Remotely Sensed High-Resolution Global Cloud Dynamics for Predicting Ecosystem and Biodiversity Distributions

Area of Science:

• Ecology
• Remote Sensing
• Climatology

Background:

• Ecological processes like reproduction and survival are influenced by cloud cover.
• Assessing cloud cover's ecological importance at fine spatial scales has been limited.
• Cloud dynamics offer potential for habitat delineation and species distribution prediction.

Purpose of the Study:

• To develop near-global, fine-grain monthly cloud frequency data.
• To analyze spatiotemporal cloud cover dynamics at unprecedented complexity.
• To demonstrate the utility of cloud-derived metrics for ecological applications.

Main Methods:

• Utilized 15 years of twice-daily Moderate Resolution Imaging Spectroradiometer (MODIS) satellite imagery.
• Developed monthly cloud frequencies at approximately 1 km spatial resolution.
• Compared cloud-derived metrics with interpolated climate data for habitat and species distribution models.

Main Results:

• Revealed previously undocumented global complexity in spatiotemporal cloud cover dynamics.
• Demonstrated significant geographic heterogeneity in cloud cover.
• Showed that direct, observation-based cloud metrics improve ecological predictions with reduced spatial autocorrelation.

Conclusions:

• Fine-grain, observation-based cloud data are essential for accurate ecological monitoring.
• Remote sensing provides a powerful tool for understanding biodiversity and ecosystem properties globally.
• Cloud-derived metrics enhance habitat and species distribution models compared to interpolated climate data.