Changes in meteorological and soil physical parameters
From November 2020 to August 2021, the daily mean air temperature and precipitation fluctuated markedly (Fig. 1a). The air temperature gradually declined after the beginning of freezing period, reaching a low of -14.49 °C on January 31, 2021 during the stable freezing period. Subsequently, the air temperature increased during the thawing period, reaching a maximum, for the complete thawing period, of 14.62 °C on July 24, 2021 (Fig. 1a). The mean air temperature <0 °C lasted for 160 days, and the mean values for air temperatures during the freezing, stable freezing, thawing, and complete thawing periods were -5.91, -7.21, 1.47, and 10.7 °C, respectively. The cumulative precipitation over the 277-day experimental period was 194.5 mm, with the majority of rainfall occurring in May, June, and July (Fig. 1a). Precipitation that occurred during the thawing and complete thawing periods accounted for 40.0% and 58.8% of total precipitation, respectively (Fig. 1a).
Meanwhile, 0–5 cm mean soil temperature and moisture calculated from automatic monitoring data revealed similar trends with air temperature and precipitation variations (Fig. 1a, b). The mean soil temperature in the 0–5 cm depth was generally higher than the corresponding air temperature, with values of -4.37, -3.41, 8.14, and 16.0 °C during the freezing, stable freezing, thawing, and complete thawing periods, respectively. The 0–5 cm soil moisture was closely linked to the amount of precipitation, with a value <5.00% during the freezing and stable freezing periods and reaching a maximum, for the complete thawing period, of 14.4% on July 26, 2021 (Fig. 1b).
Dung decomposition and nutrient release
During the freezing and decomposition period in winter, the moisture of YD and TSD sharply decreased from 83.5% to 8.09% and from 50.5% to 4.88%, respectively (Tables 1 and 2). The moisture of YD was significantly higher (P < 0.05) than that of TSD across the freezing, thawing, and complete thawing periods (Table 2). YD moisture across the thawing and complete thawing periods was significantly higher (P < 0.05) than that across the freezing and stable freezing periods, whereas TSD moisture across the complete thawing period was significantly higher (P < 0.05) than that across other freezing–thawing periods (Table 2). YD dry matter content (400−563 g/kg) was significantly greater (P < 0.05) than that of TSD (22.4−46.6 g/kg) throughout the entire decomposition period. TSD dry matter content slowly decreased over the first three freezing–thawing periods, and approximately half of the dry matter was lost (P < 0.05) across the complete thawing period (Table 2).
NH4+-N and NO3−-N concentrations for YD greatly decreased from initial values of 1242 and 15.8 mg/kg to 73.0 and 3.22 mg/kg, respectively, at the end of the freezing period. The corresponding values for TSD decreased from 641 and 3.88 mg/kg to 115 and 2.77 mg/kg, respectively (Tables 1 and 2). Furthermore, the YD NH4+-N concentration was lower (P < 0.05) than that of TSD and significantly decreased (P < 0.05) across the stable freezing and thawing periods (Table 2). This differed from the significant increase (P < 0.05) in TSD from the freezing to stable freezing period, followed by a significant decrease (P < 0.05) across the thawing and complete thawing periods (Table 2). NO3−-N concentration did not differ significantly (P > 0.05) across the four freezing–thawing periods for YD, whereas the value for TSD significantly decreased and lower (P < 0.05) than that for YD across the complete thawing period (Table 2). Similarly, the AP concentration of YD and TSD simultaneously decreased from 344 and 419 mg/kg at the beginning of the experiment to 129 and 371 mg/kg at the end of the freezing period, respectively (Tables 1 and 2). The AP concentration of TSD was significantly higher (P < 0.05) than that of YD across the freezing, stable freezing, and thawing periods; however, the amount of AP in TSD significantly decreased (P < 0.05) across the thawing and complete thawing periods (Table 2). In contrast, the AP concentration of YD increased across the stable freezing period relative to the freezing period, with the value significantly decreased (P < 0.05) during the decomposition process in the complete thawing period (Table 2).
Soil moisture and nutrient dynamics
The SM values of 2.93% and 5.37% for CK at 0–5 and 5–10 cm depths across the complete thawing period, respectively, were significantly greater (P < 0.05) than those across the freezing and stable freezing periods (Fig. 2a, b). YD, TSD, YU, and TSU deposition failed to significantly increase 0–5 and 5–10 cm SM across the freezing period but statistically increased (P < 0.05) 0–5 cm SM across the thawing period. Furthermore, YD treatment resulted in significantly higher (P < 0.05) 0−5 cm SM across the complete thawing period (Fig. 2a). The dung and urine treatments showed a similar trend: SM at 0–5 cm depth across the thawing and complete thawing periods was significantly higher (P < 0.05) than those across the freezing and stable freezing periods (Fig. 2a), while the SM at 5−10 cm depth that was impacted by dung and urine deposition exhibited differential change characteristics (Fig. 2b).
Soil NH4+-N and NO3−-N concentrations for the CK treatment varied across the four freezing–thawing periods, with the highest NH4+-N concentration at 0−5 and 5−10 cm depths reached 4.71 and 4.64 mg/kg, respectively, across the complete thawing and stable freezing periods (Fig. 3a, b). The corresponding highest NO3−-N concentration at 0−5 and 5−10 cm depths reached 3.26 and 1.16 mg/kg, respectively, across the thawing period (Fig. 4a, b). YD treatment significantly increased (P < 0.05) soil NH4+-N concentration at 0–5 cm depth across the stable freezing period, and TSD treatment significantly increased (P < 0.05) NH4+-N concentration at 0–5 and 5–10 cm depths and NO3−-N concentration at 5–10 cm depth across the thawing period (Fig. 3 and 4). In contrast, urine deposition significantly increased (P < 0.05) soil NH4+-N and NO3−-N concentrations, especially across the freezing period. The peak value of soil NH4+-N concentration at 0−5 and 5−10 cm depths was 127 and 70.0 mg/kg for YU treatment and 89.4 and 36.0 mg/kg for TSU treatment, respectively (Fig. 3a, b). The corresponding soil NO3−-N concentration at 0−5 and 5−10 cm depths reached 90.5 and 38.0 mg/kg for YU treatment and 29.6 and 6.80 mg/kg for TSU treatment, respectively (Fig. 4a, b). The complete thawing period led to a significant increase (P < 0.05) in NH4+-N concentration compared to the value throughout the thawing period (Fig. 3). The corresponding NO3−-N concentration treated with YU significantly decreased (P < 0.05) from the freezing to the complete thawing period, but the value treated with TSU exhibited fluctuations during the entire experimental period (Fig. 4).
Soil AP concentration of CK treatment at 0–5 and 5–10 cm depths ranged from 1.76 to 4.08 mg/kg and 0.78 to 1.64 mg/kg, respectively, across the four freezing–thawing periods (Fig. 5). The lowest values at 0–5 and 5–10 cm depths simultaneously occurred across the thawing period (Fig. 5a, b). The deposition of YD and YU significantly increased (P < 0.05) AP concentration at 5–10 cm depth relative to the CK across the freezing period and decreased (P < 0.05) the concentration at 0–5 cm depth across the stable freezing period (Fig. 5a, b). The TSD treatment significantly increased (P < 0.05) soil AP concentration at 0−5 cm depth across the thawing and complete thawing periods, and the increase in concentration at 5–10 cm depth occurred across the complete thawing period (Fig. 5a, b). Soil AP concentration at 0–5 and 5–10 cm depths that were affected by YU and FTCs significantly decreased (P < 0.05) from the freezing to complete thawing periods (Fig. 5a, b), with the lowest value at 5–10 cm depth that was treated by YD and TSU simultaneously occurred across the thawing period (Fig. 5b).
Soil N2O flux and its cumulative emissions
Soil N2O emission varied with the extend of the freezing-thawing period, and the N2O fluxes occurred during thawing and complete thawing periods were detected higher and largely fluctuated than other freezing-thawing periods (Fig. 6a). The highest soil N2O flux for the CK, YD, TSD, YU, and TSU treatment reached 13.3, 20.1, 20.4, 24.2, 14.1 mg N/m2 h that respectively experienced 166, 197, and 206 days after excreta deposition (Fig. 6a). Cumulative N2O emissions for YD (492.8 g N/ha) and YU (510.4 g N/ha) treatments across the four freezing-thawing periods were measured significantly greater (P < 0.05) than that for other treatments, and amounts of soil N2O emission that affected by YD and YU treatments seemed largely derived from the thawing and completed thawing periods with high soil mineral N concentrations and concentrated precipitation compared to the CK and Tibetan sheep excreta treatments (Fig. 6b). In contrast, TSD and TSU treatments were considered to slightly alter soil N2O flux and its cumulative emissions as the weak effects of dung and urine in affecting soil mineral N concentration, nitrification, and denitrification processes (Fig. 6a, b).
Relationship between excreta and soil nutrient parameters
Two-way ANOVA revealed that experimental factors, including treatments, freezing–thawing periods, and their interactions significantly (P < 0.001) influenced soil NH4+-N and NO3−-N, whereas the AP was significantly influenced (P < 0.001) by the freezing–thawing period (Table 3). The SM was positively correlated with NH4+-N across the thawing period (r = 0.443, P < 0.05) and negatively correlated with AP across the freezing period (r = -0.519, P < 0.01), stable freezing period (r = -0.710, P < 0.01), and complete thawing period (r = -0.642, P < 0.01) (Table 4). Soil NH4+-N was positively correlated with NO3−-N across the freezing period (r = 0.928, P < 0.01), stable freezing period (r = 0.896, P < 0.01), and complete thawing period (r = 0.880, P < 0.01). It was also positively correlated with AP across the freezing period (r = 0.428, P < 0.05) and thawing period (r = 0.664, P < 0.01) (Table 4). In contrast, soil NO3−-N was only positively correlated with AP across the freezing period (r = 0.418, P < 0.05) and with TN across the freezing (r = 0.484, P < 0.01) and thawing periods (r = 0.454, P < 0.05) (Table 4). The SOC was positively correlated with TN (r = 0.957, P < 0.01), NH4+-N (r = 0.445, P < 0.05), NO3−-N (r = 0.373, P < 0.05), and AP (r = 0.678, P < 0.01) across the thawing period (Table 4).