9 °C), and the precipitation is less than 200 mm [26]. The North China Plain has a warm, semi-humid continental monsoon climate with mean annual temperature ranging from 8 °C to 15 °C [27]. Annual precipitation is extremely variable, ranging from 300 to 1000 mm, with an average of about 500 mm selleck products in North China [28]. The main cropping system is an annual winter wheat–summer maize rotation in North China. In South China, the mean annual temperature and annual precipitation are above 15 °C

and 800 mm, respectively, and double rice cropping and rice–wheat or rice–rape rotation system dominate in South China. The experimental durations of > 5 years of CA were grouped into four categories: 1–5, 5–10, 10–15, and > 15 years. Annual crop yield data were used to compare the CA effect sizes as affected by experimental durations. To compare the differences in CA effect sizes between climate patterns, annual precipitation, mean annual temperature, and aridity indexes in the tested areas were divided into three categories each: < 400, 400–600, and > 600 mm, < 5, 5–10, and > 10 °C, and < 1, 1–1.25, and > 1.25, respectively [29]. The effect size (Li) was calculated as the natural logarithm of the response ratio (R), which is the crop yield under CA practices (NT, CTSR, and NTSR) divided

by that under CT. Studies lasting several years or seasons were represented by several observations as annual and seasonal yield, respectively, in the data set [15]. Studies were weighted by observation numbers: Wi = n where Wi is the weight for the effect Daporinad cell line size from the ith paired trial and n is the number of observations. Mean effect sizes were estimated as ∑(Li × Wi) / ∑ Wi, with Li denoting the effect size from the ith paired trial, and Wi as defined above. The data were analyzed using MetaWin 2.1 software [30]. Bias-corrected 95% confidence intervals (CIs) were calculated for each mean effect size by a bootstrapping procedure (4999 iterations) [31]. To ease interpretation, the results in ln R were

back-transformed and reported as percentage changes under CA relative to CT ([R − 1] × 100). Means were considered to be significantly Oxymatrine different from one another if their 95% CIs did not overlap, and were significantly different from zero if the 95% CIs did not contain zero [31]. Positive mean effect sizes indicate an increase in crop yield caused by CA, whereas negative values indicate a decrease. The overall and actual effects of the specific CA practices are presented in Fig. 2. Taking all specific practices as an overall effect, CA significantly increased crop yield by 4.6% compared to CT (Fig. 2). However, there were large differences in specific effect sizes among the CA practices (P < 0.05). The yield gains of CTSR and NTSR were 4.9% and 6.3%, respectively, whereas there was no significant effect in NT compared to CT. The longer the experimental duration of CA, the higher was the magnitude of the increase in crop yield (P < 0.01, Fig. 2).