Based on the panel data of daily meteorological stations and winter wheat yield in Henan Province from 2000 to 2023, this study comprehensively used the Mann–Kendall trend test, wavelet coherence analysis (WTC), and other methods to reveal the temporal and spatial evolution of extreme precipitation and its multi-scale stress mechanism on grain yield. The results showed the following: (1) Extreme precipitation showed the characteristics of ‘frequent fluctuation-gentle trend-strong spatial heterogeneity’, and the maximum daily precipitation in spring (RX1DAY) showed a significant uplift. The increase in rainstorm events (R95p/R99p) in the southern region during the summer is particularly prominent; at the same time, the number of consecutive drought days (CDDs > 15 d) in the middle of autumn was significantly prolonged. It was also found that 2010 is a significant mutation node. Since then, the synergistic effect of ‘increasing drought days–increasing rainstorm frequency’ has begun to appear, and the short-period coherence of super-strong precipitation (R99p) has risen to more than 0.8. (2) The spatial pattern of winter wheat in Henan is characterized by the three-level differentiation of ‘stable core area, sensitive transition zone and shrinking suburban area’, and the stability of winter wheat has improved but there are still local risks. (3) There is a multi-scale stress mechanism of extreme precipitation on winter wheat yield. The long-period (4–8 years) drought and flood events drive the system risk through a 1–2-year lag effect (short-period (0.5–2 years) medium rainstorm intensity directly impacted the production system). This study proposes a ‘sub-scale governance’ strategy, using a 1–2-year lag window to establish a rainstorm warning mechanism, and optimizing drainage facilities for high-risk areas of floods in the south to improve the climate resilience of the agricultural system against the background of climate change.

Global climate change is exerting a growing impact on agricultural ecosystems. Accurately assessing the spatiotemporal dynamics of agricultural ecosystem respiration and its response mechanisms to climate has therefore emerged as a critical issue in agricultural carbon cycle research and climate change response. It should be noted that the ‘agro-ecosystem’ referred to in this study covers two major types: one is the farmland agro-ecosystem dominated by crop planting (such as farmland, orchard and other artificial management systems), and the other is the grassland agro-ecosystem dominated by herbaceous plants and managed by humans (such as grazing grassland and mowing grassland). Remote sensing technology provides a new way to break through the limitations of traditional ground observation by virtue of its advantages of large-scale and continuous monitoring. Based on the CiteSpace bibliometric method, this study focused on the key time window of 2021–2025, systematically searched the core collection of Web of Science, and finally included 222 related literature. This period marks the initial stage of the rise and rapid development of this interdisciplinary field, enabling us to capture the formation of its knowledge structure and the evolution of its research paradigm from the source. Through the quantitative analysis of this literature, it aims to reveal the research hotspots, development paths and frontier trends in this field. The results show that China occupies a dominant position in this field (135 articles). The evolution of research shows a three-stage development characterized by “technology-driven-method fusion-system coupling,” which is divided into the initial development period (2021–2022), the rapid growth period (2023–2024) and the deepening development period (2025) (because 2025 has not yet ended, this stage is a preliminary discussion). Keyword clustering analysis identified 13 important research directions, including machine learning (# 0 clustering), permafrost (# 1 clustering) and carbon flux (# 2 clustering). It is found that the deep integration of artificial intelligence and remote sensing data is promoting the transformation of research methods from traditional inversion to intelligent modeling. At the same time, the attention to alpine grassland and other ecosystems also reflects the trend that the research frontier extends to the interaction zone between the agricultural ecosystem and the natural environment. Future research should prioritize three key directions: building multi-scale monitoring networks, developing “grey box” models that integrate mechanisms and data fusion, and evaluating the carbon emission reduction efficiency of agricultural management practices. These efforts will provide a theoretical basis for carbon management and climate adaptation in agricultural ecosystems, as well as scientific and technological support for achieving global agricultural sustainable development goals (specifically, SDG13 on climate action and SDG15 on terrestrial ecosystem conservation).

This study used bibliometric methods to systematically analyze the development trend, knowledge structure and evolution path of the field of “quantitative research on agricultural soil respiration based on machine learning” from 2021 to 2025, and further explored its implications for agricultural soil carbon sinks. Based on 966 articles included in the core collection of Web of Science, this paper comprehensively uses tools such as Biblioshiny, CiteSpace and VOSviewer to carry out multi-dimensional analysis from the aspects of annual publication trends, international and institutional cooperation networks, keyword clustering and emergent evolution. It is found that this field has shown phased evolution characteristics of “technology-driven mechanism deepening–application expansion” in the past five years. At the beginning of the 5-year period of research, the introduction of machine learning methods and model verification were the core, then gradually expanding to multi-algorithm comparison, environmental factor coupling mechanisms and multi-source data fusion. Recently, the field has focused on regional-scale simulation, uncertainty quantification and model interpretability research. Keyword clustering identifies three thematic clusters—machine learning algorithm and model optimization, environmental driving factors and process mechanism, and remote sensing fusion and regional application—which form a knowledge system of “method–mechanism–application” collaborative evolution. The national cooperation network presents a pattern of “Asia-led, China–US dual-core, and European connectivity”. China dominates in scientific research output, and the United States plays a key role in international cooperation. This study further points out that the development of this field provides important methodological support and a scientific basis for accurate assessment, intelligent management and carbon neutralization decision-making for agricultural soil carbon sinks. Based on the above findings, future research should focus on the development of intelligent models of mechanisms and data fusion, the construction of multi-source data assimilation and uncertainty assessment frameworks, the expansion of global diversified agricultural system cases, and the promotion of an open and shared international scientific research cooperation ecology. This study provides empirical evidence and a direction reference for academic development, scientific research layout, carbon sink management and international collaboration in this field.

Hunan Province from 2000 to 2023 is the study area. Based on NOAA precipitation data and county-level rice yield statistics in Hunan Province, the Mann–Kendall test, extreme precipitation indices, and wavelet analysis examine the spatial and temporal evolution characteristics of extreme precipitation and its multi-scale impact on rice yield. The results show that the extreme precipitation in Hunan Province showed a stable pattern of fluctuation, and the main extreme precipitation indexes had no significant change trend. The spatial distribution showed a pattern of “high value in central-northern Hunan and stable in southern Hunan”, and the precipitation was concentrated in June–August. The rice yield showed the characteristics of “stable increase in the core area, intensified fluctuation in the transition area, and continuous shrinkage in the marginal area”, and the Dongting Lake Plain was a high-yield and stable area. Multi-scale analysis shows significant coupling between extreme precipitation and yield: in the 4–8-year cycle, the peak value of precipitation lags behind the response of 1–2 years, and changes synchronously in a short period. The response of rice to extreme precipitation showed a threshold-type nonlinear characteristic. Moderate wetting was beneficial to stable yield, while the yield decreased significantly when the intensity or continuous precipitation exceeded the threshold. Hunan’s rice system has strong climate resilience but requires a multi-scale climate-adaptive agricultural system via engineering, technology, and policy for long-term stability and sustainable grain production.

Understanding how the spatiotemporal variability of precipitation affects ecosystem respiration (RE) is central to carbon–climate feedback in climate-smart agriculture, yet remains unresolved for the alpine agricultural region of the southern Qinghai–Tibet Plateau, where flux observations are sparse. Using 25 years (2000–2024) of monthly gridded climate and remote sensing data for the Yarlung Zangbo River Basin and Its Two Tributaries Basin, we developed a flux tower-constrained reference–respiration (R_ref) environment-matching model in which R_ref varies with the enhanced vegetation index (EVI) and land surface temperature (LST) to correct the Lloyd–Taylor parameterization. The correction reduced the RE root mean square error by 54.8% (1.04 → 0.47 gC·m⁻²·month⁻¹) and eliminated systematic bias (+0.80 → −0.001) relative to an independent gridded RECO product. We then constructed a multidimensional index of precipitation variability (intra-annual concentration, interannual variability, long-term trend, spatial clustering) and combined random forest, spatial regression, lag analysis, and structural equation modeling (SEM) to disentangle direct and indirect pathways from precipitation variability to RE. The central finding is an indirect-conduction mechanism: precipitation concentration (PCI) affects RE almost entirely through vegetation productivity (PCI → GPP → RE, indirect effect −0.676) rather than directly (direct effect +0.076, opposite in sign), because low temperature and high soil water holding capacity buffer the immediate soil moisture response. The basin functions as a net carbon source (mean NEP = −0.550 gC·m⁻²·month⁻¹) with a significant warming-driven interannual RE increase (Sen’s slope = 0.0025 yr⁻¹, p = 0.022) that is independent of the stable precipitation total. The framework offers a transferable paradigm for carbon flux attribution in alpine regions under sparse observation.

This study combined bibliometric analysis and qualitative analysis to examine 652 publications retrieved from the Web of Science Core Collection between 1 January 1989 and 25 May 2026. Since the publications from 2026 cover only the period from 1 January to 25 May, these records were excluded from analyses of annual publication volume and citation trends. Analysis using CiteSpace 7.0 and VOSviewer 1.6.20 reveals a shift in focus from regional phenomenon studies to comprehensive exploration of how irrigation water quality regulates soil carbon cycling mechanisms. Salinity, ion composition, and dissolved organic matter influence carbon flux by altering soil physicochemical properties, microbial activity, and the balance between autotrophic and heterotrophic respiration. Overall, salinity primarily suppresses soil respiration through osmotic stress and ion toxicity, while the reclaimed water effect exhibits significant context dependence. Current controversies center on the initial effects of dissolved organic carbon, salinity thresholds, and their coupling mechanisms with temperature sensitivity (Q₁₀). Future research should enhance the integrated application of isotopic tracing, microbiomics, and process modeling to improve understanding and prediction of how irrigation water quality affects soil carbon cycling mechanisms.