Joint analysis of the iron emission in the optical and near-infrared spectrum of I Zw 1

Constraining the physical conditions of the ionized media in the vicinity of an active supermassive black hole (SMBH) is crucial to understanding how these complex systems operate. Metal emission lines such as iron (Fe) are useful probes to trace the gaseous media's abundance, activity, and evolution in these accreting systems. Among these, the FeII emission has been the focus of many prior studies to investigate the energetics, kinematics, and composition of the broad-emission line region (BELR) from where these emission lines are produced. In this work, we present the first simultaneous FeII modeling in the optical and near-infrared (NIR) regions. We use CLOUDY photoionization code to simulate both spectral regions in the wavelength interval 4000-12000 Angstroms. We compare our model predictions with the observed line flux ratios for IZw1 - a prototypical strong FeII-emitting active galactic nuclei (AGN). This allows putting constraints on the BLR cloud density and metal content that is optimal for the production of the FeII emission, which can be extended to IZw1-like sources, by examining a broad parameter space. We demonstrate the salient and distinct features of the FeII pseudo-continuum in the optical and NIR, giving special attention to the effect of micro-turbulence on the intensity of the FeII emission.


Introduction
Active Galactic Nuclei (AGNs) are harbored within the central cores of galaxies and host active supermassive black holes (SMBHs) at their very centers.The radiation originates from the heating of material that is drawn toward the SMBH and subsequently triggers ionization within the gas and metal-rich environment surrounding these systems [1,2].The appearance of AGNs can vary based on the observer's perspective due to their intricate geometry [3,4].This phenomenon also becomes evident in their observed spectra.
AGNs are further classified in terms of the types of emission line profiles they demonstrate in their observed spectrum.Sources displaying a combination of broad (permitted and semi-permitted) as well as narrow forbidden emission line profiles are categorized as Type I sources, while those lacking broad emission lines are classified as Type-II [1,[5][6][7].This study exclusively concentrates on Type-I sources and their emissions.The origin of the broad emission lines can be traced back to a region located around 0.01 -0.1 parsecs away from the central ionizing source.This measurement is derived through the widely recognized technique of reverberation mapping, which has been applied to a substantial sample of over 200 AGNs to date [8][9][10][11][12][13][14].This region, known as the broad line region (BLR), contributes significantly to the emissions observed in a typical Type-I AGN spectrum [10,11,.Notably, among the salient BLR emission features is the FeII emission, encompassing a wavelength range from ultraviolet (UV) to near-infrared (NIR) [37][38][39].
The emission from FeII is remarkable in the Type-I AGN spectra, primarily due to the aggregation of an extensive array of transitions, above 344,000 in number.The accumulation of the lines mimics a continuum and is hence known as a pseudo-continuum in an AGN spectrum [38,[40][41][42].The large number of emissions stemming from the FeII ion play a pivotal role in characterizing the energy distribution within the BLR -approximately 25% of the total emission originating in the BLR is attributed to the FeII lines [37].Succeeding investigations have confirmed the utility of FeII emission as a surrogate for quantifying metal abundances in the BLR of AGNs across a broad range of redshifts, contributing to insights into the evolutionary trends of metallic elements within these galaxies [4,[43][44][45][46].
Another crucial outcome from studies centered on FeII pertains to the development of emission templates specific to FeII.From a purely observational standpoint, the pioneering work of Boroson and Green [47] marked the inception of extracting FeII emission from a quintessential AGN FeII emitter, I Zw 1, which itself has a long and rich history [40,[48][49][50][51][52][53][54].The FeII template was meticulously constructed by isolating and excluding all emission lines except in the FeII.This template-driven approach continues to be employed across various studies, serving as the foundation for quantifying the optical intensity of FeII.Boroson and Green [47] played a pioneering role in comprehending FeII emission.In the same article, the authors established a robust association between the strength of FeII and various attributes of both the Broad Line Region (BLR) and the Narrow Line Region (NLR).Employing Principal Component Analysis (PCA) -a dimensionality reduction technique, they identified multiple correlations between observed spectroscopic parameters.Eigenvector1 (EV1) was the paramount correlation space among them.The EV1 from their study shows a strong correlation between the optical FeII emission within the spectral range of 4434 -4684Å, (centered at 4570 Å), the peak intensity of [OIII]λ5007 and the Full Width at Half Maximum (FWHM) of broad component of the Hβ emission profile.This led to the well-established FWHM(Hβ) vs. R FeII connection 1 , that is the well-known optical plane of the quasar main sequence [4,24,26,[55][56][57][58]. Within this framework, FeII emission not only plays a crucial role in discerning the underlying factors driving EV1 [4] but also serves as a link between line strength and the emitting gas physics for a diverse population of Type-1 AGNs along the EV1 plane.
Regarding FeII templates, Vestergaard and Wilkes [59] progressed in the UV domain by extending the template methodology into the ultraviolet spectrum using a high-quality HST/FOS spectrum for I Zw 1. Subsequent efforts have since created several templates, effectively characterizing UV-optical FeII emission across large samples of AGNs [60][61][62][63][64]. Kovačević et al. [64] devised an innovative optical FeII template through a semi-empirical approach, combining the observed FeII emission with theoretical predictions based on allowed transitions between energy levels for the ion.By meticulously measuring individual FeII multiplet groups, they achieved a better overall agreement and were successful in getting an improved estimate for the R FeII .Reproducing FeII emission in sources similar to I Zw 1 has been challenging, as empirical templates struggle to replicate specific features in observed spectra.In a recent development, Park et al. [65] introduced a new template based on the Mrk 493 spectrum to improve the modeling of FeII emission in I Zw 1-like sources.
Significant strides in predicting near-infrared (NIR) FeII spectra were accomplished by Sigut and Pradhan [66] and Sigut and Pradhan [41] through their utilization of the Lyα fluorescence excitation mechanism to elucidate FeII emission within the wavelength range of, 8500 -9500 Å.This Lyα process exclusively governs the stimulation of energy levels reaching up to 13.6 eV, giving rise to the emission lines observed within this spectral segment.This assertion was corroborated by observations conducted by Rodriguez-Ardila et al. [67].A semi-empirical template for NIR FeII emissions was later developed by Garcia-1 R FeII is usually the ratio of the integrated FeII emission within 4434 -4684Å to the broad Hβ emission.
It is important to note here that the NIR spectrum exhibits FeII emission lines that are either isolated or semi-isolated, in contrast to the UV-optical region (see Figure 1).This feature provides a distinct advantage by enabling a more precise and reliable determination of the FeII emission line properties.The FeII lines around the 1-micron region at λ9997, λ10502, λ10863 and λ11127 are the most intense among the FeII lines [67,70,71].
Recent research [69] has highlighted the intrinsic correlation between FeII emission in the optical and near-infrared spectral regions.Due to the intricacies of FeII ion's behavior and the challenges associated with modeling its emission, researchers have been prompted to explore alternative approaches for investigating the FeII emitting gas [72,73].Observational and photoionization modeling works have been indicating that less complex ions, such as CaII triplet (λλ8498Å, 8542Å, and 8662Å, also known as CaT), and OI lines at λ8446 and λ11287, can serve as proxies for studying the spatial distribution of FeII emitting gas [67,69,[72][73][74].
The FeII emission is dependent on the BLR temperature and gas density, and also on other parameters such as composition and Brownian motion within the BLR cloudhigh temperature (∼5000 -10000 K), relatively high density (log n H = 10 -12 cm −3 ), supersolar abundances and micro-turbulence of the order of ∼100 km s −1 have been shown to positively impact the production of the FeII emission, especially in the UV and optical [4,44,75,76].In a recent study [77], we assessed the NIR FeII emission in I Zw 1 under low column density regime (10 22 cm −2 ).In this current work, we make a comprehensive analysis of the influence of micro-turbulence on NIR emission, over a broad range of cloud densities and metal content, in conjunction with the optical region.
We aim to model the FeII emission to understand the line formation and the nature of the physical conditions in AGN with strong FeII emission, exploring how changing metal content and micro-turbulence effectively increases the FeII optical and NIR emissions simultaneously.For this purpose, we investigate three cases of micro-turbulence velocities within the ionized media, i.e., 0, 10, and 100 km s −1 and their impact on the optical and NIR FeII emission (based on the previous study by Panda [74] which focused only on recovering the optical FeII emission).The inclusion of the micro-turbulence parameter has been shown to augment the FeII emission line intensity in the optical and UV regions, due to the increase in the overall random, Brownian motion of the ions within the cloud [46,76,78].This is the first study that tests the effect of micro-turbulence and metal content on the production of NIR FeII emission, along with the optical FeII emission.
The paper is organized as follows: Section 2 presents the observational spectroscopic data in the optical and NIR for I Zw 1. Section 3 describes the extraction of the FeII emission in the optical and NIR regime and estimation of the corresponding FeII strengths from the observed spectra.In Section 4, we demonstrate the modeling setup using the photoionization code CLOUDY [79].Results are highlighted in Section 5, followed by a brief discussion in Section 6.We summarize our findings from this study in Section 7.
The I Zw 1 spectrum has a rich historical record of UV, optical, and NIR studies [47,59,64,68].Recent reverberation mapping estimated the BLR's distance of the lineemitting BLR from the central ionizing source, R BLR =37.2 light-days [81], a crucial parameter in our model that allows to minimize the degeneracies in our models.
To perform the optical and NIR analysis, we use the reduced and available spectra of the I Zw 1 from previous works [67][68][69]71].The NIR spectrum was observed using the 3.2 m IRTF telescope (NASA Infrared Telescope Facility) at Mauna Kea, Hawaii-USA in 2000 (PI: Rodríguez-Ardila).It employed the SpeX spectrograph in cross dispersion mode (SXD) covering the wavelength interval 0.8 -2.4 microns, and photometric bands zJHK, with a spectral resolution, R = 2000 corresponding to the 0.8" × 15" slit.Rodriguez-Ardila et al. [67] and Riffel et al. [71] presented in detail the reduced spectrum and related observational information.The optical counterpart region was obtained from the published work of Rodriguez-Ardila et al. [67].The optical spectrum was obtained at the CASLEO Observatory (Complejo Astronómico el Leoncito -San Juán, Argentina), employing the REOSC spectrograph in long slit mode covering the range 3500 -6800 Å with resolution is 0.10 Å pixel −1 .

Optical and NIR FeII Templates
The FeII pseudo-continuum is comprised of a multitude of permitted FeII lines spanning from the UV to the NIR spectral regions, necessitating the use of templates for accurate modeling.In the optical domain, we used the empirical FeII template by Boroson and Green [47] derived from the spectrum of I Zw 1, the archetypal Narrow-Line Seyfert 1 (NLS1) galaxy.In the NIR range, we employed the semi-empirical template from Garcia-Rissmann et al. [68].These models encompass approximately 1915 lines, within the wavelength range of 8000 -11600 Å.The FeII spectrum was synthesized using the template, where first we widened it based on the width of the 10502 Å line, and then scaled the template to match with the observed spectrum.We use the 10502 Å line as it is the most isolated FeII line in this region, thus allowing for accurate measurement of the overall FeII intensity.The best-fit template was determined by minimizing the chi-square value at, 10502 Å. Figure 1 illustrates the spectral fitting and component decomposition for the optical and NIR region for I Zw 1.The estimation of flux uncertainty is based on the standard deviation of the best-fitted model, considering the spectrum within 3σ of the lines of interest.For the ratios, we employed error propagation to estimate their uncertainties (see Table 1).In NLS1s, BLR emission line profiles are best represented using Lorentzian, while those of Type-I AGN with broader lines follow Gaussian shapes [83].We fit the emission lines using the LMFIT library from Python, employing Lorentzian profiles to represent the BLR components, where the χ 2 minimizes the residual left after subtraction of the fitted profile.For lines emitted by the Narrow-line Region (NLR), we employ only Gaussian components.This procedure enabled us to measure the integrated flux, FWHM, and centroid position of the lines of interest for this study.
From these spectra and template fits, we obtain the optical FeII and NIR intensities and their respective ratios to the nearest Hydrogen line, where for the optical FeII we obtained R 4570 = 1.62±0.06.Our findings are similar to previous works (∼1.47 [7]).For R 1µ m , we observe a discrepancy between our findings and the results reported by Marinello et al. [69] where they recover a ratio, R 1µm = 1.81±0.08.It is crucial to clarify that the value of 1.81 is derived from the study by Marinello et al. [69], and not from our current investigation.This value, from our analysis, is much smaller.We refer the readers to Table 1.The noted disparity is attributed to differing methodologies employed in the two studies.Specifically, Marinello et al. [69] estimated measurements corresponding to Paα and utilized the theoretical lines' ratio from Garcia-Rissmann et al. [68] to re-scale to Paβ intensity.In contrast, our approach involves the estimation of the flux of the broad component of Paγ.We opted for the utilization of Paγ rather than Paβ, as the latter is situated within a telluric spectral region [see e.g., 39].
We investigated how the ratio between Paγ and Paβ evolves under varying gas physical conditions from our CLOUDY simulations (see Appendix A1) instead of using the theoretical value for Paβ/Paγ (i.e., 0.8531).In Figure A1, we highlight the dependence of the ratio on the cloud density as a function of the a range of metal content that is assumed in this work using CLOUDY photoionization code.The value of the Paβ/Paγ  ratio varies between 1.05 -3.38.We assume a fixed column density (=10 24 cm −2 ) with no micro-turbulence for this analysis.We then derive the R 1µm estimates based on the calculated minimum and maximum value of the ratios obtained from these simulations which we detail in the next section.We re-scale the Paγ observed flux to get the flux for Paβ (for the minimum and maximum values of their ratio) and estimate the minimum and maximum values for the intensity of FeII 1-micron lines.Consequently, we obtained a range of values for R 1µm = 0.46±0.47,and 1.48±0.15,respectively.

Photoionization modelling
The range of the physical conditions, used in this work, was based on previous approaches [73,74,77].We employ the default model for the FeII ion in CLOUDY, i.e., the one defined by Verner et al. [38], which includes 371 levels ranging in energy up to 11.6 eV, and 68,635 transitions to keep the consistency with previous studies in low column density regime [77].
We carry out numerical simulations using CLOUDY v17.03 [79].To constrain the number of free parameters to three, we adopt the following fixed values: the luminosity at 5100 Å , L 5100 = 3.19×10 44 erg s −1 from Kaspi et al. [84], the BLR radius determined via reverberation mapping, R BLR = 37.2 light-days, based on Huang et al. [81], and a cloud column density N H = 10 24 cm −2 as concluded from earlier studies involving the recovery of the optical FeII emission in this source [73,85].Also, we adopt the spectral energy distribution for I Zw 1 from Panda et al. [73] as the radiation field in our simulations.Thus, the remaining free parameters are: (i) the cloud mean hydrogen density, n H which ranges between 10 9 -10 14 cm −3 and (ii) the metal content in the cloud in the range of 0.1 ≤ Z ≤ 10 (in solar units), estimated using the GASS10 module [86].Moreover, we incorporate two scenarios for internal random motions within the BLR clouds, specifically micro-turbulence (V turb ) of 10 and 100 km s −1 .The micro-turbulence in the BLR is caused by random motions of photons within the ionized gas, which is likely triggered by magnetic fields in the confined clouds [87,88] and acts as a secondary contributor to the overall line width, as has been shown in previous studies (e.g., Kollatschny and Zetzl [89]).We assume the values for the micro-turbulent velocity based on previous works, which were successful in modeling the FeII emission in the UV region for I Zw 1 about 10-30km s −1 [75].In addition, Panda et al. [76], Panda [85] find that a micro-turbulence velocity between 10-100 km s −1 can reproduce the FeII emission in the optical region.Additionally, as we aim to compare the effect of using micro-turbulence in the gas, we perform simulations in the absence of micro-turbulence in the gas (V turb = 0 km s −1 ).This results in a total of 609 model combinations: V turb (3) × n H (29) × Z (7).

Results
We derive from CLOUDY models the optical FeII and the NIR FeII ratio, denoted as R 4570 and R 1µm , respectively.The R 4570 represents the ratio of the flux of optical FeII multiplets 37, 38, within the range 4434 -4684 Å to the flux of Hβ broad component [7,47].The R 1µm is calculated as the ratio of the combined fluxes from the four NIR prominent isolated FeII lines at wavelengths λ9997, λ10502, λ10863, and λ11127 to the flux of Paβ broad component [67].Rodriguez-Ardila et al. [67] and Marinello et al. [69] demonstrate that the optical and NIR intensities are correlated, indicating that both emissions should have origin and excitation mechanisms in common.
In the optical FeII emission, the micro-turbulence effect has been extensively studied [74,75], but not in the NIR.In this work, for the first time, we study the micro-turbulence effect applied to the NIR FeII region.The results are shown in Figure 2 for the column density scenario of 10 24 cm −2 for optical (upper panels) and NIR (lower panels) spectral regions.The diagnostic diagrams in the same figure show the outcomes for V turb values of 0, 10, and 100 km s −1 , for R 4570 and R 1µm intensities as a function of the local cloud density.In each of the panels in Figure 2, we observe how the intensities vary with different metal content levels in the BLR clouds depicted using the color gradient.We further highlight the observed ratios, R 4570 and R 1µm , and their associated uncertainties, obtained by fitting the observed spectra, using the gray and blue shaded regions, respectively.

No micro-turbulence
Our result without micro-turbulence (i.e., at 0 km s −1 ) is shown in the left panels of Figure 2, where we compare the simulated results from CLOUDY with the observed values for optical and NIR, R 4570 = 1.62±0.06and R 1µm = 0.46-1.48,respectively, represented by the gray and blue bands 2 .Here, the widths of the gray bands account for the uncertainties associated with the ratios.The results show that we can reproduce R 4570 with a metal content above 3 Z ⊙ and with a local hydrogen density between 10 10.75 -10 12.50 cm −3 .Moreover, from the NIR results, to reproduce R 1µm , it is necessary to have a local hydrogen densities range from 10 9.00 to 10 11.50 cm −3 , and a metal content above 3 Z ⊙ .These results demonstrate a comparable physical parameter range to reproduce the FeII strengths for I Zw 1 in the optical and NIR regimes within our setup.

Applying micro-turbulence
Previous studies [73,75] demonstrated that the value obtained for R 4570 in high metal content simulations can be achieved by increasing the micro-turbulent velocity.To investigate how changing metallicity and micro-turbulence can increase both optical and NIR emissions simultaneously, we include two cases of micro-turbulent velocity, at 10 and 100 km s −1 , and compare it with the default case (i.e., with no micro-turbulence).We expect that the micro-turbulence in the gas increases the Brownian motion within the cloud and allows for the FeII ions to receive preferentially more ionizing photons, consequently leading to an increase in the intensity of the FeII lines.
The results using micro-turbulence of 10 and 100 km s −1 are shown in Figure 2 in the middle and right panels, respectively.Our results for the optical region reveal that the FeII at 4570Å strength increases with the increase in micro-turbulence, as expected from previous works [73,74].
On the other hand, the results in the NIR show a different behavior compared to the ones in the optical, as presented in Figure 2. Considering a single case with n H =10 12 cm −3 , solar metallicity, and V turb = 10 km s −1 (see Figure 2 lower middle panel), R 1µm is approximately ∼0.2, whereas at 100 km s −1 , it dropped to ∼0.05 (see Figure 2 lower right panel).In contrast, the case with no micro-turbulence produces an R 1µm value of ∼0.1 (see Figure 2 lower left panel).Notably, the absence of micro-turbulence results in lower R 1µm values compared to the scenario with V turb = 10km s −1 , as expected, based on the results obtained in the optical regime.Although, for metallicities above solar, R 1µ m values overall decrease when compared to scenarios without micro-turbulence.The R 1µ m appears to be suppressed and fails to increase as expected from the optical results.For example, when we consider the highest metallicity value of 10Z ⊙ (see Figure 2), and V turb = 10 km s −1 , and 100km s −1 the maximum FeII ratio in the NIR is, respectively, ∼0.8, and ∼0.6, whereas in the zero micro-turbulence case, this ratio approaches ∼1.2.
To comprehend this intriguing behavior from our CLOUDY simulations, Figure 3 provides a comparison of the FeII pseudo-continuum in the spectral region 9500-11500 Å for a fixed local hydrogen density (10 10.5 cm −3 ) for three scenarios: without micro-turbulence, 2 the symmetric errors for 1-micron values are 0.15 and 0.47, respectively, also highlighted in Table 1.V turb = 10 and 100 km s −1 .We note that the pseudo-continuum of V turb = 10 km s −1 exhibits less intense lines relative to the case without micro-turbulence.The case with 100 km s −1 shows even weaker FeII lines.Although, for the two cases with V turb = 10 and 100 km s −1 , we observe a boosting of lines , specifically between 9600-9900 Å, which were not prominent in the spectrum in the zero micro-turbulence case.
It is important to emphasize that the intensity of the NIR FeII is characterized by the ratio of four isolated emissions to Paβ line, whereas the ratio in the optical is composed of blended FeII multiplets (m37, m38) normalized by the Hβ [64,90].Hence, to understand the variations in the R 1µm , we need to study also the behavior of the Paβ line across the parameter space, especially under the influence of microturbulence.We notice that the luminosity of the Paβ emission significantly exceeds that of the NIR FeII lines.In contrast, in the optical range, both the Hβ and the optical FeII lines exhibit a similar increase in luminosity with the density, as shown in Figures 4 and 5.
Moreover, the 1-micron lines exhibit two significant excitation mechanisms, collisional and Lyman-α fluorescence, whereas the FeII bump at 9200 Å is exclusively influenced by the Lyman-α fluorescence mechanism [67,69].This distinction is an important indicator of the Lyman-α fluorescence mechanism.In our work, we note that the 9200 Å region remains unaffected by the enhancement of micro-turbulence in the models.

Discussion
The specific influence of micro-turbulence on the NIR emission of FeII is particularly intriguing in comparison to its well-documented impact in the optical region.The optical models suggest that when micro-turbulence is introduced in the gas, it enhances the intensity of the FeII lines.Surprisingly, our simulations indicated a contrary trend in the NIR region, where the R 1µm displays lower values when microturbulence is considered than when micro-turbulence is absent.
The discrepancy between our observations and CLOUDY predictions may raise intriguing questions regarding the true impact of micro-turbulence in the near-infrared region.Initially, it was anticipated that the increased micro-turbulence would result in enhanced photon absorption by FeII ions, leading to an increase in the emission in this spectral region as well, as re-affirmed by the results from the FeII modeling in the optical region.This leads us to question whether micro-turbulence potentially influences the Lyman-α absorption by the FeII ion, thereby affecting NIR FeII emission [66][67][68][69].A comprehensive investigation is required, which is beyond the scope of our current work.
We attribute the unexpected effects observed in the 1-micron lines to a physical limitation, where we assume a fixed value for the column density as a stopping criterion for the simulations.If we allow this parameter to be free, this can lead to a complex interplay between the column density and the metal content in the ionized gas, as found in Panda [74] in the case of the optical FeII emission.If this parameter is ≳ 10 24.5 cm −2 , one would have a cloud where the optical depth is ≳ 1 where scattering effect becomes significant [see 74, for more details].The limit at 10 24 cm −2 serves to constrain the cloud's physical condition, drawing inspiration from prior works [74,75,91] wherein the radiative calculations are made under the optically thin regime.Thus, the non-monotonic behavior in the NIR may be ascribed to the collective influence of the microturbulence and cloud column density.In light of these intricate findings, further dedicated investigations are warranted to unravel the exact mechanisms at play, that includes also the advances in the available FeII atomic datasets.
Coming back to the issue of the FeII flux ratio in the NIR obtained from the observed spectrum and under the physical conditions proposed in our study, our model is not able to replicate the ratio from Marinello et al. [69], which is R 1µm ∼1.81.To reproduce this value, a metal content exceeding 10Z ⊙ would be required.This would mean that the optical and NIR FeII emissions require quite different metal compositions, even though they are formed in close vicinity to each other.This latter aspect is confirmed from the correlation observed for the FWHMs for FeII in the optical and NIR [69].Nevertheless, exploring scenarios with significantly elevated metal content may be a possibility in future research of the NIR FeII emission, especially for high-accreting sources.
Furthermore, it is worthy to emphasize the importance of investigating how the hydrogen line ratios within the BLR of an AGN vary under different physical conditions.This inquiry will significantly improve the precision of measuring the FeII ratio in the NIR for quasars with a non-reliable Paβ line (see Marinello et al. [39,69] for other examples).A future study should be considered, as it has the potential to further refine our understanding of AGN environments and improve the accuracy of critical astrophysical measurements.

Conclusions
In this study, we aimed at understanding the line formation and physical conditions in a strong FeII emitter AGN, I Zw 1.We investigated how changes in micro-turbulence and metal content impact the optical and NIR emissions of FeII simultaneously.By introducing micro-turbulent velocities within the line-emitting cloud, of 10 and 100 km s −1 , employing a range of cloud densities (10 9 -10 14 cm −3 ), and metal content from sub-solar to 10 times solar inspired by previous works [13,73], we explored their combined effect on the recovery of the FeII intensities.Notably, this research is the first to examine the impact of micro-turbulence in the NIR regime, especially for the FeII emission, using photoionization model predictions and corroborating them with observed flux ratios obtained from archival spectra in optical and NIR for this source.While our findings align with prior findings made in the optical regime, we observed contrasting behavior in the NIR when micro-turbulence is introduced.Nonetheless, there is an overall agreement within the parameter ranges studied in the two wavelength regimes.A set of physical conditions can simultaneously reproduce optical and NIR FeII intensities, with n H = 10 10.75 -10 11.50 cm −3 , and a metal content ranging from 5 Z ⊙ ≲ Z ≲ 10 Z ⊙ .
We reproduced the R 1µm value for I Zw 1 simultaneously in the optical and NIR.This result is significant not only for this source but is applicable also for I Zw 1-like AGNs.The ability to replicate R 1µm values across these wavelengths is a novel accomplishment.In conclusion, a comprehensive future analysis of excitation mechanisms and FeII models holds significant promise for further insights into this complex study.

Figure 1 .
Figure 1.Three panels depicting spectra of galaxy I Zw 1.The (a) panel shows the optical spectrum (the observed spectrum is shown in black) with Hβ emission line fits: green represents the outflow component, yellow is the BLR component, and orange-dashed indicates a faint narrow component.The blue curve represents the optical FeII template.The (b) displays the near-infrared spectrum in black with the fit of the semi-empirical Garcia-Rissmann et al.[68] template in red, highlighting four main NIR FeII lines used to estimate the 1-micron intensity.In addition, (c) panel shows Paγ-HeI line blend with HeI, where the solid line in orange represents the BLR components, while the dashed lines imply the NLR components.

Figure 2 . 1 .
Figure 2. Diagnostic diagrams from simulations depicting R 4570 intensities vs. n H for a column density of 10 24 cm −2 .The upper panels (a, b, and c) illustrate different

Figure 4 .
Figure 4. Micro-turbulence case of 10 km s −1 .Top panel: The y-axis represented the luminosity of the 4570 Å bump vs. local hydrogen density, and the luminosity of Hβ line vs. local hydrogen density.The colors corresponded to the metal content cases.Down panel: The colors and the x-axis represent the same meaning.The values on the y-axis differ from NIR quantities, respectively, the sum of the luminosity of the 4 FeII 1-micron lines, and Paβ luminosity.

Figure 5 .
Figure 5. Same from the previous figure caption, but for micro-turbulence case of 100 km s −1 .

Table 1 .
Values measured from the near-infrared and optical spectra of I Zw 1, where the fluxes are in units of 10 −14 erg s −1 cm −2 Å −1 .The errors for 1-micron values are respectively 0.15 and 0.47, derived from the values of 1.48 and 0.46.The flux errors are also reported side-by-side with their respective measured values.