Development of GIS Products and Analysis Plugins for Hayabusa2 Observational Data

1. Introduction

Hayabusa2 is equipped with a variety of scientific instruments and landers. The observational data obtained from them are available via Planetary Data System (PDS) by NASA and Data Archives and Transmission System (DARTS) by JAXA. However, the status of higher-level, analysis-ready product development and the data formats are different across instruments, making multi-instrument analysis challenging and often requiring specialized knowledge. To address this, the JAXA Hayabusa2# ^[1] International Visibility Enhancement Project (https://hayabusa2visibility.jaxa.jp) is developing Geographic Information System (GIS)-format data products to promote the usability of Hayabusa2 observational data. By formatting the data for use in GIS software, this approach is expected to enable multi-instrument analysis while lowering the barrier to data utilization and analysis.This presentation introduces GIS products and an experimental QGIS plugin designed for data analysis.

2. Development of the GIS products

This study developed two GIS-compatible data formats: raster (GeoTIFF) and vector (GeoPackage). These formats are broadly supported by standard GIS software, such as QGIS and ArcGIS, and can be easily viewed and edited. As both formats contain geographic coordinate information, geospatial analyses through comparisons between different observation dates and instruments at the same location are possible. In addition, users can analyze the data quickly and easily by utilizing built-in GIS tools, such as raster calculators and spatial statistics, without the need for additional coding. A raster data (GeoTIFF) has a regular digital picture like a JPEG or a normal TIFF file. Besides the pixel data, it includes the georeferenced information such as location, scale, projection, and coordinate system.  Vector data consist of geometries (lines, polygons, etc.) and attribute information. We employ this format to NIRS3^[2,3]: Attribute table can store various types of information—such as observation conditions and reflectance values at each wavelength —associated with corresponding footprint. This structure enables users to examine detailed information for each observation footprint and extract data subsets based on specific observation conditions. Figure 1 illustrates how a NIRS3 footprint outlined in red corresponds to the records selected in the attribute table, demonstrating the linkage between spatial features and tabular data. The NIRS3 GIS product was generated using geometry information calculated with SPICE kernels and L2 products stored in FITS format. For the data conversion process, Python libraries such as Shapely, GeoPandas, and Rasterio^[4] were used.

Figure 1. Vector product examples of NIRS3 footprints. 

3. Development of plugin for data analysis

QGIS ^[5] is a widely used open-source GIS application that offers many powerful features. It is particularly used for the visualization and analysis of Earth satellite data. In this study, we apply QGIS to the analysis of small body data, thereby making it easier for a broad range of researchers to work with spacecraft observation data. We present the experimental development of a QGIS plugin based on GIS-formatted NIRS3 data which has three functions, spectral profile visualization, spectral slope calculation, and map data generation. Providing the tool as QGIS plugin enables users to easily analyze observation data, without specialized knowledge of QGIS or Python programming.

4. GIS Analysis Example

Expected use cases for the GIS product and plugin functions are presented below.

・Filtering and Extraction of Observational Data

Observation data that meet specific conditions can be extracted. This function enables users to extract the specific condition data and examine the distribution and correlation of these data with terrain features. This can be performed using existing GIS software.

・Visualization of spectral profile

We developed a plugin function to visualize the spectral profile at specific locations and compare profiles across multiple user-selected points (Fig. 2).

・Calculate spectral slope and map data generation

We developed plugin functions to calculate spectral slope and generate global maps by averaging values such as spectral slope and reflectance, derived from overlapping footprints, helping users visually assess overall trends across the surface. As an example, Figure 3 shows the resulting spectral slope map calculated from 2.1 µm to 2.5 µm using the least squares method.

Figure 2. Plugin example of spectral profile visualization

Figure 3. Plugin example for spectral slope calculation and map generation

5. Future work

We are preparing for the release of the GIS-format data and the distribution of the plugin, as well as developing GIS-format data for other instruments and exploring enhancements to the plugin’s functionality. Furthermore, we hold Hayabusa2 data analysis workshops for further reaching out to users. These ongoing efforts aim to promote broader and more effective use of the Hayabusa2 observation data.

Acknowledgements: This work is supported by the JAXA Hayabusa2# International Visibility Enhancement Project.

References: [1] Hirabayashi, M. et al., 2021, Advances in Space Research 68, 1533-1555. [2] Iwata, T. et al., 2017, Space Science Reviews 208, 317. [3] Kitazato, K. et al., 2019, Science 364(6437), 272. [4] Python libraries such as GeoPandas (Jordahl et al., 2023), Shapely (Gillies et al., 2023), and Rasterio (Gillies et al., 2023) were used in the data conversion process. [5] QGIS Development Team. (2024). QGIS Geographic Information System. Open Source Geospatial Foundation Project. http://qgis.org