Services

McDonald Data Sciences specializes in the following ecology and wildlife analyses:

Capture-Recapture Analysis (CMR)

Survival and abundance estimation via open population, closed population, spatially-explicit, Multi-state, and Robust design models.

Distance Analysis

Density and abundance estimation via line transect and point transect distance analyses. In addition to point estimates, these analyses can produce density maps.

Environmental Surveys

Design and analysis of long-term monitoring studies. Monitoring studies involve spatial, temporal, and site designs. Modern spatial designs draw Spatially balanced samples via Balanced Acceptance Sampling (BAS). Temporal designs usually involve Rotating Panels of sample sites.

Habitat Modeling

Home range estimation. Tracking and path estimation from radio telemetry data. GPS fix-rate compensation. Resource selection functions. Habitat suitability and maps ( polar bear, walrus ).

Survival Analysis

Survival rates via capture-mark-recapture ( California spotted owl, Northern spotted owl, polar bear, ) and proportional hazards models ( sage grouse, mule deer, ). Proportional hazards modeling in wildlife almost always involves using the Anderson-Gill counting process formulation to allow for staggered entry and time-varying covariates.

Regression Analysis

Regression analyses are the heart of many analyses in wildlife and ecology because they compare responses among covariates. Regression analyses involve use of generalized linear models (GLM) generalized additive models (GAM) and ordinary analysis of variance (AOV). Variations on regression include Incomplete block experiments, Quantile regression, and Spline models.

Computer intensive methods

Many wildlife analyses involve bootstrapping, permutation, Monte Carlo simulation, and Bayesian Markov Chain Monte Carlo method (MCMC).

Database services

McDonald Data Sciences offers a number of data base construction, population, maintenance, and reporting services, including hierarchical database design SQL programming postgreSQL database construction and programming, Spatial postgreSQL for GIS servers, and R interfaces.

R programming

McDonald Data Sciences commonly houses client data, analyses, and reports in R Packages. These R packages can contain Shiny interfaces and R markdown reports. In common use are ggplot graphics, tidyverse syntax git version control, and client delivery via gitHub or gitLab.

Common questions

Data format

McDonald Data Sciences accepts and can use data in practically any format. If you have data in a previously unseen format, Dr. McDonald will get excited. Common data formats include MS Excel workbooks, Comma Separated Values (CSV), MS Access, SQL, and shapefiles. Common delivery methods include email attachment, Google Drive, and Box folders.

Code

McDonald Data Sciences uses R for almost everything including data manipulation, database access, graphics, and report generation. The R code we produce for clients is property of the client and as such can be delivered upon request. This allows clients to re-run code and reports on a routine basis.

Report Formats

McDonald Data Sciences delivers reports in any format that a client requests. Commonly, we deliver final and one-off reports as PDF files. Methods and Results destined for other reports are commonly delivered in MS Word format. Dr. McDonald is well versed in LaTex and can collaboratively edit online documents
(e.g., Google docs or Overleaf).

Workshop Formats

McDonald Data Sciences can customize workshops and teaching materials to specific audiences. Dr. McDonald commonly teaches 3-day, 1-day, and half-day workshops. Live on-line workshops are also a possibility.