The grazed grasslands of Inner Mongolia offer a critical research setting for studying the impacts of long-term grazing on soil, hydraulic, vegetative, and meteorological variables, and potentially contribute to examine threshold responses to grazing activity or climate dynamics in colder Eurasian temperate grasslands. The dataset consists of long-term observations of soil temperature and moisture, as well as other related parameters across three scales: pedon, field, and catchment scale. This includes: i) At the pedon scale, data collection was conducted on five sites: long-term grazing exclusion since 1979 (IJG79), short-term grazing exclusion since 1999 (IJG99), continuous grazing (CG), heavy year-round grazing (HG), and moderate winter grazing (WG), from May 2004 to August 2008. Profiled soil moisture at depths of 5, 20, and 40 cm was continuously monitored using theta-probes, while soil temperature at depths of 2, 8, 20, 40, and 100 cm was monitored using Platinum ground temperature probes. Since 2016, newly automated monitoring instruments were also used for continuous monitoring of soil temperature and moisture at depths of 10, 30, 50, 70, and 100 cm at the IJG79, IJG99, and CG sites. ii) At the field scale, during the growth period from 2004 to 2008, a regular sampling grid (about 100 points) was established in all five sites using differen- tial GPS and UTM systems. Soil water content, water drop penetration time, shear strength, and hydraulic conductivity were measured once per week/month. At the beginning, soil organic carbon concentration, bulk density, soil texture, and plant parameters were also taken at each grid point. iii) At the catchment scale, a field sampling scheme was designed, using land use and soil type as stratification variables. A to- tal of 30 sampling points were selected. At each sampling point, detailed soil surveys were conducted to measure soil profile characteristics, including soil colour, texture, structure, and chemical elements. Additionally, some soil hydrological properties were recorded on site. This dataset offers criti- cal insights into the factors influencing livestock carrying ca- pacity in Mongolian grasslands. The integration of these data types can substantially enhance our understanding and man- agement of these ecosystems. (c) 2024 The Author(s). Published by Elsevier Inc. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/)