Get familiarized with current state-of-art opportunities offered by open source tools, especially ILWISPy, using both Earth Observation (EO) images and derived products.
The way currently remote sensing images, derived products and data is provided has changed the way how this information is being processed. Traditionally a stand-alone system was used with several graphical user interface-based software tools to process and analyse the data. For efficient processing nowadays, information from the cloud can be accessed, and processing is done using capabilities, like the use of a virtual research environment (VRE), provided by these cloud-based service providers. Once the analysis has been completed in these cloud VRE’s, the final products are downloaded to be integrated with existing local data sources for final analysis and visualization.
The course starts with an overview of the various operational EO satellite data and information access services and catalogues. This is followed by an overview of how ILWISPy operations can be used for interactive data science and scientific computing in conjunction with other relevant Python packages. Data applicable for various application domains will be covered. Finally, an assignment based on a topic of interest, applying the knowledge gained and tools introduced, will complete the course.
For whom is this course
This course is intended for professionals and researchers/lecturers with a background in hydrology, meteorology or water resources and environmental sciences in general, who would like to know more about the available Earth Observations data and products while simultaneously upgrading their multi-temporal data acquisitions and processing/analysis skills, using open Python based geo-information processing software tools like ILWISPy to supplement their Graphical User Interface (GUI) based GIS and EO software skills. The course offers possibilities to change from local data processing to cloud-based data processing and visualization of the results using common GUI software tools. Also, professionals already familiar with a GUI-based version of ILWIS are welcome and will experience the seamless transition to the new scripting-based geo-processing capabilities offered. The use of the open and free tools, in combination with course materials provided are a good starting point for lecturers at centres of higher learning to introduce these within their own curricula.
During the course participants will be familiarized with ample opportunities to apply ILWISPy functionality using a wide and readily available range of Earth Observation (EO) images and products provided in open and online data repositories.
The course starts with an overview of the changing geo-processing landscape, the role of ILWISPy and an introduction in Python and Jupyter Notebooks. This is followed by the various Data Access Services through which satellite products and derived geospatial data are made publicly available.
During the following weeks examples are provided on how to apply ILWISPy Geo-information and Image Processing operations in conjunction with Python tools for retrieval, processing and analysis of various types of remote sensing images and relevant environmental data from different open data access catalogues. Examples are state-of-the-art EO data sets, like those obtained by the Sentinel suite of satellites, Earth Observation or model-derived products like vegetation indices, precipitation, evapotranspiration, soil moisture, water level, reanalysis from numerical weather prediction models, etc.
The course closes with an assignment selected by the participant. The results of these assignments will be compiled and made available to all participants, including the code developed, which can be used for further independent studies, use within relevant curricula, etc. after the course terminates.
What will be achieved
The course objectives are both having a focus on the use of Python and ILWISPy for interactive data science and scientific computing, to be conducted locally as well as through a cloud-based virtual research environment next to the current capabilities offered by open cloud-based data repositories containing time series Earth Observation images and analysis-ready products.
With regard to the use of Python and ILWISPy for interactive data science and scientific computing, upon completion of this course, your can:
- Use Python scripting in combination with Jupyter Notebooks and required packages for Remote Sensing and GIS data (pre-) processing in the cloud;
- Use of ILWISPy scripting for interactive data science and scientific computing;
- Retrieve the results obtained to their local processing system and conduct further single or time series analysis in conjunction with local datasets;
- Import and pre-process time-series geospatial data having different data formats and conduct data conversions to numpy arrays/panda data frames and their use in advanced machine learning tools, e.g. like those offered by scikit-learn;
- Process EO data and products from online catalogues into relevant information.
With regard to open cloud-based data repositories, upon completion of this course, you can:
- Select a number of relevant open image and data repositories for their application domain;
- Retrieve the Meta Data relevant for the products of interest and derive the required information for image or data scaling, coordinate and projection information;
- Access the Earth Observation (EO) images and data products from a suite of online repositories, using Application Programming Interfaces (API);
- Retrieve selected data for use in the allocated virtual research environment or local system.
Why this course?
Recent technological developments for access, processing and analysis revolutionised the procedures currently available for efficient scientific computing in environmental applications. This course aims at the double objective of introducing the participants to the online data availability and processing capability of petabytes of free EO multi-source and multi-temporal data and products which can be applied for different environmental studies.
The course tackles the data, means and tools required for the market of spatial development of today in any kind of initiative requiring the input of Earth Observation and free online environmental data catalogues. The focus on applications relevant for environment, hydrology, food security, climate and disaster assessments supports for example, the SDGs related to food security and climate (SDGs 1, 2, 3, 13 and 15). The course is designed to positively impact your possibilities in the job market both in the GOs and NGOs sectors.
Online learning, what is it like
The course has been designed for 8 weeks. Each week you complete one module of the course, which consists of lectures, demo’s and exercises to bring into practice what was learnt. The study load of the entire course is 70 hours distributed in a way that you need to spend a minimum of 8 hours per week during the first seven weeks of the course. In the final week, an individual assignment having a study load of 14 hours, has to be completed and submitted online.
Hardware and software requirements
HARDWARE MINIMUM REQUIREMENTS:
- Laptop/PC with MS Windows 10, Intel(R) Core(TM) i5, 32 GB RAM, 50 GB free disc space or equivalent.
Free and open software tools are at the core of this course, next to Python and the Jupyter Notebook site package, other essential Python site-packages, ILWIS version 3.8.6 will be used to visualise results. Participants will be required to (freely) register for several online data portals, like Google Earth Engine and Google Drive, the Copernicus Climate Data Store and, depending on the assignment, developed eventually for other open data access portals, like the FAO WaPOR, the Copernicus Space Component Data Access Portal, the Copernicus Open Access Hub and the Copernicus Open Data Access Portal at EUMETSAT.
A Certificate of Attendance is issued upon full attendance during online lectures and successful submission of the individual assignment. The assignment will be graded (pass mark is 60) and should be submitted in the form of a Jupyter notebook, including markdown explaining the code as well as the code cells themselves.