About PEST

PEST (Parameter ESTimation) is a general-purpose, model-independent, parameter estimation and model predictive uncertainty analysis package developed by Dr. John Doherty. PEST is the most advanced software readily available for calibration and predictive error analysis of groundwater, surface water, and other environmental models. Using PEST you can:

• apply advanced and efficient regularization techniques in calibrating your models to extract maximum information content from your data
• undertake linear and nonlinear predictive error analysis of model outputs
• simultaneously parameterize several models using multiple datasets
• accommodate heterogeneity using advanced spatial parameterization
• combine PEST with stochastic field generation to efficiently explore pre- and post-calibration uncertainty
• conduct parallel model optimization runs across PC or UNIX networks
• compare the worth of different proposed data acquisition strategies in reducing model predictive error thereby optimizing resources allocated to such tasks
• quantify the contributions to model predictive error made by different parameter types
• establish the irreducible uncertainty of a model prior to calibrating that model
• quantify the reduction in predictive uncertainty accrued through model calibration
• use a model for quantitative hypothesis-testing
• provide risk analysis input to the decision-support process

Traditional methods of parameter estimation and uncertainty analysis based on a handful of zones do not provide the flexibility to extract vital information from expensive data, or to quantify the benefits and risks associated with different environmental management options.

 Some of the advances made over the last few years in PEST to overcome these problems include:

• The combination of regularized inversion with the use of pilot points as a spatial parameterization device
• The unique and extremely efficient “SVD-assist” regularized inversion methodology which combines subspace and Tikhonov methods. Inversion can be carried out with model-run efficiencies comparable with traditional approaches - but with hundreds of parameters
• The combination of regularized inversion with stochastic field generation to produce calibration-constrained parameter fields with high levels of model run efficiency
• The use of Pareto methods to compare the merits of different management strategies
• The use of linear methods to assess the relative worth of different strategies for future data aquisition

PEST is accompanied by a large number of utility programs which support its use with commonly used models (such as MODFLOW, MT3D, SEAWAT, FEFLOW, MICROFEM and others).

Other software automates PEST file set-up for complex parameter estimation and uncertainty analysis problems involving one or multiple models. Further components of the PEST suite implement the global SCE and CMA optimisation algorithms which can be employed where high levels of model nonlinearity and/or poor numerical behaviour make use of gradient based optimisation methods difficult.