Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

On-Demand Earth System Data Cubes

David Montero, César Aybar, Chaonan Ji, Guido Kraemer, Maximilian Söchting, Khalil Teber, Miguel D. Mahecha

TL;DR

Condo, an open-source Python tool designed for easy generation of AI-focused ESDCs, is introduced, requiring only central coordinates, spatial resolution, edge size, and time range.

Abstract

Advancements in Earth system science have seen a surge in diverse datasets. Earth System Data Cubes (ESDCs) have been introduced to efficiently handle this influx of high-dimensional data. ESDCs offer a structured, intuitive framework for data analysis, organising information within spatio-temporal grids. The structured nature of ESDCs unlocks significant opportunities for Artificial Intelligence (AI) applications. By providing well-organised data, ESDCs are ideally suited for a wide range of sophisticated AI-driven tasks. An automated framework for creating AI-focused ESDCs with minimal user input could significantly accelerate the generation of task-specific training data. Here we introduce cubo, an open-source Python tool designed for easy generation of AI-focused ESDCs. Utilising collections in SpatioTemporal Asset Catalogs (STAC) that are stored as Cloud Optimised GeoTIFFs (COGs), cubo efficiently creates ESDCs, requiring only central coordinates, spatial resolution, edge size, and time range.

On-Demand Earth System Data Cubes

TL;DR

Condo, an open-source Python tool designed for easy generation of AI-focused ESDCs, is introduced, requiring only central coordinates, spatial resolution, edge size, and time range.

Abstract

Advancements in Earth system science have seen a surge in diverse datasets. Earth System Data Cubes (ESDCs) have been introduced to efficiently handle this influx of high-dimensional data. ESDCs offer a structured, intuitive framework for data analysis, organising information within spatio-temporal grids. The structured nature of ESDCs unlocks significant opportunities for Artificial Intelligence (AI) applications. By providing well-organised data, ESDCs are ideally suited for a wide range of sophisticated AI-driven tasks. An automated framework for creating AI-focused ESDCs with minimal user input could significantly accelerate the generation of task-specific training data. Here we introduce cubo, an open-source Python tool designed for easy generation of AI-focused ESDCs. Utilising collections in SpatioTemporal Asset Catalogs (STAC) that are stored as Cloud Optimised GeoTIFFs (COGs), cubo efficiently creates ESDCs, requiring only central coordinates, spatial resolution, edge size, and time range.
Paper Structure (9 sections, 3 equations, 4 figures, 1 table)

This paper contains 9 sections, 3 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Framework
  3. ESDC characterisation
  4. ESDC construction
  5. Bounding box calculation
  6. ESDC creation
  7. Attributes writing
  8. Showcase
  9. Conclusions

Figures (4)

  • Figure 1: cubo's ESDC characterisation. Note that in this representation, $t_s$ and $t_e$ represent values from the temporal dimension, while $\phi$ and $\lambda$ represent coordinates associated with the spatial centre of the cube.
  • Figure 2: Overview of the workflow for building an ESDC inside cubo. This diagram presents a high-level summary of the ESDC construction process.
  • Figure 3: Examples of ESDCs generated using cubo. Each row in the figure depicts ESDCs' example timesteps. The first column shows the parameters used and the location of the central coordinates of the ESDC. The first row displays an ESDC generated from Landsat-8 Collection 2 Level 2 data, capturing a fire event in Indonesia. The second row shows an ESDC constructed from Sentinel-1 Radiometrically Terrain Corrected (RTC) data, highlighting a flood event in Pakistan. The third row presents an ESDC based on MODIS 17A2HGF gap-filled 8-day GPP data, depicting a forest area in Germany. The final row features an ESDC created using data from the ESA CCI LC product, focused on Brasil.
  • Figure 4: Example of ESDCs aligned with a common spatio-temporal grid. The displayed MODIS datasets are a) FPAR (15A2H), b) Red SR (09A1), c) Enhanced Vegetation Index (EVI, 13Q1), d) NIR SR (09A1), e) Normalised Difference Vegetation Index (NDVI, 13Q1), f) Nighttime LST (11A2), g) Daytime LST (11A2), h) LAI (15A2H), and i) GPP (17A2HGF). The ESDCs were rendered using lexcubesoechting2023lexcube.