Introduction | Building a deme graph | Choosing the neighborhood | Genetic covariance from biallelic (SNP) data | Pairwise F_ST | Covariance from multivariate or microsatellite data | Microsatellite allele calls | From covariance to distance | Directional edge covariates | Gradient of a scalar potential | Projection of a vector flow field | Quick reference | Summary of key functions | See also | References

What is multiScaleR? | Installation | The Core Workflow | A Complete Example | Load the package and data | Prepare inputs with kernel_prep() | Fit an initial model | Optimize scales of effect | Examine results | Visualize kernel decay and covariate effects | Diagnose the likelihood surface | Project the fitted model to the landscape | Quick-Reference: Full Workflow | Learn More

Background | The scale-of-effect problem | What multiScaleR does | A Conceptual Road Map | Distance-weighted Effects | Choosing a kernel | Preparing Data | Explore Data | kernel_prep | Performance and memory for large datasets | Analysis | multiScale_optim | Kernels | Model Selection | Why standard AIC is not enough | Optimization with unmarked | Poisson Count Model | Project model | Binomial Occurrence Model | Other model classes | Other Functions & Features | kernel_dist | plot_kernel | sim_rast | sim_dat

What surface texture covariates measure | Surface metric reference | A note on scale | Load example data | Define derived covariates | Prepare model data and fit | Optimize and interpret | Project covariates to rasters | Roughness surfaces without a fitted model | Distribution-shape and geometry surfaces | Optimizing the scale of roughness | Weighted roughness: the scale of effect of heterogeneity | Practical guidance | Quick reference | Summary of key functions | See also

Why explicit covariate specifications? | Landscape metric reference | Composition metrics | Edge metrics | Adjacency metrics | Information theory metrics | Class-level metrics | Load example data | Define derived covariates | Prepare model data | Optimize the kernel covariate | Project derived covariates to rasters | Producing landscape metric rasters without a fitted model | Separating composition from configuration | Optimizing landscape metric radii | Practical guidance | See also | References

Build aligned annual maps and observations | Prepare one pooled set of covariates | Fit one relationship across years | Project the fitted relationship onto each annual map | Quick reference

Introduction to Spatial Projection | From fitted model to landscape map | The extrapolation problem | What clamping does | This vignette | Demonstrating Clamping Under Controlled Extrapolation | Conceptual setup | Step 1: Simulate a known landscape and response surface | Step 2: Impose a biased sampling design | Step 3: Fit the multiscale model | The Extrapolation Problem | Seeing extrapolation in practice | Understanding and Applying Clamping | How clamping works | The pct_mx argument: tuning the guardrail | Comparing different clamping settings | Quantifying prediction error: inside vs. outside the sampled domain | Key Takeaways | Practical guidance