Towards effective sampling for nutrients’ predictive mapping in farm-scale crop management

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00027049%3A_____%2F23%3AN0000071" target="_blank" >RIV/00027049:_____/23:N0000071 - isvavai.cz</a>

Alternative codes found

RIV/00027049:_____/23:N0000063

Result on the web

—

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Towards effective sampling for nutrients’ predictive mapping in farm-scale crop management

Original language description

Plant nutrition and balanced fertilisation require highly accurate field information with limited sampling. Therefore, the optimisation of sampling schemes is crucial. The aim of the study is to compare the widely used sampling designs combined with variable sample size for supervised prediction of common soil macronutrients. Conditioned Latin Hypercube Sampling (cLHS), Feature Space Coverage Sampling using k-means (FSCS) and Simple Random Sampling (SRS) were compared. The influence of sampling scheme and sample size on the accuracy of predicted nutrient maps was investigated in a real case study of a field (35 ha) with heterogeneous soil properties. A total of 200 training points were placed in 6 networks: cLHS and FSCS with 10, 30 and 60 samples each, corresponding to 1 sample per 3, 1 and 0.5 ha respectively. For the numerical experiment with the varying sampling rates, all 200 training samples were interpolated into a set of nutrient maps, which were treated as an error-free dataset for both calibration and validation samples included in the predictive modelling. Sampling networks of variable sample size from 2 to 60 were generated using the 80 SRS, cLHS and FSCS combined with a pragmatic suite of environmental covariates. Each network of each method was automatically generated 100 times using the same algorithm settings. These were used for covariate-driven models. The performance of the models was monitored. The results show the advantage of using the FSCS, which shows both less variation in the accuracy of the prediction compared to SRS and cLHS, and better results under sparse sampling.

Czech name

—

Czech description

—

Type

O - Miscellaneous

CEP classification

—

OECD FORD branch

10511 - Environmental sciences (social aspects to be 5.7)

Project

<a href="/en/project/QK21010247" target="_blank" >QK21010247: Management optimization of unbalanced fields by means of digital soil mapping and soil moisture changes monitoring in order to stabilize the achievable yield</a><br>

Continuities

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Publication year

2023

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů