Baikal-GVD (Baikal Gigaton-Volume Detector) is the largest neutrino telescope in the Northern Hemisphere, intended for detecting fluxes of ultrahigh-energy neutrinos from astrophysical sources.
{PHOTO:ID=3396}
The Baikal-GVD Neutrino Telescope is a spatial array of light-sensitive elements (optical modules). The facility is a complex of independent clusters. Each cluster functions as an autonomous high-energy neutrino detector, with individual control and trigger systems for data acquisition and processing. This architecture enables building up the facility seamlessly by adding new clusters, without having to rebuild the entire installation.
Each cluster comprises 8 vertical strings — cables with optical modules (photodetectors), control and positioning modules. Peripheral strings of the cluster are 60 m away from the central one.
Each string carries 36 optical modules vertically spaced 15 m from each other. The string is divided into three sections, each of which includes 12 optical modules and one section master module (SMM). The uppermost optical module of a string is located at a depth of 700 m, and the bottommost one at 1240 m. Strings are anchored to the lake bed. Bundles of buoys keep the strings in a vertical position.
This cluster configuration ensures high detection efficiency for both muon tracks and cascades.
Expeditions to construct the neutrino telescope are usually scheduled twice a year: in winter and in summer.
Winter expeditions begin in late February and continue until the first ten days of April. During this period, new clusters of the telescope are deployed and old ones are upgraded.
Summer expeditions last throughout August. During this time, the winter construction season of the facility is prepared at the Shore Station by repairing the equipment and developing the infrastructure of the project.
The acoustic positioning system continuously monitors the position of all optical modules. Signals are transmitted via coaxial cables to master modules of the section. After digitization and time stamping, data are sent to the Cluster Centre. Cluster Centres are connected to the Shore Centre by fibre-optic cables. These same cables are used to control the facility remotely.
All clusters are synchronized with each other, which enables intercluster correlation of events during reconstruction of muon tracks and cascades.
Raw data from optical and master modules of sections are transmitted via fibre-optic cables first to the Cluster Centre, and then to the Shore Station.
The cluster control program processes primary data and classifies them into three different types of records: master records, service records, and monitor records.
Master records contain digitized data from optical modules with metadata, such as time stamps, optical module addresses, trigger information, etc.
Service records contain information about static and dynamic configuration of the cluster: number of clusters, IP addresses of telescope elements, and the layout of sections and strings. Dynamic configuration data describe current session numbers and the trigger system in use.
Monitor records comprise operating parameters of optical modules: PMT voltage, trigger thresholds, temperature inside the module, and signal amplitude distributions.
One cluster generates approximately 15 GB of data per day. The data are compressed and transmitted to the servers of the Joint Institute for Nuclear Research in Dubna.
The data are processed by the BARS program (Baikal Analysis and Reconstruction Software). At the core of BARS is the MARS software package of the MAGIC Collaboration (used with permission). It, in turn, is built on the ROOT software package.
BARS is designed to run on the Linux operating system and consists mainly of C++ classes and ROOT macros, each of which performs specific tasks:
— unpacks and reads raw data files,
— sorts and filters different types of records,
— distributes data by cluster sections or time stamps,
— extracts information about pulses,
— reconstructs unified events from time-coincident signals.
The result of one macro is usually a ROOT file that contains the input data for the next one, thereby forming a chain of processing tasks. This chain represents a complete data processing cycle. For the daily processing of new data, this cycle is launched automatically. It is managed by a specialized Python-based automation system.
