Computing

• Parallel Computing: Local parallel computing via R, Matlab, Mathematica, etc. is a useful way to save time and make use of those multi-core laptops.
• Cloud/Cluster Computing: Large simulations or analyses are best done in the cloud or through HPC clusters.
• scratch, slurm & pbs scripting
• Github/Git

Modeling

• Model Development: Ranging from creating models for specific systems to developing new extensions of existing frameworks.
• Model Types: ODEs, PDEs, large food web and pollination dynamics on networks, IBMs or agent based models, spatial population genetics model.
• Computational & Numerical Analysis: Linear stability analysis, bifurcation analysis, sensitivity analysis involving single and multiple attractors, Lyapunov exponential analysis to study chaos, fractal dimensions (Kaplan–Yorke dimension), phase space analysis of multiple attractors, etc.
• Graphical Analysis: Volume of irregular shapes.

Network Analysis

• Analysis & Metrics: Motif analysis & detection, network incoherence measurements, network assembly, cluster analysis, nestedness metrics, Jaccard index, connectance, path lengths, and other standard metrics.
• Network Visualization: creating visual depictions of networks for manuscript figures, to aid in analysis, and help convey research to general audiences.
• igraph & Matlab

Machine Learning

• Models:
• Supervised Learning: Neural networks (Example application), random forests (Example application), CART (Example application), regression analysis, etc.
• Unsupervised Learning: k-means clustering, Hierarchical clustering
• Predictive Modeling: Managing and reduction of OOTB error, examining confusion matrices, parameter tuning (caret), cross validation techniques, creating receiver operating characteristic (ROC) curves and measuring AUC.
• Feature Analysis: Mean Accuracy Decrease to measure feature importance. PDPs, ICE plots, ALE plots, and H-statistic to meausure feature effects and interactivity. Currently working on manuscript to extend the reach of the explanatory power of Random Forests in analyzing ecological models.

Statistics

• Regressions: linear models, mixed models, generalized linear models, generalized linear mixed models, generalized additive models, beta regressions, partial least squares regressions, loess regressions, etc.
• Time Series Analysis: Dynamic Time Warping, historical time series (e.g., animal abundance measurements, temperature patterns, rainfall), various smoothing techniques.
• Spatial: Working with GIS and raster data, measuring spatial auto-correlation (Moran’s I), Ripley's K-Function.

Databases/Data Warehouses

• Beyond gathering field data for urban ecology projects (see Resources), I have made use of multiple large databases/data warehouses for research projects. These may be useful to others:
• GBIF the Global Biodiversity Information Facility
• COMPADRE plant matrix demographics - Data & Analysis
• Great Lakes Fisheries Council

Lab Work & Project Management

• Personnel mentoring, management, scheduling, & training.
• Individual and team based development of best practices for data management and storage.
• Bee specimen collection, cleaning, and identification techniques.
• Stakeholder engagement.
• Media engagment (see CV for listing).
• Professional pothole avoider with experience all around the southeastern MI area.