Root forks and link length influence spatial distribution and resource retrieval of plant roots. The trade-off between root forks and link length has a significant effect on ecological adaptation of root architecture. This study used the MaChangtan grassland in eastern SuNan country on the northern slope of Qilian Mountains in Northwest China (38°47'22.1"-38°52'11.4" N, 99°45'11.7"-99°57'41.2" E) as a study field. The altitude is 2610-2632 m, the average annual temperature varies within a range of 1-2℃, the annual precipitation is 270-350 mm, the climate is continental and vertical, the soil is composed primarily of mountain chestnut soil, and the dominant native vegetation is mainly perennial grasses and forbs. In August 2014, the density, height, coverage, and species of the plants in the community were recorded, which were from 1 m×1 m grids in four kinds of degradation gradients in six replicates, SC-900 Soil Compaction Meter was used to measure soil compaction (at 0-10 cm, 10-20 cm, and 20-30 cm) in each grid (five replicates), and the average compaction of the entire 0-30 cm depth was used to represent the degradation gradient. According to detailed community study, population characteristics of Stipa krylovii and grassland stocking rate (2004-2013), we defined as:no degradation (> 90%), light degradation (98%-115%), moderate degradation (120%-131%), and heavy degradation (135%-156%). Then, the method of whole root system excavation was used to dig soil cores (30×30×50 cm) from 30 cm×30 cm grids in all four kinds of degradation gradients (240 in all), and sieved it (bore diameter=0.25 mm) to clean and select Potentilla acaulis in a nearby river. Meanwhile, we used stratified sampling (0-50 cm), where each layer of soil was 10 cm, with an auger (diameter=4 cm), to collect soil in every layer and the total depth. In the laboratory, the roots divided from P. acaulis were scanned with Win-RHIZO to measure its root forks and link length, and the biomass of different plots was measured after being dried in an oven (at 105℃ for 12 h). Similarly, soil water was measured by oven drying (at 84℃ for 24 h) and measuring weight loss. Density, height, coverage, aboveground biomass, and soil moisture content of the grassland community and the root forks of the P. acaulis population declined with increasingly degraded natural grassland, while the density, height, coverage, aboveground and belowground biomass, root-shoot ratio, root link length of the P. acaulis population, and soil compaction of the grassland community increased. Root forks and link length were negatively correlated with one another (P < 0.05), although the exact relationship varied along the degraded alpine grassland. A highly significant negative correlation (P < 0.01) was found between root forks and link length of P. acaulis population in undegraded and severely degraded grasslands, with a less significant correlation (P < 0.05) in lightly and moderately degraded grasslands. The P. acaulis root architecture construction of root forks decreases and link length increases with increasing degradation of natural grassland, reflecting the trade-off of biomass allocation between different plant parts.