Dataset Information

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

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 (R 2) 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.

SUBMITTER: Zhang Z

PROVIDER: S-EPMC6656960 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Zhang Zhou Z Jin Yufang Y Chen Bin B Brown Patrick P

Frontiers in plant science 20190718

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. Howe ...[more]

PMID: 31379888

Similar Datasets

Project description:Agricultural productivity is subject to various stressors, including abiotic and biotic threats, many of which are exacerbated by a changing climate, thereby affecting long-term sustainability. The productivity of tree crops such as almond orchards, is particularly complex. To understand and mitigate these threats requires a collection of multi-layer large data sets, and advanced analytics is also critical to integrate these highly heterogeneous datasets to generate insights about the key constraints on the yields at tree and field scales. Here we used a machine learning approach to investigate the determinants of almond yield variation in California's almond orchards, based on a unique 10-year dataset of field measurements of light interception and almond yield along with meteorological data. We found that overall the maximum almond yield was highly dependent on light interception, e.g., with each one percent increase in light interception resulting in an increase of 57.9 lbs/acre in the potential yield. Light interception was highest for mature sites with higher long term mean spring incoming solar radiation (SRAD), and lowest for younger orchards when March maximum temperature was lower than 19°C. However, at any given level of light interception, actual yield often falls significantly below full yield potential, driven mostly by tree age, temperature profiles in June and winter, summer mean daily maximum vapor pressure deficit (VPDmax), and SRAD. Utilizing a full random forest model, 82% (±1%) of yield variation could be explained when using a sixfold cross validation, with a RMSE of 480 ± 9 lbs/acre. When excluding light interception from the predictors, overall orchard characteristics (such as age, location, and tree density) and inclusive meteorological variables could still explain 78% of yield variation. The model analysis also showed that warmer winter conditions often limited mature orchards from reaching maximum yield potential and summer VPDmax beyond 40 hPa significantly limited the yield. Our findings through the machine learning approach improved our understanding of the complex interaction between climate, canopy light interception, and almond nut production, and demonstrated a relatively robust predictability of almond yield. This will ultimately benefit data-driven climate adaptation and orchard nutrient management approaches.

Project description:Concerns over the availability of honeybees (Apis mellifera L.) to meet pollination demands have elicited interest in alternative pollinators to mitigate pressures on the commercial beekeeping industry. The blue orchard bee, Osmia lignaria (Say), is a commercially available native bee that can be employed as a copollinator with, or alternative pollinator to, honeybees in orchards. To date, their successful implementation in agriculture has been limited by poor recovery of bee progeny for use during the next spring. This lack of reproductive success may be tied to an inadequate diversity and abundance of alternative floral resources during the foraging period. Managed, supplementary wildflower plantings may promote O. lignaria reproduction in California almond orchards. Three wildflower plantings were installed and maintained along orchard edges to supplement bee forage. Plantings were seeded with native wildflower species that overlapped with and extended beyond almond bloom. We measured bee visitation to planted wildflowers, bee reproduction, and progeny outcomes across orchard blocks at variable distances from wildflower plantings during 2015 and 2016. Pollen provision composition was also determined to confirm O. lignaria wildflower pollen use. Osmia lignaria were frequently observed visiting wildflower plantings during, and after, almond bloom. Most O. lignaria nesting occurred at orchard edges. The greatest recovery of progeny occurred along the orchard edges having the closest proximity (80 m) to managed wildflower plantings versus edges farther away. After almond bloom, O. lignaria nesting closest to the wildflower plantings collected 72% of their pollen from Phacelia spp., which supplied 96% of the managed floral area. Phacelia spp. pollen collection declined with distance from the plantings, but still reached 17% 800 m into the orchard. This study highlights the importance of landscape context and proximity to supplementary floral resources in promoting the propagation of solitary bees as alternative managed pollinators in commercial agriculture.

Project description:Nitrogen is the most limiting nutrient for turfgrass growth. Instead of pursuing the maximum yield, most turfgrass managers use nitrogen (N) to maintain a sub-maximal growth rate. Few tools or soil tests exist to help managers guide N fertilizer decisions. Turf growth prediction models have the potential to be useful, but the currently existing turf growth prediction model only takes temperature into account, limiting its accuracy. This study developed machine-learning-based turf growth models using the random forest (RF) algorithm to estimate short-term turfgrass clipping yield. To build the RF model, a large set of variables were extracted as predictors including the 7-day weather, traffic intensity, soil moisture content, N fertilization rate, and the normalized difference red edge (NDRE) vegetation index. In this study, the data were collected from two putting greens where the turfgrass received 0 to 1,800 round/week traffic rates, various irrigation rates to maintain the soil moisture content between 9 and 29%, and N fertilization rates of 0 to 17.5 kg ha-1 applied biweekly. The RF model agreed with the actual clipping yield collected from the experimental results. The temperature and relative humidity were the most important weather factors. Including NDRE improved the prediction accuracy of the model. The highest coefficient of determination (R2) of the RF model was 0.64 for the training dataset and was 0.47 for the testing data set upon the evaluation of the model. This represented a large improvement over the existing growth prediction model (R 2 = 0.01). However, the machine-learning models created were not able to accurately predict the clipping production at other locations. Individual golf courses can create customized growth prediction models using clipping volume to eliminate the deviation caused by temporal and spatial variability. Overall, this study demonstrated the feasibility of creating machine-learning-based yield prediction models that may be able to guide N fertilization decisions on golf course putting greens and presumably other turfgrass areas.

Project description:BackgroundThe objective of this study is twofold. First, ascertain the important variables that predict tomato yields from plant height (PH) and vegetation index (VI) maps. The maps were derived from images taken by unmanned aerial vehicles (UAVs). Second, examine the accuracy of predictions of tomato fresh shoot masses (SM), fruit weights (FW), and the number of fruits (FN) from multiple machine learning algorithms using selected variable sets. To realize our objective, ultra-high-resolution RGB and multispectral images were collected by a UAV on ten days in 2020's tomato growing season. From these images, 756 total variables, including first- (e.g., average, standard deviation, skewness, range, and maximum) and second-order (e.g., gray-level co-occurrence matrix features and growth rates of PH and VIs) statistics for each plant, were extracted. Several selection algorithms (i.e., Boruta, DALEX, genetic algorithm, least absolute shrinkage and selection operator, and recursive feature elimination) were used to select the variable sets useful for predicting SM, FW, and FN. Random forests, ridge regressions, and support vector machines were used to predict the yield using the top five selected variable sets.ResultsFirst-order statistics of PH and VIs collected during the early to mid-fruit formation periods, about one month prior to harvest, were important variables for predicting SM. Similar to the case for SM, variables collected approximately one month prior to harvest were important for predicting FW and FN. Furthermore, variables related to PH were unimportant for prediction. Compared with predictions obtained using only first-order statistics, those obtained using the second-order statistics of VIs were more accurate for FW and FN. The prediction accuracy of SM, FW, and FN by models constructed from all variables (rRMSE = 8.8-28.1%) was better than that from first-order statistics (rRMSE = 10.0-50.1%).ConclusionsIn addition to basic statistics (e.g., average and standard deviation), we derived second-order statistics of PH and VIs at the plant level using the ultra-high resolution UAV images. Our findings indicated that our variable selection method reduced the number variables needed for tomato yield prediction, improving the efficiency of phenotypic data collection and assisting with the selection of high-yield lines within breeding programs.

Project description:BackgroundVirtually all currently available microRNA target site prediction algorithms require the presence of a (conserved) seed match to the 5' end of the microRNA. Recently however, it has been shown that this requirement might be too stringent, leading to a substantial number of missed target sites.ResultsWe developed TargetSpy, a novel computational approach for predicting target sites regardless of the presence of a seed match. It is based on machine learning and automatic feature selection using a wide spectrum of compositional, structural, and base pairing features covering current biological knowledge. Our model does not rely on evolutionary conservation, which allows the detection of species-specific interactions and makes TargetSpy suitable for analyzing unconserved genomic sequences.In order to allow for an unbiased comparison of TargetSpy to other methods, we classified all algorithms into three groups: I) no seed match requirement, II) seed match requirement, and III) conserved seed match requirement. TargetSpy predictions for classes II and III are generated by appropriate postfiltering. On a human dataset revealing fold-change in protein production for five selected microRNAs our method shows superior performance in all classes. In Drosophila melanogaster not only our class II and III predictions are on par with other algorithms, but notably the class I (no-seed) predictions are just marginally less accurate. We estimate that TargetSpy predicts between 26 and 112 functional target sites without a seed match per microRNA that are missed by all other currently available algorithms.ConclusionOnly a few algorithms can predict target sites without demanding a seed match and TargetSpy demonstrates a substantial improvement in prediction accuracy in that class. Furthermore, when conservation and the presence of a seed match are required, the performance is comparable with state-of-the-art algorithms. TargetSpy was trained on mouse and performs well in human and drosophila, suggesting that it may be applicable to a broad range of species. Moreover, we have demonstrated that the application of machine learning techniques in combination with upcoming deep sequencing data results in a powerful microRNA target site prediction tool http://www.targetspy.org.

Project description:Background and purposePatients with ischemic stroke frequently develop hemorrhagic transformation (HT), which could potentially worsen the prognosis. The objectives of the current study were to determine the incidence and predictors of HT, to evaluate predictor interaction, and to identify the optimal predicting models.MethodsA prospective study included 360 patients with ischemic stroke, of whom 354 successfully continued the study. Patients were subjected to thorough general and neurological examination and T2 diffusion-weighted MRI, at admission and 1 week later to determine the incidence of HT. HT predictors were selected by a filter-based minimum redundancy maximum relevance (mRMR) algorithm independent of model performance. Several machine learning algorithms including multivariable logistic regression classifier (LRC), support vector classifier (SVC), random forest classifier (RFC), gradient boosting classifier (GBC), and multilayer perceptron classifier (MLPC) were optimized for HT prediction in a randomly selected half of the sample (training set) and tested in the other half of the sample (testing set). The model predictive performance was evaluated using receiver operator characteristic (ROC) and visualized by observing case distribution relative to the models' predicted three-dimensional (3D) hypothesis spaces within the testing dataset true feature space. The interaction between predictors was investigated using generalized additive modeling (GAM).ResultsThe incidence of HT in patients with ischemic stroke was 19.8%. Infarction size, cerebral microbleeds (CMB), and the National Institute of Health stroke scale (NIHSS) were identified as the best HT predictors. RFC (AUC: 0.91, 95% CI: 0.85-0.95) and GBC (AUC: 0.91, 95% CI: 0.86-0.95) demonstrated significantly superior performance compared to LRC (AUC: 0.85, 95% CI: 0.79-0.91) and MLPC (AUC: 0.85, 95% CI: 0.78-0.92). SVC (AUC: 0.90, 95% CI: 0.85-0.94) outperformed LRC and MLPC but did not reach statistical significance. LRC and MLPC did not show significant differences. The best models' 3D hypothesis spaces demonstrated non-linear decision boundaries suggesting an interaction between predictor variables. GAM analysis demonstrated a linear and non-linear significant interaction between NIHSS and CMB and between NIHSS and infarction size, respectively.ConclusionCerebral microbleeds, NIHSS, and infarction size were identified as HT predictors. The best predicting models were RFC and GBC capable of capturing nonlinear interaction between predictors. Predictor interaction suggests a dynamic, rather than, fixed cutoff risk value for any of these predictors.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets