The spatial extent of plantation forests continued to increase after the Grain for Green Project (GGP) started in January 2002 in China, with more than 20 thousand km² forests developed from 1999 to 2008 throughout the entire country. From 2012 to 2014, the purpose of Beijing's plain reforestation project was to increase the spatial extent of the forest area in the administrative sub-units that make up the city of Beijing with a plains habitat. In the GGP, one third of the farmland in this region has been replanted with trees to increase the extent of the green area and improve the environment. However, considerable controversy remains related to whether afforestation can effectively result in the conservation of biological diversity. The objective of this study is to explore the effects of afforestation on the ground-dwelling spider community and evaluate the effects of the GGP on the ecosystem and biodiversity during afforestation. From March to September 2014 and 2015, pitfall traps were used to collect ground-dwelling spiders every 10 days in habitats that had been returned to forests from farmland for 2 and 3 years, respectively, in Shunyi District, Beijing. Ground-dwelling spider communities were compared between two adjacent habitats, a former non-crop habitat and an afforested habitat. A total of 18 pitfall traps were placed in the same randomly selected pattern in both habitats. During the investigations, we collected a total of 1465 individuals from eight families, 18 genera and 32 species in 2014, and a total of 2186 individuals from 10 families, 25 genera and 45 species in 2015. Afforestation within the study sites had a positive effect on ground-dwelling spider diversity over the first 3 years. The species of ground-dwelling spiders present changed with afforestation; total species richness increased and some of the farmland spiders were not found in afforested habitats. In 2014, species richness and a diversity index of ground-dwelling spiders in former non-crop habitats were significantly higher than in afforested habitats (P < 0.01). In 2015, the diversity index in former non-crop habitats was still significantly higher than in afforested habitats (P < 0.01), while species richness in former non-crop habitats was significantly higher than in afforested habitats (P < 0.05). An evenness index in the two habitats was not significantly different in both 2014 and 2015. In the second year after afforestation, the peak of the ground-dwelling spider activity density in former non-crop habitats occurred in mid-June, which was 1 month earlier than that in the afforested habitat. In the third year after afforestation, the ground-dwelling spider activity density exhibited similar trends in both habitats; that is, in both habitats the peak of activity density occurred at the same time in early July. The results show that in the first years of habitat alteration, the ground-dwelling spider community in the afforested habitat was still in a stage of community reconstruction, and the community's stability increased year by year. Returning farmland to forests influenced the original farmland ecosystem through changes in the vegetation type, structure and management practices. As the forests mature, additional effects on spider species at the site are expected because of changes in light conditions, microclimate, vegetation and so on. Therefore, the effects of returning farmland to forests on the ground-dwelling spider community will need further study over time. The aim of this study is to determine whether afforestation contributed to enhancing biodiversity and conservation. Studying the process of change in biological communities after habitat alteration is of great theoretical significance.