Nonhomogeneous hidden Markov models (NHMMs) provide a relatively simple fra
mework for simulating precipitation at multiple rain gauge stations conditi
onal on synoptic atmospheric patterns. Building on existing NHMMs for preci
pitation occurrences, we propose an extension to include precipitation amou
nts. The model we describe assumes the existence of unobserved (or hidden)
weather patterns, the weather states, which follow a Markov chain. The weat
her states depend on observable synoptic information and therefore serve as
a link between the synoptic-scale atmospheric patterns and the local-scale
precipitation. The presence of the hidden states simplifies the spatio-tem
poral structure of the precipitation process. We assume the temporal depend
ence of precipitation is completely accounted for by the Markov evolution o
f the weather state. The spatial dependence of precipitation can also be pa
rtially or completely accounted for by the existence of a common weather st
ate. In the proposed model, occurrences are assumed to be conditionally spa
tially independent given the current weather state and, conditional on occu
rrences, precipitation amounts are modeled independently at each rain gauge
as gamma deviates with gauge-specific parameters. We apply these methods t
o model precipitation at a network of 24 rain gauge stations in Washington
state over the course of 17 winters. The first 12 yr are used for model fit
ting purposes, while the last 5 serve to evaluate the model performance. Th
e analysis of the model results for the reserved years suggests that the ch
aracteristics of the data are captured fairly well and points to possible d
irections for future improvements.