Effective site-specific management requires an understanding the influence of soil and weather on yield variability. Our objective was to examine the influence of soil, precipitation, and temperature on spatial and temporal corn grain yield variability.  The study site (10 by 250 -m in size) was located in Jaboticabal, São Paulo State, on a Rhodic Hapludox. Corn yield (planted with 0.9-m spacing) was measured in 100 4.5x10-m cells along four parallel transects (25 observations per transect) over six growing seasons between 2001 and 2010. Soil chemical and physical properties, daily precipitation, relative humidity, and temperatures were measured.  Data were analyzed with regression and geostatistical techniques. Corn yield was positively correlated with base saturation (0.43 < r < 0.66).  Additionally, there was a positive regression relationship between the number of days with rainfall  and the correlation coefficient between base saturation (r² = 0.85) and relative humidity (r² = 0.94). The semivariance at the first lag (i.e., 5-m) had a strong negative relationship with January days with precipitation (r = 0.94) indicating that less frequent precipitation in critical months increases small-scale spatial variability. Only a small fraction of the spatial and temporal yield variability could be modeled as a function of base saturation, clay, soil potassium, and total number of days with precipitation (r² = 0.24).  However, the relationship between climate variables and the correlation coefficient suggested that soil base saturation became more limiting when water was less limiting.  Further, climate change that reduces the frequency of precipitation at critical periods during the growing season may increase small-scale spatial variation in corn yield.