Adding funding proposals for Cycle 4. One for astroquery and one for specutils work.

Author: NuriaLorente

finance/proposal-calls/cycle4/AAO-MQ-astroquery.md

@@ -0,0 +1,95 @@
+### Title
+Cycle 4 Funding: Making Australian astronomical datasets accessible 
+
+### Project Team
+AAO Research Data & Software, Macquarie University  (rds.org.au) 
+
+The foundation of the AAO Research Data & Software (RDS) team is built upon over 
+40 years of supporting astronomical observatories in Australia. This software 
+development experience ranges from controlling telescopes and their instruments, 
+to building data reduction pipelines, to developing and implementing tools that 
+store, manage, and analyse data. The RDS team aims to be at the forefront of 
+research software and data engineering. We look for the best technology to solve
+a problem, but at the same time realise that software is not innovative or useful
+if it cannot be maintained, extended, or scaled. We appreciate the benefits that
+a holistic view of a software ecosystem can bring, but also recognise that a 
+"deep dive" might be necessary to solve a complex problem. And while our historical
+strength has an astronomical flavour, our skills and expertise are increasingly 
+recognised across research disciplines. Some of our current and recent projects 
+include the Data Central Science Platform, including Virtual Observatory services; 
+VLT MAVIS Data Reduction pipeline design and development; our state-of-the-art 
+Pipelines as a Web Service (PAWS) system; and our recently completed ESO VLT 
+Pipelines development and maintenance project.
+
+We propose the following individuals for this project, although other team members 
+can be involved if needed:
+
+* James Tocknell
+* Anthony Horton
+* Brent Miszalski
+* Nuria Lorente
+* Simon O'Toole
+
+Although this is our first proposed contribution to astropy as a team, some of
+our members have made individual contributions over the years. Moreover, given
+the use that we make of astroquery in our Data Central science platform
+components in particular, we are keen to further contribute to astroquery
+infrastructure, not just through this proposal, but also in the longer term.
+
+### Project Description
+The astroquery package has become a popular touchstone for observational 
+astronomers to find data, being used by more than 4000 github repositories (The 
+Astropy Collaboration et al. 2022, 935, 167). Apart from astroquery.casda, 
+Australian astronomical data are conspicuously absent from astroquery. In this 
+project we address this shortcoming by adding access to the Anglo-Australian 
+Telescope (AAT) Archive and several other Australian large survey datasets hosted
+by Data Central (https://datacentral.org.au): images, catalogues and spectra. 
+Enhancing the visibility and accessibility of Australian datasets is critically 
+important to complement multi-wavelength and multi-messenger observations from 
+upcoming Southern Hemisphere facilities such as the Vera Rubin Observatory (VRO), 
+Cerenkov Telescope Array (CTA) and the Square Kilometre Array (SKA). 
+
+While many of the datasets we plan to add are accessible via VO services, many 
+astronomers find VO services very difficult to interact with. Adding them to their 
+own dedicated astroquery module would alleviate their anxieties about using these
+services. The AAT archive, on the other hand, is not currently accessible via a 
+VO service and this work would allow for robust programmatic access for the 
+first time. 
+
+We also propose to address a longstanding problem in the preparation of multi-object 
+spectroscopy (MOS) observations by adding the Cone of Darkness algorithm 
+(Lorente 2014, ASPC 485, 95) to select blank sky positions. This automates a 
+tedious task in a robust manner for any input astroquery catalogue. 
+
+
+### Project / Work
+1. Data Central catalogues, images and spectra are accessible, respectively, via 
+TAP, SIA and SSA services hosted by Data Central. We plan to provide access via 
+a dedicated astroquery.datacentral module. The module would also include an API 
+to access the AAT archive, also hosted by Data Central. The AAT archive hosts data 
+from several instruments, the most prominent of which are MOS observations taken 
+with the 2dF and 6dF facilities, amassing vast quantities of spectroscopic data. 
+Via improvements to the archive API endpoints, we plan to provide access to the 
+individual fibre tables of MOS observations, providing unprecedented access to 
+this rich spectroscopic database. In the case of 2dF-AAOMega observations, users 
+may then request automatic reductions of queried spectra via our Pipeline As a 
+Web Service (PAWS) system.
+
+1. Addition of the Cone of Darkness algorithm involves porting the C++ code and 
+tests to astroquery such that any input catalogue (also obtained with astroquery) 
+may be used to select sky positions. High proper motion stars may be excluded 
+with an optional GAIA DR3 query.
+
+
+### Approximate Budget
+This funding application is to cover staff time. We estimate 420 hours of effort
+are required to complete this project, which at an hourly rate of USD$145/hour 
+gives a budget of USD$60,900.
+
+This effort estimate will be refined, and the minimum useful time to carry out
+some of the proposed work will be determined and added here during the iteration
+period.
+
+
+### Period of Performance
+We expect the period of performance to be two years.

finance/proposal-calls/cycle4/AAO-MQ-specutils.md

@@ -0,0 +1,81 @@
+### Title
+Cycle 4 Funding: Realising the full potential of specutils via an extensible
+interface for community contributed algorithms
+
+### Project Team
+AAO Research Data & Software, Macquarie University  (rds.org.au) 
+
+The foundation of the AAO Research Data & Software (RDS) team is built upon over 
+40 years of supporting astronomical observatories in Australia. This software 
+development experience ranges from controlling telescopes and their instruments, 
+to building data reduction pipelines, to developing and implementing tools that 
+store, manage, and analyse data. The RDS team aims to be at the forefront of 
+research software and data engineering. We look for the best technology to solve
+a problem, but at the same time realise that software is not innovative or useful
+if it cannot be maintained, extended, or scaled. We appreciate the benefits that
+a holistic view of a software ecosystem can bring, but also recognise that a 
+"deep dive" might be necessary to solve a complex problem. And while our historical
+strength has an astronomical flavour, our skills and expertise are increasingly 
+recognised across research disciplines. Some of our current and recent projects 
+include the Data Central Science Platform, including Virtual Observatory services; 
+VLT MAVIS Data Reduction pipeline design and development; our state-of-the-art 
+Pipelines as a Web Service (PAWS) system; and our recently completed ESO VLT 
+Pipelines development and maintenance project.
+
+We propose the following individuals for this project, although other team members 
+can be involved if needed:
+
+* James Tocknell
+* Anthony Horton
+* Brent Miszalski
+* Nuria Lorente
+* Simon O'Toole
+
+### Project Description
+The latest version of specutils includes useful common analysis tasks for
+astronomical spectra. 
+To realise its potential as a fully-fledged contributor to scientific workflows,
+spectutils would greatly benefit from the addition of a generic capability for
+advanced users to supply their own tailored algorithms (i.e. code) to be used in place of
+the defaults. 
+We would like to add this capability to specutils, allowing users to apply
+advanced analysis and algorithms to specutils spectra via a regularlised
+interface.  This interface would be analagous to how astroquery provides a
+regularised interface to access VO and other web services.  It will provide the
+opportunity for the community to contribute their own algorithms, e.g. relating
+to large community spectroscopic surveys, or other generic methods. This would go
+a long way towards providing a sustainable central repository for analysis
+algorithms outside of traditional legacy environments (e.g. IRAF).  In this
+project we propose implementing the generic interface and will provide an example
+implementation of cross-correlation of spectra using Fast Fourier Transforms.
+
+### Project / Work
+1. We plan to introduce into specutils a regularised framework for user contributed
+algorithms.  We have experience in contributing to specutils loaders, used by
+our team to ingest spectra into Data Central.
+
+1. An example contributed algorithm will be provided, namely a Fast Fourier
+Transform (FFT) cross-correlation.  This consists of two steps:
+
+   2. Interpolating the spectra and templates to a common (log) wavelength binning;
+   2. Computing the FFTs, inverting the combined result, and shifting the output
+such that it aligns with the redshift bins.  We will adapt existing code
+developed by our team when porting the MARZ redshifting code 
+(Hinton et al. 2016, A&C, 15, 61) from Javascript to Python. The code has been
+thoroughly tested and validated, addressing the shortcomings of the original codebase.
+
+### Future Work
+We are keen to make substantial contributions to specutils in the future.
+
+### Approximate Budget
+This funding application is to cover staff time. We estimate 630 hours of effort
+are required to complete this project, which at an hourly rate of USD$140/hour 
+gives a budget of USD$88,200.
+
+This effort estimate will be refined, and the minimum useful time to carry out
+some of the proposed work will be determined and added here during the iteration
+period.
+
+
+### Period of Performance
+We expect the period of performance to be two years.