Rapid and accurate leaf area index (LAI) determination is important for monitoring daylily growth, yield estimation, and field management. Because of low estimation accuracy of empirical models based on single-source data, we proposed a machine-learning algorithm combining optical and microwave remote-sensing data as well as the random forest regression (RFR) importance score to select features. A high-precision LAI estimation model for daylilies was constructed by optimizing feature combinations. The RFR importance score screened the top five important features, including vegetation indices land surface water index (LSWI), generalized difference vegetation index (GDVI), normalized difference yellowness index (NDYI), and backscatter coefficients VV and VH. Vegetation index features characterized canopy moisture and the color of daylilies, and the backscatter coefficient reflected dielectric properties and geometric structure. The selected features were sensitive to daylily LAI. The RFR algorithm had good anti-noise performance and strong fitting ability; thus, its accuracy was better than the partial least squares regression and artificial neural network models. Synergistic optical and microwave data more comprehensively reflected the physical and chemical properties of daylilies, making the RFR-VI-BC05 model after feature selection better than the others ( r = 0.711, RMSE = 0.498, and NRMSE = 9.10%). This study expanded methods for estimating daylily LAI by combining optical and radar data, providing technical support for daylily management.

As global climate change intensifies and economic transformation progresses, the agro-pastoral ecotone of northern China faces dual challenges of stopping ecological degradation and enhancing farmers’ livelihoods. Yunzhou District in Shanxi Province represents a typical ecologically fragile area, where the daylily industry contributes significantly to improving livelihood resilience. This study categorized farmers into three types based on their dependence on daylily income: major-job farmers (50–90% income from daylily), sole agriculture farmers (≥90%), and side-job farmers (<50%). Using questionnaire survey data and the optimal parameter-based geographical detector method, we evaluated and compared the livelihood resilience levels of these farmer types and identified their key explanatory factors. The results showed that (1) major-job farmers exhibited the highest livelihood resilience index (0.165), followed by sole agriculture farmers (0.152), whereas side-job farmers exhibited the lowest (0.138); (2) significant differences in livelihood resilience existed across farmer types (p < 0.05); and (3) health status was a common key factor across all types, while factors such as traffic accessibility, policy awareness, social security, and information acquisition capability exhibited differential effects among groups. These findings provide empirical evidence to guide targeted livelihood interventions and sustainable transitions in the agro-pastoral ecotone.

Rapid and accurate leaf area index (LAI) determination is important to monitoring daylily growth, yield estimation, and field management. Because of low estimation accuracy of empirical models based on single-source data, we proposed a machine-learning algorithm combining optical and microwave remote-sensing data, and the random forest regression (RFR) importance score to select features. A high-precision LAI estimation model for daylilies was constructed by optimizing feature combinations. The RFR importance score screened the top five important features, including vegetation indices land surface water index (LSWI), generalized difference vegetation index (GDVI), normalized difference yellowness index (NDYI) and backscatter coefficients VV and VH. Vegetation index features characterized canopy moisture and color of daylilies, and the backscatter coefficient reflected dielectric properties and geometric structure. Selected features were sensitive to daylily LAI. The RFR algorithm had good anti-noise performance and strong fitting ability; thus, its accuracy was better than the partial least squares regression and artificial neural network models. Synergistic optical and microwave data more comprehensively reflected the physical and chemical properties of daylilies, making the RFR-VI-BC05 model after feature selection better than the others. This study expanded methods for estimating daylily LAI by combining optical and radar data, providing technical support for daylily management.