Abstract: Abstract: With the stability of arable land's quantity, monitoring land quality has become a high priority research for understanding the effects of the dynamic change of arable land on food security in China, as well as the layout method on monitoring arable land quality change. However, there are integrated factors such as climate, terrain, soil, access to irrigation, rural road, trade-off between input and output, which affect arable land quality change over time and space. We propose a new monitoring framework titled factors' combination, which includes such factors affecting arable land quality as natural conditions (e.g., climate, soil, geomorphology), the utilization of level (e.g., farmland infrastructure, land management, land use coefficient), income level (e.g., land use structure and mode, the input and output of arable land, land economic coefficient), and reference cropping system to form a monitoring reference arable land unit. We illustrate this new method using the Dianqian plateau mountain area as a case study. Spatial overlay analysis of main factors and geostatistics method using GIS were employed to test this method. Specific steps of factors' combination method are as follows: 1) we preliminarily determine the number of monitoring reference samples according to the type of factors' combination; 2) on the basis of the proportion of arable land area at each grade accounting for the total area, we then revise the number of monitoring samples and supplement monitoring samples for those gradations which have relatively few monitoring samples; 3) based on GIS analysis results, if the same factors' combination distributes in the different space positions of second zone and meets the requirements of monitoring sample, multiple figure spots of the factors' combination will be kept at the same time, and eventually figure spot of the grading unit will be determined; 4) given overlay the map spot of grading unit and the national standard sample and the provincial standard sample respectively, we take the national standard sample or provincial standard sample as the monitoring sample for those overlaying parts; then convert the remaining figure spots of grading into a point as the monitoring sample, and determine the final number of monitoring samples, spatial location and its source; 5) we build up model on representative index of area of monitoring sample and adopt the geo-statistical method to carry on the representative test for monitoring sample to optimize the monitoring sample. The results show that 144 monitoring reference sample units include 7 from the national standard sample, 44 from provincial standard sample and 93 from arable land grading unit, and they were selected as a whole for monitoring arable land quality in Dianqian plateau mountain area using our new method of factors' combination. The distribution of the selected monitoring reference land units not only provide samples to monitor the arable land quality change in the second zone of national scale, but also meet the requirements of statistical science and representative of area. Layout method for monitoring sample point of arable land quality level using factors' combination, can provide reference for building up the whole country's dynamic monitoring systems, and offer technical support to achieve comprehensive management of quantity and quality of arable land at national level in China.