Data management and the configuration of analysis environments represent critical and time-consuming phases in the research lifecycle.
The IPSES Virtual Research Environment (VRE) was developed to overcome these challenges, offering researchers an integrated infrastructure that bridges the gap between data discovery and processing.
The VRE is an advanced service, available to registered users, designed to streamline the computational workflow and facilitate reproducible analysis.
VRE Functionality and Architecture
The IPSES Virtual Research Environment is based on JupyterLab, an interactive web application for creating and sharing computational documents (known as "notebooks"), code, and data. In practice, the VRE provides each user with a personal, multi-session, and persistent instance of this environment.
This workspace, accessible via a browser, serves as a private computational laboratory where it is possible to integrate data from the IPSES portal with external datasets provided by the user. The persistence of the environment ensures that files, notebooks, scripts, and custom Python libraries remain available between sessions, ensuring the continuity and reproducibility of the research work.
The VRE offers a powerful, flexible, and reproducible computing environment. Its main features include:
-
Integration with the IPSES Portal: A direct interface allows users to transfer data references (bookmarks) selected in the IPSES GUI directly into the VRE workspace, eliminating the need for manual downloads.
-
Programmatic Data Access: The VRE provides tools that facilitate interaction with the platform's services, allowing for the programmatic retrieval of data associated with bookmarks for use in custom scripts.
-
Analysis and Visualization Tools: The environment is pre-configured with standard libraries for data analysis (pandas) and visualization (plotly, ipyleaflet), enabling the rapid generation of plots and interactive maps.
-
Extensible Environment: Users have the ability to install additional Python libraries via pip, customizing the environment to meet specific analysis requirements and ensuring the flexibility needed for various research methodologies.
Workflow:
-
Data Selection: The user identifies and selects datasets of interest through the IPSES GUI, saving them as bookmarks.
-
Transfer to the Analysis Environment: Using the "Share" function in the GUI, the data references are made available within the VRE.
-
Accessing the VRE: The user accesses their JupyterLab instance from their personal area.
-
Data Processing: Using a Python script in a Jupyter Notebook, the user loads the data from the bookmarks via the features provided by the platform. The data can be combined with other datasets (e.g., local CSV files uploaded to the user space).
-
Analysis and Visualization: The data is processed and analyzed. Results can be visualized, for example, by generating station distribution maps or time-series plots.