Conditioning Machine Learning Models to Adjust Lowbush Blueberry Crop Management to the Local Agroecosystem.

Serge-Étienne Parent,Jean Lafond,Noura Ziadi,Léon Etienne Parent,Maxime C Paré

doi:10.3390/plants9101401

Serge-Étienne Parent, Jean Lafond + Show 3 more

Open Access

https://doi.org/10.3390/plants9101401

Copy DOI

Abstract

Agroecosystem conditions limit the productivity of lowbush blueberry. Our objectives were to investigate the effects on berry yield of agroecosystem and crop management variables, then to develop a recommendation system to adjust nutrient and soil management of lowbush blueberry to given local meteorological conditions. We collected 1504 observations from N-P-K fertilizer trials conducted in Quebec, Canada. The data set, that comprised soil, tissue, and meteorological data, was processed by Bayesian mixed models, machine learning, compositional data analysis, and Markov chains. Our investigative statistical models showed that meteorological indices had the greatest impact on yield. High mean temperature at flower bud opening and after fruit maturation, and total precipitation at flowering stage showed positive effects. Low mean temperature and low total precipitation before bud opening, at flowering, and by fruit maturity, as well as number of freezing days (<−5 °C) before flower bud opening, showed negative effects. Soil and tissue tests, and N-P-K fertilization showed smaller effects. Gaussian processes predicted yields from historical weather data, soil test, fertilizer dosage, and tissue test with a root-mean-square-error of 1447 kg ha−1. An in-house Markov chain algorithm optimized yields modelled by Gaussian processes from tissue test, soil test, and fertilizer dosage as conditioned to specified historical meteorological features, potentially increasing yield by a median factor of 1.5. Machine learning, compositional data analysis, and Markov chains allowed customizing nutrient management of lowbush blueberry at local scale.

Highlights

Lowbush blueberry species (Vaccinium angustifolium Ait. and, to some extent, V. myrtilloides Michx.)are North American wild ericaceous species growing in upland acid sandy soils
Berry yields from experimental plots ranged between 0.6 and 13.8 Mg ha−1 in our data set and were more dispersed than lowbush blueberry yields published in other studies conducted in Maine, Québec, the Canadian Atlantic provinces, and Estonia (Figure 4)
The first Bayesian linear regression with a gaussian response investigated the effects of leaf nutrients, soil nutrients, soil pH, NPK dosage and seasonal weather indices over 2-years on yields of lowbush blueberry

Summary

Introduction

Lowbush blueberry species (Vaccinium angustifolium Ait. and, to some extent, V. myrtilloides Michx.)are North American wild ericaceous species growing in upland acid sandy soils. Lowbush blueberry species (Vaccinium angustifolium Ait. and, to some extent, V. myrtilloides Michx.). The province of Québec, Canada, is among the world leaders in the production of lowbush blueberry [1]. Berry yields vary widely between 0.6 [2] and 8.9 Mg ha−1 [3], indicating high risk of production failure. Lowbush blueberry is managed over 2-year cycles where vegetative (or pruning) and fruit-bearing (or fruit-harvesting) years alternate. Flower bud initiation occurs during the vegetative year and impacts on crop productivity. Plants 2020, 9, 1401 during the fruit-bearing year [4]. Fruit set depends on the number of flowers, pollination success, edaphic and managerial conditions, year, and clone [5], as well as nesting habitats of pollinators [6]

Objectives

Methods

Results

Discussion

Conclusion