Michael Simmler

Switzerland has a wide diversity of soil types that are subject to pressures of varying intensity and scale. Localised pressures such as compaction, sealing or heavy metal accumulation threaten both short‐ and long‐term soil fertility and food production. Large‐scale trends such as the intensive use of fertilisers or continued cultivation of organic soils can have effects on the ecosystem functioning of an entire region. Many of these pressures are concentrated on the Swiss plateau, which covers only about one third of the country and contains both the densest population and the most productive soils. We here review the main soil challenges of Switzerland and examine to which extent these challenges are addressed by different legal frameworks. Soil erosion, contamination as well as nutrient availability and use efficiency are all best described in the scientific literature and regulated by policies. In contrast, for several pressing soil challenges such as soil organic carbon loss, biodiversity loss or soil compaction only limited policy instruments are in place. We also report on the key findings of the 5‐year EJP SOIL project that aimed to address several of these soil challenges in research projects of different scale. We discuss the key findings relevant for Switzerland, and outline their implications for various stakeholder groups, including researchers, policy makers, farmers, land managers and advisors. This not only enhances the value of the research outputs for stakeholders but also demonstrates how a research programme of this scale can directly benefit different national stakeholder groups.

Multispectral imaging satellites such as Sentinel‐2 are considered a possible tool to assist in the mapping of soil organic carbon (SOC) using images of bare soil. However, the reported results are variable. The measured reflectance of the soil surface is not only related to SOC but also to several other environmental and edaphic factors. Soil texture is one such factor that strongly affects soil reflectance. Depending on the spatial correlation with SOC, the influence of soil texture may improve or hinder the estimation of SOC from spectral data. This study aimed to investigate these influences using local models at 34 sites in different pedo‐climatic zones across 10 European countries. The study sites were individual agricultural fields or a few fields in close proximity. For each site, local models to predict SOC and the clay particle size fraction were developed using the Sentinel‐2 temporal mosaics of bare soil images. Overall, predicting SOC and clay was difficult, and prediction performances with a ratio of performance to deviation (RPD) > 1.5 were observed at 8 and 12 of the 34 sites for SOC and clay, respectively. A general relationship between SOC prediction performance and the correlation of SOC and clay in soil was evident but explained only a small part of the large variability we observed in SOC prediction performance across the sites. Adding information on soil texture as additional predictors improved SOC prediction on average, but the additional benefit varied strongly between the sites. The average relative importance of the different Sentinel‐2 bands in the SOC and clay models indicated that spectral information in the red and far‐red regions of the visible spectrum was more important for SOC prediction than for clay prediction. The opposite was true for the region around 2200 nm, which was more important in the clay models.

Accelerometers are sensors proven to be useful to analyze the lying behavior of cows. For reasons of algorithm transparency and control, researchers often prefer to use their own data analysis scripts rather than proprietary software. We developed the triact R package that assists animal scientists in analyzing the lying behavior of cows from raw data recorded with a triaxial accelerometer (manufacturer agnostic) attached to a hind leg. In a user-friendly workflow, triact allows the determination of common measures for lying behavior including total lying duration, the number of lying bouts, and the mean duration of lying bouts. Further capabilities are the description of lying laterality and the calculation of proxies for the level of physical activity of the cow. In this publication we describe the functionality of triact and the rationales behind the implemented algorithms. The triact R package is developed as an open-source project and freely available via the CRAN repository.

Current evidence suggests that frequent exposure to situations in which captive animals can solve cognitive tasks may have positive effects on stress responsiveness and thus on welfare. However, confounding factors often hamper the interpretation of study results. In this study, we used human-presented object-choice tests (in form of visual discrimination and reversal learning tests and a cognitive test battery), to assess the effect of long-term cognitive stimulation (44 sessions over 4–5 months) on behavioural and cardiac responses of female domestic goats in subsequent stress tests. To disentangle whether cognitive stimulation per se or the reward associated with the human–animal interaction required for testing was affecting the stress responsiveness, we conditioned three treatment groups: goats that were isolated for participation in human-presented cognitive tests and rewarded with food (‘Cognitive’, COG treatment), goats that were isolated as for the test exposure and rewarded with food by the experimenter without being administered the object-choice tests (‘Positive’, POS treatment), and goats that were isolated in the same test room but neither received a reward nor were administered the tests (‘Isolation’, ISO treatment). All treatment groups were subsequently tested in four stress tests: a novel arena test, a novel object test, a novel human test, and a weighing test in which goats had to enter and exit a scale cage. All treatment groups weretested at the same two research sites, each using two selection lines, namely dwarf goats, not selected for production traits, and dairy goats, selected for high productivity. Analysing the data with principal component analysis and linear mixed-effects models, we did not find evidence that cognitive testing per se (COG–POS contrast) reduces stress responsiveness of goats in subsequent stress tests. However, for dwarf goats but not for dairy goats, we found support for an effect of reward-associated human–animal interactions (POS–ISO contrast) at least for some stress test measures. Our results highlight the need to consider ontogenetic and genetic variation when assessing stress responsiveness and when interacting with goats.

Free-stall cubicles are designed so that cows do not defecate in the bedding material, while still providing comfortable lying places. However, fixed cubicle elements, such as the partition and neck rail, restrict available movement space and may hinder cows from performing natural lying down and standing up movement patterns. Although there are various types of cubicle partitions and configurations that differ in design, dimensions, or materials, there is no method other than direct observations to assess their effect on cow welfare. The aim of this research was to explore which atypical lying down and standing up behaviours could be suitable to be detected using body-mounted accelerometers and machine learning. Three leg- and one head-mounted accelerometer set to record at 20 Hz were fitted to 48 lactating dairy cows (Brown Swiss and Holstein × Swiss Fleckvieh). Lying down and standing up events were simultaneously assessed through video observations. Time series classification algorithms were used for model development using a nested 5-fold stratified cross-validation, with individual cows considered as non-overlapping groups. The most effective classifiers were the MiniRocket time-series transform and the deep learning architecture InceptionTime. Because of its speed and accuracy, MiniRocket was selected for further analyses. Atypical behaviours were detected with balanced accuracies between 0.56 and 0.74. Atypical behaviours performed during standing up events, namely Hesitant head lunge, Sideways head lunge, and Crawling backwards, appear to be the most promising candidates for accelerometer-based detection. These behaviours were best detected, and further analysis with MiniRocket indicated that more training data will likely result in increased model performance. Detection accuracies for these atypical behaviours did not decrease when down sampling the acceleration signal until a sampling frequency of 5 Hz. The proposed behaviours could be used for the development of an automated detection system to improve efficiency and promote objectivity in the evaluation procedure of dairy cow housing installations with regard to animal welfare.

Cognitive research in long-lived species commonly involves using the same animals in different experiments. It is unclear whether the participation in cognitive tests can notably alter the performance of individuals in subsequent conceptually different tests. We therefore investigated whether exposure to cognitive tests affects future test performance of goats. We used three treatment groups: goats with long-term exposure to human-presented object-choice tests (for visual discrimination and reversal learning tests + cognitive test battery), goats that were isolated as for the test exposure but received a reward from the experimenter without being administered the object-choice tests, and goats that were isolated but neither received a reward nor were administered the tests. All treatment groups were subsequently tested in two conceptually different cognitive tests, namely a spatial A-not-B detour test and an instrumental problem-solving test. We tested dairy goats, selected for high productivity, and dwarf goats, not selected for production traits, each at the same two research sites. We did not find notable differences between treatments with respect to the goats' detour or problem-solving performance. However, high variation was observed between the research sites, the selection lines, and among individuals, highlighting potential pitfalls of making accurate comparisons of cognitive test performances.

Contrafreeloading (CFL) is the phenomenon when animals work for a resource although an identical resource is available for free. Possible explanations for CFL are that animals seek context for species-specific behaviours or to control their environments. We investigated whether goats show CFL and whether breeding for productivity traits has altered its occurrence. In a manipulation task, we compared two selection lines: 27 Nigerian dwarf goats, not bred for productivity traits, and 30 dairy goats, bred for high milk yield. Over 10 trials, each goat could perform one of three behaviours: not participating in the trial, feeding for free from an open door, or opening a sliding door for a feed of similar value. The results were analysed using an Item Response Tree (IRTree) generalized linear mixed model (GLMM). The fitted probabilities to participate were > 0.87 over all trials in both selection lines. For dwarf goats, the probability of choosing the closed door, and thereby demonstrating CFL, increased from 0.30 in Trial 1 to 0.53 in Trial 10. For dairy goats, this probability was constant at approximately 0.43. Unlike dwarf goats, dairy goats were faster to approach the closed compared to the open door. Overall, our results suggest that both selection lines were similarly interested in CFL.

The analysis of multispectral imagery (MSI) acquired by unmanned aerial vehicles (UAVs) and mobile near-infrared reflectance spectroscopy (NIRS) used on-site has become increasingly promising for timely assessments of grassland to support farm management. However, a major challenge of these methods is their calibration, given the large spatiotemporal variability of grassland. This study evaluated the performance of two smart farming tools in determining fresh herbage mass and grass quality (dry matter, crude protein, and structural carbohydrates): an analysis model for MSI (GrassQ) and a portable on-site NIRS (HarvestLab^TM 3000). We compared them to conventional look-up tables used by farmers. Surveys were undertaken on 18 multi-species grasslands located on six farms in Switzerland throughout the vegetation period in 2018. The sampled plots represented two phenological growth stages, corresponding to an age of two weeks and four to six weeks, respectively. We found that neither the performance of the smart farming tools nor the performance of the conventional approach were satisfactory for use on multi-species grasslands. The MSI-model performed poorly, with relative errors of 99.7% and 33.2% of the laboratory analyses for herbage mass and crude protein, respectively. The errors of the MSI-model were indicated to be mainly caused by grassland and environmental characteristics that differ from the relatively narrow Irish calibration dataset. The On-site NIRS showed comparable performance to the conventional Look-up Tables in determining crude protein and structural carbohydrates (error ≤ 22.2%). However, we identified that the On-site NIRS determined undried herbage quality with a systematic and correctable error. After corrections, its performance was better than the conventional approach, indicating a great potential of the On-site NIRS for decision support on grazing and harvest scheduling.