Practical demonstration: Operation applications

In this session, participants will develop the capacity to effiently summarise and map spatio-temporal dynamics represented in geospatial datasets over regions of interest using open-source tools. Both raster and vector data will be in applied. These capacities are useful for research, as well as for ’distilling’ complex datasets for policy- and other decision-making purposes.

Visit site

Miao Zhang and Hongwei Zeng - Chinese Academy of Sciences

Remote sensing is an effective means of crop monitoring, but requires in situ data for calibration and validation. However, the traditional way of in situ data collection is known to be time-consuming, laborious, and expensive. Additionally, the commonly used harvesting method for actual yield measurement is destructive and inefficient, further limiting the acquisition of ground truth yield data.To overcome these limitations, crowd-sourced data acquisition methods using cell phones have emerged as the most efficient way on a large scale. During this hands-on training program focuses on two specific applications - GVG and FieldWatch. GVG enable efficient collection of crop type information and FieldWatch allows non-destructive sensing of crop yields based on AI and computer vision techniques. Training on the installation, registration, and using of the APPs will be provided. Both Apps are free for all.

The GEO Knowledge Hub is the digital repository for storing and sharing the knowledge produced by the GEO community and its various initiatives. This session aims to introduce the GEO Knowledge Hub technologies and concepts and how the community can use the digital repository to share and preserve knowledge.

In this session, we will demonstrate the practical application of F.A.I.R. principles, through the encoding of heterogeneous in situ data into standardized NetCDF files. We will show how the use of standardized and widely supported formats and the integration of metadata facilitate data manipulation and open the way to new tools and uses (QC, visualization, ...). The tools we developed and demonstrate are open source. Although this work focuses mainly on solar irradiation data, it can be applied to other sectors to encode and normalize your own data.

This session aims at showing participants how to analyze crop or rangelands conditions of a country using the ASAP online system (i.e. understanding its automatic warnings derived from coarse resolution MODIS NDVI, CHIRPS rainfall and JRC crop water satisfaction index data and zooming to the field level with 10m resolution S2 data). Participants will need an access to internet to be able to explore ASAP.We will also show how the JRC standalone software CST (Crop Statistical Tool) can be used for forecasting crop yield through the analysis of the (linear) relationships between yield statistics and yield indicators (e.g. derived from ASAP at the 10 daily timestep) at region level during the crop season.

The Digital Earth Africa (DE Africa) workshop will cover the introduction to the DE Africa platform https://www.digitalearthafrica.org, a continental platform, which provides analysis ready free and open earth observation data to help in decision making across all sectors from agriculture, water, and natural resource management. DE Africa provides a routine, reliable and operational service, using Earth observations to deliver decision-ready products enabling policy makers, scientists, the private sector and civil society to address social, environmental and economic changes on the continent and develop an ecosystem for innovation across sectors.At the end of the workshop, participants will be empowered to be able to manage their natural resources towards achieving national development goals, and Sustainable Development Goals (SDGs).

The GEOGloWS ECMWF Streamflow Model is a Hydrologic Model which provides forecasted and historically simulated river dischargeThe GEOGloWS global streamflow forecasting service allows local stakeholders to focus on solving water management problems such as flooding, drought, and water/food security issues by providing the water intelligence they need to make decisions. It also benefits the global economy by providing water intelligence to sectors that need to make high-risk investment decisions such as the insurance and reinsurance industries.

Petabytes of Earth observation data are now open and free, making the full extent of image archives available for researchers and experts. Remote sensing experts can now track environmental change using satellite image time series. Using image time series, analysts make best use of the full extent of big Earth observation data collections, capturing subtle changes in ecosystem health and condition and improving the distinction between different land classes.This hands-on session introduces ’sits’, an open-source R package for land use and land cover classification of big Earth observation data using satellite image time series. Users build regular data cubes from cloud services such as Amazon Web Services, Microsoft Planetary Computer, Brazil Data Cube, Swiss Data Cube, NASA Harmonised Landsat-Sentinel, and Digital Earth Africa. The “SITS” API provides functions for measuring quality of training samples, classifying cubes using machine learning and deep learning, and post-processing using Bayesian smoothing. It also includes spatial uncertainty measures, support for active learning, methods for ensemble prediction and for accuracy measures with best practices. Thus, “sits’ provides and end-to-end solution to image time series analysis.

Visit site