Estimating indigenous nitrogen supply (INS) by measurement of crop N uptake in N omission plots for site-specific N management is not feasible on a routine basis because it involves destructive plant sampling and plant tissue analysis, which is time-consuming and expensive. The objective of this study was to determine the amount of INS and develop a method to estimate it using soil testing in the North China plain (NCP). On-farm experiments at 229 sites were conducted from 2003 to 2005 in seven key winter wheat (Triticum aestivum L.)/ summer maize (Zea mays L.) production regions of the NCP. The mean INS during the wheat-growing season was129 kg N ha -1 with a range from 62 to 212 kg N ha -1 , and it varied from 69 to 202 kg N ha -1 with a mean of 142 kg N ha -1 during the maizegrowing season. Considering all sites, the variability of INS was not simulated by initial soil N min or apparent N mineralization (N organic ) alone, while together they could explain about 38% and 60% of INS during the wheat and maize-growing seasons, respectively. During the wheat-growing season, mean N organic was 63 kg N ha -1 , and 59% and 33% of its variation could be explained by SOM in high-yielding regions (mean yield, 7.6 t ha -1 ) and low-yielding regions (mean yield, 5.3 t ha -1 ), respectively. Mean N organic during the maize-growing season was 109 kg N ha -1 , 22% of which could be explained by SOM across all sites. An average of 40% and 42% of INS variation could be explained by both SOM and initial soil N min content during the wheat and maize-growing seasons, respectively. We conclude that the accuracy of estimating crop N requirement for site-specific N management will be increased by using initial soil N min and SOM.