assigned.
In order to facilitate the computation of the lineshape we perform a parametric fit of the production cross section of a zerowidth scalar boson as a function of its virtuality (with ) for a centerofmass energy of TeV. To guarantee agreement of the fitted cross section with the actual cross section at a level better than , we split the range of interesting scalar boson mass values into three intervals:

in the range , cf. Tab. 3, we find
(11)
where . The fits of Eqs. (10)(12) can be used as the kernel of the convolution in eq. (8).
4 Validity of the EFT approach
So far, we have assumed that the effective theory of eq. (1) furnishes an accurate description of the gluonscalar interaction. However, this assumption may be challenged if the scalar couples to light coloured particles. To investigate this effect, we will consider a scalar of mass GeV which couples to a toplike quark of mass . In this scenario, we can compute the crosssection exactly through NLO in perturbative QCD. We can then compare this prediction with the crosssection derived with the effective theory of eq. (1).
The red line in Fig. 7 shows the percent difference of the two predictions,
(13) 
as a function of for a scalar with a mass of GeV. While in the region the effective field theory is inadequate, for larger values of the EFT description becomes accurate very quickly. Indeed, for the discrepancy with respect to the exact result is already below the theoretical uncertainty of QCD origin (of about ).
Let us now define a rescaled effective theory crosssection,
(14) 
The green line of Fig. 7 shows the relative difference
(15) 
between the rescaled effective theory prediction for the crosssection and the exact NLO cross section. We notice that the rescaled effective theory crosssection reproduces the exact cross section very well (as it has also been the case for the Standard Model Higgs boson [30, 31]).
We conclude that, while for the study of scalars which couple to relatively light particles one should resort to a calculation within a specific model, also in such situations the effective theory computation can be utilised as a means to compute with a good accuracy the corresponding QCD factor with respect to the exact LO cross section.
5 Topquark contributions
So far we have considered the case where the interaction between the heavy scalar and the Standard Model is mediated exclusively by the dimensionfive operator in eq. (1). In many extensions of the Standard Model new scalars may also couple directly to the quarks of the third generation as well as the and bosons. However, the absence of resonances in toppair production and fourlepton production suggests that these couplings should be small. It may be nonetheless useful to estimate the contribution to the inclusive scalar cross section due to the heavy SM quarks and gauge bosons for the purpose of setting precise experimental constraints on their couplings to . In this section, we study the effects due to the top quark, which in many scenarios are expected to give the most important contribution.
For light scalar masses, , contributions due to the coupling of the top quark can be simply taken into account through NLO in perturbative QCD by using an appropriate Wilson coefficient . For heavier scalars with masses around and above the toppair threshold, however, it is not justified theoretically to integrate out the topquark. For this reason we study in the following the effect of modifying the Largangian in eq. (1) by including a direct Yukawa interaction between the scalar and the top quark, i.e., we consider the Lagrangian
(16) 
where is the ratio of the Yukawa coupling of the scalar and the Yukawa coupling of the Higgs boson to the top quark and denotes the ratio of the Wilson coefficients in this theory and the SM with the topquark integrated out.
The production cross section now depends on the values of and :
(17) 
It does not take too much effort to show that the NLO cross section can be cast in the form
(18) 
The cross section with no Yukawa coupling, , is known through NLO. Hence, we can improve the previous expression by including all NLO corrections to the terms proportional to ,
(19) 
Component  value[fb]  (theory) []  (pdf+) []  

7 TeV  69.72  7.0  
55.59  19.52  6.95  
61.71  22.69  6.94  
152.6  22.1  6.92  
8 TeV  111.4  6.1  
89.37  19.18  6.23  
98.92  22.3  6.22  
245.3  21.71  6.2  
13 TeV  496.9  4.0  
404.6  18.3  4.5  
442.7  21.3  4.4  
1108  20.7  4.4  
14 TeV  609.7  3.8  
497.3  18.21  4.26  
543.  21.14  4.2  
1361  20.57  4.21 
Although we have taken into account QCD corrections within the EFT framework through NLO , the result for the cross section is formally only NLOaccurate because we are missing finitemass effects beyond NLO. We therefore estimate the uncertainties on the NLO crosssection as
(20) 
where
(21) 
is the relative change of the gluonfusion crosssection in the effective theory from NLO to NLO. Note that in this way make the assumption that the crosssection components that are not known beyond NLO will not have a worse perturbative convergence than in the effective theory. Finally, we enlarge this uncertainty further by
(22) 
which measures the scheme dependence of the topquark contribution at NLO.
Equation (19) is our best prediction for the gluonfusion production cross section of a generic scalar . We recall that the values of as a function of the scalar boson mass can be read off from Tabs. 25 in Appendix A. The NLO cross sections () are reported in Tabs. 714. As an illustration, we present the complete set of terms entering in eq. (19) for a scalar of mass in Tab. 1.
The following fits are a good approximation to the central values of the cross sections introduced in this section. The fits are functions of valid for a centerofmass energy of TeV and PDF4LHC15 parton distribution functions.

In the range , we find
(23) (24) 
In the range , we find

In the range , we find

In the range , we find
(25) (26) (27) 
In the range , we find
(28) (29) (30)
6 Conclusion
In this paper we have presented the most precise predictions for the production of a CPeven scalar produced in gluon fusion at the LHC. We assume that the coupling of the scalar to the gluons can be described in an effective field theory approach. This enables us to provide precision QCD corrections at NLO to the production cross section at the LHC. We find that for a scalar with a mass of GeV the NLO corrections yield a theoretical uncertainty of when we choose the central renormalization and factorization scales to be and follow the prescriptions layed out in ref. [19].
If we assume that the effective theory Wilson coefficient coupling of the CPeven scalar and the gluons is generated by a heavy toppartner, we can estimate the validity of the effective description as a function of the heavy fermion mass. We find that in the threshold region, when the mass of the toppartner is about half the mass of the scalar, the corrections can be as large as . However, from direct searches the low mass ranges for toppartners are tightly constrained, so that at least for the discussion of a GeV scalar this configuration is unlikely. For heavier fermion masses the effective theory description performs very well and is already accurate at when then fermion mass is about times the scalar mass.
Additionally, we have considered the case that the new scalar couples directly to the SM top quark. In this case we need to take into account the top corrections to the cross section, which are unfortunately only available through NLO in QCD, and they thus come with uncertainties of up to . We incorporate the top corrections to the scalar production cross section as a function of the Yukawa coupling of the top quark to the scalar, allowing for a modeldependent choice of the parameters.
In summary, we have provided the ingredients to endow predictions for the production of a CPeven scalar at the LHC with the most precise QCD corrections available. The only assumptions that have been made in the calculation are that the scalar couples to the gluons through an effective theory like operator. By determining the Wilson coefficient for the scalar coupling to the gluons as well as the Yukawa coupling to the top within a concrete model, it is now possible, using the numbers provided here, to easily incorporate QCD corrections through NLO into the cross section predictions for the LHC. While the results presented in this article concern strictly the production of a CPeven scalar, we can use the same techniques to compute the crosssection for resonance production of different CP/spin types which may be phenomenologically relevant (see, for example, ref. [32]). This will be the subject of future works.
Acknowledgements
CD, EF and TG are grateful to the KITP, Santa Barbara, for the hospitality during the final stages of this work. We are grateful to Riccardo Barbieri and Giuliano Panico for useful dicussions. This research was supported in part by the National Science Foundation under Grant No. NSF PHY1125915, by the Swiss National Science Foundation (SNF) under contracts 200021165772 and 200020162487 and by the European Commission through the ERC grants “pertQCD”, “HEPGAME” (320651), “HICCUP”, “MathAm” and “MC@NNLO” (340983).
References
 [1] ATLAS collaboration, G. Aad et al., Observation of a new particle in the search for the Standard Model Higgs boson with the ATLAS detector at the LHC, Phys. Lett. B716 (2013) 1–29, [1207.7214].
 [2] CMS collaboration, S. Chatrchyan et al., Observation of a new boson at a mass of 125 GeV with the CMS experiment at the LHC, Phys. Lett. B716 (2012) 30–61, [1207.7235].
 [3] ATLAS collaboration, Search for resonances decaying to photon pairs in 3.2 fb of collisions at = 13 TeV with the ATLAS detector, ATLASCONF2015081.
 [4] CMS collaboration, CMS Collaboration, Search for new physics in high mass diphoton events in protonproton collisions at 13TeV, CMSPASEXO15004.
 [5] P. A. Baikov, K. G. Chetyrkin, A. V. Smirnov, V. A. Smirnov and M. Steinhauser, Quark and gluon form factors to three loops, Phys. Rev. Lett. 102 (2009) 212002, [0902.3519].
 [6] T. Gehrmann, E. W. N. Glover, T. Huber, N. Ikizlerli and C. Studerus, Calculation of the quark and gluon form factors to three loops in QCD, JHEP 06 (2010) 094, [1004.3653].
 [7] M. Höschele, J. Hoff, A. Pak, M. Steinhauser and T. Ueda, Higgs boson production at the LHC: NNLO partonic cross sections through order and convolutions with splitting functions to NLO, Phys. Lett. B721 (2013) 244–251, [1211.6559].
 [8] C. Anastasiou, S. Buehler, C. Duhr and F. Herzog, NNLO phase space master integrals for twotoone inclusive cross sections in dimensional regularization, JHEP 11 (2012) 062, [1208.3130].
 [9] C. Anastasiou, C. Duhr, F. Dulat and B. Mistlberger, Soft triplereal radiation for Higgs production at N3LO, JHEP 1307 (2013) 003, [1302.4379].
 [10] C. Anastasiou, C. Duhr, F. Dulat, F. Herzog and B. Mistlberger, Realvirtual contributions to the inclusive Higgs crosssection at , JHEP 12 (2013) 088, [1311.1425].
 [11] Y. Li and H. X. Zhu, Single soft gluon emission at two loops, JHEP 11 (2013) 080, [1309.4391].
 [12] C. Anastasiou, C. Duhr, F. Dulat, E. Furlan, T. Gehrmann, F. Herzog et al., Higgs boson gluon–fusion production at threshold in NLO QCD, Phys. Lett. B737 (2014) 325–328, [1403.4616].
 [13] C. Anastasiou, C. Duhr, F. Dulat, E. Furlan, T. Gehrmann, F. Herzog et al., Higgs boson gluonfusion production beyond threshold in NLO QCD, JHEP 03 (2015) 091, [1411.3584].
 [14] F. Dulat and B. Mistlberger, RealVirtualVirtual contributions to the inclusive Higgs cross section at N3LO, 1411.3586.
 [15] C. Duhr, T. Gehrmann and M. Jaquier, Twoloop splitting amplitudes and the singlereal contribution to inclusive Higgs production at NLO, JHEP 02 (2015) 077, [1411.3587].
 [16] Y. Li, A. von Manteuffel, R. M. Schabinger and H. X. Zhu, Softvirtual corrections to Higgs production at NLO, Phys. Rev. D91 (2015) 036008, [1412.2771].
 [17] C. Anastasiou, C. Duhr, F. Dulat, F. Herzog and B. Mistlberger, Higgs Boson GluonFusion Production in QCD at Three Loops, Phys. Rev. Lett. 114 (2015) 212001, [1503.06056].
 [18] C. Anzai, A. Hasselhuhn, M. Höschele, J. Hoff, W. Kilgore, M. Steinhauser et al., Exact NLO results for qq H + X, JHEP 07 (2015) 140, [1506.02674].
 [19] C. Anastasiou, C. Duhr, F. Dulat, E. Furlan, T. Gehrmann, F. Herzog et al., High precision determination of the gluon fusion Higgs boson crosssection at the LHC, JHEP 05 (2016) 101, [1602.00695].
 [20] C. Anastasiou, C. Duhr, F. Dulat, E. Furlan, F. Herzog and B. Mistlberger, Soft expansion of doublerealvirtual corrections to Higgs production at NLO, JHEP 08 (2015) 051, [1505.04110].
 [21] Higgs CrossSection Working Group, BSM Higgs production cross sections at = 13 TeV (update in CERN Report4 2016), .
 [22] M. J. Dolan, J. L. Hewett, M. Krämer and T. G. Rizzo, Simplified Models for Higgs Physics: Singlet Scalar and Vectorlike Quark Phenomenology, 1601.07208.
 [23] Particle Data Group collaboration, S. Eidelman et al., Review of particle physics. Particle Data Group, Phys. Lett. B592 (2004) 1–1109.
 [24] V. P. Spiridonov and K. G. Chetyrkin, Nonleading mass corrections and renormalization of the operators m psibar psi and g**2(mu nu), Sov. J. Nucl. Phys. 47 (1988) 522–527.
 [25] K. G. Chetyrkin, B. A. Kniehl and M. Steinhauser, Decoupling relations to O (alphas**3) and their connection to lowenergy theorems, Nucl. Phys. B510 (1998) 61–87, [hepph/9708255].
 [26] Y. Schroder and M. Steinhauser, Fourloop decoupling relations for the strong coupling, JHEP 01 (2006) 051, [hepph/0512058].
 [27] J. Butterworth et al., PDF4LHC recommendations for LHC Run II, J. Phys. G43 (2016) 023001, [1510.03865].
 [28] S. Dulat, T. J. Hou, J. Gao, M. Guzzi, J. Huston, P. Nadolsky et al., The CT14 Global Analysis of Quantum Chromodynamics, 1506.07443.
 [29] NNPDF collaboration, R. D. Ball et al., Parton distributions for the LHC Run II, JHEP 04 (2015) 040, [1410.8849].
 [30] R. V. Harlander and K. J. Ozeren, Finite top mass effects for hadronic Higgs production at nexttonexttoleading order, JHEP 11 (2009) 088, [0909.3420].
 [31] A. Pak, M. Rogal and M. Steinhauser, Finite top quark mass effects in NNLO Higgs boson production at LHC, JHEP 02 (2010) 025, [0911.4662].
 [32] G. Panico, L. Vecchi and A. Wulzer, Resonant Diphoton Phenomenology Simplified, 1603.04248.
Appendix A Reference tables for the production cross section of a CPeven scalar through topquark loops
Here we report the values of the NLO production cross section for a scalar of mass at the 13 TeV LHC, in an effective theory where the top quark has been integrated out. The mass of the top quark is chosen to be 162.7 GeV. We also show the two components that enter this result (cf. eq. (19), with , , ), i.e. the squared SM Wilson coefficient and the matrixelement . The theory error is computed from the variation of the common renormalization and factorization scale in the range . To it we add linearly the uncertainty from missing NLO PDFs, evaluated following the method of sec. 2, and from the truncation of the threshold expansion. We refer to ref. [19] for an explanation of how the latter is computed. We use the PDF set PDF4LHC14 [27] and derive the combined PDF+ error following the indications of the PDF4LHC working group.