Widespread nitrate contamination of groundwater in agricultural areas poses a major challenge to sustainable water resources. Efficient analysis of nitrate fluxes across large regions also remains difficult. This study introduces a method of characterizing nitrate transport processes continuously across regional unsaturated zones and groundwater based on surrogate, machine-learning metamodels of an N flux process-based model. The metamodels used boosted regression trees (BRTs) to relate mappable variables to parameters and outputs of a “vertical flux method” (VFM) applied in the Fox-Wolf-Peshtigo (FWP) area in Wisconsin. In this context, the metamodels are upscaling the VFM results throughout the region, and the VFM parameters and outputs (collectively referred to as “nitrate flux”) are the BRT metamodel response variables: VFM_fcN, nitrate (NO3−) source concentration factor (which determines the local NO3− input concentrations); VFM_travel_time_yrs, unsaturated zone travel time; NO3_WT_mgL_1980−2020, NO3− concentration at the water table in 1980, 2000, and 2020 (three separate metamodels); and Zss_N_ext_depth, NO3− “extinction depth”, the eventual steady state depth of the nitrate front. The metamodels were trained using 129 wells within an active MODFLOW model area of the FWP and 58 mappable predictor variables from a geographic information system, resulting in training and cross-validation testing R2 values of 0.52 – 0.86 and 0.22 – 0.38, respectively. The provided metadata file describes all 58 predictor variables considered in metamodel development, whereas the ascii predictor variable grids comprise those in the final metamodels. Metamodel outputs (ascii prediction grids) were compiled as maps of the above metamodel response variables. Relationships between predictor variables and outputs were generally consistent with expectations, e.g. with greater source concentrations and NO3− at the groundwater table in areas of intensive crop use and well drained soils. Shorter unsaturated zone travel time in poorly drained areas indicated possible preferential flow through clay soils and a tendency for fine grained deposits to collocate with areas of shallower water table. Numerical estimates of groundwater recharge may have been a proxy for N input and redox conditions in the northern FWP, which had shallow predicted NO3− extinction depth. The metamodel results provide proof-of-concept for regional estimation of NO3− transport processes in a statistical metamodel framework based on mappable GIS input variables.