Threshold Model Research Articles

Air pollution levels enhance green bond investment: green preference and environmental perception thresholds

Environmental change is a crucial factor influencing green investment, and air pollution affects green assets through green preference. This study analyzes the impact of air pollution levels on the green bond using the threshold model. Since investors have physiological thresholds for perceiving air pollution, green preferences change investor behavior once air pollution reaches a certain threshold. The empirical results confirm that at low air pollution levels, weak physical stimuli are not sufficient to induce strong environmental perceptions, resulting in an insignificant role for green preference. However, when air pollution exceeds the threshold, the effect of air pollution becomes significantly positive due to enhanced environmental perceptions. Additionally, major crisis events (e.g. COVID-19) may weaken the role of green preference by distracting investors, leading to an insignificant impact of air pollution on green bonds after the crisis. These findings help enterprises to achieve green transformation and sustainable development through green debt financing.

Digital inclusive finance, green technological innovation, and carbon emissions from a spatial perspective

Based on the panel data of 276 prefecture-level cities in China from 2011 to 2020, this study explores the impact of digital inclusive finance (DIF) on carbon emissions and the intrinsic mechanism of green technological innovation from a spatial perspective by constructing a spatial econometric model, a mediating effect model, and a threshold model. The results show that DIF significantly inhibits carbon emissions, exhibiting a spatial spillover effect. The transmission mechanism from a spatial perspective shows that green technological innovation plays a partial mediating role between DIF and carbon emissions, with the mediating effect accounting for approximately 59.47%. The heterogeneity analysis suggests that the impact of DIF on the reduction of carbon emissions is more pronounced in large and medium-sized cities and eastern regions. Further discussion reveals that the carbon reduction effect of DIF is also influenced by green technological innovation and industrial structure upgrading, showing threshold effects with marginal decreases and gradual increases, respectively.

Is renewable energy development endangering power supply reliability?

The development of renewable energy is indispensable to promoting the low-carbon transition of power systems. Nevertheless, it also brings uncertainty to the reliability of power systems. Herein, the panel model and panel threshold model are established based on the provincial data in China from 2012 to 2020. The results reveal that the negative effect of renewable energy development (RED) on power supply reliability (PSR) is gradually lessening. If the development of renewable energy is a rational way, power supply reliability can be improved. Additionally, energy-exporting regions bear more risks of RED than energy-importing regions. If the coal prices are stable and natural disasters are manageable, the RED can enhance the PSR. However, if they are not stable or controllable, a high proportion of renewable energy in the power system could cause even more severe problems with PSR. Based on these critical results, some suggestions are made to promote the formation of a new power system.

How Does the Digital Transformation Affect the Carbon Emissions of Manufacturing Enterprises in China? The Perspective of Green Technology Innovation

With the continuous intensification of competition among major countries and the resurgence of anti-globalization trends, countries around the world have strengthened their strategic deployment for digital transformation in the manufacturing industry. Digital development enables enterprises to transform and upgrade, and the digital transformation and green transformation of enterprises have a historic intersection. This study uses the panel data of 1900 A-share listed companies to explain the theoretical mechanism of digital transformation to promote carbon emission reduction from the perspective of green technology progress. In addition, the threshold model is used to analyze the nonlinear relationship. The results show that the digital transformation has significantly reduced the carbon emission level of Chinese manufacturing enterprises, and the conclusion is still true after the instrumental variable estimation and robustness test. Heterogeneity analysis found that the carbon emission reduction in state-owned enterprises is more obvious. Green technology innovation capability shows the technology dividend effect, which is an important way for digital transformation to promote carbon emission reduction. This study expands the research on the sustainable development of China’s manufacturing industry and provides guidance for policy makers and business decision makers.

Application of General Unified Threshold Models to Predict Time-Varying Survival of Mayfly Nymphs Exposed to Three Neonicotinoids

Pesticide exposure patterns tested in laboratory bioassays often do not match real-world pesticide exposure profiles in edge-of-field waterbodies. Toxicokinetic–toxicodynamic (TKTD) models are therefore increasingly used, as they allow for predictions of the toxic effects under actual time-variable field exposures. The TKTD models from the General Unified Threshold models of Survival (GUTS), for example, are considered ready for use by regulators for calculating the survival rates for any time-variable exposure profile. However, questions remain regarding their predictive power for compounds showing increased toxicity over time, such as neonicotinoid insecticides. The aim of the present study was therefore to compare the GUTS-predicted 28 d toxicity values of three neonicotinoids (imidacloprid, clothianidin, and thiamethoxam) for the common New Zealand mayfly genus Deleatidium spp. with those observed in a previously published study. Overall, the GUTS modeling results underestimated the toxicity values derived experimentally. From the three neonicotinoids, clothianidin showed the best fit between the estimated and observed 28 d LC50 (median-lethal-concentration) values. Shortcomings of the modeling exercise, future research needs, and implications for the application of GUTS models in regulatory risk assessment are discussed.

An Investigation of a Biomimetic Optical System and an Evaluation Model for the Qualitative Analysis of Laser Interference Visual Levels.

To objectively quantify the level of visual interference induced by lasers, we developed a biomimetic optical system designed to emulate human vision. This system is based on an optical model of the eye and synthetic imaging principles, allowing it to generate biomimetic optical images that closely mimic human visual perception. Upon exposure to a 532 nm laser, biomimetic optical images were captured under various ambient lighting conditions. By employing a contrast threshold model for human visual target detection and grayscale hierarchy analysis, we devised an evaluation model to quantify the levels of laser-induced visual interference. The bionic images obtained from our experiments, in conjunction with the constructed model, enabled us to assess the degree of laser-induced visual interference. Our results indicate that this system can effectively substitute the human eye when testing laser imaging effects, with the generated bionic images achieving up to 90% concordance with human vision. The proposed evaluation model facilitates the quantitative analysis of laser-induced visual impairment. This apparatus and evaluation model hold significant promise for the precise quantification of laser-induced visual interference levels.

The influence effect of regional carbon emission reduction under the perspective of fiscal decentralization: Government intervention or market mechanism?

AbstractBased on the perspective of fiscal decentralization, the study focuses on 30 provinces in China and employs various econometric models including the threshold model, spatial econometric model, mediation model, and regulation model. The research findings indicate that fiscal decentralization has a double‐threshold effect on government intervention, market mechanisms, and regional carbon emission reduction. Both government intervention and market mechanisms have inhibiting effects on carbon emission, with significant coefficients of GOVI and MARM at the 1% level. The cooperation between government intervention and the market mechanism effectively limits carbon emissions. Government intervention facilitates regional carbon emission reduction through the construction of new infrastructures and energy structure transformation, yielding a significant intermediary effect. The market mechanism is positively regulated through green finance and technology innovation to promote regional carbon emission reduction. Moreover, government intervention enables the market to achieve carbon emission reduction more effectively, especially in areas with a higher degree of government intervention. Continuous improvement and upgrading of regional and national carbon markets are essential to attain the carbon peak and carbon neutrality goals. Furthermore, attracting more participants to these markets for emission control subjects is necessary to enhance the effectiveness of government–market coordination.

Does emerging country's urbanization redefine the environmental Kuznets hypothesis for carbon emissions? A novel perspective on national and subnational differences

Emerging countries are experiencing rapid urbanization. Examining the link between economic growth and carbon emissions in developing nations as well as the effects of urbanization on this relationship at the national and subnational levels are the goals of this research. Thus, to investigate if China's economic development and carbon emissions align with the environmental Kuznets curve (EKC) hypothesis, this study adopts the FMOLS model and the fixed effects model. Furthermore, this study uses urbanization as a threshold variable and examines how the influence of economic development on carbon emissions would alter as a result of urbanization. The findings demonstrate an inverted U-shaped link between China's economic development and carbon emissions and the relationship is compatible with the EKC hypothesis. Although there is variability at the subnational level, most of the regions follow the inverted U-shaped connection. The carbon emissions inflection point in the western area is larger than the inflection points in the eastern and central regions. According to the threshold model, as urbanization rises, the increase in carbon emissions brought on by economic expansion declines. The eight economic areas of China differ in how the link between economic growth and carbon emissions change in the process of urbanization. But compared to the initial stage of urbanization, the last stage of urbanization in most countries lessens the impact of economic expansion on carbon emissions. To decrease carbon emissions, we must not only encourage the development of cleaner technologies and boost energy consumption efficiency, but also create carbon-reduction strategies that are specific to the growth characteristics of various areas.

Risk Factor Analysis and Genetic Parameter Estimation for Pre-Weaning Mortality Traits in Boer, Spanish, and Crossbred Goat Kids.

The objectives of this study were to evaluate fixed risk factors associated with PWM and to estimate genetic parameters for PWM. A total of 927 birth records from a mixed population of purebred and crossbred Boer and Spanish goats born between 2016 and 2023 at the International Goat Research Center (IGRC) were used for this study. Four binary traits were studied: D0-3 (death within 3 days after birth), D4-60 (death between 4 and 60 days), D61-90 (death between 61 and 90 days), and D0-90 (death within 90 days). Logistic regression models were used to evaluate the risk factors associated with PWM traits. Bayesian threshold models and Gibbs sampling were used to estimate the genetic parameters. Birth weight, season, litter size, sex, dam age, breed, and heterosis were found to be significantly associated with at least one of the PWM traits. Heritability estimates were 0.263, 0.124, 0.080, and 0.207, for D0-3, D4-60, D61-90, and D0-90, respectively. The genetic correlations between the studied traits ranged from 0.892 (D0-3 and D0-90) to 0.999 (D0-3 and D61-90). These results suggest that PWM in goats is influenced by both non-genetic and genetic factors and can be reduced by management, genetic selection, and crossbreeding approaches.

How can restricted farmland development rights be compensated to stimulate farmland protection? A provincial exploration in China

Stimulating farmland protection (FP) is crucial to ensure food security and environmental well-being. Some countries, including China, are seeking to incorporate transferable development rights (TDR) to realize economic incentives for FP. This study developed an indicator – planning regulation intensity on farmland (PRIF) – to describe the restricted farmland development rights (FDRs) and designed a compensation-oriented farmland transferable development rights (FTDR) mechanism. The threshold model and two-way fixed effect model were employed to unlock the transfer direction, quantity, and economic value of FDRs. Taking Hubei Province in China as the study area; the empirical research was conducted and demonstrated the necessity and feasibility of implementing the compensation-oriented FTDR mechanism in China. This study addresses the issue of unfair distribution of FDRs due to planning regulations and provides a promising economic-incentive policy instrument for curbing farmland loss. The FTDR mechanism considering PRIF that we propose enables the extension of TDR to countries that implement authority-exerted planning regulations.

Catastrophic cascade of failures in interdependent hypergraphs.

The failures of individual agents can significantly impact the functionality of associated groups in interconnected systems. To reveal these impacts, we develop a threshold model to investigate cascading failures in double-layer hypergraphs with interlayer interdependence. We hypothesize that a hyperedge disintegrates when the proportion of failed nodes within it exceeds a threshold. Due to the interdependence between a node and its replica in the other layer, the disintegrations of these hyperedges could trigger a cascade of events, leading to an iterative collapse across these two layers. We find that double-layer hypergraphs undergo abrupt, discontinuous first-order phase transitions during systemic collapse regardless of the specific threshold value. Additionally, the connectivity measured by average cardinality and hyperdegree plays a crucial role in shaping system robustness. A higher average hyperdegree always strengthens system robustness. However, the relationship between system robustness and average cardinality exhibits non-monotonic behaviors. Specifically, both excessively small and large average cardinalities undermine system robustness. Furthermore, a higher threshold value can boost the system's robustness. In summary, our study provides valuable insights into cascading failure dynamics in double-layer hypergraphs and has practical implications for enhancing the robustness of complex interdependent systems across domains.

Describing the impact of psychological safety on risk prevention: A threshold model construction

BACKGROUND: Like other things, psychological safety has a too-much-of-a-good-thing effect on the risk prevention process. There are warranted systematic and theoretical discussions on both positive and negative influences of psychological safety, which provide useful guides to handling potential risks for workers. OBJECTIVE: This study focuses on the psychological safety analysis of work safety for workers in high-risk industries through the construction of a theoretical model, illustrating that a higher level of psychological safety is not better. METHOD: This study explores the impact of workers’ psychological safety on the potential risk of work safety by conducting a comprehensive analysis of relevant literature and employing inductive reasoning and analogy methodology. RESULTS: We systematically reviewed the research background and several hypotheses of psychological safety, described the relationship between psychological safety and risk prevention from three levels, and constructed the threshold initial, special, and extended model of psychological safety to reveal the importance of workers’ psychological safety when handling potential risks. CONCLUSION: This research effectively describes the influence of high, moderate, and low levels of psychological safety on the risk prevention of work safety, which provides a psychological model guide to developing measures for the improvement of work safety management.

Health and economic growth: new evidence from a panel threshold model

Health and economic growth: new evidence from a panel threshold model

Development of Digital Financial Inclusion in China's Regional Economy: Evidence from Panel Threshold Models

This study aims to investigate the effect of digital financial inclusion and air pollution on economic growth for 31 Chinese provinces between 2003 and 2022 using Panel Threshold Auto-Regressive (PTAR) and Panel Smooth Transition Auto-Regression (PSTAR) models. The results show that there is a nonlinear link between digital financial inclusion and economic growth in China. For PTAR, the LnDFII thresholds are 4.264 (i.e., DFII = 71.094), and for PSTAR are 4.563 (i.e., DFII = 95.871). Below these thresholds, digital financial inclusion significantly boosts economic growth by 0.061 and 0.063 in the PTAR and PSTAR models, respectively. However, above these thresholds, the positive impact diminishes, with coefficients dropping to 0.015 and 0.004 in the PTAR and PSTAR models, respectively. Additionally, both models indicate that digital financial inclusion positively affects reducing air pollution, thereby potentially fostering economic growth. Hence, authorities should strategically implement digital technologies and strengthen collaborative efforts at the regional level to maximize these benefits.  

Impact of Green Infrastructure Investment on Urban Carbon Emissions in China

Given the increasingly severe global climate change, the reduction in urban greenhouse gas emissions has become the common goal of all nations. As a widely concerned sustainable development strategy, green infrastructure investment (GII) aims to reduce urban carbon emissions, improve the efficiency of resource utilization, and improve environmental quality. However, the construction cycle of green infrastructure is long, and the construction process itself may produce carbon emissions; so, the final effect of GII on urban carbon emissions is unclear, which deserves our in-depth study. Further, is this effect having a time-lag effect? Is there only a simple linear relationship between GII and urban carbon emissions? Based on panel data from 235 Chinese cities from 2006 to 2019, this study conducted an econometric regression analysis using time-lag-effect and threshold-effect models. The results showed the following: (1) GII had a negative inhibitory effect on urban CO2 emissions. Adding one unit to the GII could reduce urban CO2 emissions by 0.032 units. (2) GII exhibited a time-lag effect on urban CO2 emissions, and the greatest reduction in CO2 emissions occurred in the third lag period. (3) GII had a threshold effect on urban CO2 emissions based on technological progress (TP). This paper used the static and dynamic panel threshold models to research separately, and obtained the corresponding regression results. In the static panel, the double threshold values for TP were 3.9120 and 6.8035. At different TP levels, GII had an inhibitory effect on CO2 emissions, but the coefficients were different. However, in the dynamic panel, the threshold value was 3.666. The threshold changed over time and the effect of GII on CO2 emissions shifted from facilitation to inhibition.

Abstract 3565: The Cox proportional hazards mixed model enables accurate estimation of the total effect somatic SNVs have on treatment outcomes

Abstract Heritability is the variation in disease susceptibility attributed to genetic variation and is a fundamental parameter for interpreting disease architecture and making predictions. In the context of germline genetics, variance component models are commonly used to estimate heritability for continuous or case-control phenotypes (under the liability threshold model, LTM). However, models for time-to-event (TTE) outcomes have been largely unstudied. This has limited the characterization of age-of-onset and treatment response phenotypes, the latter particularly relevant for clinical studies. Here, we propose and evaluate a Cox proportional hazard Mixed Model (COXMM) to enable accurate estimation for TTE phenotypes with censoring. We apply these methods to Real World data from the Profile cohort from Dana-Farber Cancer Institute, a large prospectively collected tumor sequencing cohort with detailed clinical annotation. We implemented an efficient COXMM with a random effect modeled by the genetic relatedness across individuals, and benchmarked its performance in extensive simulations. We simulated phenotypes according to generative models reflecting either LTM or TTE models. For TTE phenotypes, data followed various Weibull distributions with independent censoring. Across TTE simulations, COXMM produced unbiased estimates while the classic case-control heritability estimator showed significant downward bias. Likewise, restricting only to cases and estimating the heritability of age-of-onset as a normalized continuous phenotype produced biased estimates, due to the artificial exclusion of controls. For example, for an age-of-onset with heritability of 0.8 and 40% of individuals in the study being cases, COXMM correctly inferred the heritability to be 0.80 +/- 0.007, compared to 0.32 +/- 0.004 for a conventional case-control model, and 0.09 +/- 0.02 for a conventional case-only age-of-onset model. In the Profile cohort, we analyzed the impact of somatic SNVs on progression free survival (PFS) and overall survival (OS) in patients with non-small cell lung cancer (N=954 and N=1796, respectively). We restricted to somatic SNVs present in at least 5% of patients. For OS, we observed that the COXMM heritability estimate was significantly larger than those computed by case-control status and using a normalized case-only age/duration (COXMM: 0.19 +/- 0.003; Case-Control: 0.07 +/- 0.2; Duration: 0.04 +/- 0.01). We observed a similar phenomenon for PFS as well (COXMM: 0.11 +/- 0.01; Case-Control: 0.03 +/- 0.4; Duration: 0.03 +/- 0.02). These results demonstrate that somatic features are most strongly associated with TTE and can be accurately estimated with COXMM. COXMM enables the efficient quantification of TTE heritability for diverse clinical phenotypes and outperforms existing methods. Citation Format: Kodi Taraszka, Alexander Gusev. The Cox proportional hazards mixed model enables accurate estimation of the total effect somatic SNVs have on treatment outcomes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3565.

Abstract 7029: Trajectory of DNA methylation in blood up to 18 years prior to lung cancer diagnosis

Abstract Purpose: Lung cancer (LC) has the highest mortality rate in South Korea. As prediction and early detection of lung cancer is pivotal to enhance survival rates, we attempted to investigate the trajectory of DNA methylation (DNAm) several years prior to lung cancer diagnosis by using prediagnostic blood samples. Methods: From National Cancer Center Health Examinee Cohort, established in 2003, we selected study participants who became diagnosed with LC several years after health examination, thereby collecting prediagnostic samples from 18 years to 1 year prior to LC diagnosis. Normal controls (n=150) were matched with patients (n=150) by sex, age, and the time of sample collection. Epigenome-wide DNAm data were obtained by using Infinium Human MethylationEPIC v2.0 Beadchip. We assumed non-linear relationship between DNAm changes and the time-to-diagnosis and estimated an optimal threshold point of non-linear relationship with a piecewise linear regression analysis at each CpG sites (CpGs). We used the ratio of M-values of patients and matched controls to establish the non-linear model composed of two linear models with a threshold point where the slope changes. Then, we performed functional enrichment analysis by each year prior to LC diagnosis using ReactomePA. Results: We identified threshold models with a threshold point where the slope changes from zero (from 18 years prior to LC up to the threshold point) to positive or negative (from threshold point to the point of LC diagnosis), which we termed “flat-to-change model”. Out of 790,048 CpGs, we identified a total of 7214 CpGs showing the “flat-to-change model”, of which 944, 789, 480, 413, 687, 542, 516, 1140, 315, 284, 327, 432, 83, 95, 74, 32, and 23 CpGs were found prior to 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, and 17 years before LC diagnosis, respectively. Functional enrichment analysis showed that various pathways were associated at different time point berfore LC diagnosis. CpGs of which DNAm was changed two years before LC diagnosis were associated with telomerase extension by telomerase involving ANKRD28/TERT/TERF2/PPP6R3/WRAP53. Transcriptional regulation by TP53 was associated with DNAm changes before both 5 and 14 years of LC diagnosis via POLR2A. Conclusions: We identified the trajectories of DNA methylation up to 18 years prior to LC diagnosis, and found that a greater number of CpG sites showed DNAm changes adjacent to the time point of LC diagnosis. CpG sites associated with specific time point before LC diagnosis may elucidate the mechanisms of tumorigenesis and also have potential for LC prediction markers. Citation Format: Yoon-Jung Choi, Ki-Jae Hong, Yeon-Su Lee, Woo Jin Kim, Yeol Kim, Sun Ha Jee, Jeongseon Kim. Trajectory of DNA methylation in blood up to 18 years prior to lung cancer diagnosis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7029.

Power transformer fault diagnosis using dynamic multiscale graph modeling and M2SGCN network based on statistical fusion

Power equipment fault diagnostics hold significant importance for the stability of power grid systems. In pursuit of this objective, this paper proposes a fault diagnosis method that utilizes dynamic multiscale graph (DMG) modeling and the multiscale multi-stream GCN(M2SGCN) network, incorporating statistical fusion. Specifically, a novel DMG modeling method is proposed to derive visibility graph data and horizontal visibility graph data from vibration signals across multiple scales. Next, a comprehensive neural network architecture named M2SGCN is established to learn global and local features simultaneously, providing a more precise representation. Subsequently, a Dempster Shafer evidence theory statistical fusion technique combined with an adaptive threshold model (DSTFusion) is utilized to integrate primary decision results for enhanced fault diagnosis accuracy. In addition, two datasets obtained from single-phase and three-phase power transformers are analyzed to demonstrate the evolution process. When compared to state-of-the-art indicators such as accuracy, precision, recall, and F1 scores, the method proposed excels in multiple aspects, successfully detecting fault states before their occurrence and achieving outstanding performance.

Development of Machine Learning Algorithm to Predict the Risk of Incontinence After Robot-Assisted Radical Prostatectomy.

Introduction: Predicting postoperative incontinence beforehand is crucial for intensified and personalized rehabilitation after robot-assisted radical prostatectomy. Although nomograms exist, their retrospective limitations highlight artificial intelligence (AI)'s potential. This study seeks to develop a machine learning algorithm using robot-assisted radical prostatectomy (RARP) data to predict postoperative incontinence, advancing personalized care. Materials and Methods: In this propsective observational study, patients with localized prostate cancer undergoing RARP between April 2022 and January 2023 were assessed. Preoperative variables included age, body mass index, prostate-specific antigen (PSA) levels, digital rectal examination (DRE) results, Gleason score, International Society of Urological Pathology grade, and continence and potency questionnaires responses. Intraoperative factors, postoperative outcomes, and pathological variables were recorded. Urinary continence was evaluated using the Expanded Prostate cancer Index Composite questionnaire, and machine learning models (XGBoost, Random Forest, Logistic Regression) were explored to predict incontinence risk. The chosen model's SHAP values elucidated variables impacting predictions. Results: A dataset of 227 patients undergoing RARP was considered for the study. Post-RARP complications were predominantly low grade, and urinary continence rates were 74.2%, 80.7%, and 91.4% at 7, 13, and 90 days after catheter removal, respectively. Employing machine learning, XGBoost proved the most effective in predicting postoperative incontinence risk. Significant variables identified by the algorithm included nerve-sparing approach, age, DRE, and total PSA. The model's threshold of 0.67 categorized patients into high or low risk, offering personalized predictions about the risk of incontinence after surgery. Conclusions: Predicting postoperative incontinence is crucial for tailoring rehabilitation after RARP. Machine learning algorithm, particularly XGBoost, can effectively identify those variables more heavily, impacting the outcome of postoperative continence, allowing to build an AI-driven model addressing the current challenges in post-RARP rehabilitation.

Impacts and Risk Assessments of Climate Change for the Yields of the Major Grain Crops in China, Japan, and Korea.

Climate change poses a high risk to grain yields. Maize, rice, and wheat are the three major grain crops in China, Japan, and Korea. Assessing the impacts and risks of climate on the yields of these grain cops is crucial. An economy-climate model (C-D-C model) was established to assess the impacts of climate factors on the grain yields in different crop areas. The peaks over threshold model based on the generalized Pareto distribution was used to calculate the value at risk and the expected shortfall, which can evaluate the yield risk of different crops. The impact ratio of climate change was employed to estimate the impacts of climate change under different climate scenarios. The main conclusions can be summarized as follows: the impacts of climate factors on grain yields and the risk vary widely across the different regions and crops. Compared to 1991-2020, climate change from 2021 to 2050 exerts positive impacts on rice and wheat, while the negative impacts on maize in the crop areas are significantly affected by climate factors. The impact ratios of climate change are larger in the SSP1-2.6 and the SSP5-8.5 scenarios than under the SSP2-4.5 scenario. These findings are useful for targeting grain yields in smaller study areas.