With the goal of identifying signatures that select specific neutrino oscillation parameters, we test the robustness of global oscillation solutions that fit all the available solar and reactor experimental data. We use three global analysis strategies previously applied by different authors and also determine the sensitivity of the oscillation solutions to the critical nuclear fusion cross section, S-17(0), for the production of B-8. Our standard results make use of the precise new measurement of S-17(0) by Junghans et al. The globally favored solutions are, in order of goodness of fit: LMA (the only solution at 2sigma), LOW, and VAC. The NC to CC ratio for SNO is predicted by the standard global analysis to be 3.45(-0.54)(+0.70)(1sigma) which is separated from the no-oscillation value of 1.0 by much more than the expected experimental error. The predicted range of the day-night difference in CC rates is 8.3(-5.6)(+5.0)(1simgma)%. A measurement by SNO of either a NC to CC ratio > 3.3 or a day-night difference > 10%, would favor a small region of the currently allowed LMA neutrino parameter space. The global oscillation solution predicts a Be-7 neutrino-electron scattering rate in BOREXINO and KamLAND in the range 0.65(-0.03)(+0.04) (1sigma) of the BP00 standard solar model rate, a prediction which can be used to test both the solar model and the neutrino oscillation theory. Only the LOW solution predicts a large day-night effect (less than or equal to 42%, 3sigma) in BOREXINO and KamLAND. For the reactor KamLAND experiment, the LMA solution predicts a charged current rate relative to the standard model of 0.44(-0.07)(+0:22)(1sigma), E-threshold = 1.22MeV. We have also evaluated the effects of including preliminary Super-Kamiokande data for 1496 days of observations.