Correction: Determining soil particle-size distribution from infrared spectra using machine learning predictions: Methodology and modeling.

[This corrects the article DOI: 10.1371/journal.pone.0233242.].

Determining soil particle-size distribution from infrared spectra using machine learning predictions: Methodology and modeling.

Accuracy of infrared (IR) models to measure soil particle-size distribution (PSD) depends on soil preparation, methodology (sedimentation, laser), settling times and relevant soil features. Compositional soil data may require log ratio (ilr) transformation to avoid numerical biases. Machine learning can relate numerous independent variables that may impact on NIR spectra to assess particle-size distribution. Our objective was to reach high IRS prediction accuracy across a large range of PSD methods and soil properties. A total of 1298 soil samples from eastern Canada were IR-scanned. Spectra were processed by Stochastic Gradient Boosting (SGB) to predict sand, silt, clay and carbon. Slope and intercept of the log-log relationships between settling time and suspension density function (SDF) (R2 = 0.84-0.92) performed similarly to NIR spectra using either ilr-transformed (R2 = 0.81-0.93) or raw percentages (R2 = 0.76-0.94). Settling times of 0.67-min and 2-h were the most accurate for NIR predictions (R2 = 0.49-0.79). The NIR prediction of sand sieving method (R2 = 0.66) was more accurate than sedimentation method(R2 = 0.53). The NIR 2X gain was less accurate (R2 = 0.69-0.92) than 4X (R2 = 0.87-0.95). The MIR (R2 = 0.45-0.80) performed better than NIR (R2 = 0.40-0.71) spectra. Adding soil carbon, reconstituted bulk density, pH, red-green-blue color, oxalate and Mehlich3 extracts returned R2 value of 0.86-0.91 for texture prediction. In addition to slope and intercept of the SDF, 4X gain, method and pre-treatment classes, soil carbon and color appeared to be promising features for routine SGB-processed NIR particle-size analysis. Machine learning methods support cost-effective soil texture NIR analysis.

Current and next-year cranberry yields predicted from local features and carryover effects.

Wisconsin and Quebec are the world leading cranberry-producing regions. Cranberries are grown in acidic, naturally low-fertility sandy beds. Cranberry fertilization is guided by general soil and tissue nutrient tests in addition to yield target and vegetative biomass. However, other factors such as cultivar, location, and carbon and nutrient storage impact cranberry nutrition and yield. The objective of this study was to customize nutrient diagnosis and fertilizer recommendation at local scale and for next-year cranberry production after accounting for local factors and carbon and nutrient carryover effects. We collected 1768 observations from on-farm surveys and fertilizer trials in Quebec and Wisconsin to elaborate a machine learning model using minimum datasets. We tested carryover effects in a 5-year Quebec fertilizer experiment established on permanent plots. Micronutrients contributed more than macronutrients to variation in tissue compositions. Random Forest model related accurately current-year berry yield to location, cultivars, climatic indices, fertilization, and tissue and soil tests as features (classification accuracy of 0.83). Comparing compositions of defective and successful tissue compositions in the Euclidean space of tissue compositions, the general across-factor diagnosis differed from the local factor-specific diagnosis. Nutrient standards elaborated in one region could hardly be transposed to another and, within the same region, from one bed to another due to site-specific characteristics. Next-year yield and nutrient adjustment could be predicted accurately from current-year yield and tissue composition and other features, with R2 value of 0.73 in regression mode and classification accuracy of 0.85. Compositional and machine learning methods proved to be effective to customize nutrient diagnosis and predict site-specific measures for nutrient management of cranberry stands. This study emphasized the need to acquire large experimental and observational datasets to capture the numerous factor combinations impacting current and next-year cranberry yields at local scale.

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

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.

Site-specific machine learning predictive fertilization models for potato crops in Eastern Canada.

Statistical modeling is commonly used to relate the performance of potato (Solanum tuberosum L.) to fertilizer requirements. Prescribing optimal nutrient doses is challenging because of the involvement of many variables including weather, soils, land management, genotypes, and severity of pests and diseases. Where sufficient data are available, machine learning algorithms can be used to predict crop performance. The objective of this study was to determine an optimal model predicting nitrogen, phosphorus and potassium requirements for high tuber yield and quality (size and specific gravity) as impacted by weather, soils and land management variables. We exploited a data set of 273 field experiments conducted from 1979 to 2017 in Quebec (Canada). We developed, evaluated and compared predictions from a hierarchical Mitscherlich model, k-nearest neighbors, random forest, neural networks and Gaussian processes. Machine learning models returned R2 values of 0.49-0.59 for tuber marketable yield prediction, which were higher than the Mitscherlich model R2 (0.37). The models were more likely to predict medium-size tubers (R2 = 0.60-0.69) and tuber specific gravity (R2 = 0.58-0.67) than large-size tubers (R2 = 0.55-0.64) and marketable yield. Response surfaces from the Mitscherlich model, neural networks and Gaussian processes returned smooth responses that agreed more with actual evidence than discontinuous curves derived from k-nearest neighbors and random forest models. When conditioned to obtain optimal dosages from dose-response surfaces given constant weather, soil and land management conditions, some disagreements occurred between models. Due to their built-in ability to develop recommendations within a probabilistic risk-assessment framework, Gaussian processes stood out as the most promising algorithm to support decisions that minimize economic or agronomic risks.

Cultivar-specific nutritional status of potato (Solanum tuberosum L.) crops.

Gradients in the elemental composition of a potato leaf tissue (i.e. its ionome) can be linked to crop potential. Because the ionome is a function of genetics and environmental conditions, practitioners aim at fine-tuning fertilization to obtain an optimal ionome based on the needs of potato cultivars. Our objective was to assess the validity of cultivar grouping and predict potato tuber yields using foliar ionomes. The dataset comprised 3382 observations in Québec (Canada) from 1970 to 2017. The first mature leaves from top were sampled at the beginning of flowering for total N, P, K, Ca, and Mg analysis. We preprocessed nutrient concentrations (ionomes) by centering each nutrient to the geometric mean of all nutrients and to a filling value, a transformation known as row-centered log ratios (clr). A density-based clustering algorithm (dbscan) on these preprocessed ionomes failed to delineate groups of high-yield cultivars. We also used the preprocessed ionomes to assess their effects on tuber yield classes (high- and low-yields) on a cultivar basis using k-nearest neighbors, random forest and support vector machines classification algorithms. Our machine learning models returned an average accuracy of 70%, a fair diagnostic potential to detect in-season nutrient imbalance of potato cultivars using clr variables considering potential confounding factors. Optimal ionomic regions of new cultivars could be assigned to the one of the closest documented cultivar.

The Ionomics of Lettuce Infected by Xanthomonas campestris pv. vitians.

Bacterial leaf spot (BLS) caused by Xanthomonas campestris pv. vitians (Xcv) places a major constraint on lettuce production worldwide. The most sustainable strategy known to date for controlling BLS is the use of resistant cultivars. The nutrient elemental signature (ionome) of ten lettuce cultivars with three levels of resistance was analyzed by inductively coupled plasma optical emission spectroscopy (ICP-OES) to determine which nutrient balances are linked to resistance to BLS, and to assess the effect of Xcv infection on the ionome. The elemental concentrations were preprocessed with isometric log-ratios to define nutrient balances. Using this approach, 4 out of 11 univariate nutrient balances were found to significantly influence the resistance of lettuce cultivars to BLS (P < 0.05). These significant balances were the overall nutritional status balancing all measured nutrients with their complementary in the dry mass, as well as balances [Mn | Zn,Cu], [Zn | Cu], and [S,N | P]. Moreover, the infection of lettuce cultivars mostly affected the lettuce ionome on the [N,S | P] balance, where infection tended to lean the balance toward the N,S part relatively to P. This study shows that nutrient uptake in lettuce can be affected by BLS infection and that nutrient status influences resistance to BLS infection.

Using a soil bacterial species balance index to estimate potato crop productivity.

The development of 'molecular-omic' tools and computing analysis platforms have greatly enhanced our ability to assess the impacts of agricultural practices and crop management protocols on soil microbial diversity. However, biotic factors are rarely factored into agricultural management models. Today it is possible to identify specific microbiomes and define biotic components that contribute to soil quality. We assessed the bacterial diversity of soils in 51 potato production plots. We describe a strategy for identifying a potato-crop-productivity bacterial species balance index based on amplicon sequence variants. We observed a significant impact of soil texture balances on potato yields; however, the Shannon and Chao1 richness indices and Pielou's evenness index poorly correlated with these yields. Nonetheless, we were able to estimate the portion of the total bacterial microbiome related to potato yield using an integrated species balances index derived from the elements of the bacterial microbiome that positively or negatively correlate with residual potato yields. This innovative strategy based on a microbiome selection procedure greatly enhances our ability to interpret the impact of agricultural practices and cropping system management choices on microbial diversity and potato yield. This strategy provides an additional tool that will aid growers and the broader agricultural sector in their decision-making processes concerning the soil quality and crop productivity.

Balance design for robust foliar nutrient diagnosis of "Prata" banana (Musa spp.).

The "Cavendish" and "Prata" subgroups represent respectively 47% and 24% of the world banana production. Compared to world average progressing from 10.6 to 20.6 t ha-1 between 1961 and 2016, and despite sustained domestic demand and the introduction of new cultivars, banana yield in Brazil has stagnated around 14.5 t ha-1 mainly due to nutrient and water mismanagement. "Prata" is now the dominant subgroup in N-E Brazil and is fertigated at high costs. Nutrient balances computed as isometric log-ratios (ilr) provide a comprehensive understanding of nutrient relationships in the diagnostic leaf at high yield level by combining raw concentration data. Although the most appropriate method for multivariate analysis of compositional balances may be less efficient due to non-normal data distribution and limited nutrient mobility in the plant, robustness of the nutrient balance approach could be improved using Box-Cox exponents assigned to raw foliar concentrations. Our objective was to evaluate the accuracy of nutrient balances to diagnose fertigated "Prata" orchards. The dataset comprised 609 observations on fruit yields and leaf tissue compositions collected from 2010 to 2016 in Ceará state, N-E Brazil. Raw nutrient concentration ranges were ineffective as diagnostic tool due to considerable overlapping of concentration ranges for low- and high-yielding subpopulations at cutoff yield of 40 Mg ha-1. Nutrient concentrations were combined into isometric log-ratios (ilr) and normalized by Box-Cox corrections between 0 and 1 which may also account for restricted nutrient transfer from leaf to fruit. Despite reduced ilr skewness, Box-Cox coefficients did not improve model robustness measured as the accuracy of the Cate-Nelson partition between yield and the multivariate distance across ilr values. Sensitivity was 94%, indicating that low yields are attributable primarily to nutrient imbalance. There were 148 false-positive specimens (high yield despite nutrient imbalance) likely due to suboptimal nutrition, contamination, or luxury consumption. The profitability of "Prata" orchards could be enhanced by rebalancing nutrients using ilr standards with no need for Box-Cox correction.

N-P Fertilization Inhibits Growth of Root Hemiparasite Pedicularis kansuensis in Natural Grassland.

Fertilization has been shown to affect interactions between root hemiparasitic plants and their host plants, alleviating damage to the hosts by parasitism. However, as a majority of studies were conducted in pot cultivation, the influence of fertilizer application on root hemiparasites and the surrounding plant community in field conditions as well as relevant mechanisms remain unclear. We manipulated soil nutrient resources in a semi-arid subalpine grassland in the Tianshan Mountains, northwestern China, to explore the links between fertilization and plant community composition, productivity, survival, and growth of a weedy root hemiparasite (Pedicularis kansuensis). Nitrogen (at a low rate, LN, 30 kg N ha-1 year-1 as urea; or at a high rate, HN, 90 kg N ha-1 year-1 as urea) and phosphorus [100 kg ha-1 year-1 as Ca(H2PO4)2⋅H2O] were added during two growing seasons. Patterns of foliar nutrient balances were described with isometric log ratios for the different plant functional groups receiving these fertilization regimes. Fertilization with LN, HN, and P reduced above-ground biomass of P. kansuensis, with above-ground biomass in the fertilization treatments, respectively, 12, 1, and 39% of the value found in the unfertilized control. Up to three times more above-ground biomass was produced in graminoids receiving fertilizers, whereas forb above-ground biomass was virtually unchanged by the fertilization regimes and forb species richness was reduced by 52% in the HN treatment. Fertilization altered foliar nutrient balances, and distinct patterns emerged for each plant functional group. Foliar [C | P,N] balance in the plant community was negatively correlated with above-ground biomass (P = 0.03). The inhibited competitiveness of P. kansuensis, which showed a much higher [C | P,N] balance, could be attributed to reduced C assimilation rather than mineral nutrient acquisition, as shown by significant increase in foliar N and P concentrations but little increase in C concentration following fertilization.

Phosphorus Over-Fertilization and Nutrient Misbalance of Irrigated Tomato Crops in Brazil.

Over the past 20 years, the use of center-pivot irrigation has increased tomato (Solanum lycopersicum L.) yields in Brazil from 42 Mg ha-1 to more than 80 Mg ha-1. In the absence of field trials to support fertilizer recommendations, substantial amounts of phosphorus (P) have been applied to crops. Additional P dosing has been based on an equilibrated nutrient P budget adjusted for low-P fertilizer-use efficiency in high-P fixing tropical soils. To document nutrient requirements and prevent over-fertilization, tissue samples and crop yield data can be acquired through crop surveys and fertilizer trials. Nevertheless, most tissue diagnostic methods pose numerical difficulties that can be avoided by using the nutrient balance concept. The objectives of this study were to model the response of irrigated tomato crops to P fertilization in low- and high-P soils and to provide tissue diagnostic models for high crop yield. Three P trials, arranged in a randomized block design with six P treatments (0-437 kg P ha-1) and three or four replications, were established on a low-P soil in 2013 and high-P soils in 2013 and 2014, totaling 66 plots in all. Together with crop yield data, 65 tissue samples were collected from tomato farms. We found no significant yield response to P fertilization, despite large differences in soil-test P (coefficient of variation, 24%). High- and low-yield classes (cutoff: 91 Mg fruits ha-1) were classified by balance models with 78-81% accuracy using logit and Cate-Nelson partitioning models. The critical Mahalanobis distance for the partition was 5.31. Tomato yields were apparently not limited by P but were limited by calcium. There was no evidence that P fertilization should differ between center-pivot-irrigated and rain-fed crops. Use of the P budget method to arrive at the P requirement for tomato crops proved to be fallacious, as several nutrients should be rebalanced in Brazilian tomato cropping systems.

Guava Waste to Sustain Guava (Psidium guajava) Agroecosystem: Nutrient "Balance" Concepts.

The Brazilian guava processing industry generates 5.5 M Mg guava waste year(-1) that could be recycled sustainably in guava agro-ecosystems as slow-release fertilizer. Our objectives were to elaborate nutrient budgets and to diagnose soil, foliar, and fruit nutrient balances in guava orchards fertilized with guava waste. We hypothesized that (1) guava waste are balanced fertilizer sources that can sustain crop yield and soil nutrient stocks, and (2) guava agroecosystems remain productive within narrow ranges of nutrient balances. A 6-year experiment was conducted in 8-year old guava orchard applying 0-9-18-27-36 Mg ha(-1) guava waste (dry mass basis) and the locally recommended mineral fertilization. Nutrient budgets were compiled as balance sheets. Foliar and fruit nutrient balances were computed as isometric log ratios to avoid data redundancy or resonance due to nutrient interactions and the closure to measurement unit. The N, P, and several other nutrients were applied in excess of crop removal while K was in deficit whatever the guava waste treatment. The foliar diagnostic accuracy reached 93% using isometric log ratios and knn classification, generating reliable foliar nutrient and concentration ranges at high yield level. The plant mined the soil K reserves without any significant effect on fruit yield and foliar nutrient balances involving K. High guava productivity can be reached at lower soil test K and P values than thought before. Parsimonious dosage of fresh guava waste should be supplemented with mineral K fertilizers to recycle guava waste sustainably in guava agroecosystems. Brazilian growers can benefit from this research by lowering soil test P and K threshold values to avoid over-fertilization and using fresh guava waste supplemented with mineral fertilizers, especially K. Because yield was negatively correlated with fruit acidity and Brix index, balanced plant nutrition and fertilization diagnosis will have to consider not only fruit yield targets but also fruit quality to meet requirements for guava processing.

Plant ionome diagnosis using sound balances: case study with mango (Mangifera Indica).

Plant ionomes and soil nutrients are commonly diagnosed in agronomy using concentration and nutrient ratio ranges. However, both diagnoses are biased by redundancy of information, subcompositional incoherence and non-normal distribution inherent to compositional data, potentially leading to conflicting results and wrong inferences. Our objective was to present an unbiased statistical approach of plant nutrient diagnosis using a balance concept and mango (Mangifera indica) as test crop. We collected foliar samples at flowering stage in 175 mango orchards. The ionomes comprised 11 nutrients (S, N, P, K, Ca, Mg, B, Cu, Zn, Mn, Fe). Traditional multivariate methods were found to be biased. Ionomes were thus represented by unbiased balances computed as isometric log ratios (ilr). Soil fertility attributes (pH and bioavailable nutrients) were transformed into balances to conduct discriminant analysis. The orchards differed more from genotype than soil nutrient signatures. A customized receiver operating characteristic (ROC) iterative procedure was developed to classify tissue ionomes between balanced/misbalanced and high/low-yielders. The ROC partitioning procedure showed that the critical Mahalanobis distance of 4.08 separating balanced from imbalanced specimens about yield cut-off of 128.5 kg fruit tree(-1) proved to be a fairly informative test (area under curve = 0.84-0.92). The [P | N,S] and [Mn | Cu,Zn] balances were found to be potential sources of misbalance in the less productive orchards, and should thus be further investigated in field experiments. We propose using a coherent pan balance diagnostic method with median ilr values of top yielders centered at fulcrums of a mobile and the critical Mahalanobis distance as a guide for global nutrient balance. Nutrient concentrations in weighing pans assisted appreciating nutrients as relative shortage, adequacy or excess in balances.

The plant ionome revisited by the nutrient balance concept.

Tissue analysis is commonly used in ecology and agronomy to portray plant nutrient signatures. Nutrient concentration data, or ionomes, belongs to the compositional data class, i.e., multivariate data that are proportions of some whole, hence carrying important numerical properties. Statistics computed across raw or ordinary log-transformed nutrient data are intrinsically biased, hence possibly leading to wrong inferences. Our objective was to present a sound and robust approach based on a novel nutrient balance concept to classify plant ionomes. We analyzed leaf N, P, K, Ca, and Mg of two wild and six domesticated fruit species from Canada, Brazil, and New Zealand sampled during reproductive stages. Nutrient concentrations were (1) analyzed without transformation, (2) ordinary log-transformed as commonly but incorrectly applied in practice, (3) additive log-ratio (alr) transformed as surrogate to stoichiometric rules, and (4) converted to isometric log-ratios (ilr) arranged as sound nutrient balance variables. Raw concentration and ordinary log transformation both led to biased multivariate analysis due to redundancy between interacting nutrients. The alr- and ilr-transformed data provided unbiased discriminant analyses of plant ionomes, where wild and domesticated species formed distinct groups and the ionomes of species and cultivars were differentiated without numerical bias. The ilr nutrient balance concept is preferable to alr, because the ilr technique projects the most important interactions between nutrients into a convenient Euclidean space. This novel numerical approach allows rectifying historical biases and supervising phenotypic plasticity in plant nutrition studies.