Large scale structures in the Universe
Lead Research Organisation:
University College London
Department Name: Physics and Astronomy
Abstract
The PhD project will map clusters and voids in the Dark Energy Survey and other data sets, and their imprint on the Cosmic Microwave Background.
Publications
Naidoo K
(2019)
MiSTree: a Python package for constructing and analysing Minimum Spanning Trees
in Journal of Open Source Software
Naidoo K
(2020)
Beyond two-point statistics: using the minimum spanning tree as a tool for cosmology
in Monthly Notices of the Royal Astronomical Society
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| ST/N50449X/1 | 30/09/2015 | 30/03/2021 | |||
| 1783821 | Studentship | ST/N50449X/1 | 30/09/2016 | 30/03/2020 | Krishna Naidoo |
| Title | If a tree falls in a forest |
| Description | A collaboration between artist from the Royal College of Arts and cosmologist from Universty College London. Pairs of artist and scientist came together to produce a small exhibition piece on their current research. Ours involved creating a a scuptural piece displaying the minimum spanning tree methodology and a video projected on a dome showing how a particular algorithm finds this data. This was combined with a musical piece which was constructed algorithmically and displayed on stars from the night sky. |
| Type Of Art | Artistic/Creative Exhibition |
| Year Produced | 2018 |
| Impact | The collaboration helped me realise the many similarities between artist and scientist as well as helped me understand my research in ways I had not seen prior. The exhibition also provided me with lots of lessons about how I should go about communicating my research to an audience that is not familiar with the technicalities of the work. |
| URL | http://ifatreefalls.rca.ac.uk |
| Description | Very broadly we wanted to determine whether a new statistics could be developed for analysing the large scale distribution of matter in the Universe. This is pivotal for future galaxy surveys that will map the positions of millions of galaxies. Our core purpose was to determine whether the cosmic web (the way in which matter clusters in a filamentary web like structure) could be captured statistically and used to determine whether what we see in galaxies matches what we see in simulations. We determined that the minimum spanning tree (MST), could be used as such a statistics and could be used with existing and more conventional statistical methods for determining parameters of the cosmological model and testing new physics. In particular we have shown in my PhD thesis that the statistics is very sensitive to neutrino mass, of keen interest to physicist as the mass of these elusive particles remains unknown. The application to real data still requires some work. We have made great progress in understanding how we can mitigate common systematic issues from galaxy surveys and have compared the MST on real and simulated data from the Baryon Oscillation Spectroscopic Survey. |
| Exploitation Route | The statistical methods used have a broad application. The minimum spanning tree statistics and the public software created for this study can be used for a wide range of studies which look to characterise point distributions. The software has been download over 9000 times with studies employing it in other areas of science, including star cluster studies. |
| Sectors | Digital/Communication/Information Technologies (including Software) Other |
| Title | Minimum Spanning Tree |
| Description | Using decision maths we have investigated the use of the Minimum Spanning Tree to try and capture topological information present in Large galaxy redshift surveys. |
| Type Of Material | Data analysis technique |
| Year Produced | 2019 |
| Provided To Others? | Yes |
| Impact | Although not in use yet, as we currently writing a paper on the subject, the minimum spanning tree is expected to improve our knowledge on the constants that define our model of the universe. |
| Description | Integrated Sachs-Wolfe for Lightcone Simulations |
| Organisation | University of Barcelona |
| Department | Institute for Space Studies of Catalonia (IEEC) |
| Country | Spain |
| Sector | Academic/University |
| PI Contribution | I have created Integrate Sachs-Wolfe maps for the MICE and Flagship lightcone simulations. This will allow for cross correlation studies to be carried out between large galaxys surveys and the cosmic microwave background (CMB). I will additionally make the c++ pipeline public so that ISW maps can be created from future lightcone simulations. |
| Collaborator Contribution | I have worked closely with Pablo Fosalba (based in Barcelona) who has provided me with the MICE and Flagship lightcone simulations. |
| Impact | We will eventually make the ISW maps and code created in this study public. |
| Start Year | 2019 |
| Description | Simulations for minimum spanning tree |
| Organisation | International School for Advanced Studies |
| Country | Italy |
| Sector | Academic/University |
| PI Contribution | We have been working on a project designed for future use on large scale galaxy redshift surveys. We have been exploring a new statistics which is designed to pick up information not currently used in cosmology. The hope is this will better constrain parameters used to describe the universe we live in. We have been exploring its application in a variety of settings, systematics consideration. Additionally a major challenge has been to tie down future observations with theoretical expectations, since the statistics can not be predicted analytically we have to resort to state of the art simulations. We plan to release a public code to allow others in the community to run a similar analysis. |
| Collaborator Contribution | Our partners (Elena Massara, Matteo Viel) have provided state of the art N-body simulations. Whilst Hector Gil-Marin has supplied us with a piece of code to calculate bispectrum measurements from data. Bispectrum are analytically and observationally difficult measurements to make so we resort to using a tried and tested code for this calculation since our aim is to simply compare methodologies. |
| Impact | We are currently writing a paper to detail our work on this new statistic and plan to use it on the existing Baryon Oscillation Spectrocopic Survey and the future Dark Energy Spectrocopic Survey. |
| Start Year | 2017 |
| Description | Simulations for minimum spanning tree |
| Organisation | Simons Foundation |
| Department | Flatiron Institute |
| Country | United States |
| Sector | Academic/University |
| PI Contribution | We have been working on a project designed for future use on large scale galaxy redshift surveys. We have been exploring a new statistics which is designed to pick up information not currently used in cosmology. The hope is this will better constrain parameters used to describe the universe we live in. We have been exploring its application in a variety of settings, systematics consideration. Additionally a major challenge has been to tie down future observations with theoretical expectations, since the statistics can not be predicted analytically we have to resort to state of the art simulations. We plan to release a public code to allow others in the community to run a similar analysis. |
| Collaborator Contribution | Our partners (Elena Massara, Matteo Viel) have provided state of the art N-body simulations. Whilst Hector Gil-Marin has supplied us with a piece of code to calculate bispectrum measurements from data. Bispectrum are analytically and observationally difficult measurements to make so we resort to using a tried and tested code for this calculation since our aim is to simply compare methodologies. |
| Impact | We are currently writing a paper to detail our work on this new statistic and plan to use it on the existing Baryon Oscillation Spectrocopic Survey and the future Dark Energy Spectrocopic Survey. |
| Start Year | 2017 |
| Description | Simulations for minimum spanning tree |
| Organisation | University of Barcelona |
| Country | Spain |
| Sector | Academic/University |
| PI Contribution | We have been working on a project designed for future use on large scale galaxy redshift surveys. We have been exploring a new statistics which is designed to pick up information not currently used in cosmology. The hope is this will better constrain parameters used to describe the universe we live in. We have been exploring its application in a variety of settings, systematics consideration. Additionally a major challenge has been to tie down future observations with theoretical expectations, since the statistics can not be predicted analytically we have to resort to state of the art simulations. We plan to release a public code to allow others in the community to run a similar analysis. |
| Collaborator Contribution | Our partners (Elena Massara, Matteo Viel) have provided state of the art N-body simulations. Whilst Hector Gil-Marin has supplied us with a piece of code to calculate bispectrum measurements from data. Bispectrum are analytically and observationally difficult measurements to make so we resort to using a tried and tested code for this calculation since our aim is to simply compare methodologies. |
| Impact | We are currently writing a paper to detail our work on this new statistic and plan to use it on the existing Baryon Oscillation Spectrocopic Survey and the future Dark Energy Spectrocopic Survey. |
| Start Year | 2017 |
| Title | MiSTree |
| Description | MiSTree is a public python software used to construct and analyse data using the minimum spanning tree method we have outlined. The software has been complete for some time but is not publicly released yet but is planned to be released on an MIT licence. |
| Type Of Technology | Software |
| Year Produced | 2019 |
| Open Source License? | Yes |
| Impact | The software is used through out our methodology paper on the use of the minimum spanning tree for research in cosmology. |
| URL | https://knaidoo29.github.io/mistreedoc/ |