Seasonal subsurface heat conduction can have a large influence on Triton's N2 frost distribution. Increasing surface thermal inertia reduces the thickness and extent of seasonal N2 frosts. If the thermal inertia of the nonvolatile substrate is greater than about 30% of the value for nonporous H2O or CO2, or if nonporous H2O or CO2 is overlaid by a porous regolith of low thermal inertia but less than a few meters thick, the northernmost latitudes visible to Voyager should have been frost-free at the time of the encounter, possibly accounting for their relatively low albedo. If the substrate in the northern hemisphere has sufficiently low albedo and/or emissivity and also has a thermal inertia comparable to that of nonporous H2O or CO2, there may be no seasonal or permanent N2 deposits in the northern hemisphere at all. Because this model, like previous ones, predicts a monotonic recession of permanent N2 deposits toward the poles and very limited seasonal N2 frost in the southern hemisphere at Voyager time, and because of new spectroscopic evidence for nonvolatile CO2 on Triton's bright southern hemisphere, we consider it possible that much of the bright material on Triton's southern hemisphere is not N2. Bright nonvolatiles in the southern hemisphere may allow seasonal N2 frosts to form there during the southern summer, possibly helping to explain Triton's spectroscopic changes during the past decade. All the models considered here predict 10-fold or greater seasonal variations in atmospheric pressure, with pressure currently increasing in high-thermal-inertia models and decreasing in models with low thermal inertia. © 1992.
