No abstract

Above-ground biomass (AGB) is a key indicator for studying grassland productivity and evaluating carbon sequestration capacity; it is also a key area of interest in hyperspectral ecological remote sensing. In this study, we use data from a typical alpine meadow in the Qinghai–Tibet Plateau during the main growing season (July–September), compare the results of various feature selection algorithms to extract an optimal subset of spectral variables, and use machine learning methods and data mining techniques to build an AGB prediction model and realize the optimal inversion of above-ground grassland biomass. The results show that the Lasso and RFE_SVM band filtering machine learning models can effectively select the global optimal feature and improve the prediction effect of the model. The analysis also compares the support vector machine (SVM), least squares regression boosting (LSB), and Gaussian process regression (GPR) AGB inversion models; our findings show that the results of the three models are similar, with the GPR machine learning model achieving the best outcomes. In addition, through the analysis of different data combinations, it is found that the accuracy of AGB inversion can be significantly improved by combining the spectral characteristics with the growing season. Finally, by constructing a machine learning interpretable model to analyze the specific role of features, it was found that the same band plays different roles in different records, and the related results can provide a scientific basis for the research of grassland resource monitoring and estimation.

BACKGROUND: The roots are the main functional organs involved in the overwintering adaptability of alfalfa (Medicago sativa). However, it is still unclear how the roots are involved in the cold resistance in the high-altitude area of the Qinghai-Tibet Plateau (QTP). In this study, three winter-surviving 2-year-old alfalfa varieties (M. sativa Zhongmeng No.1, M. sativa Chiza No.1, and M. sativa Gongnong No.1) planted at two different altitudes (2812 m and 3109 m) in the northeast edge of the QTP were used to explore the cold-resistance mechanism. RESULTS: At low altitudes (2812 m), the overwintering rate, taproot length, root area, root surface area, and root average diameter, plant height, fresh yield and hay yield of M. sativa Zhongmeng No.1 were significantly higher (P < 0.01) than for the other two varieties. At high altitude (3109 m), lateral root length, number of lateral roots, main root dry weight, and lateral root dry weight of M. sativa Chiza No.1 were higher (P < 0.01) than the other two varieties. At low and high altitudes, the activities of peroxidase and catalase were higher (P < 0.05) in M. sativa Chiza No.1 during post-winter and pre-winter respectively. At low altitude, higher soluble sugar (P < 0.05) and proline (P < 0.01) contents were recorded during the pre-and post-winter periods. Membership function analysis showed that M. sativa Zhongmeng No.1 has the strongest cold resistance. The structural equation model showed that the overwintering rate of alfalfa was mainly affected by the morphological characteristics of roots and the physiological characteristics of roots, with contribution rates of 0.54 and 0.75 respectively, and the physiological characteristics of roots had the greatest effect on the overwintering rate.CONCLUSIONS: This study is of great significance to effectively solve the overwintering of alfalfa, the lack of high-quality legume forage resources, and promote the development of animal husbandry in the alpine areas of the QTP.