Aggregation - Grouping data by common or similar variables.

Assisted migration - The human assisted movement of trees or tree seeds; generally as a response to climate change or for more favorable growing conditions 

Assisted migration prescription - Plans or recommendations for non-local seed sourcing for reforestation or afforestation. 

Basal area - Cross sectional area of a tree stem. 

Compound Topographic Index (CTI) - How likely water is to accumulate at a given location, based on slope and upslope area.

Deep Neural Networks (DNNs) - A type of machine learning algorithm inspired by the structure and function of the human brain. They consist of layers of interconnected nodes (or "neurons"), which process and learn patterns from large amounts of input data.

Digital Elevation Model (DEM) - A raster grid representing the elevation of the terrain surface at each pixel. 

Discriminant Analysis - A statistical method used to classify data into distinct categories, in this case, reclassifying ecosystems based on climate data.

Ecosystem delineation - The classification of landscapes based on ecological characteristics such as vegetation, climate, and geography. 

Exposure - How exposed a location is to environmental factors like wind and solar radiation, based on its position relative to surrounding terrain. 

Forest Mangement Area - A designated region where forestry activities are planned and regulated.

Hillshade - Shaded relief map created from a DEM, simulating how sunlight hits the terrain.

Hurdle Model - A hurdle model is a two-part statistical model used for count data with lots of zeros. In the case of species frequency, it first asks whether the species is present at a site, and then, if it is, it models how frequently it occurs.

Interpolated points - Data points estimated between known values to create a smoother, more continuous representation.

Land cover classes - Categories that describe different types of vegetation or land use, such as forests, grasslands, or wetlands.

Mean Absolute Error (MAE) - The average of the distance from the mean (as an absolute value) of each data point. Basically you take every value, subtract each one from the mean individually. Then you take the average of these differences.

Overfitting - Overfitting occurs when a model learns patterns that are too closely tied to the training data, including noise or random fluctuations, rather than general patterns that apply to new data. This can lead to excellent performance on the training set but reduced accuracy when the model is applied to unseen data.

Percent basal area - proportion of the total basal area of a certain species over the total basal area of all tree species in an area.

Percent increase in mean squared error - How much the model’s error increases when a specific variable’s values are randomly permuted (shuffled), breaking its relationship with the response   

Predictive modelling - Using statistical and machine learning techniques to forecast future trends based on past and present data.

Predictor variables - Environmental factors (such as temperature, precipitation, or soil type) used in models to estimate species distributions.

RandomForest model - A machine learning technique that improves predictions by combining multiple decision trees. Decision trees are a set of predictions at each step a predictor variable is used to split the data, and the final result is based on averaging (for numerical data) or voting (for classification), improving accuracy and reducing errors.

Rasterize - converting data into a grid-based format where each cell represents a specific value, such as climate or species frequency. This allows for continuous spatial analysis and easier extrapolation across large areas.

Sample Plot - A defined area for measuring forest attributes including species composition, basal area (per tree), canopy cover, and more. Permanent sample plots are those that are  used for return sampling (i.e. a plot is set up, and remeasured every 10 years)

Sky island - A high-elevation ecosystem isolated from surrounding lowlands, often home to unique plant and animal species.

Species frequency - The amount of a certain tree species compared to all trees in an area (often same as percent basal area). 

Species frequency map - Map showing species frequency measured as a percentage per ecosystem. (Darker colors = higher proportion of said species).

Species provenance - Species provenance is the particular genotype or phenotype of a population of a species from a particular geographic area of its range. For example, a lodgepole pine from Utah has a different provenance and different climate suitabilities than a lodgepole pine from Alaska or northern British Columbia. 

Topographic Position Index (TPI) - how high or low a location is relative to its surrounding area. 

Training data - The dataset used to "teach" a predictive model by providing examples of known conditions

Target data - The dataset that the trained model predicts frequency values onto.

Vegetation Continuous Field - A product developed by NASA using satellite imagery to determine global vegetation and tree cover percentages.