2024 ARCHIVES|Workshops
We’re pleased to offer the following high-quality half-day workshops on Wednesday, April 3, 2024 from 1:30 – 5:30 PM. Thank you to the workshop organizers for their efforts to provide these additional educational opportunities with an emphasis on a specific skill, technique, or process.
W-01: GRID Sampling: GIS, GPS, Fieldwork & Analysis with ML/AI and beyond
Overview:
GRID sampling presents a core scheme in Landscape Ecology to obtain information on 'patterns and processes'. It can be done in GIS, with GPS and in the remote field and rugged terrain. Often it involves Remote Sensing imagery, also. In times of Machine Learning (ML)/AI, the cloud and ensemble models even more options are on the table, beyond 'just' scale, extend, pixel size, trapping webs, probability, imputation, shape and smoothed prediction surfaces. This workshop will present an overview of representative sampling, how GRIDs fit into that picture, and why and how it matters for science linking directly to real-world outcome, often to be achieved in-time, globally. Examples from climate, biodiversity, the humanities and landscape categories will be used for an elaboration.
Intended Audience: Landscape Ecologists
Contact: Falk Huettmann, Professor, University of Alaska Fairbanks, [email protected]
W-02: Metacoupling: A New Interdisciplinary Frontier for Global Sustainability
Overview:
Global sustainability challenges, such as biodiversity loss, climate change, and landscape fragmentation, are increasingly influenced by local and distant forces. Factors like globalization, environmental changes, disasters, natural processes, social unrest, war, and many other human activities connect humans and nature worldwide. These interconnections are made possible through various processes, including animal migration, species invasion, human migration, disease spread, sound/noise transmission, transfer of pollutants and wastes, trade of goods and products, flows of ecosystem services, environmental and hydrological flows, foreign investment, technology transfer, water transfer, and tourism. They form metacoupling -- human-nature interactions within a system and across adjacent and distant systems. Metacouplings have profound implications for sustainability as they can transform landscape structure, function, pattern, process, and dynamics. For example, farmers convert forest landscapes for food production to meet demands from local populations and those in adjacent and distant places. To help integrate and understand various interconnections and feedback comprehensively and systematically, the metacoupling framework has been developed. In this workshop, we will introduce the framework, present applications of the framework, and conduct hands-on exercises. Workshop participants will have opportunities to apply the framework to various case studies.
Intended Audience: The target audience encompasses attendees at any career stage (e.g. students, postdoctoral scholars, professors, resource managers) and with a variety of interests, such as landscape change, climate change, natural resource policy and governance, biodiversity, ecology, and landscape patterns (e.g. connectivity) and processes (e.g. disturbance, dispersal, migration).
Contact: Jianguo (Jack) Liu, University Distinguished Professor, Michigan State University, [email protected]
Co-organizers: Logan Hysen, Nan Jia - Center for Systems Integration and Sustainability, Michigan State University
W-03: Using NEON Airborne Remote Sensing Data to Address Ecological Questions at Scale in Python
Overview:
The National Ecological Observatory Network (NEON) is a multi-decadal continental-scale observatory designed to collect and synthesize data at field sites across the US to study the impacts of climate change, land use change, and invasive species on natural resources and biodiversity. The observatory is designed to provide a wealth of standardized datasets that characterize plants, animals, soils, microbes, nutrients, freshwater, and atmosphere from 81 terrestrial and aquatic sites for the next three decades to enable a better understanding of how the US ecosystems are changing. As one of five data collection subsystems, NEON's Airborne Observation Platform (AOP) acquires highly calibrated, co-registered sub-meter to meter-scale hyperspectral imagery, discrete and waveform lidar, and digital photography that are optimally suited to analyze landscape-scale ecosystem change. However, there are a few barriers to entry in using AOP data, including large data volumes, custom data formats, and a dearth of open-source tools for working with geospatial data. This interactive workshop would begin with an introduction to NEON and the AOP remote sensing data and provide participants with hands-on coding experience to ingest and analyze high-resolution AOP hyperspectral, lidar, and camera data in Python. We will present examples of ecological analyses (machine learning-based land cover classification, modeling plant foliar traits, detecting changes in vegetation structure following fires, etc.) that bring together high-resolution remote sensing datasets and co-located in-situ terrestrial observation data. Participants will leave with knowledge of how to access NEON data and resources to investigate ecological questions and a suite of open-source Python tools that can be leveraged to further their own research interests.
Intended Audience: The workshop is intended for students and professionals interested in using remote sensing data to study ecological processes. Participants should have a basic knowledge of remote sensing. Some experience coding in a scientific programming language (e.g., Python, R, JavaScript) is recommended, but not required.
Contact: Shashi Konduri, Remote Sensing Scientist, NEON/Battelle, [email protected]
Co-organizers: Bridget Hass, Kate Murphy, John Musinsky, Tristan Goulden - NEON/Battelle
Overview:
GRID sampling presents a core scheme in Landscape Ecology to obtain information on 'patterns and processes'. It can be done in GIS, with GPS and in the remote field and rugged terrain. Often it involves Remote Sensing imagery, also. In times of Machine Learning (ML)/AI, the cloud and ensemble models even more options are on the table, beyond 'just' scale, extend, pixel size, trapping webs, probability, imputation, shape and smoothed prediction surfaces. This workshop will present an overview of representative sampling, how GRIDs fit into that picture, and why and how it matters for science linking directly to real-world outcome, often to be achieved in-time, globally. Examples from climate, biodiversity, the humanities and landscape categories will be used for an elaboration.
Intended Audience: Landscape Ecologists
Contact: Falk Huettmann, Professor, University of Alaska Fairbanks, [email protected]
W-02: Metacoupling: A New Interdisciplinary Frontier for Global Sustainability
Overview:
Global sustainability challenges, such as biodiversity loss, climate change, and landscape fragmentation, are increasingly influenced by local and distant forces. Factors like globalization, environmental changes, disasters, natural processes, social unrest, war, and many other human activities connect humans and nature worldwide. These interconnections are made possible through various processes, including animal migration, species invasion, human migration, disease spread, sound/noise transmission, transfer of pollutants and wastes, trade of goods and products, flows of ecosystem services, environmental and hydrological flows, foreign investment, technology transfer, water transfer, and tourism. They form metacoupling -- human-nature interactions within a system and across adjacent and distant systems. Metacouplings have profound implications for sustainability as they can transform landscape structure, function, pattern, process, and dynamics. For example, farmers convert forest landscapes for food production to meet demands from local populations and those in adjacent and distant places. To help integrate and understand various interconnections and feedback comprehensively and systematically, the metacoupling framework has been developed. In this workshop, we will introduce the framework, present applications of the framework, and conduct hands-on exercises. Workshop participants will have opportunities to apply the framework to various case studies.
Intended Audience: The target audience encompasses attendees at any career stage (e.g. students, postdoctoral scholars, professors, resource managers) and with a variety of interests, such as landscape change, climate change, natural resource policy and governance, biodiversity, ecology, and landscape patterns (e.g. connectivity) and processes (e.g. disturbance, dispersal, migration).
Contact: Jianguo (Jack) Liu, University Distinguished Professor, Michigan State University, [email protected]
Co-organizers: Logan Hysen, Nan Jia - Center for Systems Integration and Sustainability, Michigan State University
W-03: Using NEON Airborne Remote Sensing Data to Address Ecological Questions at Scale in Python
Overview:
The National Ecological Observatory Network (NEON) is a multi-decadal continental-scale observatory designed to collect and synthesize data at field sites across the US to study the impacts of climate change, land use change, and invasive species on natural resources and biodiversity. The observatory is designed to provide a wealth of standardized datasets that characterize plants, animals, soils, microbes, nutrients, freshwater, and atmosphere from 81 terrestrial and aquatic sites for the next three decades to enable a better understanding of how the US ecosystems are changing. As one of five data collection subsystems, NEON's Airborne Observation Platform (AOP) acquires highly calibrated, co-registered sub-meter to meter-scale hyperspectral imagery, discrete and waveform lidar, and digital photography that are optimally suited to analyze landscape-scale ecosystem change. However, there are a few barriers to entry in using AOP data, including large data volumes, custom data formats, and a dearth of open-source tools for working with geospatial data. This interactive workshop would begin with an introduction to NEON and the AOP remote sensing data and provide participants with hands-on coding experience to ingest and analyze high-resolution AOP hyperspectral, lidar, and camera data in Python. We will present examples of ecological analyses (machine learning-based land cover classification, modeling plant foliar traits, detecting changes in vegetation structure following fires, etc.) that bring together high-resolution remote sensing datasets and co-located in-situ terrestrial observation data. Participants will leave with knowledge of how to access NEON data and resources to investigate ecological questions and a suite of open-source Python tools that can be leveraged to further their own research interests.
Intended Audience: The workshop is intended for students and professionals interested in using remote sensing data to study ecological processes. Participants should have a basic knowledge of remote sensing. Some experience coding in a scientific programming language (e.g., Python, R, JavaScript) is recommended, but not required.
Contact: Shashi Konduri, Remote Sensing Scientist, NEON/Battelle, [email protected]
Co-organizers: Bridget Hass, Kate Murphy, John Musinsky, Tristan Goulden - NEON/Battelle