Filename
stringlengths 22
64
| Paragraph
stringlengths 8
5.57k
|
|---|---|
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | The CFB 1M-0.5k mode on the ATCA Compact Array Broadband Backend (CABB) was used to simultaneously ob- serve all four ground state OH lines in zoom bands centred on the line rest frequencies (a single zoom band was used for the main lines, centred at 1666 MHz). This provided a raw channel width of 0.09 km s–1. The 1.5D array resulted in a synthesised beam size of ∼ 30(cid:48)(cid:48) at 1.6 GHz. The total observ- ing time for all 15 sources was 50 hours. |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | contained more radio frequency interference (RFI) than the satellite-line observations. Flagging this RFI resulted in sys- tematically larger synthesised beams for the main-line obser- vations, and hence lower continuum brightness temperatures in the main lines (see Table 2). This would not affect the peak optical depths measured in our analysis as they are derived from a ratio of Tb and Tc which are equally affected by this increase in synthesised beam. The visibilities were inverted using a Brigg’s visibility weighting robustness parameter of 1 (Briggs 1995), corresponding to roughly natural weighting. The velocity spectrum at the location of the brightest contin- uum pixel was selected for further analysis. A linear baseline was fit to these velocity spectra to determine the background continuum brightness temperature Tc, which was subtracted to produce line brightness temperature (Tb) spectra. These were then converted to optical depth (τν) spectra, assuming –τν – 1). The rms noise levels of the optical that Tb = Tc(e depth spectra ranged from 0.006 to 0.023, and are outlined in Table 2. |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | Method practicalities and limitations In this section we discuss practical details and limitations of the methods used in this work. We will also discuss the process of Gaussian decomposition used to obtain our results. This will include details of our use of AMOEBA, an automated Bayesian Gaussian decomposition algorithm developed primarily for this dataset. AMOEBA is described extensively in Petzler, Daw- son, and Wardle (2021), and this section will provide addi- tional details on its use in this work. |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | Ta = Tb(cid:15)eff , where (cid:15)eff is an effective beam efficiency parameter. This parameter accounts for the efficiency of the main beam and the sidelobes as they overlap with the background continuum source. Previous surveys of HI (GALFA-HI Peek et al. 2011) apply a single value of (cid:15)eff , found by averaging the convolu- tion of the beam efficiency with continuum source size over the whole survey. The Millennium survey used a similar ap- proach, adopting an effective beam efficiency of 0.9. Though the OH observed in our data from Arecibo is likely to be less smoothly distributed than the HI of the Millennium survey, in the absence of exact information about that distribution we adopt the same effective beam efficiency of 0.9. This may lead to an underestimation of the brightness temperatures Tb and hence our derived excitation temperatures Tex, likely by no more than 10%. Our derived optical depths would be unaf- fected. |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | Table 1. Summary of sightlines observed by the Arecibo telescope included in this work. aSource names are given along with the original Arecibo bproject designation and the galactic longitude and latitude. cSources with detections are indicated ‘Y’ and those without are indicated ‘N’. Source names indicated with asterisks were excluded from analysis due to contamination of off-source pointings as described in the text. The brightness temperature of the back- ground continuum Tbg at each of the four OH ground-rotational state transitions are given along with the rms noise of the optical depth τσ and expected brightness temperature spectra Texp ff . |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | Table 2. Detailed information for the continuum sources coinciding with the sightlines observed by the ATCA examined in this work and their optical depth sensitivities. ∗Central frequency of zoom band (MHz). The systematically lower brightness temperatures in the central band are a result of the slightly larger synthesized beam at this frequency (see text). Notes: 1. HII region near-side, radio recombination line in brackets, 2. HII region far-side, 3. Extragalactic, 4. Nearby HII region. References: aCaswell and Haynes (1987), bLockman (1989), cPetrov et al. (2006), dCondon et al. (1998), eGray (1994), f Wink, Altenhoff, and Mezger (1982), gHelfand and Chanan (1989), hGriffith and Wright (1993). Sources with detections are indicated ‘Y’ and those without are indicated ‘N’. |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | Our method of generating the OH expected brightness temperature spectra differed slightly from the method used for HI observations described by Heiles and Troland (2003a), in that we did not interpolate between the off-source point- ings to determine Texp, but rather simply averaged the off- source brightness temperature spectra. This choice was made because (for a majority of sightlines) there were not signif- icant differences between the features seen in the individual off-source brightness temperature spectra. |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | As noted in the Observations section the on-off method assumes that the OH optical depths and excitation tempera- tures, and the diffuse background continuum brightness tem- perature are the same in both the on-source and all the off- source positions. If one or more of these assumptions is in- correct – i.e. if the OH gas varies in optical depth or excita- tion temperature across the on- and off-source pointings or if there is additional continuum behind any of the off-source positions – then the averaged off-source spectra will not be a good estimation of the expected brightness temperature spec- trum of the on-source pointing. For the majority of sources presented in this work (for which the individual off-source pointings were available), there was little noticeable differ- ence between the individual off-source spectra surrounding each on-source pointing before the background continuum Tbg had been subtracted. Any variation in the OH gas or continuum between the off-source pointings in these cases is therefore likely to be small. This is in contrast to the findings of Liszt and Lucas (1996) who note inconsistencies between the absorption (‘on-source’) and emission (‘off-source’) spectra of OH. |
Processed_GNOMES_II:_Analysis_of_the_Galactic_diffuse_molecu.txt | More generally, our assumption that the foreground OH gas is uniform across the on- and off-source pointings (for both our on-off and our ATCA observations) is also limited by the fact that molecular gas is clumpy on sub-parsec scales (below the resolution of our observations). Engelke and Allen (2019) addressed this issue, as well as the presence of unre- solved structure in the bright background continuum source. This is a difficult problem to solve directly without higher resolution observations, but the overall consequence appears to be that our measurements of optical depth may represent lower limits rather than their true values. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | Abstract—Integrating on-chain and off-chain data storage for decentralised and distributed information systems, such as blockchain, presents specific challenges for providing transparency of data governance and ensuring data integrity through stakeholder engagement. Current research on blockchain-based supply chains focuses on using on-chain governance rules developed for cryptocurrency blockchains to store some critical data points without designing tailored on-chain governance mechanisms and disclosing off-chain decision-making processes on data governance. In response to this research gap, this paper presents an integrated that coordinates supply chain data governance stakeholders with inter-linked on-chain and off-chain governance to disclose on-chain and off-chain rules and decision-making processes for supply chain transparency and integrity. We present a Proof-of- Concept (PoC) of our integrated data governance approach and suggest future research to strengthen scaling up and supply chain- based use cases based on our learnings. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | developed for cryptocurrency blockchains to store critical data points [6]. The application of blockchain to a multi- stakeholder supply chain can generate a large amount of data and accordingly increase the cost of data storage [7]. This creates demand for more cost-effective off-chain data storage. However, a proper disclosure of off-chain decision- making processes is often still lacking, which can jeopardise data transparency, compliance and integrity expectations. This paper reports ongoing research, which seeks to address integrated data this research gap by developing an governance framework that coordinates stakeholders with inter-linked on-chain and off-chain supply chain data. It enables users to disclose on-chain and off-chain rules and decision-making processes to the wider community via the Smart Trade Network (STN), which is a proof of authority Ethereum Virtual Machine compatible blockchain network. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | Blockchain is a decentralised and distributed ledger, which highlights the necessity of achieving consensus from various stakeholders on how to record and store transaction information on the shared ledger [1], [2]. This requires a proper governance mechanism to coordinate stakeholders with data injection, storage, sharing and access. Blockchain- based systems are generally governed by a set of rules to achieve a certain fairness, transparency, assurance, and stakeholder engagement. While there is a lack of commonly accepted definition associated with blockchain governance [3], De Filippi and Mcmullen [4] defined blockchain governance as either the governance of a blockchain system by infrastructure or the governance of a blockchain infrastructure system. They describe blockchain governance as either ‘on-chain’ or ‘off-chain’ patterns [4]. On-chain governance enables decision-making processes to be finalised within the blockchain system, which they consider to be formal, fair, and transparent, whereas off-chain governance finalises decision-making away from the blockchain system and this is considered informal and hidden most of the time [3]. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | Following this introduction section, we review related work on blockchain governance, and data transparency and integrity (II). This is followed by the description of our work- in-progress of an integrated data governance framework (III) before presenting the pilot use case implementation and evaluation results. We conclude with our learnings from our pilot use case and suggest future research areas. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | and Mcmullen [8], on-chain governance mechanisms have formal and rigid coding structures that restrict their flexibility to enable the system to react to unforeseen circumstances quickly and smoothly. Differently, off-chain governance has relatively informal and unstructured formats [8] and can complement the weakness of on-chain governance. Bitcoin that use a and Ethereum are combination of offline coordination and online code modifications to implement update changes. However, they are criticised for allowing miners and developers to play the role in coordinating and achieving consensus between stakeholders in off-chain governance [14]. Therefore, an integrated on-chain and off-chain governance mechanism needs to be developed with stakeholders in ways that not only maintain the transparency and efficiency of on-chain governance rules, but also have the capability to balance the different needs and interests of stakeholders. Our study aims to develop and test an integrated blockchain governance model to coordinate on-chain and off-chain mechanisms for blockchain-based supply chains. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | Transparency and integrity are two key concepts related to trust [15], [16] in multi-tier supply chains that involve interdependent actors from production to consumption. Supply chain transparency refers to the disclosure of information to supply chain stakeholders about an agent’s operations and how products are distributed to consumers [17]. Differently, supply chain integrity refers to adherence to good practice of being honest in supply chain activities and showing a consistent dedication to maintaining integrity in supply chain processes and flows [18]. Though they are different concepts, supply chain transparency is regarded as a prerequisite to supply chain integrity. This is supported by Stohl et al. [19] who argued that the transparency of decision processes and behaviours to others could force actors and companies to behave with integrity because of accountability. A transparent supply chain requires information disclosure to supply chain stakeholders and the collection and sharing of relevant information. At root, it is about reducing the extent of information asymmetry [2], [13]. Technology plays a role in addressing these issues. However, companies are worried about their private and confidential information [17]. Blockchain technology has its transparent, immutable, secure, and decentralised attributes and has the potential to address these issues. Recent research has explored the role of blockchain technology on supply chain transparency. Sunny et al. [20] found that better supply chain transparency can be achieved through blockchain-based traceability. Rao et al. [21] reported that blockchain plays an important role in improving supply chain transparency from two serialised data projects. Blockchain can also improve the integrity of data with immutable ledgers that effectively prohibits the alteration or deletion of recorded information. integrity and This allows data users authenticity. However, blockchain does not safeguard against intentional efforts to upload false and corrupted data, nor does it guarantee against unintentional error [13], [20]. Therefore, there is a need for transparency in decision-making processes on data entry and validation to ensure supply chain integrity, which is the focus of this paper. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | This paper proposes an integrated approach that enables the interaction between on-chain and off-governance, which leads to an integrated data governance framework as shown in Figure 1. The framework not only provides a way to ensure the transparency and efficiency of off-chain governance rules but also supports the achievement of flexibility in on-chain and off-chain governance. Both on-chain and off-chain governance are achieved through multi-sig smart contracts deployed on the STN network that can interact with the Ethereum Mainnet and Polygon/Matic network. The Polygon/Matic network is a scaling blockchain framework for developing Ethereum-compatible blockchain networks [22]. The off-chain governance leverages the smart contract created on the Polygon/Matic network to ensure the transparency of consensus-based decision-making in off- chain governance so as to improve supply chain transparency and integrity. Although these smart contracts could also be deployed on the STN network itself, we chose to deploy these off-chain governance support functions on the Matic network to explore its bridging functions to the Ethereum mainnet, enabling users to move tokens from one to the other, as well as the impact of variable transaction costs when the multisig governance contracts are deployed and executed. The STN network, by way of contract, does not have ‘variable’ fees per se, which makes it more suitable for stable cost-based business case development for industry application. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | The on-chain governance approach deployed on the STN network uses the primary multisig protocol built on top of the Ethereum Virtual Machine compatible blockchain. Multisig wallets are implemented as a smart contract to validate the data state updates and storage when data points are proposed on the network for validation following the pre-defined multisig protocol. The on-chain consensus protocol implemented by the multisig smart contract defines the rules, including the signatory group, the number of signatories and the process of validation, for writing data to the blockchain and validating data in the context of a multi-stakeholder supply chain [2]. The on-chain consensus protocol also defines the rights and obligations of each group member in creating and joining a signatory group to propose and accept/reject changes to the on-chain consensus protocol. The on-chain consensus protocol is being resolved using multisig “voting” on the Polygon/Matic network with a dedicated multisig mechanism. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | The off-chain governance is also deployed on the STN network but using secondary multi-sig protocols on the Polygon/Matic network, which offers a higher level of transparency of off-chain decision-making processes compared with traditional informal and unstructured off- chain governance [8]. Our proposed off-chain governance approach is also embedded with the multisig smart contract function to ensure transparent consensus vis-a-vis transaction information. The multisig smart contract built on the Polygon/Matic network not only identifies the users who are involved in finalising transaction rules and policies but can for example also record the process of how users agree on which data need to be validated and stored on the STN blockchain. This improves the transparency of off-chain off-chain decision-making governance approach interacts with on-chain governance that equips the on-chain governance with flexible capability to be responsive to unforeseen circumstances. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | This section reports the design architecture of the Proof- of-Concept (PoC) and its implementation in an Australian beef supply chain. The on-chain governance mechanism is designed with the multisig smart contract user interface on the STN Blockchain that enables supply chain actors to set on-chain transaction rules. The off-chain governance mechanism is built with the multisig smart contract user interface on the Polygon/Matic network that allows supply chain members to finalise off-chain rules. While on-chain and off-chain PoCs are built on different blockchain networks, they are linked by MetaMask and designed with compatibility on the STN network. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | The proof of concept for on-chain governance was built to track data about cattle which were registered on farms and whose subsequent changes of state were measured with transactions confirmed via a multisig smart contract protocol to ensure data integrity at the entry point. Fig. 2 shows the developed user interface for implementing on-chain multisig protocol. The smart contract was built as multisig wallets, which allows users to propose and agree to data state updates. Users can use the interface to set contracts, add and remove signature group members as well as change signing requirements. This user interface also allows users to check approved and pending transactions. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | Fig. 3 gives a view of the data and cattle assets that are registered and measured by supply chain members within the system, which can offer a better understanding of how data is captured and presented in this PoC use case after applying the multisig smart contract protocol. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | The proof of concept for off-chain governance was built to track decision-making processes in finalising rules and policies that are used to establish the protocols to govern on- chain transactions and off-chain storage. A simplified view of the designed user interface for implementing off-chain multisig protocol is shown in Fig. 3. The smart contract was built with STNBI Wallets on the Polygon/Matic network, which allows supply chain members to propose and agree on transaction governance rules. Supply chain members can use the interface to set multisig names, add/remove signers, change multisig requirements, transfer and mint tokens. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | The application of our proposed off-chain governance on the Polygon /Matic network is demonstrated with an illustrative use case. Fig. 5 shows the history of decision- making processes in finalising governance rules and policies within the system. Guided by the multisig smart contract protocol, the decision-making process of off-chain governance rules and policies is recorded and disclosed with critical actions, including participants, confirmation, and description of activities. It offers a higher level of transparency of off-chain decision- making processes in this PoC use case. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | This paper presents an integrated data governance framework that leverages multisig smart contracts in both on- chain and off-chain governance mechanisms to improve supply chain transparency and integrity from the data point of view. This paper furthers the conversation about the integration of on-chain and off-chain blockchain governance, especially the disclosure of off-chain decision-making processes on data governance, to strengthen supply chain transparency and integrity. We explore the applicability of for our practical implementation with the proof-of-concept design and implementation. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | While our study is still work-in-progress, it makes both technical theoretical contributions. The technical and contributions include the development of multisig smart contracts and the development of a compatible network to accommodate two different blockchain networks for an integrated on-chain and off-chain governance approach. This paper also advances blockchain governance for supply chain transparency and integrity management from a theoretical point of view. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | Our current design and proof-of-work bridges two different blockchain networks for on-chain and off-chain integration using MetaMask, which is one of the areas that need further investigation. Our further research will focus on cross-chain interoperability and design and evaluate cross- that can enable more effective chain architecture communications between independent blockchain networks and consider the potential trade-off between practical implementation with supply chain stakeholders will be included in the next stage of the project. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | The authors thank our study participants, project partners, and additional team members working on the larger Smart Trade Networks program of research, including Assoc. Prof. Felicity Deane and Santiago Del Valle. This project was supported by funding from the Future Food Systems CRC Ltd, funded under the Commonwealth Government CRC industry-led Program. The CRC Program collaborations between the community. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | to agree on common knowledge in blockchain-enabled supply chains,” Computer Networks, vol. 200, p. 108536, Dec. 2021. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | systematic review of challenges and opportunities,” Renewable Sustainable Energy Rev., vol. 100, pp. 143–174, Feb. 2019. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | Out: The Precarious Link between IoT and Blockchain in Food Supply Chains,” Journal of Industrial Information Integration, p. 100261, Aug. 2021. |
Processed_Integrating_On-chain_and_Off-chain_Governance_for_.txt | transparency through blockchain-based traceability: An overview with demonstration,” Comput. Ind. Eng., vol. 150, p. 106895, Dec. 2020. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | We present the first results from the HST Archival Legacy project “SKYSURF.” As described in Windhorst et al. (2022), SKYSURF utilizes the large HST archive to study the diffuse UV, optical, and near-IR backgrounds and foregrounds in detail. Here we utilize SKYSURF’s first sky-surface bright- ness measurements to constrain the level of near-IR diffuse Extragalactic Background Light (EBL) in three near-IR filters (F125W, F140W, and F160W). This is done by comparing our preliminary sky measurements of > 30, 000 images to Zodiacal light models, carefully selecting the darkest images to avoid contamination from stray light. Our sky-surface brightness measurements have been verified to an accuracy of better than 1%, which when combined with systematic errors associated with HST, results in sky brightness uncertainties of ∼ 2 − 4% (cid:39) 0.005 MJy/sr in each image. When compared to the Kelsall et al. (1998) Zodiacal model, an isotropic diffuse background of ∼ 30 nW m−2 sr−1 remains, whereas using the Wright (1998) Zodiacal model results in no discernible diffuse background. Based primarily on uncertainties in the foreground model subtraction, we present limits on the amount of diffuse EBL of 29 nW m−2 sr−1 , 40 nW m−2 sr−1 , and 29 nW m−2 sr−1 , for F125W, F140W, and F160W respectively. While this light is generally isotropic, our modeling at this point does not distinguish between a cosmological origin or a Solar System origin (such as a dim, diffuse, spherical cloud of cometary dust). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Because of the difficulty of characterizing the foreground signal of Earth’s atmosphere, observational attempts at constraining the EBL level directly are primarily done with space missions, such as COBE (e.g., Puget et al. 1996; Fixsen et al. 1998; Dwek & Arendt 1998; Hauser et al. 1998; Finkbeiner et al. 2000; Cambr´esy et al. 2001; Sano et al. 2020), Spitzer (Dole et al. 2006), HST (Bernstein et al. 2002; Bernstein 2007), IRTS (Matsumoto et al. 2005, 2011), and AKARI (Matsuura et al. 2011; Tsumura et al. 2013). These observations have large errors and are often discrepant with each other because of the limited number of observations and the difficulty of subtracting the instrumental, Zodiacal, Galactic, and astrophysical foregrounds (Cooray 2016). Regardless, these direct measurements consistently arrive at EBL levels of ∼ 20 − 50 nW/m2/sr, significantly above the predictions from galaxy counts of ∼ 10 nW/m2/sr (e.g. Driver et al. 2011; Andrews et al. 2018). Recent advances have been made with the CIBER experiment (Matsuura et al. 2017; Korngut et al. 2022), and Pioneer and New Horizons missions (Matsumoto et al. 2018; Lauer et al. 2021, 2022) that aim to better subtract the Zodiacal foreground, using Ca absorption features and by leaving the solar system, respectively. These observations find EBL levels closer to expectations, but they still identify a significant diffuse signal and represent a relatively small number of measurements. A parallel indirect approach, using observations of attenuated γ-rays, also finds values in line with predictions from galaxy counts (H. E. S. S. Collaboration et al. 2013). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | While the presence of diffuse EBL may diminish as new measurements better constrain foreground levels, many astrophysical sources have been hypothesized as contributing to it. The large population of recently-identified Ultra- Diffuse Galaxies in clusters (Impey et al. 1988; van Dokkum et al. 2015) and the field (Dalcanton et al. 1997; Leisman et al. 2017) represents one possible source of diffuse light, although many more unidentified UDGs would have to be present to contribute significantly to the EBL (Jones et al. 2018). Diffuse light in the outskirts of galaxy halos (IGL) may contribute as well (Conselice et al. 2016), although a number of studies (e.g. Ashcraft et al. 2018; Borlaff et al. 2019; Cheng et al. 2021) find that halo light, or light in galaxy outskirts, only represents 15% of the luminosity of bright galaxies. Alternatively, significant levels of difficult-to-detect diffuse intracluster (Bernstein et al. 1995) or intragroup light (Mihos et al. 2005) may contribute to the diffuse EBL. More exotic explanations, such as light from reionization (Santos et al. 2002; Cooray et al. 2004; Kashlinsky et al. 2004) have been put forward as well. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The SKYSURF project, introduced in Windhorst et al. (2022), aims to better understand the EBL level with the large volume of archival HST observations using a two-pronged approach. First, it will use HST’s remarkable stability and precision as an absolute photometer to conduct precise sky brightness measurements for over 200, 000 HST images. Second, it will use the depth and large volume probed by those images to search for possible sources of diffuse EBL. For the full motivation and overview of the SKYSURF project, and an overview of its methods, see Windhorst et al. (2022); we refer to this paper as SKYSURF-1 throughout. In this paper, we describe the first results of SKYSURF surface-brightness measurements at 1.25, 1.4, and 1.6 microns. In Section 1.1 we further outline the diffuse foreground sources necessary to consider for SKYSURF’s EBL constraints, in Section 2, we briefly describe our measurement procedure, Sections 3 presents our results, Section 4 includes a discussion of those results, and Section 5 summarizes our conclusions. Throughout we use Planck cosmology (Planck Collaboration et al. 2016): H0 = 66.9 km s−1 Mpc−1 , matter density parameter Ωm=0.32 and vacuum energy density ΩΛ=0.68. When quoting magnitudes, our fluxes are all in AB-magnitudes (hereafter AB-mag), and our SB-values are in AB-mag arcsec−2 (Oke & Gunn 1983) or MJy/sr, using flux densities Fν = 10−0.40(AB−8.90 mag) in Jy. Further details on the flux density scales used are given in Fig. 10 and the Table footnotes in § 3. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Galactic Light (DGL), discrete stellar and extragalactic light, and diffuse EBL. The Zodiacal Light (ZL) is the main foreground in most HST images, and SKYSURF will measure and model it as well as possible with available tools. All stars in our galaxy (except the Sun) and all other galaxies are beyond the InterPlanetary Dust Cloud (IPD), so the ZL is thus always referred to as a “foreground”. Similarly, the Diffuse Galactic Light, caused by scattered star-light in our Galaxy, can be a background (to nearby stars), or a foreground (to more distant stars and all external galaxies). Most objects in an average moderately deep (AB(cid:46)25–26 mag) HST image are faint galaxies close to the peak in the cosmic star-formation history at z(cid:46)2 (e.g. Madau & Dickinson 2014). Most of the Extragalactic Background Light (EBL) therefore comes from distant galaxies and AGN, and is thus referred to as a “background”. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Before SKYSURF can quantify and model these astronomical foregrounds and backgrounds, it needs to address the main contaminants, which are residual detector systematics, orbital phase-dependent straylight from the Earth, Sun, and/or Moon, and the WFC3/IR Thermal Dark signal. Instrumental and stray light contaminants, as well as the contribution of discrete objects to the SKYSURF EBL constraints, are discussed in SKYSURF-1. Below, we discuss the diffuse Zodiacal, Galactic, and Extragalactic foregrounds in more detail. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | By far, the brightest component of the sky brightness is Zodiacal Light from the InterPlanetary Dust (IPD) cloud, i.e., from distances less than 5 AU, representing over 95% of the photons with 0.6-1.25 µm wavelengths in the HST archive (see Fig. 10). Given its extremely diffuse nature, as well as its time variability, it has been a challenge to understand in detail; observations with all-sky space missions such as COBE/DIRBE are required to fully model it. For example, the Kelsall et al. (1998) and Wright (1998) Zodiacal models use the COBE/DIRBE data to model the Zodiacal emission, considering multiple dust components scattering sunlight toward Earth. The absence of an all-sky optical survey means that such modeling cannot be done in the optical to a similar extent; most authors simply assume that the Zodiacal spectrum is a Solar, or slightly reddened Solar spectrum (e.g., Leinert et al. 1998). Future SKYSURF studies will utilize its UV-to-optical database to improve constraints on the Zodiacal spectrum, but here we only consider observations with wavelengths similar to COBE/DIRBE wavebands for which a detailed Zodiacal model is obtainable. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The darkness of the night sky, “Olbers’ Paradox”, was one of astronomy’s oldest mysteries: an infinite and infinitely old Universe full of stars and galaxies would have a sky as bright as the surface of an average star. The resolution of this “paradox” — an expanding Universe of finite age — is, of course, the central tenet of Big Bang cosmology, where the galaxy surface density is a finite integral over the galaxy luminosity function and the cosmological volume element (Driver et al. 1995; Metcalfe et al. 1995; Odewahn et al. 1996; Tyson 1988). Because of their very steep observed number counts, Kuiper Belt Objects (KBOs) can also appear to violate Olbers’ Paradox, producing an apparently diverging sky integral when the smallest objects are taken into account (e.g., Kenyon & Windhorst 2001). To not exceed the observed the ZL sky-SB, the counts of KBOs at distances (cid:38)40 AU must turn over from the non-converging power-law slope γ(cid:39)0.6 dex/mag observed for R(cid:46)27 mag (Fraser et al. 2014) to a converging slope flatter than γ=0.4 dex/mag at R-band fluxes of AB(cid:38)45–55 mag, in combination with a limited volume over which KBOs occur (Kenyon & Windhorst 2001). Assuming albedos of a few percent (e.g., Kenyon & Luu 1999) and a physical size distribution of N (r) ∝ r−β, such a slope change in the KBO number counts implies that the size-slope of unresolved Solar System debris at ∼40 AU must flatten from β(cid:39)4 for larger objects to β(cid:46)3.25–3.5 for objects with sizes r∼0.05–5 m. A flattening of the size distribution of the planetesimal population with radii r(cid:38)10 km from β(cid:39)4 to β(cid:46)3.5 is consistent with simulations for the debris population with r(cid:46)1 km, which suggest that collisions with r(cid:46)100 m objects tend to produce debris rather than larger planetesimals (Kenyon & Luu 1999; Kenyon & Bromley 2004, 2020). It is also consistent with ground-based observations of KBOs with r(cid:46)50 km (e.g., Fuentes et al. 2009; Shankman et al. 2013), and with New Horizons (NH) crater counts on Pluto and Charon, which suggest a flattening of the KBO count slope for r(cid:46)1 km (e.g., Singer et al. 2019). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | To refine these constraints across the Kuiper Belt, SKYSURF will measure the panchromatic Zodiacal foreground in the ecliptic plane in places where other foregrounds are small. Better SB-limits on the small KBO population may constrain the slope of the KBO counts, and hence the total Kuiper Belt mass at 35–50 AU. Time-tagged monitoring of the sky-SB in the Ecliptic may also yield constraints to the integral of Plutinos in Neptune’s L4 and L5 Lagrange points, which have moved significantly in Ecliptic Longitude (lEcl) during the 32-year HST mission. Kelsall et al. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Diffuse Galactic Light (DGL) in the UV–optical is mainly caused by scattered light or reflection nebulae from early- type (O and B) stars, scattered by dust and gas in the Interstellar Medium (ISM). The DGL is thus a strong function of Galactic coordinates (lII , bII ). SKYSURF’s SB-measurements may thus also constrain the DGL at low Galactic latitudes (bII (cid:46)20–30◦), although these fields are very likely not useful for background galaxy counts. The All-sky Infrared Astronomical Satellite (“IRAS” Soifer et al. 1984; Helou & Walker 1985), COBE/DIRBE (Kelsall et al. 1998; Schlegel et al. 1998), Planck (Planck Collaboration et al. 2016), Wide-field Infrared Survey Explorer (WISE; Wright et al. 2010), and AKARI (Tsumura et al. 2013) maps in the near to far-IR help identify Galactic infrared cirrus and regions of likely enhanced Galactic scattered light. Possible high spatial frequency structures in the DGL appear in deep ground-based images of low Galactic latitude at SB-levels of B(cid:39)26–27 mag/arcsec2, and at much fainter levels sometimes also at high Galactic latitudes (e.g., Szomoru & Guhathakurta 1998; Guhathakurta & Tyson 1989). While not a main goal of SKYSURF, the DGL needs to be estimated and subtracted in order to better estimate the levels of the ZL and the EBL at higher Galactic latitudes, as discussed in § 3. Panchromatic HST constraints on the DGL in the Galactic plane (|bII |(cid:46)20◦) are interesting in their own right and are a byproduct of SKYSURF. We refer to § 3.5 for the DGL levels we subtract from any diffuse light levels implied by the comparison between our HST sky-SB measurements and the ZL models. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | An overview of the SKYSURF database and our sky measurement procedure can be found in SKYSURF-1. Further details on the multiple sky measurements procedures, as well as the full results of the sky surface brightness measure- ments across the entire SKYSURF database will come in O’Brien et. al. (2022, in preparation). For context, we give a brief overview of the database and methods here. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | First, the HST archive was searched for images taken with its wide-band filters, excluding grism images, quad/linear ramp or polarizing filters, subarray images, time-series, moving targets, or spatial scans. This resulted in 249, 861 images that made up the initial database. Further cuts on target selection, HST orbital phase, and exposure time will be conducted to avoid possible contamination and minimize measurement errors. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | To measure the sky background of these images, the SKYSURF team tested multiple sky-measurement algorithms on realistic simulated images to identify the most robust method of estimating the uncontaminated sky background. All algorithms that were tested had an accuracy of better than 0.2% for flat images, and slightly worse for images with gradients (Fig. 8 in SKYSURF-1). At this point, it is worth identifying the general philosophy of the SKYSURF program as to identify the Lowest Estimated Sky (LES) value — defined as the lowest sky-SB in an image — as the fiducial sky measurement. While electronic errors within the cameras can introduce either positive or negative errors in sky estimation, errors deriving from contamination (i.e. stray light from nearby bright sources like the Earth and the Sun or thermal emission from the telescope) are more common and more significant. To make full use of our large dataset, we aimed to develop and use algorithms that are the most robust across our database, which contains a wide variety of images. The full results with the most robust algorithms will be presented in O’Brien et. al. (2022, in preparation); here we present the first results using an initial estimation done by fitting a Gaussian to the sigma-clipped image (described as method 2 in O’Brien et. al. (2022, in preparation) and SKYSURF-1). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | for the F160W filter. The systematic uncertainties come from Bias/Darkframe subtraction ((cid:46) 1.0%), the global flat field correction (0.5 − 2%), zeropoint accuracy (∼ 1.5%), and thermal dark subtraction (∼ 0.2% for F125W, 0.5% for F140W, and ∼ 3.8% for F160W). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | For the final analysis of 249,861 SKYSURF images, we expect (cid:46)50% to be usable for sky-SB measurements. Although these images are not completely randomly distributed on the sky, they on average provide ∼4400 sky-SB measurements in each of the 28 broad-band SKYSURF filters. In this section, we will use two complementary analyses of the HST sky-SB estimates to make our first assessment of available Zodiacal Light models, identify any diffuse light that may be present, and check on the consistency of our methods. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The results from both methods will be compared to the Kelsall et al. (1998) and Wright (1998) models, which predict the ZL brightness as a function of sky position and time of the year. Both Kelsall et al. (1998) and Wright (1998) models are fit to COBE/DIRBE measurements at 1.25 − 2.2 µm. Kelsall et al. (1998) is a physical model that contains multiple dust components, whereas the Wright (1998) model is a more parametric model normalized at 25 µm to ensure 0 residual diffuse light at the ecliptic poles. Because their ZL model predictions are anchored to the COBE/DIRBE 1.25–2.2 µm data, we will limit our analysis in this paper to the SKYSURF WFC3/IR filters F125W, F140W, and F160W. We will deal with the uneven sky-sampling of the HST data by comparing the HST sky-SB data with the corresponding ZL model predictions. Again, our premise throughout is that the lowest estimated sky-SB values measured amongst the HST images in each direction will be the least affected by HST systematics or discrete foreground objects, and therefore be closest to the true sky-SB in that direction. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The first approach uses the Lowest Estimated Sky-SB values from the HST images. Both the HST LES-data and the Kelsall et al. (1998) model predictions are fit with analytic functions as a function of Ecliptic Latitude (bEcl) in the darkest parts of the Galactic sky. These fits will be referred to as the Lowest Fitted Sky-SB (“LFS”) method. To avoid regions with significant DGL, the LFS method will first select the LES-data and model predictions as a function of Galactic Latitude (bII ), to identify the darkest regions of the Galactic sky. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Next, the LFS method will identify the lowest sky-SB as a function of Ecliptic Latitude (bEcl) to constrain the ZL+EBL sky-SB in each direction (see Fig. 2). For |bII |(cid:38)20◦, where the DGL contribution is lower, the LFS fits provide analytical functions describing the lowest sky-SB as a function of Ecliptic Latitude for both the HST data and the model predictions in the same directions of the sky. The limitation of the LFS method is that not all sky- SB measurements are done at constant Sun Angles (SA; defined as the Sun-HST-target angle), which ranges from SA(cid:39)85–180◦ at the Ecliptic to SA=90◦ at the Ecliptic poles. Although many HST observations are scheduled around SA∼ 90◦, many others are done with higher solar elongations for which the Zodiacal sky-SB is lower (the Zodiacal sky-SB reaches a minimum in the Ecliptic at Solar Elongations of 120–150◦ (Leinert et al. 1998)). This method will thus focus on observations with SA∼150◦ in the Ecliptic Plane and SA∼90◦ at the Ecliptic Poles. However, because the analysis is conducted on the Zodiacal models in parallel, this is not expected to bias our results. In particular, this method aligns with the SKYSURF philosophy that most sources of error are positive, and thus the lowest sky values are likely the most accurate. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The second method more closely follows the actual selection of the COBE/DIRBE data, on which both the Kelsall et al. (1998) and Wright (1998) models were based. The COBE/DIRBE data were measured at Sun Angles SA(cid:39)94±30◦ (e.g., Leinert et al. 1998). The HST data are observed over a range of Sun Angles, but a significant fraction is also observed at SA(cid:39)90±10◦, i.e., over a Sun Angle range similar to, but somewhat narrower than that of the COBE/DIRBE data. Hence, our second method will only select the HST LES-data and COBE/DIRBE-based model predictions in the Sun Angle range of SA(cid:39)90±10◦. This “SA90 method” has the advantage of the selected HST data being more directly comparable to the COBE/DIRBE based models, but because of their SA-selection, it may also have somewhat higher levels of (unrecognized) Earthshine. The HST data from the SA90 method may thus be systematically somewhat higher than the minimum Zodiacal sky-SB level that is traced with the LFS-method. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Figure 1. All SKYSURF F125W, F140W, and F160W sky-SB measurements vs. Galactic Latitude for the HST data [left sub-panels] and the Kelsall et al. (1998) COBE/DIRBE models [right sub-panels]. The upper left plot is for F125W, the upper right plot is for F140W, and the lower left plot is for F160W. All bottom sub-panels are enlargements of top sub-panels. Because of the ∼60◦ tilt of the Galactic plane with respect to the Ecliptic, the darkest sky-SB occurs for 20◦(cid:46)|bII |(cid:46)60◦, highlighted by the red columns in all bottom sub-panels. Fields with |bII |(cid:46)20◦ have significant DGL, and are ignored in the final sky-SB analysis. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | as we are simply looking for a possible diffuse excess above the Kelsall et al. (1998) and Wright (1998) models, these approaches work well. Future SKYSURF analysis will investigate stray-light contamination, as well as the structure of offsets between SKYSURF sky values and model predictions, in more detail. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Figure 2. All SKYSURF F125W, F140W, and F160W sky-SB measurements vs. Ecliptic Latitude for HST data (left sub- panels) and the Kelsall COBE/DIRBE models (right sub-panels). The orange and blue sech functions and error wedges outline the darkest ∼1% of the sky-SB measurements (magnify the PDF figure as needed to see this). Bottom sub-panels give enlargements of the top sub-panels. As in Figure 1, the upper left plot is for F125W, the upper right plot is for F140W, and the lower left plot is for F160W. The short-dashed blue line represents the upper envelope to the Kelsall et al. (1998) model predictions, and the long-dashed orange line the correspondingly scaled upper envelope to the HST data that does not suffer excessive DGL or straylight, as described in § 3.1 and Table 1. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | In this section, we present our first SKYSURF results from 34,412 images observed in the WFC3/IR filters F125W, F140W, and F160W. Figs. 1 and 2 show the sky-SB in F125W, F140W, and F160W as a function of Galactic Latitude and Ecliptic Latitude. In these Figures, we simply attempt to find the minimum sky-SB signal in the darkest parts of the sky. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Figure 3. All SKYSURF sky-SB measurements vs. Ecliptic Latitude for HST data (left sub-panels) and the Kelsall COBE/DIRBE models (right sub-panels), but only for the darkest Galactic regions with 20◦(cid:46)|bII |(cid:46)60◦ (see Fig. 1). As in Figure 1, the upper left plot is for F125W, the upper right plot is for F140W, and the lower left plot is for F160W. The orange and blue sech functions and error wedges outline the darkest ∼1% of the sky-SB measurements (magnify the PDF figure as needed to see this). Bottom panels give enlargements of the top panels. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | and the fact that most of the relevant information is at the low-end of the SB-range in all these Figures, the bottom panels in Fig. 1cd provide enlargements of the top panels in Fig. 1ab. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The WFC3/IR ZPs used in the F125W, F140W, and F160W filters are 26.232, 26.450, 25.936 AB-mag, respectively, for an object with 1.000 e−/pixel/sec. Fig. 4 shows the HST WFC3/IR F125W 1 and the COBE/DIRBE J-band total system responses 2 compared to the Solar spectrum in Fν (e.g., Arvesen et al. 1969) 3, which is fairly flat across both these filters. From this, we calculate that for a Solar type spectrum like the ZL that the ∆(HST data–Kelsall COBE/DIRBE model) flux is –0.0061 AB-mag due to the small J-band filter differences. This was calculated three independent ways: using integration in Fλ, pysynphot, and black body interpolation between the two very similar filters, resulting in a scaling factor of HST/Kelsall = 1.00557±0.0008. That is, for an SED with a Zodiacal spectrum, the HST 1.25 µm fluxes will be ∼0.56% brighter than in the COBE/DIRBE J-band filter. Hence, we will multiply the Kelsall et al. (1998) model predictions, which are based on COBE/DIRBE observations, by 1.00557 to bring them onto exactly the same J-band flux scale as the HST WFC3/IR F125W filter for a Solar type spectrum. ZL model predictions for the HST WFC3/IR F140W and F160W filters were derived by interpolation between the Kelsall et al. (1998) COBE/DIRBE J-band and K-band predictions using the slope of the slightly reddened near-IR Zodiacal spectrum of Aldering (2001), with uncertainties that include the errors in the Kelsall et al. (1998) model. While HST and COBE are at different orbits, MSISE-90 Upper Atmospheric models of the Earth 4 list that the mean atmospheric pressure is 2.27×10−7 Pa at 540 km and 1.04×10−8 Pa at 885 km, so it is unlikely that the differences in altitudes between HST and COBE contribute significantly to systematic differences in sky-SB levels between the two missions. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Figure 4. A comparison of the HST F125W and COBE/DIRBE J-band filters, and the to Solar spectrum in Fν , which is fairly flat across these filters. For a Zodiacal spectrum, the expected J-filter flux difference between HST WFC3/IR measurements and the COBE/DIRBE models due to small differences in their respective filter efficiency curves corresponds to –0.0061 AB-mag. We apply this ZP difference to our diffuse light estimates in the Tables of § 3.1. Note that the Geocoronal 1.083 µm He II line is essentially suppressed in the wing of both filters. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Because of the ∼60◦ inclination of the Galactic plane with respect to the Ecliptic, the darkest sky-SB occurs for 20◦(cid:46)|bII |(cid:46)60◦ and not at the Galactic poles. Fields with |bII |(cid:46)20◦ have significant DGL, and are ignored in the analysis of § 3.1–3.4. Fig. 2 shows all HST WFC3/IR F125W, F140W, and F160W sky-SB measurements as in Fig. 1, but now plotted vs. Ecliptic Latitude. The orange and blue sech functions and their error wedges outline the dimmest 1% of the sky-SB measurements as described below. Fig. 3 show the SKYSURF F125W, F140W, and F160W sky-SB values vs. Ecliptic Latitude as in Fig. 2, but only for the darkest Galactic regions with 20◦(cid:46)|bII |(cid:46)60◦ as selected from Fig. 1. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | According to these authors, the sech model provides a better fit to the vertical or z-direction SB-distribution of flattened or ellipsoidal light-distributions seen edge-on than cosine, Gaussian, exponential, single, or squared hyperbolic secant functions. The IPD cloud has a number of modeled components that Kelsall et al. (1998) identify as “Cloud”, “Bands”, and “Ring” around the Sun, within which the Earth orbits. These Zodiacal components have a ratio of their size in the Ecliptic plane to their vertical Ecliptic Height of approximately 4:1, i.e., a rather flattened or “edge-on” distribution as viewed from the Earth. As we will see, sech-functions describe the vertical ZL distribution as a function of Ecliptic Latitude as observed from the Earth remarkably well. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Inspired by the work that resulted in Eq. 1, we will use sech-type functions to describe the LFS as a function of Ecliptic latitude bEcl. While the actual dependence of ZL brightness with bEclmay be more complicated than Eq. 1 in reality (notably having a significant Sun Angle dependence as discussed below), we find that Eq. 1 is a good description of the dimmest 1% of the sky-SB values for both the HST sky-SB measurements and the Kelsall et al. (1998) model predictions. Furthermore, this fitting procedure allows us to focus on the lower envelope of measurements, which we assume are the least affected by straylight. By repeating the same fitting procedure on the lowest 1% of the Kelsall et al. (1998) model predictions, which predict the ZL brightness for the same direction and at the same time of the year as the HST sky-SB measurements, we can search for any systematic offset between HST measurements and the Kelsall et al. (1998) predictions. This offset could be, an additional unrecognized Thermal Dark component (§ 3.3), a dim spherical or mostly spheroidal Zodiacal component not present in the model, or a dim spherical diffuse EBL component, or some combination of these possibilities. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Here, a3 is the plateau value that the sech function attains when bEcl reaches ±∞. Next, a1 is a constant that captures the maximum vertical amplitude that the sech function reaches at bEcl=0◦ above this plateau. Last, a2(cid:39)19.5◦ measures the effective thickness of the Zodiacal disk (or “vertical scale height”) as seen edge-on from HST. Coefficient a4 in Eq. 1 is a constant that converts the SB in MJy/sr to AB mag arcsec−2, and is not used in the linear flux density representation of Eq. 2–3. The best estimate parameters of the sech constants a1, a2, and a3 are given in Table 1 for both the lower envelope to the HST data and the Kelsall models at 1.25–1.6 µm. The upper and lower sech envelope a2 values are best determined from F160W measurements, which have the best statistics, so we adopt the same a2 values and their errors for the F125W and F140W filters in Table 1, which seem to bound the Kelsall et al. (1998) model predictions well for the F125W and F140W measurements. These sech functions are indicated by the bottom orange and blue lines plus their uncertainty wedges in Fig. 2-3, respectively. The main result we are after in Table 1 is the (boldfaced) difference in the bottom envelopes (or a3-values) between the HST data and the ZL models5. Because the best fit a1 and a2 values turn out to be very similar in Table 1 for both the HST data and the ZL models, we adopt the differences in a3-values as a direct measure of the HST-ZL model differences. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The first four lines of Table 1 also list the same a1–a3 parameters (and their estimated uncertainties) for the upper envelope to the Kelsall models in the right-most panels, and for the HST data in the left-most sub-panels of Fig. 2-3 (upper blue and orange dashed lines, respectively). The sech upper envelope to the Kelsall et al. (1998) models was directly estimated from the predictions in Fig. 2-3, which show a very good empirical sech-type fit to the upper envelope of the Kelsall et al. (1998) model values. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | between the value at 90◦ and ∞ is < 2% for our fits, and this detail does not affect our fitting procedure regardless. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The amplitude of the upper envelope to the HST data was scaled upward using the (HST–Kelsall) difference from the lower envelopes in Fig. 2 and Table 1. The orange dashed lines indicating the upper envelopes to the HST data in Fig. 2a thus provide another way to identify HST exposures with excessive sky-SB, which could be due to several reasons: (a) targets with higher DGL; (b) large nearby galaxy targets, such as the LMC or M31; or (c) exposures with higher straylight levels, including those that got too close to the Earth’s limb. The presence of such images is most noticeable in the F160W filter. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Fig. 5 shows a comparison of SKYSURF’s F125W, F140W, and F160W sky-SB measurements from the HST data to the Kelsall COBE/DIRBE models as a function of Ecliptic Latitude (the top sub-panels show all data, and bottom sub-panels only for the darkest Galactic regions at 20◦(cid:46)|bII |(cid:46)60◦ as selected from Fig. 1). The left sub-panels give the HST/Kelsall model flux density ratio, while the right sub-panels give the linear flux density difference between the HST data and the Kelsall COBE/DIRBE models for the same subsample. In the top sub-panels of Fig. 5, the orange sech functions in Eq. 2–3, and their error wedges outline the darkest sky-SB measurements from Fig. 2. The bottom sub-panels of Fig. 5 give enlargements of the top sub-panels, and show a significant Ecliptic Latitude dependence of the HST/Kelsall model flux ratios, suggesting that the difference between the bottom envelopes of the HST data and the Kelsall models are not due to a flux density scale issue. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The green wedges in the bottom right panels of Fig. 5 indicate our best estimate of the ∆(HST–Kelsall) offsets. For each filter, these linear flux density differences between the bottom envelopes of the HST data and the Kelsall models are fairly constant for |bII |(cid:38)20◦ and well above zero, suggesting a somewhat wavelength-dependent constant linear offset between the bottom envelopes of the HST data and the Kelsall models. For |bII |(cid:46)20◦, the differences between the data and model have more scatter, suggesting that complex and subtle adjustments to the Kelsall model in the ecliptic plane may be required. We thus discard all data with |bII |(cid:46)20◦ to estimate the LFS difference between the HST data and Kelsall models. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The LFS values from Fig. 1 are summarized in Table 1. For example, Table 1 shows that the plateau value a3 of the sech function in Eq. 2–3 that best captures the LFS values at high Ecliptic Latitudes in the F125W filter amounts to a3(HST) = +0.108±0.005 MJy/sr, which best fits the lowest ∼1% of the sky-SB values, while for COBE/DIRBE model predictions for the same sky pointings and filters, observing day of the year, and Sun Angles, the Kelsall et al. (1998) model predicts a lowest ∼1% envelope with sech parameter a3(COBE) = +0.093±0.006 MJy/sr. The most likely HST–Kelsall difference from Fig. 5d is thus ∼(0.108–0.093*1.0056)(cid:39)+0.0145±0.008 MJy/sr, which includes the correction for the –0.0061 mag ZP difference between the HST F125W and COBE/DIRBE J-band flux scales. Similar but somewhat larger values are listed in Table 1 for the F140W and F160W filters, where the Kelsall et al. (1998) models were interpolated between the COBE/DIRBE predictions at 1.25 and 2.2 µm following the discussion in § 3.2. This interpolation also results in somewhat larger a3 errors for the lower envelope to the Kelsall et al. (1998) model predictions in the F140W and F160W filters in Table 1 (see §3.2), and in somewhat larger errors of ∼0.009 MJy/sr in the F140W and F160W HST–Kelsall difference signal listed in Table 1. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | following Aldering (2001), who adopted a power-law slope α=0.730 for wavelengths 0.61(cid:46)λ(cid:46)2.20 µm. Hence, in our analysis, we will use Eq. 4 to represent the Zodiacal spectrum for 0.61(cid:46)λ(cid:46)2.20 µm. Fig. 6a shows the spectral index distribution N(α) when interpolating the Kelsall et al. (1998) Zodiacal sky-SB prediction in the COBE/DIRBE J and K-band filters for all HST pointings in the F160W filter (which is very similar to the distribution of slopes for all HST pointings in the F140W filter). The resulting median spectral index and its 1σ range is α=0.713±0.023, consistent with the value adopted by Aldering (2001)’s power-law approximation of Eq. 4 to within the error. We verified through numerical integration that the power-law interpolation in Eq. 4 produces a (cid:46)2% error in the prediction of the reddened Zodiacal spectrum at 1.4–1.6 µm wavelengths, compared to the Kelsall et al. (1998) model that was fit to the COBE/DIRBE 1.25 and 2.2 µm data and interpolated to 1.4–1.6 µm. This is folded into the error budget of Table 1, resulting in somewhat larger a3 errors for the lower envelope to the Kelsall et al. (1998) model predictions in the F140W and F160W filters. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Figure 5. SKYSURF F125W, F140W, and F160W sky-SB measurements from the HST data compared to the Kelsall COBE/DIRBE models [in Myr/sr] vs. Ecliptic Latitude bEcl. Data in the bottom sub-panels is restricted to the darkest Galactic regions with 20◦(cid:46)|bII |(cid:46)60◦ (see Fig. 1) and zoomed into the most relevant range. Left sub-panels give the HST F125W/Kelsall 1.25 µm model flux density ratio, while the right sub-panels give the linear flux difference between the HST data and the Kelsall COBE/DIRBE models for the same subsample. The orange sech functions and error wedges outline the darkest sky-SB measurements from Fig. 2. The green wedges in the bottom sub-panels illustrate the HST-Kelsall difference between sech fits from Figure 3, highlighting our best estimate of the ∆(HST data – Kelsall COBE/DIRBE model) of 0.0145 ±0.007 MJy/sr for F125W, 0.025 ±0.010 MJy/sr for F140W, and 0.048 ±0.010 for F160W (Table 1). The bottom right sub-panels best capture the linear SB-difference between the HST sky-SB observations and the COBE/DIRBE Kelsall 1.25 µm model (magnify the PDF figure as needed to see the difference). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Possibly the most significant source of uncertainty regarding our measurement of the near-IR diffuse light is the level of WFC3 Thermal Dark signal. Based on onboard temperature measurements and emissivity calculations, the WFC3 IHB lists the IR Thermal Dark signal levels as 0.052 e−/pixel/sec, 0.070 e−/pixel/sec, and 0.134 e−/pixel/sec for the F125W, F140W, and F160W filters, respectively (Dressel 2021). However, modest changes in HST component temperatures (±2.5 K) can impact the TD signal at a level comparable to the diffuse signal. For example, Fig. 6b shows how much changing the overall telescope temperature can affect the TD signal. A sequel paper (Carleton et al.2022b, in preparation), will explore the TD signal as a function of orbital phase and HST component temperatures in more detail. Here, we show a preliminary analysis constraining the TD signal in SKYSURF data by fitting the spectral energy distribution (SED) of the near-IR sky with a Zodiacal component and a temperature-dependent thermal signal. We queried the HST archive for IR images that were taken of the same target within two days of each other, such that the overall Zodiacal sky-SB level does not change substantially. We further identified image sets where at least one image was in the WFC3/IR F125W filter, and another in either the F098M, F105W, F110W, F125W, F127M, F139M, F140W, F159M, and/or the F160W filter. We then ran the adjusted calibration program for the individual WFC3/IR ramps, as described in SKYSURF-1, and measured the minimum sky-SB levels in these images. Based on the orbital phase-dependent straylight constraints in Fig. 10 of SKYSURF-1, we only selected those WFC3/IR exposures in the above filters that have minimal stray light in order to better estimate the most likely TD levels. This resulted in a sample of over 500 useful images in these filter pairs, predominantly from the BORG pure-parallel program PID 12572 (PI: M. Trenti). By dividing the sky value in each filter’s image by the sky in the associated F125W filter taken in that same direction, we construct a spectral energy distribution of the Zodiacal sky. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | WFC3/IR filters, which — when subtracted from the above data — imply a Zodiacal spectrum with a somewhat steeper power-law slope in Fig. 6a–6d. The best χ2 fit occurs for α=0.66 and ∆T = –1.15 K, which we adopt in the Tables of § 3.4 as our nominal TD case. Non-linearities in the Zodiacal spectrum have a relatively small impact on the implied thermal background. For example, adding a ∼ 7% bump in the spectrum from 1.4 − 1.6 microns, similar to what is seen in Matsuura et al. (2017), changes the best fit slope to 0.69, and the ∆T to –2.72K (which is consistent with our estimated uncertainties of ∼ 2K). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | The results are shown in Fig. 6c–6d for this range of ∆T and α-values, with their associated range in Thermal Dark signal values given in Fig. 6b. The cases shown in Fig. 6b–6d bracket the likely range in telescope ambient temperature values (Appendix A). This results in a plausible range of F125W–F160W Thermal Dark signal values, with the most plausible ones subtracted from any diffuse sky-SB signal in § 3.4. The error range resulting from the TD signal predictions is summarized in Fig. 11 and bracket the range of ∆T temperature variations that the above analysis implies (see § 3.4). |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | In Fig. 10 and Fig. 11, we compute and plot our limits to any diffuse light at 1.25-1.6 µm as follows. Summarizing Fig. 5, Table 1 suggests average offsets of the HST LFS-values minus the Kelsall et al. (1998) COBE/DIRBE model predictions of 0.0145, 0.025, and 0.048 MJy/sr at the effective wavelengths of the F125W, F140W, and F160W filters, respectively. Below we will convert these differences to our limits on diffuse light. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | First, we need to subtract the true WFC3/IR Thermal Dark signal, which has not yet been subtracted in any of the processing. Here, we cannot simply use the F125W thermal foreground of 0.052 e−/pixel/sec from Table 7.11 in the WFC3 IHB (Dressel 2021), as it is larger than our 1.25 µm SB upper limit. The reason that the IHB thermal foreground is higher is that it includes a modeled Thermal Dark signal from the instrument housing, which is subtracted during dark-frame removal. All SKYSURF’s WFC3/IR images have been dark-frame subtracted, and so our modeled Thermal Dark signal values do not contain the instrument housing contribution. The Thermal Dark signals predicted with synphot (in units of e−/pixel/sec) for the plausible range in the temperatures of the HST optical and instrument components across a typical orbit are listed in the first set of three columns of Table 2 for the F125W, F140W, and F160W filters, respectively. With the WFC3/IR pixel scale and zeropoints of Sec. 4 of SKYSURF-1, these are converted to equivalent sky-SB values in units of MJy/sr and nW m−2 sr−1. The conversion factors needed for these calculations are also given in the footnotes of Table 1–Table 3. These TD values are subtracted from the net HST data–Kelsall model differences listed in boldface on the bottom line of Table 1, which are repeated on the top line of Table 2. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | To give a specific example, for the nominal temperature difference of ∆T = (T–Tref ) = –1.15 K (§ 3.3), the Thermal Dark value in the F125W filter is predicted to be 0.00399 e−/pixel/sec, which corresponds to 0.00123 MJy/sr. This value is subtracted from the HST–Kelsall difference of 0.0145 MJy/sr in F125W listed in Table 1 to arrive at the net signal of 0.0133 MJy/sr listed in Table 2 (2nd column for F125W) or 32.1 nW m−2 sr−1 (3rd column for F125W). To be conservative, we quote the values derived in the 3rd column for each filter in Table 2 (in nW m−2 sr−1) as upper limits, given the uncertainties in the ∆T to be used for the TD subtraction, the absolute errors in the Kelsall et al. (1998) model (footnote e of Table 1), as well as the uncertainties in the discrete eEBL (§ 3.4.2–3.4.3) and the DGL (§ 3.5), which still need to be subtracted. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | One of the strengths of the SKYSURF experiment is that it is very effective at removing discrete object light from our diffuse EBL constraint. As discussed in SKYSURF-1, the median SKYSURF exposure is complete down to a limit of ∼ 26 mAB, whereas most discrete extragalactic light comes from galaxies between 17 − 22 mAB. Here we describe the magnitude of this discrete object light for context with other diffuse EBL measurements. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Similarly, in SKYSURF-1 we find that the F160W sky-SB integral of objects detected to AB(cid:46)30 mag amounts to a total sky-SB of 1.813 × 10−26 W/Hz/m2/deg2 or 11.68 nW m−2 sr−1. The fraction of these integrals that comes from discrete objects detected to AB(cid:46)26.5 mag is 10.74 nW m−2 sr−1 in the F125W filter, and 11.31 nW m−2 sr−1 in the F160W filter, respectively. Hence, to AB(cid:46)26.5 mag even the average shallow single HST/WFC3 exposures in the F125W and F160W filters already resolve and detect (cid:38)96.6–96.8% of the total discrete EBL, respectively. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | Many published direct EBL measurements — or upper limits — do include the full discrete iEBL+eEBL signal above, since these methods traditionally measure the total diffuse+discrete galaxy light. By the nature of our SKYSURF methods, we have already removed almost all of the discrete iEBL signal, except for the last ∼0.4–0.6 nW m−2 sr−1 that comes from unresolved objects with AB(cid:38)26.5 mag (see § 3.4.3). Other direct EBL limits should appear higher than our diffuse light limits in part because their values include the discrete EBL signal of 11.11–11.68 nW m−2 sr−1 at 1.25–1.6 µm, while our SKYSURF method already has subtracted (cid:38)96.7% of the discrete EBL signal from the typical 500 sec HST WFC3/IR exposures. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | While the discrete EBL down to ∼ 26 mAB is already automatically excluded from the diffuse EBL limits, we do need to subtract from the upper limits in Tables 2–3 the expected eEBL sky-integral of galaxies beyond the detection limits of the typical short F125W, F140W and F160W exposures in which the HST sky-SB measurements were made. In SKYSURF-1, we showed that for typical exposure times of texp(cid:39)500 sec the WFC3/IR detection limit is AB (cid:46)26.5 mag for compact objects in the F125W filter. For similar median exposure times, this detection limit is about 0.3 mag shallower in the F160W filter (see Table 1 and Fig. 10 of Windhorst et al. 2011). Hence, we assume that all objects with JAB (cid:38)26.5 mag or HAB (cid:38)26.2 mag have been undetected in SKYSURF’s individual ∼500 sec WFC3/IR F125W or F160W exposures, respectively, and so their sky-integral is still included in the diffuse sky-SB measurement. We will therefore estimate and subtract it here. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | This incompleteness correction was also applied to the F160W counts, accounting for the fact that the F160W catalogs have ∼0.3 mag lower sensitivity per unit time. This is justified by the similarity of the J- and H-band versions of Fig. 11 of SKYSURF-1, and as shown in Fig. 10 of Windhorst et al. (2011). Fig. 2 of SKYSURF-1 showed that 75% of the discrete EBL is already reached for objects with AB(cid:46)22.0 mag in the F125W filter, so in essence, this procedure corrects the faintest 25% of the EBL integral for SB-incompleteness of objects known to exist in deeper HST images. The potential impact of very low-SB discrete objects that are beyond the SB-limits of all HST images including the HUDF — and thus not captured by Eq. 8 — will be discussed in § 4. |
Processed_SKYSURF:_Constraints_on_Zodiacal_Light_and_Extraga.txt | aThe effective central wavelengths used for the WFC3/IR F125W, F140W, and F160W filters are λc = 1.2364 µm, 1.3735 µm, and 1.5278 µm, or central frequencies of 2.4248 × 1014 Hz, 2.1827 × 1014 Hz, and 1.9622 × 1014 Hz. b The second row of the a3 parameter gives its estimated errors in parentheses. The estimated errors in a1 and a2 from Eq. 2–3 are not independent from the error in a3, and is of the same order. Hence, only the error in a3 is listed, which is most relevant for estimating the resulting diffuse sky-SB limits in the bottom five rows. c The estimated values of a2 are approximately the same for all three filters F125W, F140W, and F160W for both the HST data and the Kelsall models to within the errors (approximately 1◦), so the same value is adopted for all filters. The a2 values are slightly narrower for the upper envelope to the Kelsall models compared to the lower-bound a2 values and were assumed to be equally narrow for the upper envelopes of those HST data where the sky-SB was not enhanced by the Earth’s limb. d[Between square brackets we list the Figure numbers, from which the sech coefficients on the lines directly above were determined]. e The Kelsall et al. (1998) COBE/DIRBE J-band model prediction has been corrected for the –0.0061 mag ZP difference between the HST F125W and COBE/DIRBE J-band flux scales. The ZL model predictions for the HST WFC3/IR F140W and F160W filters were derived by interpolation between the Kelsall et al. (1998) J- band and K-band predictions. The errors in the (HST–Kelsall) differences in MJy/sr are derived in quadrature from the a3 fitting errors in the previous rows. Kelsall et al. (1998) reported errors in their ZL model of 15 nW m−2 sr−1 at 1.25 µm and 6 nW m−2 sr−1 at 2.2 µm, respectively (see their Table 7). We propagate these also into the errors of our adopted HST–Kelsall differences at 1.25–1.6 µm in nW m−2 sr−1 (bottom row; see also Table 2), which correspond to ∼47–18% errors in these differences at 1.25–1.6 µm, respectively. f The units in these last two rows were converted from MJy/sr to nW m−2 sr−1, using multipliers of 2425, 2183, and 1962 (=10−11 c/λc), respectively, yielding the upper limit to the total diffuse light in nW m−2 sr−1. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | Abstract. The Mock LISA Data Challenges are a program to demonstrate LISA data-analysis capabilities and to encourage their development. Each round of challenges consists of several data sets containing simulated instrument noise and gravitational-wave sources of undisclosed parameters. Participants are asked to analyze the data sets and report the maximum information about source parameters. The challenges are being released in rounds of increasing complexity and realism: in this proceeding we present the results of Challenge 2, issued in January 2007, which successfully demonstrated the recovery of signals from supermassive black-hole binaries, from 20,000 overlapping Galactic white-dwarf binaries, and from the extreme–mass-ratio inspirals of compact objects into central galactic black holes. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | The Laser Interferometer Space Antenna (LISA), a NASA and ESA space mission to detect gravitational waves (GWs) in the 10−5–10−1 Hz range [1], will produce time series consisting of the superposition of the signals from millions of sources, from our Galaxy to the edge of the observable universe. Some of the signals (such as those from extreme–mass-ratio inspirals, or EMRIs) are very complex functions of the physical parameters of the sources; others (such as those from Galactic white- dwarf binaries) are simpler, but their resolution will be confused by the presence of many other similar signals overlapping in frequency space. Thus, data analysis is integral to the LISA measurement concept, because no source can be identified without first carefully teasing out its individual voice from the noisy party of each data set. Understanding data analysis is therefore important to demonstrate that LISA can meet its science requirements, and to translate them into decisions about instrument design. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | group, although some searches locked on strong secondary maxima of the source parameter probabilities. More important, Challenge 1 helped set the playing field and assemble the computational tools for the more realistic Challenge 2. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | Challenge 2, issued in Jan 2007 with results due at the end of Jun 2007, raised the bar by proposing three complex subchallenges. Data set 2.1 contained a full population of Galactic binary systems (about 26 million sources). Data set 2.2 contained a different realization of the Galaxy, plus 4–6 MBH binary inspirals with single-interferometer signal-to-noise ratios (SNRs) between 10 and 2000 and a variety of coalescence times, and five EMRIs with SNRs between 30 and 100. Last, five more data sets (denoted 1.3.1–5, since they were actually released at the time of Challenge 1) contained a single EMRI signal over instrument noise alone. See [4] for more details about the signal models and the exact source content of the data sets. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | Thirteen collaborations (comprising all the researchers listed as participants in the byline of this article, and most task force members) submitted a total of 22 entries, including a proof-of-principle analysis for stochastic backgrounds performed on data set 2.1. Altogether, Challenge 2 successfully demonstrated the identification 20, 000 Galactic binaries, the accurate estimation of MBH inspiral parameters, of and the positive detection of EMRIs. In the rest of this paper, we describe some of its highlights. All the solutions submitted by participating groups, together with technical write-ups of their methods and findings, can be found at the URL www.tapir.caltech.edu/~mldc/results2. A few groups are also contributing descriptions of their work to the proceedings of this conference. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | GLIG A collaboration of research groups at institutions in the UK, United States and New Zealand developed a Reversible-Jump Markov Chain Monte Carlo (RJ MCMC) code that can sample models with different numbers of sources; for lack of time, however, they only submitted parameter sets for the verification binaries. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | MTJPL The Montana State–JPL collaboration used a Metropolis–Hastings Monte Carlo (MHMC) code that ran separately for overlapping frequency bands and for different hypotesized numbers of sources; model comparison was then used to determine the most probable number of sources in each band. The collaboration submitted parameter sets for 19,324 sources for data set 2.1, and 18,461 sources for data set 2.2. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | statistic, template-bank–based matched-filtering search using a hierarchical scheme that enforced trigger coincidence between TDI observables, followed by a coherent follow-up using noise-orthogonal TDI combinations [11]. They submitted parameter sets for 1777 sources for data set 2.1, and 1737 sources for data set 2.2. submitted parameters for 3862 sources in data set 2.1. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | statistic, template-bank–based matched-filtering search using a hierarchical scheme that enforced trigger coincidence between TDI observables, followed by a coherent follow-up using noise-orthogonal TDI combinations [11]. They submitted parameter sets for 1777 sources for data set 2.1, and 1737 sources for data set 2.2. submitted parameters for 3862 sources in data set 2.1. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | Figure 1. Correlation analysis of Challenge-2.1 Galactic-binary catalogs (histogram, with bin fractions given on logarithmic scale). Left panel: associated by correlation; right panel: associated by Doppler metric. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | into the data sets cannot be recovered because they are (relatively) too weak, we do not have a precise estimate of how many sources should be recoverable. Thus, the notion of false dismissal is not well defined. To make matters worse, the notion of false positive is also ill defined, because a single recovered source can provide a good fit to the blended signal from several injected sources, which may well be the “right” answer with the knowledge we have, since it is the best fit to the data with the smallest numbers of parameters. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | The task force devoted considerable time to the analysis of Galactic-binary searches, and we do not have space here to describe all the treatments that we applied to the data. Instead we will limit our report to techniques that pair up individual recovered sources with individual sources from the challenge key, with the understanding that this will overestimate the number of false positives, and say nothing about false dismissals. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | One way to proceed is to associate the reported and injected sources that have the strongest signal correlation, limiting the search to the bright (SNR > 2) injected sources that could in principle have been found: in the left panel of figure 1 we show the distribution of correlations generated with this procedure for data set 2.1. Detections with the highest correlations can be considered “safest,” while those with the lowest correlations probably represent spurious associations. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | Another procedure is to associate the reported and injected sources that minimize the Doppler metric that spans the frequency–sky-location subspace of the full parameter space, and automatically maximizes correlation over the extrinsic parameters (amplitude, polarization, inclination, initial phase): the right panel of figure 1 shows the resulting distribution of correlations. The UTB entry, which includes frequency and sky position but not the extrinsic parameters, can only be plotted this way. Generally, this is a softer criterion, and all searches do better by it (especially the PrixWhelanAEI entry, whose long-wavelength approximation for the LISA response is prone to extrinsic-parameter errors). |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | Figure 2. Recovered SNRs and intrinsic and extrinsic parameter errors for Challenge-2.1 Galactic-binary catalogs (histogram). True sources and templates are associated by correlation, except for the UTB catalog, for which they are associated by Doppler metric. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | in the (extrinsic) orientation angles are larger. The φ0 graph for the PrixWhelanAEI suggests a systematic error in the definition of initial phase. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | Altogether, these challenges demonstrated a solid capability in analyzing signals from the Galaxy and resolving a large number of binaries. As we mentioned, deciding how well they were recovered is not an easy question to answer, because of the difficulty of defining (at least operationally) a notion of identity for recovered sources. These problems deserve careful attention in the future. |
Processed_Report_on_the_second_Mock_LISA_Data_Challenge.txt | MTAEI Cornish and Porter used an MHMC matched-filtering search with a frequency-annealed scheme where shorter, lower-frequency templates are used in the initial phases of the search and then progressively extended. |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.