Matt Strimas-Mackey

Matt Strimas-Mackey

Data Scientist

Cornell Lab of Ornithology

I write code and build tools to turn biodiversity data into conservation and science insights. As part of the eBird Status and Trends team at the Cornell Lab of Ornithology, I help develop species distribution models to estimate the occurrence, abundance, and population trends of bird species at a high spatiotemporal resolution using data generated by the citizen-science project eBird. I work on all facets of this project, from model testing and development to visualization of model results.

To ensure eBird data are widely and correctly used in science and conservation, I produce data products, build open source tools, develop educational resources, and teach workshops.


  • Bird Conservation
  • Geographic Information Science
  • Geovisualization
  • Citizien Science Data
  • R


  • MSc Zoology, 2016

    University of British Columbia

  • BSc Ecology, 2012

    Univesity of Guelph

  • MSc Physics, 2006

    University of Toronto

  • BSc Physics, 2005

    University of Toronto


Extracting eBird Data From a Polygon

A short post demonstrating how to use the R package auk to extract eBird data from within a study area defined by a polyon or Shapefile.

Raster Summarization in Python

As part of the eBird Status & Trends team, I often find myself having to quickly summarize large rasters across layers. For examples, we might summarize weekly relative abundance layers for a single species across the weeks to estimate year round abundance or, for a given week, we might summarize across species to produce a species richness layer.

Efficient Block Processing of Rasters in R

In this post, I examine how the size of blocks impacts the efficiency of block processing raster data in R

Processing Large Rasters in R

Investigating how raster data are processed in R to improve the efficiency of processing large raster files.

Lotka-Volterra Predator Prey Model

Analyzing dynamical systems in R. Using the Lotka-Volterra predator prey model as a case-study, I use the R packages deSolve and FME to solve a system of differential equations and perform a sensitivity analysis.

Fishnets and Honeycomb: Square vs. Hexagonal Spatial Grids

Exploring the benefits of hexagonal grids relative to square grids for spatial sampling and analysis, and generating hexagonal grids in R.


eBird Status & Trends

Linking bird sightings from eBird with habitat information from satellites to model when and where birds occur. This project estimates the distribution, abundance, and trends of bird populations at high spatial and temporal resolution across the entire Western Hemisphere.

eBird Best Practices

This open source book outlines a set of best practices for using eBird data to estimate species distributions in R. It covers accessing and pre-processing eBird data, preparing satellite-derived habitat variables for use as model covariates, and best practices for model encounter rate, occupancy, and abundance using eBird data.


This R package provides access to eBird Status and Trends data for over 600 North American bird species. These data include weekly estimates of relative abundance and occurrence at high spatial resolution across the entire Western Hemisphere.


An R package to access and analyze bird observation data from eBird, the largest biodiversity-related citizen science project in the world. Observations can be extracted based on a variety of taxonomic, spatial, and temporal filters. auk also includes tools for importing and processing eBird data in R.


The prioritizr R package provide a flexible interface for building and solving systematic conservation prioritization problems using integer linear programming (ILP). It supports a broad range of objectives, constraints, and penalties that can be used to custom-tailor prioritization problems to the specific needs of a conservation planning exercise. For more details, watch this talk by lead developer Jeff Hanson.