Strategy of statistical model selection for precision farming on-farm experiments

Thöle, Heinrich; Richter, Christel; Ehlert, Detlef

doi:10.1007/s11119-013-9306-9

Cited by 18 publications

(12 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…However, these two types of models require many parameters, some of which are difficult to measure, and the practicability of the two models is deficient for a highly decentralized agricultural production pattern. While based on crop fertilization rate and yield effects, unary and multivariate statistical models have the advantages of simplicity and practicality and have been widely studied and popularized 4,5,10,24 . But, it is unfortunate that this polynomial model has problems such as bias error and multicollinearity 7 , which leads to a significantly lower modelling success rate.…”

Section: Discussionmentioning

confidence: 99%

Improving Rice Modeling Success Rate with Ternary Non-structural Fertilizer Response Model

Zhang

Chen

et al. 2018

Sci Rep

View full text Add to dashboard Cite

Fertilizer response modelling is an important technical approach to realize metrological fertilization on rice. With the goal of solving the problems of a low success rate of a ternary quadratic polynomial model (TPFM) and to expand the model’s applicability, this paper established a ternary non-structural fertilizer response model (TNFM) based on the experimental results from N, P and K fertilized rice fields. Our research results showed that the TNFM significantly improved the modelling success rate by addressing problems arising from setting the bias and multicollinearity in a TPFM. The results from 88 rice field trials in China indicated that the proportion of typical TNFMs that satisfy the general fertilizer response law of plant nutrition was 40.9%, while the analogous proportion of TPFMs was only 26.1%. The recommended fertilization showed a significant positive linear correlation between the two models, and the parameters N0, P0 and K0 that estimated the value of soil supplying nutrient equivalents can be used as better indicators of yield potential in plots where no N or P or K fertilizer was applied. The theoretical analysis showed that the new model has a higher fitting accuracy and a wider application range.

show abstract

Section: Discussionmentioning

confidence: 99%

Improving Rice Modeling Success Rate with Ternary Non-structural Fertilizer Response Model

Zhang

Chen

et al. 2018

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Yield mapping is one of the most widely-used precision agriculture techniques [11][12][13]. Most of these datasets are characterized by a non-normal distribution due to the presence of errors and outliers and can be misleading if used for decision making processes.…”

Section: Introductionmentioning

confidence: 99%

A Precision Agriculture Approach for Durum Wheat Yield Assessment Using Remote Sensing Data and Yield Mapping

et al. 2019

View full text Add to dashboard Cite

The availability of big data in agriculture, enhanced by free remote sensing data and on-board sensor-based data, provides an opportunity to understand within-field and year-to-year variability and promote precision farming practices for site-specific management. This paper explores the performance in durum wheat yield estimation using different technologies and data processing methods. A state-of-the-art data cleaning technique has been applied to data from a yield monitoring system, giving a good agreement between yield monitoring data and hand sampled data. The potential use of Sentinel-2 and Landsat-8 images in precision agriculture for within-field production variability is then assessed, and the optimal time for remote sensing to relate to durum wheat yield is also explored. Comparison of the Normalized Difference Vegetation Index(NDVI) with yield monitoring data reveals significant and highly positive linear relationships (r ranging from 0.54 to 0.74) explaining most within-field variability for all the images acquired between March and April. Remote sensing data analyzed with these methods could be used to assess durum wheat yield and above all to depict spatial variability in order to adopt site-specific management and improve productivity, save time and provide a potential alternative to traditional farming practices.

show abstract

“…For these simulations an isotropic spherical covariance model including nugget effect was used (Eq. [2]), which, along with the exponential model, is one of the models used in literature to represent both soil and yield properties (Webster and Oliver, 2007, Thöle et al, 2013, Richter et al, 2015).

C_{nug + sph} (h) = {\begin{matrix} σ_{nug}^{2} + σ_{psill}^{2} \frac{3 h}{2 a} + \frac{1 h^{s}}{2 a^{s}} & 0 \leq h \leq a \\ 0 & h > a \end{matrix}}

where

C_{nug + sph} (h)

is the spatial covariance function; a is the range of spatial dependence,

σ_{nug}^{2}

is the non‐spatially structured variance or nugget, and

σ_{psill}^{2}

is the spatially structured variance or partial sill.…”

Section: Methodsmentioning

confidence: 99%

Experimental Designs and Estimation Methods for On‐Farm Research: A Simulation Study of Corn Yields at Field Scale

et al. 2019

View full text Add to dashboard Cite

On‐farm experimentation using Precision Agriculture technology enables farmers to make decisions based on data from their fields. Results from on‐farm experiments depend on the experimental design and statistical analyses performed. Detailed information about the accuracy of the treatment effect estimates, and Type I error rates of hypothesis testing under different spatial structure scenarios attained by alternative experimental designs and analysis is required to improve on‐farm research experiments. Three thousand yield data sets were drawn from 15 random fields simulated by unconditional Gaussian geostatistical simulation technique and were modeled by applying 10 experimental designs and three estimation methods with experimental units ranging from 138 to 9969 m2. No effect of spatial structure, experimental design, and estimation methods was observed on overall mean yield and treatment bias. Unaddressed changes of nugget/sill ratio and range of variogram had a significant effect on estimator efficiency and accuracy with Type I error rates above the nominal rate, which increased with higher spatial autocorrelation. Spatial methods were robust to changes in spatial structure regardless of the design. Randomization of treatment increased the uncertainty of model estimators. In general, the accuracy of treatment effect estimates increased with the number of replications of smaller size. The opposite trend was observed between those estimates and the size of the plots. Analyses showed that the best designs for testing the overall treatment effect in two‐treatment experiments would be split‐planter, strip‐plots, and chessboard because of their size and number of experimental units. Core Ideas Spatial autocorrelation increases grand mean estimator variance in any design or method. Spatial autocorrelation reduces treatment effect estimator efficiency if not modeled. Spatial autocorrelation increases Type I error if not modeled. Designs with small experimental units (strip plots or chessboard) performed better.

show abstract

Strategy of statistical model selection for precision farming on-farm experiments

Cited by 18 publications

References 20 publications

Improving Rice Modeling Success Rate with Ternary Non-structural Fertilizer Response Model

Improving Rice Modeling Success Rate with Ternary Non-structural Fertilizer Response Model

A Precision Agriculture Approach for Durum Wheat Yield Assessment Using Remote Sensing Data and Yield Mapping

Experimental Designs and Estimation Methods for On‐Farm Research: A Simulation Study of Corn Yields at Field Scale

Contact Info

Product

Resources

About