Browsing by Person "Perdana-Decker, Sari"

Now showing 1 - 2 of 2

Development of a robust sensor calibration for a commercially available rising platemeter to estimate herbage mass on temperate seminatural pastures
(2024) Werner, Jessica; Salazar‐Cubillas, Khaterine; Perdana-Decker, Sari; Obermeyer, Kilian; Velasco, Elizabeth; Hart, Leonie; Dickhöfer, Uta
Rising platemeters are commonly used in Ireland and New Zealand for managing intensive pastures. To assess the applicability of a commercial rising platemeter operating with a microsonic sensor to estimate herbage mass with its own equation, the objectives were (i) to validate the original equation; (ii) to identify possible factors hampering its accuracy and precision; and (iii) to develop a new equation for heterogeneous swards. A comprehensive dataset (n = 1511) was compiled on the pastures of dairy farms. Compressed sward heights were measured by the rising platemeter. Herbage mass was harvested to determine reference herbage availability. The adequacy of estimating herbage mass was assessed using root mean squared error (RMSE) and mean bias. As the adequacy of the original equation was low, a new equation was developed using multiple regression models. The mean bias and the RMSE for the new equation were overall low with 201 kg dry matter/ha and 34.6%, but it tended to overestimate herbage availability at herbage mass < 500 kg dry matter/ha and underestimate it at >2500 kg dry matter/ha. Still, the newly developed equation for the microsonic sensor-based rising platemeter allows for accurate and precise estimation of available herbage mass on pastures.
Modelling nitrogen use and excretion in dairy cattle herds grazing temperate, semi-natural grasslands
(2025) Perdana-Decker, Sari; Dickhöfer, Uta
Grazing-based dairy cattle systems exhibit several benefits, such as preserving biodiverse grassland habitats, improving animal welfare, or turning grassland protein into human-edible protein. However, grazing-based diets are prone to greater nitrogen (N) losses via urine than balanced stall-fed diets, leading to a greater risk for N emissions. Strategies for improving the N use in grazing-based systems are predominantly investigated on homogenous clover-ryegrass pastures with high yields and nutritional quality. In contrast, grazing-based systems reliant on less external inputs (e.g., synthetic fertilisers or concentrates) using semi-natural grassland as main feed source received less attention. The present thesis addressed the knowledge gap on the N use of such low-input grazing-based systems by adapting an existing dynamic, process-based herd model (i.e., the LIVestock SIMulator, LIVSIM) for simulating animal performance and N use and excretion of dairy herds. For this, a broad dataset was gathered on nine commercial organic dairy cattle farms in Baden-Württemberg during two grazing periods (2019, 2020). This dataset fulfilled two purposes: firstly, to get a basic understanding on N use and excretion of dairy cows under low-input grazing conditions (study 1); secondly, to serve as reference dataset for adapting and evaluating LIVSIM for such production systems (studies 2 and 3). The reference dataset represented the wide range of grazing and production factors found on commercial farms in South Germany using semi-natural grasslands for grazing. The dataset applied for study 1 covered n = 323 individual animal observations with mean (± one standard deviation) milk production, dry matter intake (DMI), and pasture DMI (PDMI) of 23.9 (± 5.35), 21.0 (± 3.21), and 11.3 (± 4.83) kg/d, respectively. Milk N use efficiency (MNE) averaged 24.7 g/100 g N intake (± 5.91), which is greater than observations in temperate, high-input grazing-based systems but lower than in cows receiving balanced stall-fed diets. Nevertheless, MNE and other indicators of N use and excretion varied greatly among farms and seasons, highlighting the need to identify the drivers for this variation. Supplement feeding had the greatest potential for manipulating the N use and excretion. Increasing shares of fresh forages as well as of hay of total supplement DMI increased N use (e.g., MNE) and decreased urinary N excretion (e.g., urinary N to creatinine ratio), while increasing shares of concentrates of supplement DMI were related to lower N losses via urine. Study 1 highlighted that using semi-natural grasslands for grazing can potentially reduce environmentally harmful N losses compared to high-input grazing systems. For future research endeavours, a modelling approach may simplify the investigation of more feeding scenarios, their interactions, different local conditions, and considering the spatial and temporal variation of pasture herbage quality and yield. Hence, studies 2 and 3 focused on adaptating LIVSIM for low-input grazing-based dairy farms. The DMI and N intake are among the most decisive factors for determining animal performance and N excretion. Therefore, a module for predicting the PDMI of cows grazing semi-natural grassland was identified in study 2, using a subset of the reference dataset (n = 233 individual animal observations). Among the thirteen tested models, behaviour-based and semi-mechanistic models specifically developed for grazing animals had the lowest prediction adequacy. Their underlying empirical equations likely did not fit the grazing and production conditions of farms employing semi-natural grasslands. Modelling performance of a semi-mechanistic model developed for stall-based feeding situations (Mertens II) with slight modifications was best (relative prediction error = 13.4%) when evaluated based on the mean observed PDMI (i.e., averaged across animals per farm and period (n = 28)). Consequently, the modified Mertens II model was integrated in LIVSIM in study 3. Additionally, the modules for energy requirements, lactation, N excretion, and herd management were adopted, and breed-specific model coefficients added to represent Simmental, Brown Swiss, and Holstein-Friesian cattle breeds. Dairy cow characteristics, herd composition, annual milk yield, and DMI were predicted accurately (i.e., with a relative difference ≤ 10 % between observed and predicted outputs for the majority of outputs). The absolute total N excretion (g/d) was underpredicted by 23 % (= relative difference between observed and predicted values) mainly due to the underprediction of urinary N excretion by 43 %. The relative differences in N excretion between farming systems, in contrast, were predicted reliably. The observed faecal, urinary, and total N excretion (in % of N intake) differed by 30, -23, and -7 %, respectively, between the two reference herds, which is similar to the respective relative differences for the predicted faecal, urinary, and total N excretion of 32, -36, and -4 %. Further model improvements should focus on increasing the prediction accuracy of N excretion and its partitioning due to the varying degree of susceptibility of faecal or urinary N to volatilisation and leaching. The scenario and sensitivity analyses further confirmed that the adapted LIVSIM plausibly simulated differences in animal performance and nutrient excretions based on differences in supplement feeds and pasture herbage. Core input and model coefficients are the dietary ME, CP, and rumen-undegradable CP concentrations, as well as the available herbage biomass on pastures, for which precise measurements are thus needed. The findings of studies 2 and 3 demonstrate that existing models can be adopted for low-input grazing-based dairy production systems. There is further potential for adapting LIVSIM for production systems beyond the ones investigated in the present study, and/or for adding more outputs (e.g., enteric methane) and scales (e.g., grassland) to better capture the multifaceted aspects determining farm sustainability.