Chill accumulation models are used to estimate the days from the end of a season until bud break in the following season. Trees have specific chill requirements that are related to the number of accumulated hours within a known temperature range. A new chill accumulation model was developed to improve the estimation of bud break date. In this model, we use chill days (C_D), which are defined as the cumulative number of hours below a pre-selected threshold temperature divided by 24 hours per day (to express as chill days). We also use "anti-Chill days" (C_A), which are defined as the cumulative number of hours above the same pre-selected threshold temperature divided by 24 hours per day. The C_D values are given a negative sign and the C_A values are given a positive sign. In the model, the negative C_D values are accumulated until they reach a pre-selected value that is identified as the chill requirement (C_R), which is also a negative number. The chill requirement is met when the SC_D is less or equal to C_R. On the following day, the model no longer adds C_D values but begins to add daily anti-chill days to C_R. On the date that C_R + SC_A is greater or equal to 0, that date is selected for bud break. The optimal value for the chill temperature threshold and for C_R are determined using trial and error until the root mean square error between predicted and observed days between harvest or leaf drop and bud break is minimized.