California Almond Yield Prediction at the Orchard Level With a Machine Learning Approach.

Zhou Zhang,Patrick Brown,Bin Chen,Yufang Jin

doi:10.3389/fpls.2019.00809

Abstract

California’s almond growers face challenges with nitrogen management as new legislatively mandated nitrogen management strategies for almond have been implemented. These regulations require that growers apply nitrogen to meet, but not exceed, the annual N demand for crop and tree growth and nut production. To accurately predict seasonal nitrogen demand, therefore, growers need to estimate block-level almond yield early in the growing season so that timely N management decisions can be made. However, methods to predict almond yield are not currently available. To fill this gap, we have developed statistical models using the Stochastic Gradient Boosting, a machine learning approach, for early season yield projection and mid-season yield update over individual orchard blocks. We collected yield records of 185 orchards, dating back to 2005, from the major almond growers in the Central Valley of California. A large set of variables were extracted as predictors, including weather and orchard characteristics from remote sensing imagery. Our results showed that the predicted orchard-level yield agreed well with the independent yield records. For both the early season (March) and mid-season (June) predictions, a coefficient of determination (R2) of 0.71, and a ratio of performance to interquartile distance (RPIQ) of 2.6 were found on average. We also identified several key determinants of yield based on the modeling results. Almond yield increased dramatically with the orchard age until about 7 years old in general, and the higher long-term mean maximum temperature during April–June enhanced the yield in the southern orchards, while a larger amount of precipitation in March reduced the yield, especially in northern orchards. Remote sensing metrics such as annual maximum vegetation indices were also dominant variables for predicting the yield potential. While these results are promising, further refinement is needed; the availability of larger data sets and incorporation of additional variables and methodologies will be required for the model to be used as a fertilization decision support tool for growers. Our study has demonstrated the potential of automatic almond yield prediction to assist growers to manage N adaptively, comply with mandated requirements, and ensure industry sustainability.

Highlights

World production of almond was 2.2 million tons in 2017, and the leading producers include United States, Australia, Spain, Iran and Italy (Food and Agriculture Organization of the United Nations [FAO], 2018)
We focused on the almond orchards located in the Central Valley of California, where we have collected the historical yield and other ancillary data from eight growers managing a range of orchards of different ages (Figure 1A)
We evaluated the model performance, by quantifying the following metrics based on the testing data for each round, including (1) the root mean squared error (RMSE), (2) the coefficient of determination (R2), and (3) the ratio of performance to interquartile distance (RPIQ), which is defined as interquartile range of the observed values divided by the RMSE (Maurel et al, 2010)

Summary

Introduction

World production of almond was 2.2 million tons in 2017, and the leading producers include United States, Australia, Spain, Iran and Italy (Food and Agriculture Organization of the United Nations [FAO], 2018). To optimize N management and ensure regulatory compliance, growers must apply N in accordance with predicted yield in each production unit, taking into account N available from all sources (fertilizer, composts and manures, irrigation water nitrogen). Yield prediction is critical for N management It can help growers make plans for the harvest, processing, and transport of the crop (ZarateValdez et al, 2015). Forecasting inter-year yield variation plays a key role in food security monitoring and market planning, and has the potential to help managing food production shocks (Iizumi et al, 2018). Estimating crop yield, has broad implications for ecology, economics, and human society, e.g., through its impact on the optimal use of inputs (irrigation water, fertilizers) and other resources (machinery, labor) on the farm (Carletto et al, 2015; Hoffman et al, 2015)

Objectives

Methods

Results

Conclusion