PastureBase Ireland: A grassland decision support system and national database

- PastureBase Ireland is a web-based grassland management decision support tool for farmers, which incorporates a dual function of decision support and the development of a centralized database to collate grassland data.
- PastureBase Ireland will enhance the decision making process around grassland management at farm level, while unlocking the potential of commercial farm grassland research.
- This web-based tool has shown a relatively high level of accuracy around annual dry matter yield with a relative prediction error of 15.4%.
- An analysis of 75 commercial farms that use the system has shown greater within farm than across farm variability in individual paddock grass growth.

PastureBase Ireland (PBI) is a web-based grassland management application incorporating a dual function of grassland decision support and a centralized national database to collate commercial farm grassland data. This database facilitates the collection and storage of vast quantities of grassland data from grassland farmers. The database spans across ruminant grassland enterprises - dairy, beef and sheep. To help farmers determine appropriate actions around grassland management, we have developed this data informed decision support tool to function at the paddock level. Individual farmers enter data through the completion of regular pasture cover estimations across the farm, allowing the performance of individual paddocks to be evaluated within and across years. To evaluate the PBI system, we compared actual pasture cut experimental data (Etesia cuts) to PBI calculated outputs. We examined three comparisons, comparing PBI outputs to actual pasture cut data, for individual DM yields at defoliation (Comparison 1), for cumulative annual DM yields including silage data (Comparison 2) and, for cumulative annual DM yields excluding silage data (Comparison 3). We found an acceptable accuracy between PBI outputs and pasture cut data when statistically analyzed using relative prediction error and concordance correlation coefficients for the measurement of total annual DM yield (Comparison 2), with a relative prediction error of 15.4% and a concordance correlation coefficient of 0.85. We demonstrated an application of the PBI system through analysis of commercial farm data across two years (2014-2015) for 75 commercial farms who actively use the system. The analysis showed there was a significant increase in DM yield from 2014 to 2015. The results indicated a greater variation in pasture growth across paddocks within farms than across farms.