AgWise / Approaches for Fertilizer Advice
Approaches Used For Fertilizer Advice
Developing fertilizer advice requires knowledge of the yield gap and the yield response to nutrient applications. AgWise has four approaches that can be used based on data availability and quality and objective of the fertilizer advisory.
QUEFTS
QUEFTS crop model is used to assess soil nutrient supply and the yield gap, based on yield data from field trials with varying nutrient applications
read more
Machine Learning
Machine learning algorithms are used to model yield as response to nutrients and biophysical factors such as soil and weather
read more
QUEFTS & Machine Learning
Machine learning algorithms model soil nutrient supply estimated by QUEFTS in response to geospatial variables at scale.
read more
Integrated Approach
This approach refines the third one by incorporating a yield ceiling for rainfed farming, simulated across multiple climate scenarios
read more
How to choose which approach to use
Data availability
The complexity and accuracy of the models depend on the availability and quality of data. While the first two approaches rely mainly on field trials data, the third and fourth approaches incorporate extensive geo-spatial and climate data.
User skills
The complexity of implementation increases from the first approach to the fourth one, with the incorporation of advanced modeling techniques and data requirements.
Accuracy expected
As the complexity increases, the accuracy and precision of fertilizer advice tend to improve, especially in estimating soil nutrient supply and yield potential under varying conditions.
Applicability
The choice of the approach depends on the availability of data, computational resources, and the specific requirements of the target audience. While simpler models may suffice for initial fertilizer advice, more complex models offer better precision and adaptability