Despite of the worldwide relevance of canola (Brassica napus L.), the agronomic information available about the crop is scarce in Argentina, even more about it yield spatio-temporal variability. This issue constrain the use of canola as winter alternative in intensificated agricultural sequences. The objectives were to i) determine canola yield spatial variability and its relationship with soil properties; ii) evaluate the effect of nutritional limitations removal on canola yield spatial variability, and iii) estimate canola yield temporal variability. We carried out two fields experiments in 2014 and 2015 seasons in a farm located in Diamante, Entre Ríos, Argentina (-32°12´44´´, -60°32´42´´). We used detailed soil maps (1:20.000) and grain yield maps to delimitate productivity zones.  In both fields two productivity zones were delimited, one of high productivity (HPZ) and one of low productivity (LPZ). We selected sampling points (10 in 2014 and 9 in 2015) distributed among productivity zones. Farmer management practice (70 kg N ha^-1 y 18 kg P ha^-1) was compared in each sampling point with a treatment with non-limiting NPS, which included 230 kg N ha^-1, 48 kg P ha^-1 and 30 kg S ha^-1. Soil samples were collected at sowing up to 0.60 m to determine chemical soil properties. Soil water content was measured up to 1 m depth at sowing, stem elongation beginning, flowering beginning, full flowering, flowering end and physiological maturity. Additionally, long-term simulations were performed using SALUS simple, in order to evaluate grain yield temporal variability. Analyzing the relationship between grain yield and soil properties we found inconsistent results between seasons. Average farmer grain yield was around 1700 kg ha^-1 in both seasons. Whereas, the non-limiting NPS grain yield was 2151 kg ha^-1 in 2014 and 3183 kg ha^-1 in 2015). Grain yield response to NPS was higher in LPZ than HPZ, particularly in 2015 season where weather conditions were better than 2014 season. However, average grain yield in non-limiting conditions was similar between productivity zones, showing a reduction in spatial variability. Long-term simulation results showed that temporal grain yield variability was higher in LPZ than HPZ, in both farmer management practice and non-limiting NPS. We also found that non-limiting NPS conditions increase grain yield temporal variability. Canola grain yield spatial variability could be reduced through fertilizations satisfying main nutritional requirements of the crop. Even though, this practice tends to increase temporal variability, the effect is not negative. The higher temporal variability is due to when there are not nutritional limitations, the crop can take advantages of the best weather conditions to reach higher yields.  Despite of the worldwide relevance of canola (Brassica napus L.), the agronomic information available about the crop is scarce in Argentina, even more about it yield spatio-temporal variability. This issue constrain the use of canola as winter alternative in intensificated agricultural sequences. The objectives were to i) determine canola yield spatial variability and its relationship with soil properties; ii) evaluate the effect of nutritional limitations removal on canola yield spatial variability, and iii) estimate canola yield temporal variability. We carried out two fields experiments in 2014 and 2015 seasons in a farm located in Diamante, Entre Ríos, Argentina (-32°12´44´´, -60°32´42´´). We used detailed soil maps (1:20.000) and grain yield maps to delimitate productivity zones.  In both fields two productivity zones were delimited, one of high productivity (HPZ) and one of low productivity (LPZ). We selected sampling points (10 in 2014 and 9 in 2015) distributed among productivity zones. Farmer management practice (70 kg N ha^-1 y 18 kg P ha^-1) was compared in each sampling point with a treatment with non-limiting NPS, which included 230 kg N ha^-1, 48 kg P ha^-1 and 30 kg S ha^-1. Soil samples were collected at sowing up to 0.60 m to determine chemical soil properties. Soil water content was measured up to 1 m depth at sowing, stem elongation beginning, flowering beginning, full flowering, flowering end and physiological maturity. Additionally, long-term simulations were performed using SALUS simple, in order to evaluate grain yield temporal variability. Analyzing the relationship between grain yield and soil properties we found inconsistent results between seasons. Average farmer grain yield was around 1700 kg ha^-1 in both seasons. Whereas, the non-limiting NPS grain yield was 2151 kg ha^-1 in 2014 and 3183 kg ha^-1 in 2015). Grain yield response to NPS was higher in LPZ than HPZ, particularly in 2015 season where weather conditions were better than 2014 season. However, average grain yield in non-limiting conditions was similar between productivity zones, showing a reduction in spatial variability. Long-term simulation results showed that temporal grain yield variability was higher in LPZ than HPZ, in both farmer management practice and non-limiting NPS. We also found that non-limiting NPS conditions increase grain yield temporal variability. Canola grain yield spatial variability could be reduced through fertilizations satisfying main nutritional requirements of the crop. Even though, this practice tends to increase temporal variability, the effect is not negative. The higher temporal variability is due to when there are not nutritional limitations, the crop can take advantages of the best weather conditions to reach higher yields.