Integrating Complementary Visualizations within a Data Order Interface: The STORM Swath-Based Analysis Tool

There are many ways to build an interface for atmospheric data. The simplest is to make files available via FTP site and allow researchers to “go wild”. This often works well for automated script retrievals where all the data is desired (e.g. near-realtime data). More robust systems add features like an order mechanism that allows custom date ranges and selection by geographic region of interest. Some enable subsetting of the files to reduce download and storage volume. Others allow for static or interactive visualization of some aspect of the data to streamline the decision-making process. When all of these features are available, developers have a responsibility to integrate them in a way that is user-friendly and reduces the burden on the researcher.

This paper will discuss these issues in the context of the STORM Swath-Based Analysis Tool (https://storm.pps.eosdis.nasa.gov/storm/Analysis.jsp), a multi-featured interface for exploring and acquiring precipitation data from 19 different satellites spanning an archive that covers the past 21 years as part of the Global Precipitation Measurement (GPM) Mission and Tropical Rainfall Measurement Mission (TRMM). Here, users can compare overflights of multiple precipitation satellites simultaneously with an interactive time series. From this they can make determinations about which swaths will best serve their research. Users can preview these overflights with two distinct visualization tools. Finally, they can then download geographically subset swaths that have precipitation rate characteristics matching their research needs.

The three visualizations involved with this tool serve complimentary purposes. The primary one, the time series of aggregate statistics for up to six different satellites, has some intrinsic interactivity. The JavaScript library it is built upon, chart.js, comes with the ability to inspect values via mouseover, to append points (necessary for the progressive loading as overflights are processed), and to swap in and out different data sets. This last aspect ensures that the user can choose from a variety of statistical values mined from the individual swaths, as well as display different combinations of the observing platforms for quick comparison.

Linked to clicking the data points on the time series are two other visualization tools: THOR Online and STORM Virtual Globe. The former has been around since the TRMM era as a tool to robustly image any field within the data products. Many of these precipitation products have various diagnostic and ancillary variables that are useful to researchers and developers. THOR Online can visualize these fields without requiring any file or application download. Even 1D and non-geospatial 2D fields can be displayed in THOR Online, enabling complete investigation of the contents of the file.

STORM Virtual Globe, on the other hand, displays only a single variable, precipitation rate. The benefit is that it does so on and above the surface of a three-dimensional globe. Again using a JavaScript library, in this case CesiumJS, individual pixel values from these data products are rendered as colored circles with a fully-interactive camera to zoom and pan around the data. Users can mouse over any of these circles and get the raw precipitation rate value. Platforms with multiple instruments in the archive, such as GPM with the GPM Microwave Imager (GMI) and the Dual-frequency Precipitation Radar (DPR) have swaths visualized simultaneously. In the case of GPM, this displays both the 2D surface precipitation rates from GMI and the 3D precipitation rates from DPR to enable better understanding of the spatial and vertical extents of the precipitation.

The paper will feature all three of these visualization applications and discuss how they complement one another to create a suite that streamlines the precipitation data acquisition process. It will highlight use cases for the Swath-Based Analysis Tool including how to discover intense rainfall events and how to find possible isolated precipitation events in otherwise dry regions.