Open-Earth-Monitor Global Workshop 2026

Leandro Parente

Leandro Parente is a senior researcher at OpenGeoHub Foundation with more than 15 years of experience in processing Earth Observation (EO) data and developing Machine Learning (ML) pipelines for producing continental and global maps.

The speaker's profile picture

Do you accept that a video-recording of your talk is published under CC-BY license via https://av.tib.eu? – yes

Sessions

10-08
18:00
45min
Accessing global multi-decade Landsat cloud-free time-series in CDSE
Leandro Parente

In this workshop, the participants will have access to harmonized, analysis-ready, gap-filled and complete Landsat global mosaics from 1997 onward in cloud-optimized GeoTIFF (COG) format (130 TB of data) in CDSE (https://browser.stac.dataspace.copernicus.eu). Spanning over 25 years and structured in 7 spectral bands (RGB, NIR, SWIR-1, SWIR-2 and thermal), this data is instrumental for long-term monitoring applications of land cover change, soil proprieties, vegetation productivity, land degradation, vegetation height and other environmental characteristics. The global mosaics were produced via the Time-Series Iteration-free Reconstruction (TSIRF) framework over the entire Global Land Analysis and Discovery (GLAD) ARD Landsat archive (https://doi.org/10.7717/peerj.18585). Participants will learn about the implemented methodologies and use several python libraries (stacstac, scikit-map) JupyterLab.

Soil, water and agriculture
Room 18
10-07
17:00
15min
Global monitoring of grassland and livestock: Current status, challenges and next steps
Leandro Parente

While forest monitoring has reached high levels of maturity, grassland ecosystems remain a critical "blind spot" in global conservation. To address this, Global Pasture Watch (GPW) has established a comprehensive baseline using 30m multi-decadal datasets (2000–2022) covering grassland extent, vegetation height, and livestock density. However, the inherent heterogeneity and rapid seasonality of these landscapes present significant current challenges for traditional pixel-based classification. To overcome these barriers, our next steps involve transitioning to next-generation machine learning models that utilize Sentinel-2 spatial-temporal embeddings. By moving beyond simple spectral signatures to rich, high-dimensional latent representations, we can better capture the nuances of managed vs. natural grasslands and monitor Gross Primary Productivity (GPP) with unprecedented precision. This evolution in our workflow aims to deliver near-real-time, actionable insights, transforming how we track land-use conversion and guide sustainable restoration across the world’s most vulnerable non-forest biomes.

Soil, water and agriculture
Room 18