ABSTRACT This paper describes the latest version of PVWATTS and how its spatial resolution was improved by a factor of 25 by using a high-resolution (e.g., 40-km by 40-km cells) spatially uniform grid of meteorological input data. Like its predecessor, version 2 is Internet accessible. The user selects a grid cell containing the desired location from an electronic map, thereby initiating a selection by PVWATTS v.2 of the nearest TMY2 station that is climatically similar, followed by an hourly performance simulation for the TMY2 station. Performance is translated back to the selected grid cell based on differences in solar radiation and temperature using previously determined data grid sets of monthly solar radiation and maximum daily temperature. 1. Introduction The original PVWATTS [1] is an Internet-accessible simulation tool for providing quick estimates of the electrical energy produced by a grid-connected crystalline silicon photovoltaic (PV) system for any of 239 locations. These locations correspond to the 239-station Typical Meteorological Year (TMY2) database [2] for the United States and its territories. Users select a location from a station map and set PV system parameters, or select default values, and PVWATTS performs an hour-by-hour simulation that provides monthly and annual alternating current (AC) energy production in kilowatts and energy value in dollars. System parameters that may be specified include size (AC rating for Standard Reporting Conditions), local electric costs, PV array type (fixed or tracking), PV array tilt angle, and PV array azimuth angle. The performance model used by PVWATTS is based on Sandia National Laboratories’ PVFORM [3], but with fewer allowed specified inputs. Before the release of version 2, if the desired location was between TMY2 stations, the PVWATTS user needed to choose between two or more stations based on which station they judged to be climatically similar, or in some cases, the nearest. In these instances, PVWATTS v.2 provides better performance estimates by the use of 40-km resolution data grid values of monthly solar radiation and maximum daily temperature to translate performance from a nearby TMY2 station to the desired grid cell. This paper explains the method of translation and describes the gridded data sets.