Are there any experiences or studies on the effect and contribution of combining different methods in crop science i.e. on-farm experiments, on-station field trials and modelling to achieve aims such as increase in productivity, higher resource-efficiency or others?
Are there any articles that compare these different approaches?
http://www.fao.org/wairdocs/ilri/x5545e/x5545e08.htm
http://www.sciencedirect.com/science/article/pii/S1573521410000461
http://cordis.europa.eu/result/rcn/174256_en.html
https://www.ilri.org/biometrics/GoodStatisticalPractice/publications/guides/topoft.html
https://www.researchgate.net/publication/231977965_REML_is_an_effective_analysis_for_mixed_modelling_of_unbalanced_on-farm_varietal_trials
Article REML is an effective analysis for mixed modelling of unbalan...
I think there is added value when combining on-farm experiments, on-station field trials and modelling, because they fed each other. Models need information from field experiments to be calibrated and be able to produce more accurate simulations. And field trials and on farm-experiments can be planed more efficiently by using the information gained from the simulations.
As far as I know there are very few articles that compare these approaches. Perhaps one of these articles might be useful for you:
Strategy of statistical model selection for precision farming on-farm experiments. Precision Agric (2013) 14:434–449
A probabilistic Bayesian framework for progressively updating site-specific recommendations. Precision Agric (2015) 16:275–296
Geostatistical Models in Agricultural Field Experiments: Investigations Based on Uniformity Trials. Agron. J. 104:91–105 (2012)
Adapting and evaluating the CROPGRO-peanut model for response tophosphorus on a sandy-loam soil under semi-arid tropical conditions. Field Crops Research 176 (2015) 71–86
Tricky to answer in a general manner. For exemple, plant breeding can fit in your question, and is a case where multi-environment trials and modeling achieved increases in farm productivity through new cultivars.
But we can assume various means to increase resources use efficiency in farms depending on the type of agriculture you consider: choosing the most suitable cultivar accounting for GxE, tailoring the management practices to farm conditions (precision agriculture), using different species together (intercropping) or even relying more and more on ecosystem services as agricultural inputs (agroecology).
While modeling (statistical or simulation) is more and more used for precision agriculture (previous answers), I assume it need to gain "trust" when moving to more biodiversity-based agricultural systems. Also, a bit of shameless promotion for our work, but I assume linking public multi-environmental networks and modeling to improve knowledge on cultivars is a place where the gains you mention can happen, but much work in still needed on the articulation between field knowledge and model-based knowledge to advise farmers.
Article A model-based approach to assist variety evaluation in sunflower crop
Moritz,
interesting question which can be answered either by No or by Yes.
My answer is a loud YES.
Modelling means here you can do agro-scientific points including
ergonomical and economical aspects in the on-farm component
up to foreign politics and even a model of World Economy in
a predator-prey (Lotka) relation, but also complex natural
processes for on-station studies. Cheers
Usually pot and station based trials produce location specific results. On-farm trials in various environments allow validation of pot/station trials and help find degree of adaptability and adoptability. Modeling, whether for location specific or wider geographical environments, requires technical and economic coefficients or parameters. Those are usually derived from station and on-farm trials as well as survey based results.
Sometimes, models may be constructed with with a combination available and hypothetical data (where no data is currently available), then results may be validated by conduction station and/or on-farm trials.
So these various data sources are complementary.
See our recently published paper on this issue. Here, we compare different approaches in agronomy... Article Re-designing organic grain legume cropping systems using sys...
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India