# A Method for Predicting Alfalfa Biomass Based on Multimodal Data and Ensemble Learning Model

## TL;DR

This paper introduces a new method using drone and LiDAR data with machine learning to accurately predict alfalfa biomass for better pasture management.

## Contribution

A novel multimodal data fusion and ensemble learning approach for high-precision alfalfa biomass prediction.

## Key findings

- The ensemble model achieved an R2 of 0.813 on the test set, outperforming single models.
- Fusion of multispectral and LiDAR data improved prediction accuracy compared to using either data source alone.
- The model showed highest accuracy during the bud emergence to early flowering stage (R2 = 0.917).

## Abstract

Accurate alfalfa biomass prediction is crucial for pasture management and sustainable livestock production. However, traditional methods often perform poorly under complex field conditions. To address the limited prediction accuracy of traditional methods under complex planting environments, this study proposes an alfalfa biomass prediction method combining multispectral and LiDAR data with ensemble learning model. Based on the multispectral images acquired by unmanned aerial vehicle (UAV) and airborne LiDAR data, the spectral features, three-dimensional structural features, and their interaction features are systematically extracted at the quadrat scale, and a high-quality modeling dataset is constructed by feature selection. Secondly, an ensemble model for alfalfa biomass prediction was constructed, which was composed of random forest, extra trees, and histogram gradient boosting. After model training, the coefficient of determination (R2) of the integrated model on the test set reached 0.813, and the root mean square error (RMSE) and mean absolute error (MAE) were 0.178 kg m−2 and 0.146 kg m−2, which were significantly better than those of similar single models. Under feature combinations, the fusion model was better than that of spectral indices only (R2 = 0.773) and LiDAR traits only (R2 = 0.576), and the model achieved the highest accuracy from bud emergence to early flowering (R2 = 0.917). The overall prediction error of the model was approximately normal distribution, and the absolute error of more than 65% of the samples was less than 0.2. However, there was still a trend of underestimation in the high biomass interval. This research showed that the multimodal data fusion and ensemble learning method could achieve high-precision prediction of alfalfa biomass, which provided reliable technical support for pasture resources monitoring and precision agriculture management.