[Objective] This study was intended to explore spatial variation of soil color parameters and reliability of the parameter being used as an alternative indicator of climate change. [Method] In this study, color analysis of the soils developed under different climatic conditions was carried out. The soil profiles studied in this paper were scattered in the range geographically from 23.79°N to 64.72°N, and from 38.52°E to 148.45°W, striding across the low latitude, mid-latitude and high latitude. Among the total of 32 soil profiles, 9 had data derived from the literature and the remaining 23 sections were collected from Fujian, the Ganzi Tibetan Autonomous Prefecture in the West Sichuan Plateau, Siberian of Russia, and Alaska of the USA. The soil profiles were distributed one each in Zhao'an, Yunxiao, Changtai, Hui'an, Putian, Fuqing, Zhangping, Quanzhou, Changting, Ninghua, Lianjiang, Minqing, Shaxian, Mingxi, Ningde, Taining, Jian'ou, Zhenghe, Songxi, Wuyishan, the Ganzi Tibetan Autonomous Prefecture, Siberian of Russia, and central Alaska. As the soil profiles were distributed extensively in such a huge region, they divesified greatly in soil type, and developed separately in a huge variety of climatic conditions, including humid subtropical climate, temperate continental climate and Qinghai-Tibet Plateau climate, etc., which differed sharply in annual mean precipitation and annual mean temperature. [Result] Results show: (1) on a small spatial scale, soil color parameters are complex in variation and not so significantly related to climate, while on a large spatial scale, soil color parameters exhibit a certain trend of varying with latitude. Redness (a*) of the soils declines with increasing latitude, while yellowness (b*) does a fitting curve with latitude quite similar to a parabola with an upward opening. The ratio of yellowness to redness (b*/a*) increases approximate linearly with increasing latitude；(2) Color parameters of the soil profiles are more or less related to climatic factors (annual mean precipitation and annual mean temperature); the coorelation coefficient of b*/a* with annual mean temperature is the highest. The coorelation coefficient of a* with annual mean precipitation is slightly higher than that of b*/a*, but a* is more likely to be affected by other factors. [Conclusion] Therefore, the soil chroma parameter b*/a* is quite sensitive to spatial variation of soil color, so is a suitable indicator of climate changes.