An olive variety screening decision-making method and system based on photosynthesis data modeling

Abstract

The application provides an olive variety screening decision-making method and system based on photosynthetic data modeling, relates to the technical field of variety screening, and comprises the following steps: obtaining multi-scale photosynthetic physiological data and fruit metabolism data of target variety fruits in a fruit development cycle; obtaining a coupling feature tensor by calculating photosynthetic product distribution coefficients and oil conversion efficiency, extracting seasonal variation coefficients and midday photosynthetic efficiency features; establishing a response relationship between photosynthetic contribution weights of different parts of a tree crown and photosynthetic parameters and leaf water state to obtain an adaptability model; constructing a causal chain model to predict oil output to obtain an oil output prediction value of the variety and an uncertainty interval thereof; and combining light, temperature and water conditions of a target planting area to analyze and obtain a recommended variety ranking list with risk assessment. The application comprehensively considers photosynthetic physiology, metabolism, crown structure and environmental adaptability, and improves the scientificity and accuracy of olive variety screening.

Description

Technical Field

[0001] This invention relates to the field of variety selection technology, and more specifically, to a method and system for olive variety selection decision-making based on photosynthetic data modeling. Background Technology

[0002] Olive is an important woody oilseed crop worldwide, and its economic output mainly depends on the oil content of the fruit and the yield per tree / unit area. The seasonal accumulation of oil content is essentially determined by the tree's photosynthetic carbon fixation capacity, the distribution strategy of photosynthetic products between source and sink, and the efficiency of the conversion of carbon assimilates into oil within the fruit. Meanwhile, olive-growing areas often face strong seasonal fluctuations in light and temperature and periodic water stress (such as spring and summer droughts and uneven irrigation). Different varieties exhibit significant differences in canopy light distribution, leaf water relationships, and intraday "midday depression" responses to high temperatures and strong light. This means that the same photosynthetic "readings" do not equate to the same oil production and yield stability. Therefore, variety selection based solely on external agronomic traits or multi-year average yield statistics often masks key physiological differences and makes it difficult to identify the risks of varieties that are "high-yielding but unstable" or "stable but not high-yielding" in advance.

[0003] Currently, existing olive variety selection techniques mainly rely on field yield observations and experience-based judgment. Growers assess a variety's adaptability by observing its growth performance and fruit yield. However, this traditional method has the following drawbacks: First, it is time-consuming, requiring three to five years of observation to draw conclusions, resulting in low efficiency and high costs; second, it is highly subjective, with inconsistent judgment standards among different observers, easily leading to misjudgments; third, it lacks quantitative indicators, making it difficult to accurately assess the yield potential of a variety; and fourth, it does not consider the influence of environmental factors, making it difficult to accurately assess the performance of varieties under different environmental conditions.

[0004] Therefore, there is an urgent need for a method and system for selecting olive varieties based on photosynthetic data modeling to solve the above-mentioned technical problems. Summary of the Invention

[0005] The purpose of this invention is to provide a method and system for olive variety selection based on photosynthetic data modeling, in order to improve the aforementioned problems. To achieve the above objective, the technical solution adopted by this invention is as follows: Firstly, this application provides a method for olive variety selection decision-making based on photosynthetic data modeling, including: Acquire multi-scale photosynthetic physiological data and fruit metabolic data of the target variety during the fruit development cycle. The multi-scale photosynthetic physiological data includes intra-diurnal net photosynthetic rate, stomatal conductance, chlorophyll fluorescence parameters, seasonal light response curves and environmental parameters. The fruit metabolic data includes carbon assimilate content and oil accumulation rate during the ripening period. Based on the multi-scale photosynthetic physiological data and the fruit metabolic data, photosynthetic-metabolic coupling features and seasonal fluctuation features are jointly extracted. Specifically, by calculating the photosynthetic product allocation coefficient and oil conversion efficiency, seasonal variation coefficient and midday photosynthetic efficiency features are extracted to obtain the variety photosynthetic-metabolic-seasonal coupling feature tensor. A joint analysis of canopy photosynthetic heterogeneity and water-photosynthetic coupling was conducted based on the photosynthetic-metabolism-seasonal coupling characteristic tensor of the variety. In this analysis, a canopy-water adaptability model of the variety was obtained by establishing the response relationship between the photosynthetic contribution weights and photosynthetic parameters of different parts of the canopy and the leaf water status. Oil production is predicted based on the fruit metabolism data, the variety canopy-water adaptability model, and the variety photosynthesis-metabolism-seasonal coupling characteristic tensor. In particular, the oil production is predicted by constructing a causal chain model of photosynthetic capacity-carbon allocation-oil accumulation, and the predicted oil production value of the variety and its uncertainty range are obtained. Variety selection decisions are made based on the predicted oil yield of the variety and the uncertainty range. Specifically, a recommended variety ranking list with risk assessment is obtained by combining the light, temperature and moisture conditions of the preset target planting area.

[0006] Secondly, this application also provides a decision-making system for olive variety selection based on photosynthetic data modeling, including: The acquisition unit is used to acquire multi-scale photosynthetic physiological data and fruit metabolic data of the target variety during the fruit development cycle. The multi-scale photosynthetic physiological data includes intra-diurnal net photosynthetic rate, stomatal conductance, chlorophyll fluorescence parameters, seasonal light response curves and environmental parameters. The fruit metabolic data includes carbon assimilate content and oil accumulation rate during the ripening period. The extraction unit is used to jointly extract photosynthetic-metabolic coupling features and seasonal fluctuation features based on the multi-scale photosynthetic physiological data and the fruit metabolic data. Specifically, by calculating the photosynthetic product allocation coefficient and oil conversion efficiency, the seasonal variation coefficient and midday photosynthetic efficiency features are extracted to obtain the variety photosynthetic-metabolic-seasonal coupling feature tensor. The analysis unit is used to perform joint analysis of canopy photosynthetic heterogeneity and water-photosynthetic coupling based on the photosynthetic-metabolism-seasonal coupling characteristic tensor of the variety. In this analysis, the canopy-water adaptability model of the variety is obtained by establishing the response relationship between the photosynthetic contribution weights and photosynthetic parameters of different parts of the canopy and the leaf water status. The prediction unit is used to predict oil production based on the fruit metabolic data, the variety canopy-water adaptability model and the variety photosynthesis-metabolism-seasonal coupling feature tensor. In this case, the oil production is predicted by constructing a causal chain model of photosynthetic capacity-carbon allocation-oil accumulation, and the predicted value of variety oil production and its uncertainty range are obtained. The screening unit is used to make variety screening decisions based on the predicted oil yield of the variety and the uncertainty range. In this case, by combining the light, temperature and moisture conditions of the preset target planting area, a recommended variety ranking list with risk assessment is obtained.

[0007] The beneficial effects of this invention are as follows: This invention establishes a photosynthesis-metabolism coupling model by integrating multi-scale photosynthetic physiological data and fruit metabolic data, which can accurately predict the oil production potential of varieties and significantly shorten the variety selection cycle compared with traditional field yield observation methods.

[0008] This invention also assesses the adaptability of varieties under different environmental conditions by constructing canopy photosynthetic contribution weights and water-photosynthetic response relationships, providing a scientific basis for regionalized variety selection. Furthermore, through causal chain models and Bayesian posterior inference, this invention quantifies the uncertainty of predictions, providing risk assessment for variety selection decisions. Moreover, through Pareto front analysis, this invention can find the optimal balance between oilseed production potential and environmental adaptability stability, providing decision-making suggestions for planting managers.

[0009] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing embodiments of the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0010] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a schematic diagram of the olive variety selection and decision-making method based on photosynthetic data modeling as described in this embodiment of the invention; Figure 2 This is a schematic diagram of the olive variety selection decision system based on photosynthetic data modeling as described in an embodiment of the present invention; Figure 3This is a comparison chart of photosynthetic product partition coefficient and oil conversion efficiency in the embodiments of the present invention; Figure 4 This is a scatter plot of the seasonal variation coefficient and the midday decrease as described in the embodiments of the present invention; Figure 5 This is a comparison chart of the light response curves for each growth stage described in the embodiments of the present invention.

[0012] In the diagram: 701, acquisition unit; 702, extraction unit; 703, analysis unit; 704, prediction unit; 705, filtering unit. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0014] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this invention, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0015] Example 1:

[0016] This embodiment provides a method for selecting olive varieties based on photosynthetic data modeling.

[0017] See Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5 The figure shows that the method includes steps S1, S2, S3, S4 and S5.

[0018] Step S1: Obtain multi-scale photosynthetic physiological data and fruit metabolic data of the target variety during the fruit development cycle. The multi-scale photosynthetic physiological data includes intra-diurnal net photosynthetic rate, stomatal conductance, chlorophyll fluorescence parameters, seasonal light response curves and environmental parameters. The fruit metabolic data includes carbon assimilate content and oil accumulation rate during the ripening period. It is understandable that olive photosynthesis in this step exhibits seasonal fluctuations and intra-diurnal variations. Seasonal fluctuations are mainly reflected in the significant changes in photosynthetic capacity with the seasons, with photosynthetic depression occurring during the high-temperature summer months. Intra-diurnal variations are mainly reflected in the fluctuations in photosynthetic rate with changes in light intensity and temperature. Therefore, it is necessary to collect multi-scale photosynthetic physiological data to comprehensively characterize the photosynthetic characteristics of the variety from multiple dimensions. Simultaneously, fruit metabolic data reflects the ability to convert photosynthetic products into oil, which is a key indicator of the variety's yield potential.

[0019] In addition, multi-scale photosynthetic physiological data in this step were acquired using a portable photosynthesis meter. Intra-diurnal data were measured at different times of the day, including early morning, morning, noon, afternoon, and evening, with at least three functional leaves measured at each time period. Seasonal data were obtained monthly during the olive growing season, measuring the light response curve, with environmental parameters acquired in real time. Fruit metabolic data were obtained by periodically collecting fruit samples to determine carbon assimilate content and oil accumulation rate, with a sampling cycle of two weeks.

[0020] Step S2: Based on the multi-scale photosynthetic physiological data and the fruit metabolic data, perform joint extraction of photosynthetic-metabolic coupling features and seasonal fluctuation features. Specifically, by calculating the photosynthetic product allocation coefficient and oil conversion efficiency, extract the seasonal variation coefficient and midday photosynthetic efficiency features to obtain the variety photosynthetic-metabolic-seasonal coupling feature tensor. It is understood that the photosynthetic-metabolic coupling characteristics in this invention reflect the ability of photosynthetic products to be allocated to the fruit and converted into oil, while the seasonal fluctuation characteristics reflect the variety's adaptability to seasonal changes. By jointly extracting these two characteristics, the varietal characteristics can be comprehensively characterized from two dimensions: yield capacity and environmental adaptability. In this step, step S2 includes steps S21, S22, and S23.

[0021] Step S21: Calculate the partition coefficient based on the carbon assimilate content in the fruit metabolism data and the net photosynthetic rate in the multi-scale photosynthetic physiological data. Obtain the partition coefficient of photosynthetic products by calculating the ratio of the fruit carbon assimilate accumulation to the leaf net photosynthetic rate. Calculate the oil conversion efficiency based on the ratio of the fruit oil accumulation to the carbon assimilate accumulation to obtain the photosynthetic-metabolic coupling characteristics. It is understandable that this step quantifies the efficiency of photosynthetic product allocation to the fruit by calculating the photosynthetic product allocation coefficient. The larger the allocation coefficient, the higher the proportion of photosynthetic products allocated to the fruit. Oil conversion efficiency reflects the ability of carbon assimilates to be converted into oil, and is a direct indicator of a variety's yield potential. The photosynthetic-metabolic coupling characteristics include both the photosynthetic product allocation coefficient and the oil conversion efficiency.

[0022] The formula for calculating the partition coefficient of photosynthetic products is shown below: ;

[0023] In the formula, The partition coefficient of photosynthetic products; The amount of carbon assimilates accumulated in the fruit; This represents the net photosynthetic rate of the leaf.

[0024] Step S22: Based on the seasonal light response curve in the multi-scale photosynthetic physiological data, seasonal fluctuation analysis is performed. By dividing the growth cycle into budding, flowering, fruit enlargement and ripening stages, the ratio of the standard deviation to the mean of photosynthetic parameters at each stage is calculated to obtain the seasonal variation coefficient. The difference between the peak and trough values ​​of photosynthetic rate at midday is extracted to obtain the decrease rate. The recovery time constant is obtained by fitting the midday photosynthetic rate recovery curve to obtain the seasonal fluctuation characteristics. It is understandable that this step is based on seasonal fluctuation analysis of seasonal light response curves in multi-scale photosynthetic physiological data. By dividing the growth cycle into four key phenological stages—budding, flowering, fruit enlargement, and ripening—the ratio of the standard deviation to the mean of photosynthetic parameters at each stage is calculated to obtain the seasonal coefficient of variation. This coefficient reflects the annual fluctuation range of photosynthetic performance of the variety throughout its development. The smaller the coefficient of variation, the more stable the photosynthetic capacity and the less it is affected by environmental changes caused by seasonal changes. At the same time, the difference between the peak and trough values ​​of photosynthetic rate at midday is extracted to obtain the decline range, and the recovery time constant is obtained by fitting the midday photosynthetic rate recovery curve. These two parameters specifically capture the severity and recovery speed of the "photosynthetic midday depression" phenomenon commonly seen in olives under high temperature and strong light conditions in summer. Varieties with large decline ranges and slow recovery are prone to accumulating photosynthetic damage when facing sustained high temperatures in the afternoon, thus affecting the total amount of carbon fixation throughout the day. Therefore, the seasonal fluctuation characteristics actually assess the photosynthetic robustness of the variety from both temporal dynamics and diurnal stress perspectives.

[0025] This step quantifies the fluctuation of photosynthetic parameters at different growth stages by calculating the seasonal coefficient of variation. A larger coefficient of variation indicates a more sensitive variety to seasonal changes. Midday photosynthetic efficiency reflects the variety's adaptability to high temperatures; a smaller decrease and a shorter recovery time constant indicate better heat tolerance. The seasonal coefficient of variation and recovery time constant represent seasonal fluctuation characteristics.

[0026] The formula for calculating the seasonal variation coefficient is shown below: ;

[0027] In the formula, This is the seasonal variation coefficient; The standard deviation of photosynthetic parameters at each stage; This represents the average photosynthetic parameters at each stage.

[0028] The fitting formula for the midday photosynthetic rate recovery curve is shown below:

[0029] in, For a moment Net photosynthetic rate, This is the baseline value when the photosynthetic rate drops to its lowest point at midday. This represents the peak value before the midday photosynthetic rate begins to recover. To restore the time constant, To restore the time calculated from the start time.

[0030] Step S23: Perform tensor construction processing based on the photosynthesis-metabolism coupling features and the seasonal fluctuation features. By stacking the two features along the feature dimension and normalizing them, the variety photosynthesis-metabolism-seasonal coupling feature tensor is obtained.

[0031] It is understood that this step involves tensor construction based on the photosynthetic-metabolic coupling characteristics and the seasonal fluctuation characteristics. By stacking the two characteristics along the feature dimensions and normalizing them using a standardization method, a tensor of the variety's photosynthetic-metabolic-seasonal coupling characteristics is obtained. The stacking operation integrates the originally scattered information into a multidimensional data structure, while normalization eliminates the impact of differences in the dimensions and orders of magnitude of each feature on subsequent analysis, ensuring that the contribution of each dimension is mathematically comparable. The resulting tensor not only contains the variety's inherent photosynthetic-metabolic transformation capacity but also embeds its response characteristics to seasonal fluctuations and diurnal stress, providing a structured and highly information-dense input foundation for subsequent canopy-water adaptation modeling and oil production prediction.

[0032] Step S3: Perform a joint analysis of canopy photosynthetic heterogeneity and water-photosynthetic coupling based on the photosynthetic-metabolism-seasonal coupling characteristic tensor of the variety. In this process, by establishing the response relationship between the photosynthetic contribution weights and photosynthetic parameters of different parts of the canopy and the leaf water status, a variety canopy-water adaptability model is obtained. It is understood that canopy photosynthetic heterogeneity in this invention reflects the differences in photosynthetic capacity of different parts of the canopy, while water-photosynthetic coupling reflects the influence of water conditions on photosynthetic efficiency. By jointly analyzing these two characteristics, the adaptability of varieties under different spatial locations and water conditions can be assessed. In this step, step S3 includes steps S31, S32, and S33.

[0033] Step S31: Based on the intra-diurnal net photosynthetic rate in the multi-scale photosynthetic physiological data, the canopy is divided into twelve light zones by dividing the canopy vertically into upper, middle and lower parts, and horizontally into four directions: east, south, west and north. The photosynthetic effective radiation and photosynthetic rate of each zone are determined, and the proportion of the photosynthetic rate of each zone to the total photosynthetic rate is calculated as the photosynthetic contribution weight, thus obtaining the distribution of canopy photosynthetic contribution weight. Understandably, this step first involves canopy partitioning based on intraday net photosynthetic rates from multi-scale photosynthetic physiological data. The olive tree canopy is vertically divided into upper, middle, and lower layers, and horizontally into four directions: east, south, west, and north, resulting in twelve light-receiving zones. Within each zone, a portable photosynthetic measurement system is used to simultaneously measure the photosynthetically active radiation and net photosynthetic rate. The proportion of each zone's net photosynthetic rate to the total net photosynthetic rate of all twelve zones is then calculated, defining this proportion as the photosynthetic contribution weight for that zone, thus obtaining the overall canopy photosynthetic contribution weight distribution. This process is necessary because olive trees have tall canopies with dense foliage, and the light intensity and microclimate conditions received by leaves at different heights and orientations vary significantly, leading to uneven actual contributions to overall tree carbon fixation across different zones. Quantifying the photosynthetic contribution weight of each zone allows subsequent analysis to focus on the key canopy areas that truly determine overall tree productivity, avoiding the biases introduced by simple averaging.

[0034] Understandably, this step quantifies the photosynthetic contribution of different parts of the tree canopy by dividing it into multiple light-receiving zones. The upper part of the canopy typically receives more light and contributes more to photosynthesis, while the lower part receives less light and contributes less. By calculating the photosynthetic contribution weight of each zone, the area with the highest photosynthetic efficiency can be identified.

[0035] The formula for calculating the photosynthetic contribution weight is shown below: ;

[0036] In the formula, The contribution weight of photosynthesis to the i-th region; Let be the net photosynthetic rate of the i-th region; This is the sum of the net photosynthetic rates across all regions.

[0037] Step S32: Based on the intraday net photosynthetic rate and seasonal net photosynthetic rate in the multi-scale photosynthetic physiological data, perform water coupling analysis. Specifically, by setting three water gradients—full irrigation, mild water deficit, and severe water deficit—fit the response curves of photosynthetic parameters and leaf water potential under different water gradients to obtain a water-photosynthetic response model. Understandably, this step involves performing water coupling analysis based on intraday net photosynthetic rate and seasonal environmental parameters from multi-scale photosynthetic physiological data. Specifically, three water gradients are set: full irrigation, mild water deficit, and severe water deficit. Leaf water potential is measured (using the pressure chamber method) under these three water conditions, and net photosynthetic rate is recorded simultaneously. Then, discrete data points showing the change in net photosynthetic rate with leaf water potential under each water gradient are curve-fitted to obtain a response curve reflecting the quantitative relationship between photosynthetic parameters and leaf water status of the variety—this is the water-photosynthetic response model. The core value of this model lies in revealing the rate and threshold of photosynthetic function degradation in different varieties under water stress. For evergreen woody oil crops like olive, summer drought is the main environmental factor limiting fruit oil accumulation; therefore, understanding the photosynthetic-water response characteristics of a variety is key to determining its adaptation to local rainfall patterns.

[0038] The formula for calculating the water-photosynthetic response relationship is shown below: ;

[0039] In the formula, Net photosynthetic rate; Leaf water potential; , , represents the fitting coefficient.

[0040] Step S33: Perform joint modeling processing based on the canopy photosynthetic contribution weight distribution and the water-photosynthetic response model. By weighting and fusing the photosynthetic contribution weight as the spatial dimension and the water response as the conditional dimension, a variety canopy-water adaptability model is obtained.

[0041] Understandably, this step uses photosynthetic contribution weights as weighting coefficients in the spatial dimension and the water-photosynthetic response model as a function mapping in the conditional dimension, obtaining a unified variety canopy-water adaptability model through weighted fusion. This model is essentially a composite function that, given any water condition (represented by leaf water potential), first calls the water-photosynthetic response model to estimate the potential net photosynthetic rate of each canopy region under the current water condition, then multiplies it by the corresponding photosynthetic contribution weights for spatial accumulation, and finally outputs the expected photosynthetic production capacity of the entire tree under the current water condition. This fusion method not only considers the spatial heterogeneity of photosynthesis within the canopy but also incorporates the differentiated impact of water stress on the photosynthetic capacity of different canopy parts, thus enabling the model to more realistically reflect the overall performance of the variety in complex field environments, providing input with both spatial accuracy and environmental sensitivity for subsequent oil yield prediction.

[0042] The unified formula for the variety canopy-water adaptability model is shown below: ;

[0043] in, Indicates the water potential of a given blade Below, the expected net photosynthetic rate of the entire tree canopy, For numbering, , and For the first The quadratic fitting coefficients of the response curve of the region.

[0044] Step S4: Based on the fruit metabolism data, the variety canopy-water adaptability model, and the variety photosynthesis-metabolism-seasonal coupling characteristic tensor, oil production is predicted. In this step, a causal chain model of photosynthetic capacity-carbon allocation-oil accumulation is constructed to predict oil production, and the predicted oil production value of the variety and its uncertainty range are obtained. It is understood that the causal chain model in this invention reveals the influence of photosynthetic capacity on carbon assimilate accumulation and the influence of carbon assimilates on oil conversion. By identifying and quantifying these two causal pathways, the oil yield potential of a variety can be accurately predicted. In this step, step S4 includes steps S41, S42, and S43.

[0045] Step S41: Based on the photosynthesis-metabolism-seasonal coupling characteristic tensor of the variety, causal path identification processing is performed. By using intraday net photosynthetic rate, stomatal conductance and chlorophyll fluorescence parameters as exogenous variables and fruit metabolism data as endogenous variables, a structural equation model is constructed to determine the first causal path of photosynthetic capacity on carbon assimilate accumulation and the second causal path of carbon assimilate on oil conversion, thus obtaining the photosynthesis-oil causal chain model. This step, understandably, begins with three exogenous variables directly reflecting the immediate photosynthetic physiological state of leaves: intraday net photosynthetic rate, stomatal conductance, and chlorophyll fluorescence parameters. It uses carbon assimilation content and lipid accumulation rate from fruit metabolic data as endogenous variables to construct a structural equation model. Structural equation modeling is a multivariate statistical method capable of handling multiple dependent variables simultaneously and allowing correlations between independent variables. Its unique application here is to distinguish and quantify the two indirect pathways of photosynthetic capacity via carbon assimilation accumulation to lipid conversion. By testing the significance and comparing the magnitudes of path coefficients, it can be determined which pathway contributes more critically to the final lipid production, thus obtaining a photosynthetic-lipid causal chain model. In this invention, different varieties may have similar total photosynthetic rates, but the efficiency of photosynthetic product allocation to the fruit carbon pool and the efficiency of conversion within the carbon pool to lipid synthesis may differ fundamentally. The causal chain model can reveal the source of this difference at the mechanistic level, providing a structured causal relationship rather than empirical correlation for subsequent accurate predictions.

[0046] The first causal pathway reflects the direct impact of photosynthetic capacity on carbon assimilate accumulation, while the second causal pathway reflects the direct impact of carbon assimilates on lipid conversion. By quantifying these two causal pathways, a quantitative relationship between photosynthetic capacity and lipid production can be established.

[0047] The basic form of the structural equation model is shown below: ;

[0048] In the formula, This is a vector of endogenous variables, including carbon assimilate content and lipid accumulation rate; This is a vector of exogenous variables, including net photosynthetic rate, stomatal conductance, and chlorophyll fluorescence parameters; This is the relationship matrix between endogenous variables; This is the matrix showing the influence of exogenous variables on endogenous variables; This is the residual vector.

[0049] Step S42: Based on the photosynthesis-oil causal chain model and the variety canopy-water adaptability model, the output prediction is performed. By inputting the variety photosynthesis-metabolism-season coupling feature tensor into the causal chain model and combining it with the adaptability model for condition correction, the predicted value of variety oil output is obtained. Understandably, this step uses the varietal photosynthetic-metabolic-seasonal coupling characteristic tensor as input data into the causal chain model. It calculates the expected carbon assimilate accumulation and oil accumulation rate layer by layer along the causal path identified in the first step. At the same time, it combines the whole-tree photosynthetic correction coefficient provided by the canopy-water adaptation model under the current water conditions to conditionally correct the intermediate results output by the causal chain model. The causal chain model itself is fitted based on ideal or average water conditions, while the canopy-water adaptation model provides spatially weighted information on the actual loss of photosynthetic capacity under different water stress levels. The combination of the two is equivalent to adapting the theoretical causal path to a real-world environment. The final predicted value of varietal oil production includes both the varietal's inherent photosynthetic-metabolic transformation potential and the limiting effect of environmental water on photosynthetic function, thus more closely reflecting actual field performance.

[0050] The formula for calculating the water stress factor is as follows: ;

[0051] in, The water stress factor is used as spatial weighting information for actual losses; that is, it is multiplied by the intermediate results obtained by fitting under ideal or average water conditions to obtain the condition-corrected predicted oil yield value for the variety. No. The relative retention coefficient of the region.

[0052] Step S43: Based on the predicted oil yield of the variety, perform uncertainty quantification processing. By using historical photosynthetic-metabolic data of the variety as prior information and measured oil accumulation data as observation data, perform Bayesian posterior inference, and use Markov chain Monte Carlo method for sampling to obtain the uncertainty range of the predicted oil yield of the variety.

[0053] Understandably, this step uses historical photosynthetic-metabolic data of varieties as prior information and current measured oil accumulation data as observational data. It integrates prior knowledge with new observational evidence through a Bayesian posterior inference framework and employs a Markov chain Monte Carlo method for extensive iterative sampling to generate uncertainty intervals for variety oil yield predictions from the posterior distribution. The advantage of the Bayesian method lies in its ability to naturally handle small sample problems and multi-source uncertainties, while Markov chain Monte Carlo sampling can efficiently approximate the shape of the posterior distribution in a complex parameter space without relying on simplified normality assumptions. Since the number of individuals for each variety is limited and annual climate fluctuations are significant, single-point predictions often lack reliability. Uncertainty intervals, however, directly provide the possible range of fluctuations in the predicted value and its credibility. This allows policymakers to distinguish between risky varieties with high average yields but large interannual fluctuations and conservative varieties with moderate yields but stable performance, providing a quantitative basis for subsequent variety ranking with risk assessment.

[0054] The prior information is derived from statistical analysis of photosynthetic-metabolic data of historical varieties, while the observed data consists of measured oil accumulation data for current varieties. A Markov chain Monte Carlo method is used for sampling to obtain the posterior probability distribution of the predicted oil yield, thereby calculating the uncertainty interval.

[0055] The basic formula for Bayesian posterior inference is expressed as follows: ;

[0056] In the formula, This is the posterior probability distribution; It is the likelihood function; For the prior probability distribution, It is proportional to the sign.

[0057] Step S5: Make variety selection decisions based on the predicted oil yield of the variety and the uncertainty range. In this step, a recommended variety ranking list with risk assessment is obtained by combining the light, temperature and moisture conditions of the preset target planting area.

[0058] It is understood that this invention, by combining the environmental conditions of the target planting area, can assess the adaptability and yield potential of varieties under specific environments. Through Pareto front analysis, it can find the optimal balance between oil yield potential and environmental adaptability stability, providing planting managers with a recommended variety ranking list with risk assessment. In this step, step S5 includes steps S51, S52, and S53.

[0059] Step S51: Perform multi-scenario simulation processing based on the predicted oil yield of the variety. Generate multiple sets of environmental parameter combinations of light intensity, temperature and water stress using the Monte Carlo method. Calculate the oil yield probability distribution of each variety under different environmental conditions to obtain the multi-scenario yield matrix of each variety. Understandably, this step generates multiple sets of environmental parameter combinations for light intensity, temperature, and water stress, each representing a possible field climate scenario. Then, the oil yield prediction model for each variety (i.e., the overall prediction function of the photosynthesis-oil causal chain model constructed in the previous step, modified by the canopy-water adaptability model) is sequentially substituted into each set of environmental parameter combinations for calculation. The oil yield value of the variety under each scenario is recorded. After statistically summarizing the calculation results for all scenarios, the probability distribution of oil yield for the variety under different environmental conditions is obtained. Arranging the probability distributions of all varieties row by row constitutes a multi-scenario yield matrix. In this invention, olive, as a perennial woody oilseed crop, has a growth cycle spanning several years, inevitably experiencing various interannual climate fluctuations. Predictions under a single climate condition alone cannot determine the long-term stability of the variety. Multi-scenario simulation precisely compensates for this deficiency, expanding prediction from point estimation to area estimation, providing a data foundation for subsequent robustness evaluation.

[0060] Step S52: Perform Pareto analysis based on the multi-scenario output matrix. Find the non-dominated solution set between the two objectives of oil production potential and environmental adaptation stability. Calculate the distance of each variety to the ideal point using Euclidean distance and sort them to obtain the candidate variety ranking. Understandably, this step seeks a set of non-dominated solutions between two objectives: oil yield potential and environmental adaptation stability. A non-dominated solution is one where no other variety is superior to it in both objectives simultaneously. In other words, increasing yield potential requires sacrificing some stability, and vice versa. This trade-off is extremely common in multi-objective decision-making. Subsequently, Euclidean distance is used to calculate the distance of each variety to the ideal point, which is the virtual optimal coordinate formed by the highest yield potential and highest stability among all varieties. The closer the distance, the closer the variety's overall performance is to the theoretical optimum in both objectives, thus obtaining a ranking of candidate varieties. This step simultaneously considers the often contradictory demands of high yield and yield stability in actual production. For olive growers, a variety with extremely high yield in a wet year but complete crop failure in a dry year may not be superior to a variety with moderate yield but stable yield year after year. Pareto analysis can precisely reveal this trade-off quantitatively.

[0061] The Euclidean distance is calculated using the following formula: ;

[0062] In the formula, The distance from the variety to the ideal point; This represents the potential for oil production. For environmental adaptability and stability; and These are the coordinates of the ideal point.

[0063] Step S53: Perform risk assessment processing based on the candidate variety ranking and the uncertainty interval. By mapping the coefficient of variation of the uncertainty interval to three levels of low risk, medium risk and high risk, and combining the predicted oil yield value and confidence interval of each variety, a comprehensive ranking is performed to obtain a recommended variety ranking list with risk assessment.

[0064] It is understandable that this step maps the coefficient of variation of the uncertainty interval (i.e., the width of the uncertainty interval divided by the mean of the predicted oil yield value) to three levels of low risk, medium risk, and high risk according to a preset threshold. The smaller the coefficient of variation, the lower the dispersion of the prediction result, the higher the confidence, and the lower the corresponding risk level. Then, the predicted oil yield value, confidence interval (i.e., the uncertainty interval obtained in step S43), and adaptability score (the inverse distance obtained from the Pareto analysis in step S52) of each variety are combined for comprehensive ranking (ranked by weighted score, where the risk level, confidence interval, adaptability score, and predicted oil yield value have weights of 0.3, 0.2, 0.2, and 0.3, respectively). Finally, a recommended variety ranking list is output. In addition to the variety name and ranking position, the risk level, confidence interval, adaptability score, and predicted oil yield value of each variety are clearly marked in the list. This list transforms the final recommendation from a simple ranking into a decision-making aid that includes risk warnings. Growers can choose varieties based on their own risk tolerance and the climate characteristics of their target planting area. For example, in areas with abundant water resources, high-yield but medium-risk varieties can be prioritized, while in drought-prone areas, low-risk but stable-yield varieties should be prioritized, thus achieving personalized recommendations tailored to local conditions.

[0065] Example 2: like Figure 2 As shown, this embodiment provides an olive variety selection decision system based on photosynthetic data modeling. See [link to relevant documentation]. Figure 2 The system includes an acquisition unit 701, an extraction unit 702, an analysis unit 703, a prediction unit 704, and a filtering unit 705.

[0066] The acquisition unit 701 is used to acquire multi-scale photosynthetic physiological data and fruit metabolic data of the target variety during the fruit development cycle. The multi-scale photosynthetic physiological data includes intra-diurnal net photosynthetic rate, stomatal conductance, chlorophyll fluorescence parameters, seasonal light response curves and environmental parameters. The fruit metabolic data includes carbon assimilate content and oil accumulation rate during the ripening period. Extraction unit 702 is used to jointly extract photosynthetic-metabolic coupling features and seasonal fluctuation features based on the multi-scale photosynthetic physiological data and the fruit metabolic data. Specifically, by calculating the photosynthetic product allocation coefficient and oil conversion efficiency, seasonal variation coefficient and midday photosynthetic efficiency features are extracted to obtain the variety photosynthetic-metabolic-seasonal coupling feature tensor. Analysis unit 703 is used to perform joint analysis of canopy photosynthetic heterogeneity and water-photosynthetic coupling based on the photosynthetic-metabolism-seasonal coupling characteristic tensor of the variety. In this analysis, a variety canopy-water adaptability model is obtained by establishing the response relationship between photosynthetic contribution weights and photosynthetic parameters of different parts of the canopy and leaf water status. The prediction unit 704 is used to predict oil production based on the fruit metabolic data, the variety canopy-water adaptability model and the variety photosynthesis-metabolism-seasonal coupling feature tensor. The oil production prediction is carried out by constructing a causal chain model of photosynthetic capacity-carbon allocation-oil accumulation to obtain the variety oil production prediction value and its uncertainty range. The screening unit 705 is used to make variety screening decisions based on the predicted oil yield of the variety and the uncertainty range, wherein a recommended variety ranking list with risk assessment is obtained by combining the light, temperature and moisture conditions of the preset target planting area.

[0067] It should be noted that the specific methods by which each module performs operations in the system described in the above embodiments have been described in detail in the embodiments related to the method, and will not be elaborated here.

[0068] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0069] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for selecting olive varieties based on photosynthetic data modeling, characterized in that, include: Acquire multi-scale photosynthetic physiological data and fruit metabolic data of the target variety during the fruit development cycle. The multi-scale photosynthetic physiological data includes intra-diurnal net photosynthetic rate, stomatal conductance, chlorophyll fluorescence parameters, seasonal light response curves and environmental parameters. The fruit metabolic data includes carbon assimilate content and oil accumulation rate during the ripening period. Based on the multi-scale photosynthetic physiological data and the fruit metabolic data, photosynthetic-metabolic coupling features and seasonal fluctuation features are jointly extracted. Specifically, by calculating the photosynthetic product allocation coefficient and oil conversion efficiency, seasonal variation coefficient and midday photosynthetic efficiency features are extracted to obtain the variety photosynthetic-metabolic-seasonal coupling feature tensor. A joint analysis of canopy photosynthetic heterogeneity and water-photosynthetic coupling was conducted based on the photosynthetic-metabolism-seasonal coupling characteristic tensor of the variety. In this analysis, a canopy-water adaptability model of the variety was obtained by establishing the response relationship between the photosynthetic contribution weights and photosynthetic parameters of different parts of the canopy and the leaf water status. Causal path identification was performed based on the photosynthetic-metabolic-seasonal coupling characteristic tensor of the variety. By constructing a structural equation model with intraday net photosynthetic rate, stomatal conductance, and chlorophyll fluorescence parameters as exogenous variables and fruit metabolic data as endogenous variables, the first causal path of photosynthetic capacity on carbon assimilate accumulation and the second causal path of carbon assimilate on oil conversion were determined, thus obtaining the photosynthetic-oil causal chain model. The yield prediction process is carried out based on the photosynthesis-oil causal chain model and the variety canopy-water adaptation model. The variety photosynthesis-metabolism-season coupling feature tensor is input into the causal chain model and combined with the adaptation model for condition correction to obtain the variety oil yield prediction value. Uncertainty quantification was performed on the predicted oil yield of the variety. Bayesian posterior inference was performed using historical photosynthetic-metabolic data of the variety as prior information and measured oil accumulation data as observation data. The Markov chain Monte Carlo method was used for sampling to obtain the uncertainty range of the predicted oil yield of the variety. Variety selection decisions are made based on the predicted oil yield of the variety and the uncertainty range. Specifically, a recommended variety ranking list with risk assessment is obtained by combining the light, temperature and moisture conditions of the preset target planting area.

2. The olive variety selection and decision-making method based on photosynthetic data modeling according to claim 1, characterized in that... Based on the multi-scale photosynthetic physiological data and the fruit metabolic data, joint extraction of photosynthetic-metabolic coupling features and seasonal fluctuation features is performed, including: The partition coefficients were calculated based on the carbon assimilate content in fruit metabolic data and the net photosynthetic rate in multi-scale photosynthetic physiological data. The partition coefficients of photosynthetic products were obtained by calculating the ratio of the accumulation of carbon assimilates in fruit to the net photosynthetic rate of leaves. The oil conversion efficiency was calculated based on the ratio of the accumulation of oil in fruit to the accumulation of carbon assimilates, thus obtaining the photosynthetic-metabolic coupling characteristics. Seasonal fluctuation analysis was performed based on the seasonal light response curves in multi-scale photosynthetic physiological data. The growth cycle was divided into budding, flowering, fruit enlargement and ripening stages. The ratio of the standard deviation to the mean of photosynthetic parameters at each stage was calculated to obtain the seasonal variation coefficient. The difference between the peak and trough of photosynthetic rate at midday was extracted to obtain the decrease. The recovery time constant was obtained by fitting the midday photosynthetic rate recovery curve, and the seasonal fluctuation characteristics were obtained. Tensor construction is performed based on the photosynthesis-metabolism coupling features and the seasonal fluctuation features. By stacking the two features along the feature dimension and normalizing them, the variety photosynthesis-metabolism-seasonal coupling feature tensor is obtained.

3. The olive variety selection decision method based on photosynthetic data modeling according to claim 1, characterized in that... Based on the photosynthetic-metabolic-seasonal coupling characteristic tensor of the aforementioned varieties, a joint analysis of canopy photosynthetic heterogeneity and water-photosynthetic coupling was performed, including: Canopy zoning was performed based on intra-diurnal net photosynthetic rate from multi-scale photosynthetic physiological data. Specifically, the canopy was divided into upper, middle, and lower sections vertically and into four directions (east, south, west, and north) horizontally, resulting in twelve light zones. The photosynthetically active radiation and photosynthetic rate of each zone were determined, and the proportion of photosynthetic rate of each zone to the total photosynthetic rate was calculated as the photosynthetic contribution weight, thus obtaining the canopy photosynthetic contribution weight distribution. Based on the intraday net photosynthetic rate and seasonal water coupling analysis in the multi-scale photosynthetic physiological data, the water-photosynthetic response model was obtained by setting three water gradients: full irrigation, mild water deficit, and severe water deficit, and fitting the response curves of photosynthetic parameters and leaf water potential under different water gradients. Based on the canopy photosynthetic contribution weight distribution and the water-photosynthetic response model, a joint modeling process is performed. By weighting and fusing the photosynthetic contribution weight as the spatial dimension and the water response as the conditional dimension, a variety canopy-water adaptability model is obtained.

4. The olive variety selection decision method based on photosynthetic data modeling according to claim 1, characterized in that... Variety selection decisions are made based on the predicted oil yield of the variety and the uncertainty interval, including: Based on the predicted oil yield of the varieties, multi-scenario simulation processing was performed. Multiple sets of environmental parameter combinations of light intensity, temperature and water stress were generated using the Monte Carlo method. The oil yield probability distribution of each variety under different environmental conditions was calculated to obtain the multi-scenario yield matrix of each variety. Pareto analysis was performed on the multi-scenario output matrix. By finding the non-dominated solution set between the two objectives of oil production potential and environmental adaptation stability, and by using Euclidean distance to calculate the distance of each variety to the ideal point for sorting, the candidate varieties were ranked. Risk assessment is performed based on the candidate variety ranking and the uncertainty interval. The coefficient of variation of the uncertainty interval is mapped to three levels: low risk, medium risk, and high risk. The predicted oil yield and confidence interval of each variety are combined for comprehensive ranking to obtain a recommended variety ranking list with risk assessment.

5. A decision-making system for selecting olive varieties based on photosynthetic data modeling, characterized in that, include: The acquisition unit is used to acquire multi-scale photosynthetic physiological data and fruit metabolic data of the target variety during the fruit development cycle. The multi-scale photosynthetic physiological data includes intra-diurnal net photosynthetic rate, stomatal conductance, chlorophyll fluorescence parameters, seasonal light response curves and environmental parameters. The fruit metabolic data includes carbon assimilate content and oil accumulation rate during the ripening period. The extraction unit is used to jointly extract photosynthetic-metabolic coupling features and seasonal fluctuation features based on the multi-scale photosynthetic physiological data and the fruit metabolic data. Specifically, by calculating the photosynthetic product allocation coefficient and oil conversion efficiency, the seasonal variation coefficient and midday photosynthetic efficiency features are extracted to obtain the variety photosynthetic-metabolic-seasonal coupling feature tensor. The analysis unit is used to perform joint analysis of canopy photosynthetic heterogeneity and water-photosynthetic coupling based on the photosynthetic-metabolism-seasonal coupling characteristic tensor of the variety. In this analysis, the canopy-water adaptability model of the variety is obtained by establishing the response relationship between the photosynthetic contribution weights and photosynthetic parameters of different parts of the canopy and the leaf water status. The prediction unit is used to predict oil production based on the fruit metabolic data, the variety canopy-water adaptability model and the variety photosynthesis-metabolism-seasonal coupling feature tensor. In this case, the oil production prediction is performed by constructing a causal chain model of photosynthetic capacity-carbon allocation-oil accumulation, and the predicted value of variety oil production and its uncertainty range are obtained. The prediction unit includes: The first prediction subunit is used to perform causal path identification processing based on the photosynthetic-metabolic-seasonal coupling characteristic tensor of the variety. By constructing a structural equation model with net photosynthetic rate, stomatal conductance and chlorophyll fluorescence parameters at the intraday scale as exogenous variables and fruit metabolism data as endogenous variables, the first causal path of photosynthetic capacity on carbon assimilate accumulation and the second causal path of carbon assimilate on oil conversion are determined, and a photosynthetic-oil causal chain model is obtained. The second prediction subunit is used to perform output prediction processing based on the photosynthesis-oil causal chain model and the variety canopy-water adaptation model. By inputting the variety photosynthesis-metabolism-seasonal coupling feature tensor into the causal chain model and combining it with the adaptation model for condition correction, the predicted value of variety oil output is obtained. The third prediction subunit is used to perform uncertainty quantification processing on the predicted oil production value of the variety. By using the photosynthetic-metabolic data of the historical variety as prior information and the measured oil accumulation data as observation data, Bayesian posterior inference is performed, and the Markov chain Monte Carlo method is used for sampling to obtain the uncertainty range of the predicted oil production value of the variety. The screening unit is used to make variety screening decisions based on the predicted oil yield of the variety and the uncertainty range. In this case, by combining the light, temperature and moisture conditions of the preset target planting area, a recommended variety ranking list with risk assessment is obtained.

6. The olive variety selection decision system based on photosynthetic data modeling according to claim 5, characterized in that, The extraction unit includes: The first extraction subunit is used to calculate the partition coefficient based on the carbon assimilate content in the fruit metabolic data and the net photosynthetic rate in the multi-scale photosynthetic physiological data. The partition coefficient of photosynthetic products is obtained by calculating the ratio of the fruit carbon assimilate accumulation to the leaf net photosynthetic rate. The oil conversion efficiency is calculated based on the ratio of the fruit oil accumulation to the carbon assimilate accumulation, thus obtaining the photosynthetic-metabolic coupling characteristics. The second extraction subunit is used to perform seasonal fluctuation analysis based on the seasonal light response curve in the multi-scale photosynthetic physiological data. By dividing the growth cycle into the budding period, flowering period, fruit enlargement period and ripening period, the ratio of the standard deviation to the mean of the photosynthetic parameters at each stage is calculated to obtain the seasonal variation coefficient. The difference between the peak and trough values ​​of the photosynthetic rate at midday is extracted to obtain the decrease rate. The recovery time constant is obtained by fitting the midday photosynthetic rate recovery curve, and the seasonal fluctuation characteristics are obtained. The third extraction subunit is used to perform tensor construction processing based on the photosynthesis-metabolism coupling features and the seasonal fluctuation features. By stacking the two features along the feature dimension and performing normalization processing, the variety photosynthesis-metabolism-seasonal coupling feature tensor is obtained.

7. The olive variety selection decision system based on photosynthetic data modeling according to claim 5, characterized in that, The analysis unit includes: The first analysis subunit is used to perform canopy partitioning based on the intra-diurnal net photosynthetic rate in multi-scale photosynthetic physiological data. Specifically, the canopy is divided into upper, middle and lower parts according to vertical layers and into four directions (east, south, west and north) according to horizontal orientation, resulting in a total of twelve light zones. The photosynthetically active radiation and photosynthetic rate of each zone are determined, and the proportion of photosynthetic rate of each zone to the total photosynthetic rate is calculated as the photosynthetic contribution weight, thus obtaining the canopy photosynthetic contribution weight distribution. The second analysis subunit is used to perform water coupling analysis based on the intraday net photosynthetic rate and the seasonal scale in the multi-scale photosynthetic physiological data. Specifically, by setting three water gradients—full irrigation, mild water deficit, and severe water deficit—the response curves of photosynthetic parameters and leaf water potential under different water gradients are fitted to obtain the water-photosynthetic response model. The third analysis subunit is used to perform joint modeling processing based on the distribution of photosynthetic contribution weights in the canopy and the water-photosynthetic response model. By weighting and fusing the photosynthetic contribution weights as the spatial dimension and the water response as the conditional dimension, a variety canopy-water adaptability model is obtained.

8. The olive variety selection decision system based on photosynthetic data modeling according to claim 5, characterized in that, The filtering unit includes: The first screening subunit is used to perform multi-scenario simulation processing based on the predicted oil yield of the variety. It generates multiple sets of environmental parameter combinations of light intensity, temperature and water stress degree through the Monte Carlo method, calculates the oil yield probability distribution of each variety under different environmental conditions, and obtains the multi-scenario yield matrix of each variety. The second screening subunit is used to perform Pareto analysis based on the multi-scenario output matrix. It searches for non-dominated solution sets between two objectives: oil production potential and environmental adaptation stability. It then uses Euclidean distance to calculate the distance of each variety to the ideal point and sorts them to obtain the candidate variety ranking. The third screening subunit is used to perform risk assessment processing based on the candidate variety ranking and the uncertainty interval. By mapping the coefficient of variation of the uncertainty interval to three levels of low risk, medium risk and high risk, and combining the predicted oil yield value and confidence interval of each variety for comprehensive ranking, a recommended variety ranking list with risk assessment is obtained.

Images