Skip to content

Data Processing Documentation

Overview

At every level of ecological organisation, environmental changes influence biodiversity. Spectral signals captured through satellite remote sensing measure these environmental changes. Consequently, remote sensing has become a vital tool for biodiversity monitoring, conservation, and policymaking. However, it is important to recognise that the interplay between biodiversity and the environment is bidirectional. While the environment undoubtedly influences biodiversity, biodiversity itself also shapes the environment. This suggests that biodiversity data likely conceals valuable insights into associated environmental conditions and fluctuations. Remarkably, the utilisation of biodiversity data to model the environment has yet to receive widespread attention. Therefore, our proposed project aims to bridge this gap by harnessing both biodiversity data and deep learning to model the environment, as represented by multispectral satellite data. Given that multispectral data provides information on environmental conditions that inform spatiotemporal predictions of biodiversity, we hypothesise that the inverse is also possible. Additionally, given the importance of functional diversity in maintaining and promoting ecosystem functioning, we plan to use our modelling framework to explore scenarios of environmental change in the absence of species belonging to particular functional groups. We will provide a foundation to analyse the feedback between biodiversity and environmental change, enhancing our ability to communicate the impacts of biodiversity loss, to understand ecosystem dynamics, and to improve monitoring capabilities related to the post-2020 Global Biodiversity Framework. We aim to harness geotagged biodiversity data in the modelling of environmental conditions represented by multispectral satellite data. Additionally, given the importance of functional diversity in maintaining and promoting ecosystem functioning, we will use our modelling framework to explore scenarios of environmental change in the absence of species belonging to particular functional groups. We will develop and test our approach in areas targeted by the North American Breeding Bird Survey (BBS), using data obtained between 2015 and 2021.

Data Sources

Our project will depend exclusively on open-access data. We will use: * Landsat (LT) surface reflectances, specifically from LT8 and LT9 sensors. * Differences data sources on the occurrence of bird species, namely BBS, GBIF, and Xeno-canto * Land cover data for all of North American (i.e. North American Environmental Atlas) * Information on functional traits for every known bird species (i.e. AVONET) * Data on species function interactions for every bird species (i.e. GloBI)

CyVerse Discovery Environment

Instructions for setting up and using the CyVerse Discovery Environment for data processing. Tips for cloud-based data access and processing.

Data Processing Steps

  1. Acquisition of Landsat acquisitions during BBS observation months, and over BBS sites
  2. Derivation of annual, multispectral, pheonological composites centered around BBS surveying periods
  3. Extraction and harmonization of species observations from BBS, GBIF, Xeno-Canto into tabular checklists
  4. Detection of causal interactions, and detection of causal effect sizes, between changes species compositions and changes multispectral data
  5. Use of step 4 to reduce list of test sites for subsequent modelling
  6. Development of GAN models to predict multispectral satellite data (step 2) based on species compositions (steps 3) for the selected test sites (step 5)
  7. Use of spectral decomposition to distinguish proportions of land cover types based on national land cover data

Using GDAL VSI

Guidance on using GDAL VSI (Virtual System Interface) for data access and processing. Example commands or scripts:

gdal_translate /vsicurl/http://example.com/data.tif output.tif

Cloud-Optimized Data

Advantages of using cloud-optimized data formats and processing data without downloading. Instructions for such processes.

Data Storage

(Information on storing processed data, with guidelines for choosing between the repository and CyVerse Data Store) Satellite data: Species observation data:

Best Practices

Recommendations for efficient and responsible data processing in the cloud. Tips to ensure data integrity and reproducibility.

Challenges and Troubleshooting

Common challenges in data processing and potential solutions. Resources for troubleshooting in the CyVerse Discovery Environment.

Conclusions

Summary of the data processing phase and its outcomes. Reflect on the methods used.

References

Citations of tools, data sources, and other references used in the data processing phase.

Last update: 2024-10-01