Soil organic matter (SOM) is an important component of soil and plays an important role in improving the soil’s physical and chemical properties. Ascertaining the spatial distribution of soil organic matter and its main controlling factors in the context of provincial scale farming is of important guiding significance for soil carbon sequestration, emission reduction and sustainable utilization. Using 257 soil profiles from the second soil survey in Shandong Province, GIS, we applied geostatistical methods to study the spatial distribution characteristics of SOM in topsoil. In addition, correlation and regression analyses were used to explore the main controlling factors over the spatial variation of SOM. The results showed that the mean amount of SOM in Shandong province ranged from 1.20–74.90 g·kg⁻¹, with a coefficient of variation of 73.52%, which is a medium level of variation. The distribution of SOM in the study area was patchy, with higher levels of organic matter in the central, eastern, and northern parts, and lower levels of organic matter in the south-west. The comprehensive explanatory ability of all factors reached 52.30%. Soil type and parent material were the main controlling factors for the spatial variability of SOM in the Shandong Province, followed by soil texture and land use type, with topography and climatic factors having relatively little influence.

Soil organic matter (SOM) plays a crucial role in promoting soil tillage, improving soil fertility and providing crop nutrients. Investigation and sampling are the premise and basis for understanding the spatial distribution of SOM. The number of sampling points will affect the accuracy of spatial variation of SOM. Therefore, it is important scientific work to determine a reasonable number of sampling points under the premise of ensuring accuracy. In this study, Kenli County, a typical area of the Yellow River Delta in China, was taken as an example to investigate the effect of different sampling points on spatial-variation expression of SOM. A total of 12 sample subsets (including 900 samples) were randomly sampled at equal intervals from the 900 sample points, using geographic information system (GIS) technology and geostatistical analyses to explore the optimal number of samples. The results showed that the SOM content in the study area had a lower-middle degree of variation. As the number of sample points decreased, the spatial distribution of SOM showed the gradual weakening of detail-characterization ability; and when the number of sample points was too small (<100), there was a wrong expression that was not consistent with the actual situation. The value of RMSE has no obvious regularity with the change of sample number. The values of both ME and ASE showed a significant inflection point when the number of samples was 150 and remained around 0 and 4 as the number of samples increased, respectively. Combined with the three indicators of ME, RMSE and ASE, collecting at least 150 samples can satisfy the spatial-variation expression of SOM, equivalent to 107 sample points within the area of 1000 km². The research results could provide important references for investigation of SOM content in areas with similar natural geographical conditions.

Soil salinization is a significant threat to agricultural production, making accurate salinity prediction essential. This study addresses key challenges in the Yellow River Delta (YRD) soil salinity inversion, including (1) determining which Landsat 8 OLI level performs better, (2) identifying the most suitable month for salinity inversion, and (3) improving model performance and identifying important variables in modeling. Thus Landsat 8 OLI images (Level-1 and Level-2) for 12 months were collected, then images having less than 10% cloud cover were selected and processed to extract spectral values. A total of 86 sampled points were processed to measure soil salinity. Using Pearson correlation and expert insights, January 15 and August 26 were identified as suitable dates for inversion. Then, seven original bands, 29 spectral indicators, and 39 derived variables which created through six mathematical transformations, were used to construct the following three models: partial least squares regression (PLSR), random forest (RF), and backpropagation neural network (BPNN). The results showed the following: (1) The Level-1 data, after FLAASH atmospheric correction, outperforms Level-2 data. (2) January is optimal for salinity inversion. (3) Among the three models, RF outperformed the others, achieving test set R² = 0.55, RMSE = 3.4, suggesting that the combination of spectral indicators and mathematically transformed variables can effectively enhance model accuracy for predicting soil salinity in the YRD. Furthermore, SWIR1, SWIR2, CLEX, second-order difference of SWIR1, and first-order difference of SWIR2 along with NIR played a key role in modeling.

The degradation of black soil cropland has occurred to varying degrees in the northern agropastoral ecotone. Crop–forage rotation is an effective way to improve soil quality, with Medicago being the preferred perennial legume. The C, N, and P stoichiometric ratios are key indicators of soil quality and organic matter composition, reflecting the status of the internal C, N, and P cycles in soil. This study aims to investigate the ecological stoichiometric ratios of Medicago grassland soils with different planting durations, explore the regulatory effects of nitrogen fertilizer on soil organic carbon (SOC), total nitrogen (TN), and total phosphorus (TP) content, and assess the impacts of these changes on the Medicago grassland ecosystem. This study was conducted on the long-term cultivated grassland core experimental platform of the Hulunber National Field Station. Based on forage yield and soil nutrient measurements, field-based observations and laboratory analyses were carried out. Medicago × varia was the study subject, with different nitrogen fertilizer treatments: CK (0 kg N ha⁻¹), N₇₅ (75 kg N ha⁻¹), and N₁₅₀ (150 kg N ha⁻¹). A randomized block design was adopted. Variance analysis, boxplot statistics, and scatterplot fitting methods were used to examine soil properties and assess the effects of nitrogen application on the C, N, and P stoichiometry of soils in established perennial Medicago grasslands. The results indicate that, based on the growth characteristics of alfalfa, soil nutrient dynamics, and its effectiveness in improving soil quality, the optimal rotation period for alfalfa in the northern agropastoral ecotone is 4–5 years, but it can also be shortened to 3 years. Soil carbon, nitrogen, and phosphorus contents are significantly influenced by the planting duration. As the planting years increase, soil carbon and nitrogen contents first increase and then decrease, while soil phosphorus content initially decreases followed by a slight increase. Soil pH gradually rises with both planting years and soil depth. Both low and high levels of nitrogen fertilizer application reduce soil organic carbon concentration (by 0.40% and 10.14%, respectively). Low nitrogen fertilizer application increases soil nitrogen concentration (by 1.50%), whereas high nitrogen fertilizer application decreases it (by 7.6%). Both nitrogen levels increase soil phosphorus concentration (by 36.67% and 35.26%, respectively). For soil from an alfalfa grassland planted for 8 years, the carbon-to-nitrogen ratio ranges from 9.08 to 9.76, the carbon-to-phosphorus ratio from 13.00 to 151.32, and the nitrogen-to-phosphorus ratio from 1.65 to 17.14. In summary, alfalfa yield is primarily influenced by the nitrogen fertilizer application rate, planting duration, stoichiometric ratios, and pH. Nitrogen fertilizer application has a positive regulatory effect on soil stoichiometric ratios. The annual yield can reach 8.94 to 10.07 tons per hectare., but phosphorus remains a limiting factor. These findings provide crucial data for understanding the impact of ecological stoichiometry on crop–forage rotation cycles, as well as optimal land use and quality improvement.