Abstract:

The aim was to clarify the distribution characteristics and driving factors of soil fertility in the production area of millet in Hebei Province, and provide a theoretical basis for the practice of millet production. This study was based on the data of cultivated land quality evaluation monitoring points in Hebei Province, the modified Nemerow index method was used to calculate the soil fertility index, and the main driving factors of soil fertility index were evaluated according to the random forest results. The analysis of variance showed that the average values of total nitrogen, organic matter, available phosphorus, available potassium and pH were 1.15 g/kg, 19.55 g/kg, 34.47 mg/kg, 165.56 mg/kg and pH 7.64, respectively. The contents of total nitrogen, organic matter, available potassium and pH were the highest in southern Hebei, which were 1.28 g/kg, 21.44 g/kg, 197.2 mg/kg and pH 8.14, respectively, and the available phosphorus content was the highest in eastern Hebei (67.16 mg/kg). The average soil fertility index of the main millet producing areas in Hebei Province was 1.21, with the primary driving factors ranking as available phosphorus>available potassium>pH>organic matter>total nitrogen. From the perspective of spatial distribution, the fourth and fifthgrade cultivated lands were predominantly located in northern and eastern Hebei, and the thirdgrade and above cultivated lands were mainly distributed in central and southern Hebei. The soil fertility index in eastern Hebei region was most strongly influenced by available potassium, whereas in southern, northern and central Hebei, it was primarily driven by available phosphorus. Correlation analysis showed that there was a significant positive correlation between total nitrogen, organic matter, available phosphorus and available potassium, and all of them were significantly positively correlated with soil fertility index. pH value was significantly negatively correlated with available phosphorus, and positively correlated with available potassium and elevation. There was a significant positive correlation between elevation and available potassium. The soil fertility in Hebei millet production area was at the upper middle level, but there were distribution differences, and there was great potential for improvement. The cultivated land of grade 3 (IFI=1.2~1.5) and above accounted for 49.46% of the total area, exhibiting a trend of lower fertility in the northwest and higher fertility in the southeast. On the whole, the available nutrients (available potassium and available phosphorus) had a stronger driving effect on soil fertility index than total nutrients, but the relationship between fertility index, nutrient content and millet yield remain unclear, necessitating further indepth research through regional experiments in the future.

Key words: cultivated land, nutrient, Nemerow index method, Kriging interpolation, Hebei millet production area, soil fertility index, random forest model, driving factors, spatial distribution