Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis
The CTBTO is seeking a contractor for software development and maintenance of the NET-VISA seismic monitoring system on a call-off basis. The contract covers a firm 36-month period with two optional 12-month extensions. Bidders must demonstrate 7 years of experience in scientific software development, Bayesian methods, machine learning, C++ and Python, database systems, and CI/CD. A team of at least three staff with 5+ years in Python/C++, SQL, and maintenance is required. Proposals are evaluated on technical criteria (minimum weighted score 75) and financial offer.
Eligibility: Bidder must be legally established in a CTBT State Signatory. Minimum 7 years corporate existence and 7 years experience in scientific software development, project management, Bayesian methods, ML, C++/Python, databases, CI/CD, containerized deployment. Team of at least 3 staff, each with 5+ years recent Python/C++, SQL, and 3+ years maintenance. Staff hours must overlap 3+ hours with CET. English fluency required. Consortiums and subcontractors permitted if all members meet eligibility.
Pipeline
Funder track record · does Comprehensive Nuclear-Test-Ban Treaty Organization (CTBTO) actually pay grants?
payout history not checked · last check failed: web search unavailable: 4 of 6 searches failed
Proven payer: at least 3 named grantees and one payout in the last 90 days. Every grantee below is quoted word for word from its source. When a source does not date a grant, the date is when that page or post was published.
Trust breakdown (3/3 checkable facts verified)
| Signal | Weight | Result | Note |
|---|---|---|---|
| deadline | 0.35 | 100% | quoted in source |
| apply link | 0.075 | 100% | link in source |
| funder | 0.075 | 100% | named in source |
| source quality | 0.15 | 50% | social or search hit |
| model confidence | 0.1 | 92% | model self-report |
Fit breakdown (against the ORG_* profile)
| Signal | Weight | Result | Note |
|---|---|---|---|
| model fit | 0.5 | 10% | model rated 10% |
| ecosystem overlap | 0.25 | 0% | none vs aleo |
| category overlap | 0.15 | 33% | research, dev_tooling, infra, data_analytics, ai vs zk, privacy, infra, dev_tooling, security, interoperability |
| tech overlap | 0.1 | 0% | python, c++, sql, oracle, postgresql, ci/cd, containerized deployment, machine learning, bayesian methods vs rust, leo, zk |
Sources (1)
| Role | Document | From | Fetched |
|---|---|---|---|
| primary | extension_of_deadline_and_clarifications_no.1_to_rfp_no-2026-0108 | Web discovery: ecosystem sweep | 2026-11-17 |
History (1)
| v | Changed | Fields |
|---|---|---|
| 1 | 2026-11-17 | created |
Extracted text
TO: All Bidders FROM: CTBTO Procurement Services Section DATE: 05 November 2026 REF.: RFP No. 2026-0108/MOGAPI TEL. NO.: +43 1 26030 6350 EMAIL: procurement@ctbto.org SUBJECT: Deadline Extension and Clarifications No. 1 to RFP 2026-0108/MOGAPI - Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis In reference to request for proposal (RFP) No. 2026-0108/MOGAPI concerning “Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis”, kindly note the following: • The deadline for the submission of proposals is hereby extended from 11 November 2026 to Friday 20 November 2026 17:00 hours, Vienna (Austria) local time. • Please find attached the Clarifications No. 1 to questions raised by interested bidders, including: o Updated Attachment 2 - Technical Evaluation Criteria document part of the RFP, which is hereby replaced with the attached amended version of this document (Rev. 04 Nov 2026) (revision highlighted in yellow for ease of reference); o New Attachment A (NET-VISA Software Description Document) and Attachment B (IDC Database Schema-ODB) which are hereby added as “Attachment A” and “Attachment B”, respectively, to Annex B Terms of Reference part of this RFP. Clarifications No. 1 and its attachments are an integral part of the RFP documents and shall be considered in the preparation and submission of proposals. We are looking forward to receiving proposals prior to the extended submission deadline on Friday 20 November 2026 17:00 hours, Vienna (Austria) local time. Sincerely, CTBTO Procurement Services Section Attachments: 1. Questions and Answers – Clarifications No. 1 2. Attachment 2 - Technical Evaluation Criteria – Rev. 4 Nov 2026 3. Attachment A – NET-VISA Software Description Document 4. Attachment B – IDC Database Schema-ODB Clarifications No. 1 - questions and answers - RFP No. 2026-0108/MOGAPI: Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis 1 Clarifications No. 1 - questions and answers # Bidders’ Question Answer 1 Daily rates across Call-off Periods. Note 1 of the Price Schedule Format states that the rates shall be firm and fixed throughout the term of the Contract, and Clause 12(c) of the Model Contract fixes the Annex C unit prices for the entire duration. However, the Price Schedule has separate daily rate columns for Call-off Periods 1, 2 and 3, and Clauses 12(a)(ii), 12(a)(iii) and 12(b)(ii) refer to a daily rate agreed for the optional extensions. May bidders quote different daily rates for the optional Call-off Periods 2 and 3 than for the firm 36-month Call-off Period 1? Yes. Bidders may quote different daily rates for Call-off Periods 1, 2 and 3. The separate rate columns in the Price Schedule Format are intended for this purpose. The rates quoted for each Call-off Period shall be firm and fixed and, once the Contract is awarded, shall not be subject to adjustment for the duration of the Contract. 2 Pricing of on-site visits. Section 8 of the Terms of Reference provides for reimbursement of return flights against supporting documents and per diem at UN DSA rates, while Clause 12(d) of the Model Contract states that the Contract Price covers all costs, including travel and allowances […] Would the Commission accept on-site visits priced as an on-site daily rate plus a fixed lump sum per trip, with proof of travel provided on request but no ticket receipts? Please refer to Note 4 of Attachment 3: Price Schedule Form. Bidders are free to determine the pricing methodology for their proposed travel costs. Bidders may propose travel costs either as firm and fixed costs or as estimated costs, in accordance with the solicitation documents. Accordingly, the Commission would accept a pricing approach based on a fixed lump-sum amount per trip, provided such costs are clearly reflected in the Financial Proposal as firm and fixed travel costs, indicating a breakdown of what is included in the lump sum and the benefits of such approach as representing a more cost- efficient quote. Where travel costs are proposed as all- inclusive lump sum(s), the provisions related to daily subsistence allowance (DSA), and reimbursement of travel costs other than DSA upon presentation of supporting documents in accordance with the Contract provisions would not apply. Clarifications No. 1 - questions and answers - RFP No. 2026-0108/MOGAPI: Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis 2 Either way, any proposed travel costs will be evaluated as part of the overall Financial Proposal in accordance with the solicitation documents best value for money assessment. 3 Minimum technical score. Section 1 (Technical Evaluation) of the Evaluation Criteria and Method states that bidders must obtain a minimum weighted score of 75 in Stage 2, while the following paragraph refers to a minimum weighted score of 69. Could you confirm which minimum applies? Kindly find attached a revised Attachment C - Rev. of 3 Nov 2026, which now refers to the correct “minimum weighted score of 75”. 4 Team Roster and domain expertise. Sections 6.1 and 6.4 of the Terms of Reference require each staff member on the Team Roster to meet all of the requirements of Section 6.4, and Section 5 (item 8) refers to subject-matter experts in seismology, hydroacoustics and infrasound. Our team provides expertise […]. We also consult external experts from time to time, but would not name them in the contract. Could you confirm that a proposal naming only our own staff on the Team Roster, without separately named domain experts or external consultants, would be compliant? The Team Roster requirements under Annex B Terms of references -Section 6.4 and the desirable requirement under Section 5, item 8, are assessed separately. The subject-matter experts referred to in Section 5, item 8, are not required to be included on the Team Roster and do not need to be employees of the Bidder. However, to enable the Commission to assess this criterion, Bidders should identify the available experts and provide information on their qualifications, areas of expertise and availability. Where no such experts are identified, no score will be awarded under Section 5, item 8. 5 Infrasound association (ToR Section 4.2) Section 4.2.1 states that "both the association of pure infrasound events and the association of infrasound detections with seismic events are not yet performing at the required operational quality level." Could the Commission indicate whether any open references or publicly available documentation describe the current state of the infrasound component of NET-VISA, beyond the NET-VISA Software Description Currently about 12.5% of events from NET-VISA are matched with the interactive analysis results. Kindly refer to the current state described in the attached NET-VISA Software Description Document and IDC Database Schema-ODB. These documents are herewith attached and added as part of Annex B- Terms of Reference part of this RFP. Clarifications No. 1 - questions and answers - RFP No. 2026-0108/MOGAPI: Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis 3 Document (Attachment A) and the publications available in the open literature (e.g., Le Bras et al., 2021)? 6 As foreseen in Section 4.1.1 for bulletin quality, will the evaluation baseline, metrics and measurable objectives for infrasound be defined in the applicable Work Order? Is the expected outcome an iterative improvement of the existing model, or may a new model be considered if the investigation shows this is justified? Yes, as specified in the Annex B - Terms of Refences, the evaluation baseline, metrics and measurable objectives for infrasound will be defined in the applicable Work Order. The Commission expects both approaches to be investigated (and, in case of success, implemented): 1. Iterative improvement of the existing model. 2. Switching to a new model if it performs better than exciting one. 7 New hydroacoustic travel-time tables and blockage maps (ToR Section 4.3) Section 4.3.1 refers to the new 3D blockage maps and travel-time tables received by the IDC for all hydroacoustic stations (hydrophone stations and T-stations). The ToR does not specify the expected size of these 3D tables. Will the Contractor be expected to implement additional data-compression or compact-storage techniques, or can the 3D tables provided by the IDC be used as they are? Are there any reference figures or constraints on the computing, memory, storage and processing-time resources that the resulting deliverable must satisfy, whether in IDC operations or in “NDC in a Box” deployments? With reference to the Annex B – Terms of Refence - 3D tables provided by the Commission can be used as they are. The Commission does not expect additional compression to be made. Regarding computational resources: • The Commission has enough servers, physical and virtual, dedicated to development, operational use, database hosting etc. • “NDC in a Box” environment is just a virtual machine, with everything (backend, frontend, database) inside, intended to be launched on a PC with limited resources. 8 Use of LocOO3D in NET-VISA (ToR Section 4.4) Source localisation could potentially be incorporated into NET-VISA at several stages: (a) training, for estimating event-related distributions; (b) preliminary event localisation; or (c) refinement with respect to the probabilistic model. The Contractor should propose and justify the integration point as part of the work. Clarifications No. 1 - questions and answers - RFP No. 2026-0108/MOGAPI: Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis 4 At which of these stages does the Commission expect LocOO3D to be integrated, or should the Contractor propose and justify the integration point as part of the work? 9 […] whether an international software firm working remotely, without an Austrian entity, is eligible and whether any on-site presence, security clearance, nationality restriction, or Austrian/EU business registration is mandatory. There is no requirement for the bidder to be registered in Austria/EU nor have presence in Austria/EU. However, in accordance with the RFP requirements, bidders (including subcontractors) shall be legally established in one of the States Signatories of the Comprehensive Nuclear-Test-Ban Treaty (CTBT), the list of which is available in the CTBTO website at www.ctbto.org under Status of Signatures and Ratifications | CTBTO. Moreover, in accordance with the requirements of the Terms of Reference (ToR), the subsequent contractor’s staff must be able to travel to Vienna, Austria as requirement by the Commission – kindly refer to sections 7 and 8 of the ToR. In addition please see response to Question 12 below. 10 Could you please clarify whether the solicitation sets mandatory minimum thresholds for corporate age, turnover, reference contracts, team certifications, or prior work with visa/border-control systems. Kindly refer to the contractor’s requirements (section 5 of the Annex B - ToR) and requirements for the Team Roster (section 6 of the Annex B ToR) and the Vendor Profile Form document part of this RFP. The Bidder shall have sufficient financial capacity to perform the contract in relation to the value of the proposed Financial Proposal. 11 Could you please clarify whether consortium or subcontractor participation is permitted. The Commission does not prohibit the participation of con- sortia (joint offers) or subcontractors. Where a proposal is submitted by a consortium, bidders should clearly identify the lead entity and the respective roles and responsibilities of the participating entities, provide a copy of the relevant consortium agreement and provide the Vendor Profile Form for each of the entities Clarifications No. 1 - questions and answers - RFP No. 2026-0108/MOGAPI: Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis 5 part of the consortium. All members of the consortium must meet the eligibility criteria of the relevant RFP, in particular, they shall all be legally established in one of the CTBT States Signatories. In the event of award to the consortium, the Commission will contract with a single entity, normally the lead entity of the consortium. Any proposed subcontractors should also be clearly identified in the proposal, where applicable. In accordance with the RFP instructions, the information shall include the names, legal status, address and qualifications of the proposed subcontractor(s), and the scope of the subcontracted services. Further, the Proposal shall provide a statement that the bidder shall be fully responsible for the performance of sub-contractors. All sub-contractors shall also be legally established in one of the CTBT States Signatories. 12 Could you please confirm whether an India-incorporated development company may bid on this assignment and whether remote delivery is acceptable, subject to any required security or access controls? Please also advise whether technical clarifications should be submitted through your procurement platform rather than by email. 1. Bidder’s eligibility: Bidders and their sub- contractors (if any) shall be legally established in one of the Comprehensive Nuclear-Test-Ban Treaty (CTBT) States Signatories. In particular, the goods and services (if any) to be rendered under the Contract shall have their origin in the States Signatories of the Comprehensive Nuclear-Test-Ban Treaty (CTBT), a list of which is available in the CTBTO website at www.ctbto.org under Status of Signatures and Ratifications | CTBTO. For purposes of this paragraph, "the origin" means the place from where the materials, goods and/or from which the services are supplied. Bidders shall foremost assess Clarifications No. 1 - questions and answers - RFP No. 2026-0108/MOGAPI: Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis 6 their eligibility and the origin of the proposed goods and/or services against this requirement. Please refer to the “Instructions for preparation and submission of Proposals” document part of this RFP. 2. Remote delivery of services: Yes, this is acceptable. Kindly refer to the Terms of Reference concerning off-site work. 3. Tender clarifications submission: These shall be submitted by email to procurement@ctbto.org and in line with Section “6. Request for Clarifications and Contacting the Commission” of the “Instructions for preparation and submission of Proposals” part of this RFP. 13 Are India-incorporated companies eligible to participate directly, or is an EU/Austrian establishment or local delivery partner required? Are consortium or specialist subcontractor arrangements permitted, and can relevant experience be demonstrated by a named specialist partner? Kindly refer to responses to Question 12 above. 14 Does the RFP require previous NET-VISA experience or specialist experience in seismic monitoring, nuclear-test monitoring, scientific algorithms or related systems? Previous NET-VISA experience is not required. Bidders may further refer to the contractor’s requirements Section 5 of the Annex B - ToR and requirements for the Team Roster Section 6 of the Annex B -ToR for the requirements. 15 What are the principal technologies, languages, architecture and call-off support expectations for the existing NET-VISA codebase? Kindly refer to the “Background” Section 2 of the Annex B ToR and to NET-VISA Software Description Document (At- tachment A) and IDC Database Schema-ODB (Attachment B) attached to these Clarifications and added as part of An- nex B ToR. Clarifications No. 1 - questions and answers - RFP No. 2026-0108/MOGAPI: Provision of Software Development and Maintenance of NET-VISA on a Call-Off Basis 7 16 Are there minimum turnover, insurance, security-clearance, personnel or on-site requirements, and is an estimated contract ceiling available? Is Basic- level UNGM registration sufficient, and should all formal questions and the proposal be submitted through the CTBTO procurement platform? Kindly refer to the contractor’s requirements Section 5 of the Annex B -ToR and response to question 10 above. For the procedure to submit requests for clarifications, kindly refer to response to question 12 above. Software Development and Maintenance of NET-VISA Ref No. in TOR QUALIFICATION REQUIREMENTS (PASS/FAIL) Requirements for the Contractor (PASS /FAIL) 5.1 The Bidder is a company/organization/institution established for at least 7 years. 5.2 The Bidder has demonstrated a minimum of 7 years of experience with all elements of the project management and software development lifecycle: eliciting and documenting business process flows, use cases, requirements, quality management plans, user acceptance testing, and end-user training. 5.3 The Bidder has demonstrated a minimum of 7 years of experience in developing, maintaining and managing software systems of comparable scientific complexity and operational criticality. The Bidder has demonstrated practical expertise in Bayesian methods, machine learning, scientific computing, large-scale data processing, performance optimization, and industrial C++ and Python development. 5.4 The Bidder has demonstrated a minimum of 7 years of experience in developing database-driven operational systems, automated testing, CI/CD, containerized deployment and maintenance of legacy scientific software. 5.5 The Bidder has confirmed that the working hours of its relevant staff (e.g. project manager and lead developers), assigned to a Software Team, will overlap at least three hours with the Commission’s working hours (9am to 5pm CET). 5.6 The Bidder has confirmed that under exceptional circumstances support will be fully available during the working hours of the Commission (9 AM – 5 PM Vienna local time (CET/CEST)), without additional costs. This would be no more than one week per year. 5.7 The Bidder has confirmed that English will be used for all written and oral communication with the Commission. 7.2 The Bidder has confirmed that they shall submit Sprint reports to document completed work, personnel involved, time spent, open issues and blockers. 6.1 The Bidder has proposed at least three (3) staff members and their CVs for the Team Roster and has confirmed that the Team Roster will be maintained throughout the duration of the Contract. Requirements for Staff and Team Roster (PASS /FAIL) Skills and experience for each staff on the Team Roster 6.4.2 Each proposed staff member has demonstrated a minimum of five years of recent professional experience within the last seven years using Python and C/C++ in substantial production-grade software projects of significant complexity. 6.4.3 Each proposed staff member has demonstrated a minimum of five years of recent professional experience within the last seven years with SQL and database programming, ideally using Oracle and/or PostgreSQL databases, including experience designing data access layers and data models for an application. 6.4.4 Each proposed staff member has demonstrated a minimum of three years of recent professional experience within the last seven years providing software maintenance and support services for complex and custom software systems. 6.4.5 Confirmation that each proposed staff member has good communication skills in written and spoken English. Skills and experience available amongst Team Roster staff 6.5 The Bidder has demonstrated that the following combined skills and experience are available in the proposed Team Roster; at a minimum, two (2) of these requirements under Section 6.5 "Skills and experience available amongst Team Roster staff" are met by each individual in the Team Roster, and all of these skills and experience are available in the Team Roster in aggregate. 6.5.1 The Bidder has demonstrated that the following skill and experience is available in the Team Roster: A minimum of three years of recent professional experience within the last seven years with CI/CD (continuous integration/continuous delivery) software deployment environments. 6.5.2 The Bidder has demonstrated that the following skill and experience is available in the Team Roster: A minimum of three years of recent professional experience within the last seven years with the elements of the software development lifecycle: documenting requirements, user acceptance testing, and end-user training. 6.5.3 The Bidder has demonstrated that the following skill and experience is available in the Team Roster: A minimum of three years of recent professional experience within the last five years as a Data Scientist, including practical experience in the development, validation and deployment of machine-learning or statistical models. 6.5.4 The Bidder has demonstrated that the following skill and experience is available in the Team Roster: Minimum of one year of recent professional experience within the last two years designing, developing or integrating agent-based or agentic software systems. No. 2.TECHNICAL REQUIREMENTS - EVALUATION CRITERIA AND METHOD Quality of the Proposal Max Points Factor Weighted score General Extent to which all aspects of the ToR have been addressed in sufficient detail and clarity. 5 4 20 General Understanding of the scope of work and the responsibilities of the Contractor 5 2 10 General The proposed Team Roster demonstrates the capacity to tackle Services described in Section 4 and is explained in sufficient detail. Curricula Vitae contain relevant information demonstrating the proposed personnel’s ability to perform the specific tasks under this Contract. The following information is provided for each staff member: Name, Role, Employed since, Type(s) of Service(s) from Section 4 in which the staff member may be involved, Curriculum Vitae 5 3 15 5.8 It is desirable that the Bidder has demonstrated access to subject-matter experts in seismology, hydroacoustic and infrasound who may be engaged when required. The Bidder has identified the available experts and provided information on their qualifications, areas of expertise and availability. 5 2 10 Experience of Team Roster Staff Max Points Factor Weighted score 6.4.2 Staff member demonstrates a minimum of five years of recent professional experience within the last seven years using Python and C/C++ in substantial production-grade software projects of significant complexity. 5 4 20 6.4.3 Each proposed staff member demonstrates a minimum of five years of recent professional experience within the last seven years with SQL and database programming, ideally using Oracle and/or PostgreSQL databases, including experience designing data access layers and data models for an application. 5 1 5 6.4.4 Each proposed staff member demonstrates a minimum of three years of recent professional experience within the last seven years providing software maintenance and support services for complex and custom software systems. 5 3 15 6.5.3 Peer-reviewed relevant publications indexed in the Web of Science Core Collection or Scopus, or published in the proceedings of conferences ranked A* in the ICORE 2026 Conference Rankings, are desirable. 5 3 15 6.5 Adequacy of the number and composition of proposed staff to ensure full coverage of all required skills and experience. Each team member shall meet at least two (2) requirements, while all requirements shall be covered collectively by the Team Roster. (6.5.1; 6.5.2; 6.5.3; 6.5.4) 5 1 5 TOTAL - Technical Evaluation 45 115 1. Technical Evaluation: The technical evaluation process will be done in two stages: 1) Stage 1: Technical proposals will first be evaluated against the mandatory requirements outlined in section 1 above, on a PASS/FAIL basis. Compliance with all mandatory requirements is required in order to pass stage 1 of the technical evaluation and to be further considered for stage 2 of the evaluation process; 2) Stage 2: The technical proposals that have passed stage 1 of the technical evaluation process, will be evaluated against the weighted criteria set forth in the evaluation matrix above. In order to pass this stage, bidders must obtain a minimum weighted score of 75 and in accordance with the scoring indicated in the “Table 2 – Scoring” below. Failing one of the criteria in Stage 2 will not result in the entire technical proposal failing; provided that the technical proposal obtains a minimum weighted score of 75 as already stated. TABLE 2 Points Scoring 0 Unsatisfactory - Response incomplete, inadequate and/or non-responsive to the criterion. Bidder does not clearly understand the criterion. 1-2 Weak - Does not meet the minimum requirements of the criterion 3 Good - Meets the minimum requirements of the criterion 4 Very good - Exceeds the minimum requirements of the criterion in some areas 5 Excellent - Exceeds the minimum requirements of the criterion in all areas 2. Financial and commercial evaluation Once the technical evaluation is finalized, the financial offers of the technically compliant bidders will be evaluated in accordance with the formula given below: X= Max Available Points * Y/Z Legend: X= points to be assigned to the offer being evaluated Y= price of the lowest priced, technically compliant offer Z= price of the offer being evaluated The weight of the technical and financial components is 60% and 40% respectively. The Contract will be awarded to the bidder who receives the highest combined score resulting from the technical and financial evaluations, subject to contractual acceptability. PASS/FAIL PASS/FAIL PASS/FAIL Attachment 2 - Rev. 4 Nov 2026 Evaluation Criteria and Method PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL Points PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL PASS/FAIL Only bidders who pass all above criteria will be considered for the point system evaluation (2nd stage) 6.4.1 Each proposed staff member has an advanced university degree in computer science, data science, mathematics, physics or a related field (with accompanying explanation). PASS/FAIL PASS/FAIL PASS/FAIL NET-VISA Software Description Document Nimar S. Arora1, Stuart Russell1,2, Heidi A. Kuzma1 April 20, 2026 1Bayesian Logic, Inc.; 2University of California, Berkeley Sponsored by the Preparatory Commission for the Comprehensive Nuclear Test-Ban Treaty Organization (CTBTO) under Contract No. 2012-2107, Contract No. 2016-2048, Contract No. 2019-1884, and Contract No. 2023-1715. 1 Abstract NET-VISA stands for NETwork processing - Vertically Integrated Seismic Analy- sis. The package is composed of a physics-based, probabilistic model and a heuristic inference algorithm designed to find the most probable set of seismic events which can explain a series of arrivals detected by a global seismic network. It has been extended to find events in any of three mediums – rock, air, and water and it supports seismic, hydro-acoustic, and infrasound sensors. The basic components of NET-VISA are a Generative Model (GM) and an Inference Algorithm (IA). The GM contains the infor- mation necessary to evaluate the probability of events and arrivals. The IA searches for the set of events that has the maximum a posteriori (MAP) probability - that is the one with highest probability given the input data. It works by proposing a set of possible events and iterating between associating arrivals and refining the time, location and magnitude of the events given the associated arrivals. This document is divided into three parts: an Introduction which gives an overview of NET-VISA and the problems it solves; a Scientific Description which details the mathematics and algorithms used by the NET-VISA package, including notes as to how it is implemented in the IDC-specific package, idcnetvisa: and a Practical Guide which gives details of the various scripts and utilities necessary to install and run idcnetvisa. A readme file is included with each distribution of idcnetvisa which describes in detail how to install the package and includes examples of how to run it. 2 Contents I Introduction 7 1 NET-VISA project 7 2 Deliverables 7 3 Statement of the problem 8 4 idcnetvisa 9 II Scientific Description 10 5 Method 10 5.1 Bayes’ rule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 5.2 Conditional independence . . . . . . . . . . . . . . . . . . . . . . . . . . 11 5.3 MAP world . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 5.4 Event and arrival scores . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 5.5 Training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 5.5.1 New stations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 5.5.2 Modifications to internal models which do not affect detections . 15 5.5.3 Fundamental changes . . . . . . . . . . . . . . . . . . . . . . . . 15 5.6 Inputs and outputs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 5.7 General program structure . . . . . . . . . . . . . . . . . . . . . . . . . . 15 6 Generative Model 17 6.1 Mathematical conventions . . . . . . . . . . . . . . . . . . . . . . . . . . 17 6.2 Probability of attributes of events and arrivals . . . . . . . . . . . . . . . 18 6.2.1 Relationship between an event and an associated arrival . . . . . 19 6.2.2 Probability of the attributes of an event . . . . . . . . . . . . . . 19 6.2.2.1 Event Medium . . . . . . . . . . . . . . . . . . . . . . . 19 6.2.2.2 Rate of Events . . . . . . . . . . . . . . . . . . . . . . . 21 6.2.2.3 Event Time . . . . . . . . . . . . . . . . . . . . . . . . . 21 6.2.2.4 Event Location . . . . . . . . . . . . . . . . . . . . . . . 21 6.2.2.5 Event Depth . . . . . . . . . . . . . . . . . . . . . . . . 22 6.2.2.6 Event Magnitude . . . . . . . . . . . . . . . . . . . . . . 22 6.2.3 Probability of the attributes of an associated arrival given an event 23 6.2.3.1 Predicted Arrival Time, Amplitude, Azimuth, Slowness 23 6.2.3.2 Event Derived Feature (EDF) sets . . . . . . . . . . . . 24 6.2.3.3 Data is Available . . . . . . . . . . . . . . . . . . . . . . 24 6.2.3.4 Detection is Made . . . . . . . . . . . . . . . . . . . . . 24 6.2.3.5 Measured Arrival Amplitude . . . . . . . . . . . . . . . 28 6.2.3.6 Hydro-Acoustic Arrival Energy . . . . . . . . . . . . . . 30 6.2.3.7 Hydro Fraction Time . . . . . . . . . . . . . . . . . . . 30 6.2.3.8 Measured Arrival Azimuth . . . . . . . . . . . . . . . . 32 6.2.3.9 Measured Arrival Slowness . . . . . . . . . . . . . . . . 32 3 6.2.3.10 Arrival Phase Label . . . . . . . . . . . . . . . . . . . . 34 6.2.3.11 Infrasound Center Frequency . . . . . . . . . . . . . . . 34 6.2.3.12 Infrasound Family Size . . . . . . . . . . . . . . . . . . 35 6.2.3.13 Infrasound Consistency . . . . . . . . . . . . . . . . . . 35 6.2.3.14 Infrasound Duration / Family Size . . . . . . . . . . . . 35 6.2.4 Probability of the attributes of false arrivals . . . . . . . . . . . . 37 6.2.4.1 Number of false arrivals. . . . . . . . . . . . . . . . . . . 37 6.2.4.2 Measured Arrival Time (false arrival) . . . . . . . . . . 37 6.2.4.3 Measured Arrival Amplitude (false arrival) . . . . . . . 39 6.2.4.4 Measured Arrival Energy (false arrival) . . . . . . . . . 39 6.2.4.5 Measured Arrival Azimuth (false arrival) . . . . . . . . 40 6.2.4.6 Measured Arrival Slowness (false arrival) . . . . . . . . 40 6.2.4.7 Arrival Phase Label (false arrival) . . . . . . . . . . . . 42 6.2.4.8 Infrasound Clutter . . . . . . . . . . . . . . . . . . . . . 42 6.2.4.9 Hydroacoustic Clutter . . . . . . . . . . . . . . . . . . . 42 6.2.5 Probability of the attributes of coda arrivals . . . . . . . . . . . . 44 6.2.5.1 Determining which arrivals in SEL3 are coda arrivals to construct a coda training set . . . . . . . . . . . . . . . 44 6.2.5.2 Detection is made (coda arrival) . . . . . . . . . . . . . 45 6.2.5.3 Measured Arrival Time (coda arrival) . . . . . . . . . . 46 6.2.5.4 Measured Arrival Amplitude (coda arrival) . . . . . . . 46 6.2.5.5 Measured Arrival Azimuth (coda arrival) . . . . . . . . 47 6.2.5.6 Measured Arrival Slowness (coda arrival) . . . . . . . . 48 6.2.5.7 Arrival Phase Label (coda arrivals) . . . . . . . . . . . . 48 6.3 Probability of a world . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 6.3.1 Probability of an event . . . . . . . . . . . . . . . . . . . . . . . . 49 6.3.2 Joint probability of a set of events . . . . . . . . . . . . . . . . . 49 6.3.2.1 Probability of the total number of events in a time period 49 6.3.2.2 Probability of a set of events . . . . . . . . . . . . . . . 49 6.3.3 Probability of an arrival . . . . . . . . . . . . . . . . . . . . . . . 49 6.3.4 Joint probability of sets of arrivals . . . . . . . . . . . . . . . . . 50 6.3.4.1 Joint probability of a set of associated arrivals . . . . . 50 6.3.4.2 Joint probability of a set of false arrivals . . . . . . . . . 51 6.3.4.3 Joint probability of a set of coda arrivals . . . . . . . . 51 6.3.5 Probability of a world . . . . . . . . . . . . . . . . . . . . . . . . 51 7 Inference Algorithm 52 7.1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 7.2 Event and Arrival scores . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 7.3 Algorithm . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 7.3.1 Birth Move . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 7.3.2 Improve-arrival . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 7.3.3 Improve Event . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 7.3.4 Death-Move . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 7.3.5 Finish . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 III Practical Guide for idcnetvisa 56 4 8 Scripts and Utilities 57 8.1 readme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 8.2 manifest.in . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 8.3 pyproject.toml . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 8.4 requirements.txt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 8.5 setup.py . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 8.6 idcshema.sql . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 8.7 makeinfraprior.py . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 8.8 populate_isc.py . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 8.9 Installed Scripts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 8.9.1 idcnetvisa_learn . . . . . . . . . . . . . . . . . . . . . . . . . 59 8.9.2 idcnetvisa_infer . . . . . . . . . . . . . . . . . . . . . . . . . 61 8.9.3 idcnetvisa_infer_bg . . . . . . . . . . . . . . . . . . . . . . . 68 8.9.4 idcnetvisa_debug . . . . . . . . . . . . . . . . . . . . . . . . . 68 8.9.5 idcnetvisa_match . . . . . . . . . . . . . . . . . . . . . . . . . 69 8.9.6 idcnetvisa_add_site . . . . . . . . . . . . . . . . . . . . . . . 73 8.9.7 idcnetvisa_merge_model . . . . . . . . . . . . . . . . . . . . 74 8.9.8 idcnetvisa_visualize . . . . . . . . . . . . . . . . . . . . . . . 74 8.9.9 idcnetvisa_visualize_parquet . . . . . . . . . . . . . . . . 74 8.9.10 idcnetvisa_runner . . . . . . . . . . . . . . . . . . . . . . . . 75 8.9.11 idcnetvisa_compare_models . . . . . . . . . . . . . . . . . . 77 8.10 Docker Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 8.10.1 NET-VISA Image . . . . . . . . . . . . . . . . . . . . . . . . . . 79 8.10.2 PREREQ Image . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 9 Learning Schedule 80 10 Model Schema 81 10.1 Arrival SNR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 10.2 Arrival SNR header . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 10.3 Arrival Amplitude . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 10.4 Arrival Azimuth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 10.5 Arrival Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 10.6 Arrival Slowness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82 10.7 Arrival Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 10.8 Event Detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 10.9 Event Magnitude . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83 10.10SNR of Noise Arrivals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 10.11Amplitude of Noise Arrivals . . . . . . . . . . . . . . . . . . . . . . . . . 84 10.12Phase of Noise Arrivals . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84 10.13Slowness of Noise Arrivals . . . . . . . . . . . . . . . . . . . . . . . . . . 84 10.14Hydro Fraction Time . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 10.15Info . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 10.16Rate of Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 10.17False Detection Rate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 10.18Phase . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85 10.19QC . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 10.20Coda Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86 5 10.20.1 Coda Arrival Azimuth . . . . . . . . . . . . . . . . . . . . . . . . 86 10.20.2 Coda Detection Probability . . . . . . . . . . . . . . . . . . . . . 86 10.20.3 Coda Arrival Phase . . . . . . . . . . . . . . . . . . . . . . . . . . 87 10.20.4 Coda Arrival SNR . . . . . . . . . . . . . . . . . . . . . . . . . . 87 10.20.5 Coda Arrival Amplitude . . . . . . . . . . . . . . . . . . . . . . . 88 10.20.6 Coda Arrival Slowness . . . . . . . . . . . . . . . . . . . . . . . . 88 10.20.7 Coda Arrival Time . . . . . . . . . . . . . . . . . . . . . . . . . . 88 10.21Site . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89 11 Parquet Data File Schema 89 11.1 Assoc Parquet File (assoc.parquet) . . . . . . . . . . . . . . . . . . . . . 89 11.2 Arrival Parquet File (arrival.parquet) . . . . . . . . . . . . . . . . . . . . 91 11.3 Site_Up Parquet File (site_up.parquet) . . . . . . . . . . . . . . . . . . 92 11.4 Event_Det Parquet File (event_det.parquet) . . . . . . . . . . . . . . . 93 11.5 Site and Phase CSV files (phase.csv and site.csv) . . . . . . . . . . . . . 93 12 Event Bulletin Database Schema 94 13 Release Management at the IDC 100 13.1 Version Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 100 13.2 Testing and Deployment . . . . . . . . . . . . . . . . . . . . . . . . . . . 101 13.2.1 Release Candidate . . . . . . . . . . . . . . . . . . . . . . . . . . 101 13.2.2 DEVLAN and OPS . . . . . . . . . . . . . . . . . . . . . . . . . . 101 13.2.3 QANDEF . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 A List of Symbols 103 B Algorithms 104 C Debugging Missed Events 106 C.1 Bulletin Matching and Missed Event Detection . . . . . . . . . . . . . . 106 C.1.1 Running the Matching . . . . . . . . . . . . . . . . . . . . . . . . 106 C.1.2 Parameter File Configuration . . . . . . . . . . . . . . . . . . . . 107 C.1.3 Understanding Match Output . . . . . . . . . . . . . . . . . . . . 107 C.2 Event Score Diagnostics . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 C.2.1 Running the Score Diagnostic . . . . . . . . . . . . . . . . . . . . 109 C.2.2 Understanding Score Diagnostic Output . . . . . . . . . . . . . . 109 6 Part I Introduction 1 NET-VISA project NET-VISA stands for NETwork processing - Vertically Integrated Seismic Analysis. It is a software package composed of a physics-based, probabilistic model and a heuristic in- ference algorithm designed to find the most probable set of events which [... truncated ...]