Machine Learning for Ecology and Sustainable Natural Resource Management

1. Front Matter

   Pages i-xxiv

   PDF

2. Part I

   1. Front Matter

      Pages 1-1

      PDF

   2. Machine Learning in Wildlife Biology: Algorithms, Data Issues and Availability, Workflows, Citizen Science, Code Sharing, Metadata and a Brief Historical Perspective

      • Grant R. W. Humphries, Falk Huettmann

      Pages 3-26

   3. Use of Machine Learning (ML) for Predicting and Analyzing Ecological and ‘Presence Only’ Data: An Overview of Applications and a Good Outlook

      • Falk Huettmann, Erica H. Craig, Keiko A. Herrick, Andrew P. Baltensperger, Grant R. W. Humphries, David J. Lieske et al.

      Pages 27-61

   4. Boosting, Bagging and Ensembles in the Real World: An Overview, some Explanations and a Practical Synthesis for Holistic Global Wildlife Conservation Applications Based on Machine Learning with Decision Trees

      • Falk Huettmann

      Pages 63-83

3. Part II

   1. Front Matter

      Pages 85-85

      PDF

   2. From Data Mining with Machine Learning to Inference in Diverse and Highly Complex Data: Some Shared Experiences, Intellectual Reasoning and Analysis Steps for the Real World of Science Applications

      • Falk Huettmann

      Pages 87-108

   3. Ensembles of Ensembles: Combining the Predictions from Multiple Machine Learning Methods

      • David J. Lieske, Moritz S. Schmid, Matthew Mahoney

      Pages 109-121

   4. Machine Learning for Macroscale Ecological Niche Modeling - a Multi-Model, Multi-Response Ensemble Technique for Tree Species Management Under Climate Change

      • Anantha M. Prasad

      Pages 123-139

   5. Mapping Aboveground Biomass of Trees Using Forest Inventory Data and Public Environmental Variables within the Alaskan Boreal Forest

      • Brian D. Young, John Yarie, David Verbyla, Falk Huettmann, F. Stuart Chapin III

      Pages 141-160

4. Part III

   1. Front Matter

      Pages 161-161

      PDF

   2. ‘Batteries’ in Machine Learning: A First Experimental Assessment of Inference for Siberian Crane Breeding Grounds in the Russian High Arctic Based on ‘Shaving’ 74 Predictors

      • Falk Huettmann, Chunrong Mi, Yumin Guo

      Pages 163-184

   3. Landscape Applications of Machine Learning: Comparing Random Forests and Logistic Regression in Multi-Scale Optimized Predictive Modeling of American Marten Occurrence in Northern Idaho, USA

      • Samuel A. Cushman, Tzeidle N. Wasserman

      Pages 185-203

   4. Using Interactions among Species, Landscapes, and Climate to Inform Ecological Niche Models: A Case Study of American Marten (Martes americana) Distribution in Alaska

      • Andrew P. Baltensperger

      Pages 205-225

   5. Advanced Data Mining (Cloning) of Predicted Climate-Scapes and Their Variances Assessed with Machine Learning: An Example from Southern Alaska Shows Topographical Biases and Strong Differences

      • Falk Huettmann

      Pages 227-241

   6. Using TreeNet, a Machine Learning Approach to Better Understand Factors that Influence Elevated Blood Lead Levels in Wintering Golden Eagles in the Western United States

      • Erica H. Craig, Tim H. Craig, Mark R. Fuller

      Pages 243-260

5. Part IV

   1. Front Matter

      Pages 261-261

      PDF

   2. Breaking Away from ‘Traditional’ Uses of Machine Learning: A Case Study Linking Sooty Shearwaters (Ardenna griseus) and Upcoming Changes in the Southern Oscillation Index

      • Grant R. W. Humphries

      Pages 263-283

   3. Image Recognition in Wildlife Applications

      • Dawn R. Magness

      Pages 285-294

   4. Machine Learning Techniques for Quantifying Geographic Variation in Leach’s Storm-Petrel (Hydrobates leucorhous) Vocalizations

      • Grant R. W. Humphries, Rachel T. Buxton, Ian L. Jones

      Pages 295-312

Ecologists and natural resource managers are charged with making complex management decisions in the face of a rapidly changing environment resulting from climate change, energy development, urban sprawl, invasive species and globalization. Advances in Geographic Information System (GIS) technology, digitization, online data availability, historic legacy datasets, remote sensors and the ability to collect data on animal movements via satellite and GPS have given rise to large, highly complex datasets. These datasets could be utilized for making critical management decisions, but are often “messy” and difficult to interpret. Basic artificial intelligence algorithms (i.e., machine learning) are powerful tools that are shaping the world and must be taken advantage of in the life sciences. In ecology, machine learning algorithms are critical to helping resource managers synthesize information to better understand complex ecological systems. Machine Learning has a wide variety of powerful applications, with three general uses that are of particular interest to ecologists: (1) data exploration to gain system knowledge and generate new hypotheses, (2) predicting ecological patterns in space and time, and (3) pattern recognition for ecological sampling. Machine learning can be used to make predictive assessments even when relationships between variables are poorly understood.  When traditional techniques fail to capture the relationship between variables, effective use of machine learning can unearth and capture previously unattainable insights into an ecosystem's complexity. Currently, many ecologists do not utilize machine learning as a part of the scientific process. This volume highlights how machine learning techniques can complement the traditional methodologies currently applied in this field.

• Quantitative ecology
• artificial intelligence
• Statistics
• data mining
• machine learning
• Wildlife biology
• natural resource management
• sampling

“If you are looking for a book that describes how ML has been used or could be used on your own spatial ecology data sets, I recommend this book. I also recommend it for readers looking for a book that provides thoughtful arguments about data analysis, in general, and where it could be moving in the future.” (John Carriger, Integrated Environmental Assessment and Management, Vol. 16 (1), 2020)

• Black Bawks Data Science Ltd, Fort Augustus, Scotland, New Zealand

  Grant Humphries

• U.S. Fish and Wildlife Service, Kenai National Wildlife Refuge, Soldotna, USA

  Dawn R. Magness

• Institute of Arctic Biology, University of Alaska - Fairbanks, FAIRBANKS, USA

  Falk Huettmann

Dr. Grant Humphries is an ecological data scientist having worked on a number of marine and terrestrial projects (mostly seabirds) around the world where machine learning tools were critical to solving complex problems. He has over a decade of experience working with machine learning tools and techniques and loves applying them in novel and interesting ways. He is the founder of Black Bawks Data Science Ltd., a small data science company based in the highlands of Scotland, where he works on building interactive, web-based decision support tools that integrate advanced modeling. He is also a penguin counter, traveling to Antarctica every year to collect data for the Antarctic Site Inventory. His spare time is dedicated to music, cooking and spending time with his two daughters: Dylan and River, and his wife, Alex.

Dr. Dawn Magness is a landscape ecologist interested in climate change adaptation, landscape planning, ecological services, and spatial modeling. She earned her M.S. in Fish and Wildlife Science at Texas A & M University and her Ph.D. in the interdisciplinary Resilience and Adaptation Program at the University of Alaska, Fairbanks. Her current projects use multiple methods to assess ecosystem vulnerability to inform strategic adaptation planning. She has conducted research on songbirds, flying squirrels, and American marten.

Dr. Falk Huettmann is a ‘digital naturalist’ linking computing and the internet with natural history research for global conservation and sustainability. He is a professor of Wildlife Ecology in the Biology & Wildlife Department and Institute of Arctic Biology at the University of Alaska Fairbanks (UAF) where he and many international students run the EWHALE lab. In his lab he studies biodiversity, land- and sea-scapes, the atmosphere, global governance, ecological economics, diseases and new approaches to global sustainability on a pixel-scale. Most of his 200 publications and 7 books are centered on Open Access and Open Source science, Geographic Information Systems (GIS), data mining and machine learning.