I have data's on both soil data (i.e. soil parameters like moisture, clay, texture, nok etc) and yield and agronomic traits of banana.
1) I wanted to know which statsitical technique would best explain the relationship between these two components i.e soil quality and agronomic data/yield.
2) I know Pearson's correaltion will explain whether they are positively/negatively correlated or no correaltion, but I wanted to know if the esteemed member's of RG could share their thoughts on other statsitical techniques.
3) if there are some suggestions, may I kindly know the steps and procedures or some references.
Thank you and regards.
Nice question . Please undertake stepdown partial regression analysis to find out which are important soil parameters ( not exactly the soil quality) governing the yield. The principal component analysis will also give you important clustering of different soil parameters vis-a-vis yield.
Anoop Kumar Srivastava thank you for excellent answer sir, will definitely look into that method. Additionally, If I include other parameters such as biochemicals of the fruit, is there any practical way to quantify and coorealte with the soil parameters sir?
You have to separate biochemical parameters versus soil parameters and yield versus soil parameters , will give better interpretation through step down partial regression analysis..You can combine both parameters for PCA and see , what hypothesis emerges...
PCA would be the best tool for that. Along with that correlation and regression too.
Soil quality is one important part of sustainable agro ecosystem management, analogous to water and air quality. Assessing soil quality may help managers identify practices that could be adapted to become more sustainable. Soil quality is one aspect of sustainable agro ecosystem management.Moreover, Soil quality indicators are physical, chemical, and biological properties, processes, and characteristics that can be measured to monitor changes in the soil. Soil quality depends on the chemical composition of the soil, the topography, the presence of living organisms, the climate, and time. Agricultural practice and history may also modify the characteristics and fertility of soil. However, biochemical parameters versus soil parameters and yield versus soil parameters, will give better interpretation through step down partial regression analysis. Therefore, standard soil fertility attributes (soil pH, organic carbon, available N, P, and K) are the most important factors in terms of plant growth, crop production and microbial diversity and function. As we know, these parameters are generally sensitive to soil management.
Rk Naresh thank you for the answer sir. I will definitely look into it. Regards
Dear @Wati Temjen, I am in agreement with @Anoop Kumar Srivastava's statements. I only want to add sth to respond your question concerning using other parameters. I am a soil biologist and during my doctoral thesis I worked with a multipled case data investigating the impacts of different environmental variables on soil fauna communities, so I selected CCA canonical correspondence analysis and did it in R program. I could get very excellent results. I highly recommend if possible you also implement it for your data.
Elaheh Daghighi thank you for the excellent input! Do you have any publication on your thesis that I can follow up with? I am new to CCA in general, so I am not sure on how to structure the components.
Dear Wati, Hereby I send you the pdf of my published doctoral thesis paper in 2017. It is about the effects of climate change on long term development of soil fauna and the results have been clearly shown by MANOVA. Best, Elaheh
- Badges
- Science topic
- Similar topics
- Social Science
- Media
- Journalism