The need for an accurate model that describes the galactic cosmic radiation, GCR, environment is becoming increasingly important in light of planned space exploration missions. There is concern about the effects of heavy ions on crew health and electronic components. Recent studies have indicated that H, He, C, O, Ne, Mg, Si and Fe nuclei contribute about 80% of the radiation dose-equivalent under realistic shielding conditions. Small uncertainties in the absolute differential flux of these particles lead to large uncertainties in the amount of shielding mass needed for crew radiation protection. The absolute flux and the associated errors have received less attention than charge composition measurements. The Naval Research Laboratory (Adams, 1986) developed a comprehensive model, called the CREME model, that summarized the then-existing GCR data in terms of empirical fits. This model has been used for GCR intensities for shielding and single event upset studies. The main weakness of this model lies in its simple treatment of the solar modulation. In this paper we describe the results of a systematic study of existing data from 1965 to 1990 to derive a more accurate GCR environment model. We have fitted the hydrogen, helium, and iron spectra from 1965 to 1990, and have shown that a consistent set of local interstellar spectra and solar modulation parameters can be developed that applies to all species. A thorough analysis of uncertainties shows that the model spectra can be defined to an absolute accuracy of ±10%. A reference spectrum at solar minimum, which will be the limiting GCR spectrum for exploratory class missions, has been defined.