Objectives Exploration of the temporal and spatial changes of soil pH in Chinese farmland and the main driving factors are of great significance in alleviating acidification, soil quality improvement and sustainable land use.

Methods Based on the national long-term farmland fertilization monitoring data from the Ministry of Agricultural and Rural Affairs (950 experiments), we analyzed the temporal and spatial variations of soil pH in upland field, paddy field and paddy-upland rotation field. The main factors affecting the change of soil pH were analyzed using boosted regression tree model.

Results The soil pH value and coefficient of variation in China followed the sequence of upland (6.74 ± 1.19 and 17.63%) > upland-paddy rotation field (6.54 ± 0.93 and 14.26%) > paddy field (5.80 ± 0.81 and 13.95%). While

the paddy soil pH (5.74 ± 0.79) was higher than that of upland-paddy field and upland field in South China. From the initial stage of monitoring (I, 1988–2000) to the middle stage (II, 2001–2010), the pH of upland and paddy soils showed a decreasing trend, and the decreasing rates were 0.065 and 0.054 units per year, respectively (P < 0.05).

From stage II to III (2001–2018), pH of upland and upland-paddy soil increased with time, and the rising rates were 0.022 and 0.016 units per year, respectively (P < 0.05), but there was no significant change in paddy soils. The soil pH of uplands in Northeast, North China, Southwest, Middle and Lower Reaches of Yangtze River decreased linearly with time (P < 0.05), while opposite trend was found in South China from stage II to III (P < 0.01). The pH of paddy soils in Southwest, Middle Reaches of Yangtze River and South China decreased linearly from stage I to III (P < 0.01), but increased linearly from stage II to III (P < 0.01) in Northeast, Southwest and Lower Reaches of Yangtze River. The pH of upland-paddy soil in Southwest China decreased linearly from stage I to III (P < 0.01), while opposite result was observed in North China, Lower Reaches of Yangtze River and South China from stage II to III (P < 0.05). Pearson’s correlation and boosted regression tree model revealed that mean annual precipitation was the most important factor driving regional soil pH change, followed by soil texture, bulk density and organic matter content. Moreover, the long-term inputs of nitrogen fertilizer in upland and the input of potassium fertilizer in paddy and upland-paddy fields also played key roles in soil pH change.

Conclusions Overall, soil pH shows a trend of decreasing from the initial stage to the middle stage in upland and paddy soil, then slow increasing from the middle stage to 2018 in upland and upland-paddy soils, except that the pH in upland soil of Northeast China shows durative decrease. The application of nitrogen fertilizer in upland and potash in paddy field has caused soil pH decrease, so reasonable nutrient management, and the soil bulk density and organic matter amelioration should be considered to alleviate the decrease of soil pH in Chinese farmlands.