Accurate information about the available standing biomass on pastures is critical for the adequate management of grazing and its promotion to farmers. In this paper, machine learning models are developed to predict available biomass expressed as compressed sward height (CSH) from readily accessible meteorological, optical (Sentinel-2) and radar satellite data (Sentinel-1). This study assumed that combining heterogeneous data sources, data transformations and machine learning methods would improve the robustness and the accuracy of the developed models. A total of 72,795 records of CSH with a spatial positioning, collected in 2018 and 2019, were used and aggregated according to a pixel-like pattern. The resulting dataset was split into a training one with 11,625 pixellated records and an independent validation one with 4952 pixellated records. The models were trained with a 19-fold cross-validation. A wide range of performances was observed (with mean root mean square error (RMSE) of cross-validation ranging from 22.84 mm of CSH to infinite-like values), and the four best-performing models were a cubist, a glmnet, a neural network and a random forest. These models had an RMSE of independent validation lower than 20 mm of CSH at the pixel-level. To simulate the behavior of the model in a decision support system, performances at the paddock level were also studied. These were computed according to two scenarios: either the predictions were made at a sub-parcel level and then aggregated, or the data were aggregated at the parcel level and the predictions were made for these aggregated data. The results obtained in this study were more accurate than those found in the literature concerning pasture budgeting and grassland biomass evaluation. The training of the 124 models resulting from the described framework was part of the realization of a decision support system to help farmers in their daily decision making.

This research aims to develop a predictive model to discriminate milk produced from a cattle diet either based on grass or not using milk mid-infrared spectrometry and the month of testing (an indirect indicator of the feeding ration). The dataset contained 3,377,715 spectra collected between 2011 and 2021 from 2449 farms and 3 grazing traits defined following the month of testing. Records from 30% of the randomly selected farms were kept in the calibration set, and the remaining records were used to validate the models. Around 90% of the records were correctly discriminated. This accuracy is very good, as some records could be erroneously assigned. The probability of belonging to the GRASS modality allowed confirmation of the model’s ability to detect the transition period even if the model was not trained on this data. Indeed, the probability increased from the spring to the summer and then decreased. The discrimination was mainly explained by the changes in the milk fat, mineral, and protein compositions. A hierarchical clustering from the averaged probability per farm and year highlighted 12 groups illustrating different management practices. The probability of belonging to the GRASS class could be used in a tool counting the number of grazing days.

This article focuses on the creation of a monitoring tool using routinely collected data from milk payment analyses. Milk samples were analyzed through Fourier Transform mid-infrared spectrometry every 1 to 3 days, and their compositions were predicted using machine learning models. Among the predicted parameters, fatty acid profiles appear to be effective indicators of animal status and management practices. In this research, these profiles were summarized using 31 fatty acids or groups of fatty acids. The methodology consists of four steps: hierarchical clustering to detect patterns in a Belgian spectral dataset (N = 774,781), interpretation of the identified seven clusters, development of predictive models applied to a Canadian dataset (N = 670,165), and validation using management information collected from Canadian farms. The identified clusters revealed significant relationships with feeding management strategies and temporal evolutions, highlighting the potential to develop automated alert systems that assist farmers and advisors in herd monitoring.

The use of remote sensing data and the implementation of machine learning (ML) algorithms is growing in pasture management. In this study, ML models predicting the available compressed sward height (CSH) in Walloon pastures based on Sentinel-1, Sentinel-2, and meteorological data were developed to be integrated into a decision support system (DSS). Given the area covered (>4000 km² of pastures of 100 m² pixels), the consequent challenge of computation time and power requirements was overcome by the development of a platform predicting CSH throughout Wallonia. Four grazing seasons were covered in the current study (between April and October from 2018 to 2021, the mean predicted CSH per parcel per date ranged from 48.6 to 67.2 mm, and the coefficient of variation from 0 to 312%, suggesting a strong heterogeneity of variability of CSH between parcels. Further exploration included the number of predictions expected per grazing season and the search for temporal and spatial patterns and consistency. The second challenge tackled is the poor data availability for concurrent acquisition, which was overcome through the inclusion of up to 4-day-old data to fill data gaps up to the present time point. For this gap filling methodology, relevancy decreased as the time window width increased, although data with 4-day time lag values represented less than 4% of the total data. Overall, two models stood out, and further studies should either be based on the random forest model if they need prediction quality or on the cubist model if they need continuity. Further studies should focus on developing the DSS and on the conversion of CSH to actual forage allowance.

The advantage of employing mid-infrared spectrometry for milk analysis in breeding lies in its ability to quickly generate millions of records. However, these records may be biased if the calibration process does not account for their spectral variability when constructing the predictive model. So, this study introduces a novel method for developing a World Representative Spectral Database (WRSD) to reduce the risks of spectral extrapolation when predicting dairy traits in new samples. Utilizing a 2-phase selection procedure that is both efficient and minimizes memory usage, we first generate a decomposition matrix via Principal Component Analysis (PCA) on a data set of 2,324,443 records. The next phase iterates spectral selection based on a location index from PCA scores, calculating spectra occurrence frequency for refined barycenter estimations. The chosen spectra's barycenter closely aligns with the entire data set, proving the efficacy of using just 3 principal components (PCs). Applied to 4 varied data sets, totaling over 21 million records, we select 583,440 spectra to represent spectral diversity, with selection rates between 2.00% and 14.88%. This selection illustrates the spectral variability across different dairy populations and data providers. Demonstrated through a hypothetical calibration set of 71 samples, the WRSD's utility for algorithm developers becomes apparent. This calibration set covers between 91.42 to 98.50% of the WRSD variability, except for the Irish data set (3.50%), indicating a need for additional samples to accurately represent Irish variability and minimize spectral extrapolation. This study offers valuable insights into the representativeness of training sets for capturing spectral variability within targeted dairy populations. While the current WRSD does not fully encompass global milk spectral diversity, its development underscores the importance of gathering more data and standardizing spectral information across spectrometer brands. Ultimately, the WRSD proves crucial not just for trait prediction but also for identifying abnormal milk samples, also marking a significant relevance in dairy science technology.

Effectively evaluating and promoting pro-grazing practices necessitates the implementation of a verification system. To address this imperative, exploration of milk composition analysis as a means to assess grazing practices has garnered substantial attention. In this study, we used component predictions from milk Fourier-transform mid-infrared (FT-MIR) spectra to construct an indicator to estimate the proportion of herbage consumed by dairy cows and another indicator to validate grazing. This approach was developed and validated using 75 estimated bulk milk analyses, each associated with 3 variables related to feeding from the same day ± 3 d, totaling 526 observations. These 3 variables are based on the occupation time, harvested and conserved herbage, and other feeds from 7 farms in Luxembourg. Hierarchical clustering facilitated the effective segregation of observations into distinct groups, with one group predominantly focused on herbage and the other group on other feeds. Leveraging partial least squares discriminant analysis trained on FT-MIR predicted milk characteristics from both groups, we successfully developed an indicator-the probability of belonging to the herbage group-with a maximum accuracy of 0.93, a sensitivity of 0.94, and a specificity of 0.93 on the Luxembourg dataset. In a partial least squares regression, the cross-validation yielded results of predicting the percentage of herbage in the diet with an error of 8.77%. Notably, the indicator relied on FT-MIR predicted components expected to reflect a diet based on herbage, such as the total of C18:1 trans fatty acids and CLA. However, it also incorporated unexpected FT-MIR predicted parameters like milk acidity parameters, citrate content, and specific proteins such as lactoferrin. Finally, the developed indicators were tested on the 5,886,364 Walloon spectra collected between 2009 and 2023, as well as 23,718 Walloon spectra between 2023 and 2025 from 72 farms known to practice grazing. The annual trends were analyzed in the context of Walloon dairy production, helping to refine the selection of better indicators. These results could contribute to practical tools for monitoring and estimating days spent on pasture. Looking ahead, future research should aim to incorporate more comprehensive data, such as precise feed compositions, to further refine our understanding of the influence of specific herbage diets on milk composition and enhance the detection of associated changes.