The research area is located in Jagdaqi District, Da Xing’anling, Heilongjiang Province, China. It is situated on the southeast slope of the Da Xing’anling Mountains at an elevation of approximately 387 m. The coordinates of the area are 50°22′29″N, 124°6′27″E, as shown in Figure 1. The terrain is generally flat with gentle undulations, characteristic of a low mountain and hilly region. The vegetation in the Jagdaqi District belongs to the cold temperate coniferous forest zone of the Da Xing’anling Mountains (Tian et al., 2009). Due to the climatic conditions, the region has a relatively sparse variety of plant species, predominantly of the Dahurian flora (Jia et al., 2021). The research area is located in the permafrost and seasonal frost transition zone and experiences a winter season from September to May of the following year (Jin et al., 2007). The frost-free period ranges from 85 to 130 days, and the climate is classified as a cold temperate continental monsoon climate with distinct seasons, characterized by long, cold winters and short summers. The stratum within a depth range of 2 m consists of loamy soil, clay, and gravel.

Field monitoring parameters include air temperature, humidity, wind speed and direction, precipitation, soil moisture and temperature. A 10-meter-high meteorological tower was installed at the monitoring site, with sensors for air temperature, humidity, wind speed, and direction placed at heights of 5 m and 10 m, respectively. The air temperature and humidity are measured using the HMP155A sensor, which is equipped with a radiation shield for outdoor installation. Wind direction and speed are monitored using the WindSonic sensor, while precipitation is measured with a T-200 series precipitation gauge. Soil temperature and volumetric water content (VWC) at depths of 0.2, 0.4, 0.6, 0.8, 1, 1.5, and 2 m in the active layer are measured using CS655 sensors. The data is collected by a CR1000X data logger at a 10-minute sampling interval. The data acquisition system is powered by solar panels and batteries, with data collection conducted through remote transmission.

Figure 2 shows the variation of daily average temperature from November 2022 to November 2023. The average annual temperatures at heights of 5 m and 10 m are 1.04°C and 1.36°C, respectively, with the highest annual temperatures reaching 25.93°C and 26.09°C, and the lowest annual temperatures dropping to −36.35°C and −34.95°C. Compared to the average annual temperature during 2001–2005 (0.15°C–0.6°C) (Zhang et al., 2010), the temperature in this region has significantly increased over the past 17 years. January is the coldest month, with an average temperature of −24.25°C, while July is the warmest month, with an average temperature of 20.64°C. There are 189 days with temperatures equal to or lower than 5°C. Therefore, the study area belongs to a severe cold region (Qin et al., 2016). The warm months (with temperatures exceeding 10°C) are from May to September, with average temperatures of 3.18°C in April and 4.38°C in October. The air-freezing index and air-thawing index are the cumulative sums of degree-days below and above 0°C, respectively, over the course of a year. In this region, the air-freezing index is −2318.95 °Cd, and the air-thawing index is 2698.52°Cd. The results of air temperature monitoring can offer valuable insights for assessing climate change, planning agricultural production, and guiding infrastructure development.

Figure 3 shows the precipitation from October 2022 to September 2023. The total precipitation from October 2022 to September 2023 is 397.1 mm. The distribution of precipitation shows that the highest rainfall occurs in July, with 109.8 mm, and the total rainfall from June to September is 307.5 mm, accounting for approximately 77.4% of the annual precipitation. Winter has lower precipitation, with a total of 44.9 mm from November to March, accounting for 11.3% of the annual rainfall, with only 1.2 mm in February. In the months of frequent temperature fluctuations, October and April, the precipitation is 8.9 mm and 23.6 mm, respectively. On the other hand, the annual rainfall days are 106, with a total of 57 days from June to September, accounting for 53.77% of the annual rainfall days (Figure 4). In addition, the maximum rainfall days in June is 18, while both December and February have the least rainfall days of 3. In terms of rainfall levels, there are only 3 days throughout the year with heavy rainfall, 6 days with moderate rainfall, and the rest are light rainfall. The maximum single-day rainfall occurred on July 10th, with a 24-hour cumulative rainfall of 46.1 mm. The above monitoring results can provide guidance for local water resource utilization, flood prevention, and disaster mitigation.

Figure 5 shows the wind speed and wind direction from October 2022 to September 2023. The northwest wind direction was the prevailing wind direction, and this wind direction frequency accounts for approximately 47% of the total annual wind direction frequency. The annual wind speed ranges from 0.045 to 10.33 m/s, with an average speed of 1.51 m/s, falling into the low wind speed category. Most of the time, the wind speed is between 0 and 4 m/s, accounting for 95% of the time. In terms of wind speed distribution, the average wind speed is highest in May at 2.18 m/s, and lowest in January at only 1.12 m/s. In terms of the trend, the average wind speed gradually increases from January to May. From June to September, the wind speed remains relatively stable, averaging between 1.3 and 1.42 m/s. The results of wind speed and wind direction monitoring can guide energy development and air quality assessment in the region.

Figure 6 illustrates the relative humidity (RH) variations from October 2022 to September 2023. During this period, the average relative humidity at 5 m and 10 m heights was 63.49% and 62.1%, respectively. Notably, the average RH between June and September was higher, fluctuating between 71.54% and 76.6%. It is important to highlight that from January to May, there was a clear negative correlation between relative humidity and wind speed, with RH progressively decreasing as wind speed increased. The lowest RH values at both heights were recorded in May, with 44.58% at 5 m and 43.38% at 10 m. This could be attributed to the low precipitation levels, dominated by solid-state precipitation, along with a significant decrease in vegetation transpiration during this period. Under such conditions, the increase in wind speed likely enhanced air convection and mixing, facilitating the quicker diffusion and evaporation of water vapor, thereby reducing the relative humidity. Additionally, in October, during frequent temperature fluctuations, the RH at 5 m and 10 m was 56.41% and 54.57%, respectively. These monitoring results can offer valuable guidance for agricultural production and improving the freeze-thaw durability of buildings in cold regions.

The research area is located in a seasonally frozen ground zone, and Figure 7 illustrates the seasonal freeze-thaw process within a 2-meter depth range. Firstly, it can be observed that the maximum seasonal freezing depth during the monitoring period was 1.93–1.99 m, signified by middle-thick seasonally frozen ground. In terms of the development of the freezing depth, as winter progressed, the freezing of the 0.2-meter layer began on November 5th, and the freezing depth gradually increased, reaching its maximum in early March, and then gradually decreased. From the 0°C isotherm line, it can be seen that the seasonally frozen ground starts to thaw from both the top and bottom directions, and completely melts on May 5th. Therefore, the process of frozen depth development in the region at depths of 0.2–2 m lasts approximately 4 months, with the melting development process lasting 2 months. In the warm season, at a depth of 2 m, the soil temperature is above 12°C for about 2.5 months, above 8°C for 5 months, and the 16°C isotherm can extend to a depth below 1.4 m. Regarding the soil temperature at a depth of 0.2 m, the shallow soil layer is warmer from June to August, with approximately 2 months of soil temperature above 20°C.

Figure 8 shows the variation of soil moisture and temperature from February 2022 to October 2023. The annual average ground temperature in the depth range of 0.2–2 m is between 6.22°C and 6.6°C, as indicated in the graph. Due to differences in depth, there are significant variations in ground temperature fluctuation characteristics across different layers. The fluctuation amplitudes of ground temperature for seven layers are 37.44, 32.19, 27.65, 24.7, 22.86, 17.73, and 13.21°C respectively. Among them, the average ground temperature in the 0.2–0.6 m soil layer during the months of May to September ranges from 15.29°C to 17.51°C. In terms of freezing duration, the freezing duration at a depth of 0.4 m is 147 days, whereas at a depth of 1.5 m, it decreases to 62 days. During the freezing development period, the freezing depth development rate ranges from 0.74 to 2.22 cm/day, with slower freezing rates observed in the 0.4–0.6 m soil layer. As temperatures drop during the freezing period, soil moisture content declines significantly as water transforms from a molecular to a crystalline ice structure. In the early stages of thawing, the moisture content in the shallow soil layers rises sharply, likely due to the infiltration of surface snowmelt, resulting in a supersaturated condition in the shallow layer.

Seasonal freeze-thaw variations in the ground are indicators of climate change, influenced by soil temperature states and moisture distribution (Luo et al., 2020). Human activities can significantly alter these seasonal freeze-thaw processes, such as changes in land use, construction activities, greenhouse gas emissions, and pollutant releases (Shiklomanov et al., 2017). Specifically, activities like farmland reclamation, grazing, and urbanization modify land use patterns, which in turn affects the type and thickness of surface cover. This change impacts the heat transfer processes at the surface, thereby altering the depth and timing of seasonal freeze-thaw cycles. Infrastructure development in cold regions (such as roads, buildings, and pipelines) alters the thermodynamic properties of the soil (Jin et al., 2023). Human-induced greenhouse gas emissions are a major driver of global warming, which directly affects seasonal freeze-thaw characteristics, leading to shorter freezing periods and increased frequency of freeze-thaw events. Pollutants generated by industrial activities and transportation, such as dust and black carbon, settle on snow or ice surfaces, reducing their albedo and accelerating melting, thereby impacting the thermal balance of permafrost regions (Jin et al., 2024). Overall, human activities change the characteristics of seasonal freeze-thaw through both direct and indirect means.