This study evaluates the impact of the moisture content of biomass on thermal efficiency and relative boiler size which directly represent the economic merits of biomass drying. A model for predicting the moisture content of bundled Leucaena (Leucocephala) trees under open environment was validated for tropical Hawaii. Cumulative precipitation and evapotranspiration (ET) are the major factors affecting the biomass moisture content change. ET was computed using Hargreave's model, which requires only temperature and solar radiation data. Integration of these models made it possible to calculate the thermal efficiency and relative boiler size when using bundled trees as a fuel under a given drying regime and for a specific geographical location. A geographic information system provided the temperature and precipitation data required for evaluating the spatial variation in boiler efficiency and size for the 1440 km(2) island of Kauai. Depending on the time of harvest, the Leucaena moisture content varied from 35 to 69% (on wet basis) following a period of 6 months of in-field drying. Boiler efficiency using fuelwood with this range of moisture content varied from 49 to 73%. Boiler relative size varied from 1.2 to 2.2 times the size required when Leucaena with 0% moisture content is used as a fuel. The spatial and temporal effects on the value of biomass were thus found to be important factors for various sites in the study area. The methods for quantifying the merit of biomass moisture management proposed in this paper demonstrate how GIS modeling can lead to appropriate decision-making capability in bioenergy. Copyright (C) 1996 Elsevier Science Ltd.