Black soil fertility evaluation method and system combined with microbial community detection

Abstract

The invention discloses a black soil fertility assessment method and system combined with microbial community detection, and relates to the technical field of soil fertility assessment, the method comprises the following steps: collecting soil basic physical and chemical property indexes of black soil, and outputting a first fertility assessment result; collecting microflora monitoring data of black soil, and outputting a second fertility evaluation result; respectively performing direct influence evaluation and indirect influence evaluation according to a preset planting scheme, and configuring a first weight and a second weight; and performing weighted fusion on the first fertility evaluation result and the second fertility evaluation result according to the first weight and the second weight, and outputting a final black soil fertility evaluation result. The technical problem that an existing black soil fertility assessment method only depends on soil basic physical and chemical property indexes, and consequently assessment results cannot comprehensively and accurately reflect complex characteristics and comprehensive fertility conditions of a black soil ecosystem is solved.

Description

Technical Field

[0001] This application relates to the field of soil fertility assessment technology, specifically to a method and system for assessing black soil fertility by combining microbial community detection. Background Technology

[0002] With population growth and rising food demand, the fertility of black soil is crucial to the sustainability and stability of agricultural production. Traditional methods for assessing black soil fertility focus on the detection and analysis of basic soil physicochemical properties, mostly determining soil fertility levels by measuring parameters such as soil organic matter content, nitrogen, phosphorus, potassium, pH value, and cation exchange capacity.

[0003] However, traditional black soil fertility assessment methods rely solely on basic soil physicochemical properties, which makes it difficult to fully reflect the complex characteristics and overall fertility status of the black soil ecosystem. They neglect the integration and application of soil microbial communities, resulting in assessment results that may fail to fully and accurately reveal the true state and long-term evolution of black soil fertility, making it difficult to meet the needs of modern agriculture for refined management and sustainable utilization of black soil. Summary of the Invention

[0004] This application provides a method and system for assessing black soil fertility by combining microbial community detection. This solves the technical problem that existing black soil fertility assessment methods rely solely on basic soil physicochemical properties, resulting in assessment results that cannot comprehensively and accurately reflect the complex characteristics and overall fertility status of the black soil ecosystem.

[0005] The technical solution to the above-mentioned technical problems in this application is as follows: In a first aspect, this application provides a method for assessing black soil fertility by combining microbial community detection, the method comprising: The basic physical and chemical properties of black soil in the target area were collected as direct evaluation data, and the first fertility evaluation result was output after black soil fertility evaluation. Microbial community monitoring data of black soil in the target area over a historical time range were collected as indirect assessment data, and black soil fertility assessment was performed to output a second fertility assessment result. Based on the pre-set planting plan for the target area, the direct and indirect impacts are evaluated, and the first and second weights are assigned according to the degree of impact. The first fertility assessment result and the second fertility assessment result are weighted and fused according to the first weight and the second weight to output the final black soil fertility assessment result.

[0006] Secondly, this application provides a black soil fertility assessment system that combines microbial community detection, including: The direct impact assessment module is used to collect basic physical and chemical properties of black soil in the target area as direct assessment data, and to output the first fertility assessment result for black soil fertility assessment. The indirect impact assessment module is used to collect microbial community monitoring data of black soil in the target area over a historical time range as indirect assessment data, and to perform black soil fertility assessment and output the second fertility assessment results. The influence weight configuration module is used to evaluate the direct and indirect influences based on the preset planting plan for the target area, and to configure the first and second weights according to the degree of influence. The evaluation result output module is used to perform weighted fusion of the first fertility evaluation result and the second fertility evaluation result according to the first weight and the second weight, and output the final black soil fertility evaluation result.

[0007] This application provides one or more technical solutions, which have at least the following technical effects or advantages: This application provides a method and system for assessing black soil fertility by combining microbial community detection. First, direct assessment data is collected and a first fertility assessment result is output, directly reflecting the contribution of the current physicochemical properties of the black soil to fertility. Simultaneously, indirect assessment data of the target area is collected and a second fertility assessment result is output, taking into account the role of the microbial community in the black soil ecosystem, thus more accurately assessing its indirect impact on black soil fertility. Next, direct and indirect impacts are evaluated separately according to a preset planting plan for the target area, with first and second weights configured based on the degree of impact, making the assessment method more targeted and practical. Finally, the first and second fertility assessment results are weighted and fused according to the first and second weights to output the final black soil fertility assessment result. This comprehensively considers the current physicochemical state of the soil and the potential fertility driven by microorganisms, outputting an assessment result that reflects the complex characteristics and overall fertility status of the black soil ecosystem, providing a more reliable scientific basis for the refined management and sustainable utilization of black soil.

[0008] Through the above technical solution, this application realizes a multi-dimensional integrated assessment of basic soil physicochemical properties and microbial community characteristics, breaking through the limitations of traditional methods that rely solely on physicochemical indicators, and significantly improving the comprehensiveness, accuracy and dynamic prediction capability of black soil fertility assessment. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart illustrating the black soil fertility assessment method combined with microbial community detection provided in the embodiments of this application. Figure 2 This is a schematic diagram of the structure of the black soil fertility assessment system combined with microbial community detection provided in the embodiments of this application.

[0011] The components represented by each number in the attached diagram are explained below: The evaluation module 11 directly affects the evaluation module 12, the evaluation module 13 indirectly affects the evaluation module 14, the evaluation weight configuration module 15 directly affects the evaluation module 16, and the evaluation result output module 17 directly affects the evaluation module 18. Detailed Implementation

[0012] This application provides a method and system for assessing black soil fertility by combining microbial community detection. This addresses the technical problem that existing black soil fertility assessment methods rely solely on basic soil physicochemical properties, resulting in assessment results that cannot comprehensively and accurately reflect the complex characteristics and overall fertility status of the black soil ecosystem.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0014] In the description of this application, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "multiple" means two or more, unless otherwise explicitly specified.

[0015] In the description of this application, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use this application. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid unnecessarily obscuring the description of this application. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0016] Example 1, as Figure 1 As shown in the embodiments of this application, a method for assessing black soil fertility by combining microbial community detection is provided, including: S10: Collect the basic physical and chemical properties of black soil in the target area as direct evaluation data, and output the first fertility evaluation result for black soil fertility assessment. The basic physicochemical properties of the soil include at least the thickness of the black soil layer, soil organic matter content, soil pH value, cation exchange capacity, alkaline nitrogen content, available phosphorus content, and available potassium content.

[0017] In this embodiment, firstly, black soil samples are collected from the target area, and the basic physicochemical properties of the soil are used as direct evaluation data. A multi-point mixed sampling method can be adopted, in which multiple sampling points are set up in the target area according to the area size and soil heterogeneity. Soil samples are collected at different soil depths such as 0-20cm and 20-40cm at each sampling point. After removing impurities such as stones and plant residues, the samples are mixed evenly to form a mixed soil sample.

[0018] Specifically, laboratory tests are conducted on mixed soil samples to obtain specific values ​​for various basic soil physicochemical properties. For example, the soil organic matter content is determined using the potassium dichromate oxidation-external heating method; the soil pH value is determined using a pH meter, with the soil sample and water mixed at a mass ratio of 1:2.5, stirred thoroughly, and allowed to stand for a period of time before measurement; the cation exchange capacity is determined using the ammonium acetate exchange method; the alkaline hydrolysis diffusion method is used to determine the alkaline nitrogen content; the available phosphorus content is determined using the Olsen method; the available potassium content is determined using the ammonium acetate extraction-flame photometry method; and the thickness of the black soil layer is measured directly using a steel ruler.

[0019] After obtaining the above-mentioned basic physical and chemical properties of the soil, a basic physical and chemical fertility evaluation system for soil is constructed to conduct a fertility assessment of black soil and output the first fertility assessment result.

[0020] Furthermore, the first fertility assessment result of the black soil is output, including: Configure a direct soil fertility assessment system, wherein the direct soil fertility assessment system includes an index weight distribution and an index value-fertility index mapping table; Based on the index value-fertility index mapping table, the first fertility index distribution is output based on the black soil layer thickness, soil organic matter content, soil pH value, cation exchange capacity, alkaline nitrogen content, available phosphorus content and available potassium content. The first fertility index distribution is weighted and fused according to the index weight distribution to obtain the first comprehensive fertility index, which is used as the first fertility assessment result.

[0021] In this embodiment, firstly, through literature review, expert consultation, and statistical analysis of historical fertility assessment data, the weight distribution of various basic soil physicochemical properties in the direct assessment system is determined. For example, considering the core role of soil organic matter in black soil fertility, its weight can be set to 0.3; the thickness of the black soil layer, as an important indicator of the scarcity of black soil resources and basic productivity, is weighted at 0.2; soil pH affects nutrient availability, and its weight is set at 0.1; cation exchange capacity reflects the soil's nutrient retention capacity, and its weight is set at 0.1; alkaline-available nitrogen, available phosphorus, and available potassium, as the main available nutrients, are assigned weights of 0.1, 0.1, and 0.1 respectively, with the sum of all weights being 1.

[0022] Simultaneously, a mapping table between indicator values ​​and fertility indices was developed. This table, based on the black soil fertility grade classification standards (e.g., extremely high, high, medium, low, extremely low), assigns a fertility index between 0 and 1 to different value ranges of each indicator. For example, for soil organic matter content, the fertility index is 1 when the content is greater than 60 g / kg; 0.8 when it is 50-60 g / kg; 0.6 when it is 40-50 g / kg; 0.4 when it is 30-40 g / kg; 0.2 when it is 20-30 g / kg; and 0.1 when it is less than 20 g / kg.

[0023] Secondly, the actual measured values ​​of various basic soil physicochemical properties are converted into corresponding first fertility indices according to the above mapping table, forming a first fertility index distribution, that is, each index has a corresponding fertility index value.

[0024] Finally, according to the preset index weight distribution, the index values ​​in the first fertility index distribution are weighted and summed. The calculation formula is: First comprehensive fertility index = (black soil layer thickness fertility index × 0.2) + (soil organic matter content fertility index × 0.3) + (soil pH value fertility index × 0.1) + (cation exchange capacity fertility index × 0.1) + (alkaline nitrogen content fertility index × 0.1) + (available phosphorus content fertility index × 0.1) + (available potassium content fertility index × 0.1).

[0025] The first comprehensive fertility index is calculated and used as the first fertility assessment result. The higher the value, the better the fertility status of the black soil based on its current physical and chemical properties.

[0026] For example, if the black soil layer thickness of a black soil sample in a target area is 50 cm, the corresponding fertility index is 0.9; the soil organic matter content is 45 g / kg, the corresponding fertility index is 0.6; the soil pH value is 6.5, the corresponding fertility index is 0.8; the cation exchange capacity is 25 cmol / kg, the corresponding fertility index is 0.7; the alkaline nitrogen content is 120 mg / kg, the corresponding fertility index is 0.7; the available phosphorus content is 20 mg / kg, the corresponding fertility index is 0.6; and the available potassium content is 150 mg / kg, the corresponding fertility index is 0.6.

[0027] The first comprehensive fertility index is calculated according to the above weights: (0.9×0.2)+(0.6×0.3)+(0.8×0.1)+(0.7×0.1)+(0.7×0.1)+(0.6×0.1)+(0.6×0.1)=0.18+0.18+0.08+0.07+0.07+0.06+0.06=0.7. Therefore, the first fertility assessment result for this area is 0.7.

[0028] S20: Collect microbial community monitoring data of black soil in the target area within a historical time range as indirect assessment data, and output the second fertility assessment results for black soil fertility assessment. Among them, the microbial community sequence, microbial community structure sequence, and microbial activity index sequence obtained from the continuous monitoring of black soil in the target area over a historical time range are used as microbial community monitoring data.

[0029] In this embodiment of the application, firstly, monitoring data of microbial communities in black soil in the target area are collected within a historical time range. The historical time range can be set to the past 3 years, 5 years or longer according to the assessment needs, in order to capture the dynamic change trend of the microbial community.

[0030] Furthermore, the microbial community monitoring data includes microbial taxa sequences, microbial community structure sequences, and microbial activity index sequences.

[0031] For microbial group sequences, high-throughput sequencing technology was used to continuously collect black soil samples at different time points, and microbial DNA was extracted and sequenced to obtain information on the species composition and relative abundance of bacteria, fungi and other microorganisms in the samples, forming a microbial group sequence dataset.

[0032] The microbial community structure sequence is characterized by calculating the α diversity index and β diversity distance. The α diversity index reflects the richness and evenness within the community, while the β diversity distance reflects the similarity or difference in community structure at different time points, thus constructing the microbial community structure sequence.

[0033] Microbial activity index sequences include soil respiration rate, microbial biomass carbon and nitrogen content, and enzyme activities such as urease, phosphatase, and sucrase. The index values ​​at each time point are obtained through corresponding laboratory measurement methods, such as the alkali absorption method for measuring soil respiration rate, the chloroform fumigation extraction method for measuring microbial biomass carbon and nitrogen, and colorimetric methods, to determine the time series data of enzyme activity.

[0034] After obtaining microbial community monitoring data, a microbial community characteristic fertility evaluation system was constructed to assess black soil fertility and output the second fertility assessment results.

[0035] Furthermore, a second fertility assessment result is output for black soil fertility assessment, including: Based on historical black soil fertility testing records, sample microbial group sequence sets, sample microbial community structure sequence sets, sample microbial activity index sequence sets, and sample black soil fertility index sets were collected. Using the sample microbial group sequence set, sample microbial community structure sequence set, and sample microbial activity index sequence set as inputs, and using the sample black soil fertility index set as supervision, a long short-term memory network is trained until convergence to generate an indirect black soil fertility assessment model. Using the black soil fertility indirect assessment model, a second comprehensive fertility index is obtained based on the indirect assessment data, which serves as the second fertility assessment result.

[0036] In this embodiment, firstly, black soil fertility monitoring records are collected, including black soil fertility indices for time points corresponding to microbial community monitoring data. Based on historical data, a sample dataset is constructed, where the input data consists of a sample microbial group sequence set, a sample microbial community structure sequence set, and a sample microbial activity index sequence set, with each sample containing observations over a time series; the output data is the corresponding sample black soil fertility index set, i.e., the actual black soil fertility index for each time point.

[0037] Secondly, because Long Short-Term Memory (LSTM) networks can effectively handle and learn long-term dependencies in time-series data, they were chosen as the basic model architecture. Data preprocessing, such as standardization and normalization, was performed on the sample microbial taxa sequence set, sample microbial community structure sequence set, and sample microbial activity index sequence set to meet the model training requirements.

[0038] Then, the preprocessed input data and the corresponding sample black soil fertility index set are input into the LSTM network. The network parameters, such as weights and biases, are continuously adjusted through the backpropagation algorithm to minimize the loss function, such as mean squared error loss, between the model's predicted values ​​and the actual sample black soil fertility index. The network is continuously trained until the model's performance on the validation set no longer improves, reaching a convergence state. At this point, a trained indirect evaluation model for black soil fertility is generated.

[0039] After model training is complete, the collected indirect assessment data of the target area—namely, the microbial taxonomic sequences, microbial community structure sequences, and microbial activity index sequences within the historical time range—is preprocessed and input into the black soil fertility indirect assessment model. The model analyzes and predicts the input microbial community monitoring data based on the learned dynamic mapping relationship between microbial community characteristics and black soil fertility, outputting a second comprehensive fertility index. This index serves as the second fertility assessment result, characterizing the indirect impact of the microbial community on black soil fertility.

[0040] For example, the steps for constructing and training an indirect assessment model for black soil fertility based on a long short-term memory network are as follows: First, data preparation involves collecting sample microbial group sequence sets, sample microbial community structure sequence sets, sample microbial activity index sequence sets, and sample black soil fertility index sets, based on historical black soil fertility test records.

[0041] Secondly, in model construction, the number of nodes in the input layer is equal to the dimension of the input features. For example, if there are 6 features in total, such as the sample microbial group sequence set, the sample microbial community structure sequence set, and the sample microbial activity index sequence set, then the input layer contains 6 nodes. Set 1-3 hidden layers, and adjust the number of nodes in each layer through experiments, such as 64, 32, etc. The activation function is ReLU. The output layer generally does not use an activation function. If the output takes 2 nodes, directly output continuous values.

[0042] Next, the model is trained, and the predicted black soil fertility index is used as the output, i.e., the second comprehensive fertility index, as the second fertility assessment result. The sample black soil fertility index set serves as the supervised label. The Adam optimizer and mean squared error loss function are used to construct the training framework. The initial learning rate is set to 0.001. A learning rate decay strategy is used to reduce the learning rate when the validation set loss no longer decreases. The batch size is set to 32, the total number of training epochs is 50, and an early stopping mechanism with a patience of 5 is introduced. When the validation set loss does not decrease for 5 consecutive epochs, the training process is automatically terminated, resulting in a trained indirect black soil fertility assessment model. This effectively avoids model overfitting while ensuring that the model reaches a convergent state.

[0043] S30: Evaluate the direct and indirect impacts based on the pre-set planting plan for the target area, and assign the first and second weights according to the degree of impact. In this embodiment, the preset planting plan for the target area refers to a crop planting plan pre-formulated based on factors such as the region's climate conditions, soil characteristics, and agricultural production goals, including the types of crops to be planted and the planting system. Since different planting plans have different priorities and impact paths on black soil fertility, the preset planting plan is used to evaluate the impact on both direct and indirect assessment data, thereby dynamically allocating the first and second weights.

[0044] Specifically, conducting a direct impact assessment refers to analyzing the direct dependence and sensitivity of the pre-set planting scheme to basic soil physicochemical properties. For example, if the pre-set planting scheme involves planting nitrogen-loving crops, the content of available nitrogen in the soil will have a higher impact on crop growth and final yield. This means that the importance of available nitrogen in the direct assessment data increases under this planting scheme, potentially leading to a corresponding increase in the first weight of the first fertility assessment result in the overall evaluation. Conversely, if the crop planted is sensitive to soil pH, the importance of soil pH in the direct assessment data will be prominent, similarly affecting the allocation of the first weight.

[0045] Furthermore, conducting indirect impact assessment refers to analyzing the indirect dependence and sensitivity of the pre-set planting scheme on the soil microbial community. For example, if the pre-set planting scheme adopts a long-term continuous cropping system, it may lead to a homogenization of the soil microbial community structure, a decrease in the abundance of beneficial microorganisms, and an accumulation of pathogenic microorganisms. In this case, the structure and activity of the microbial community are more critical to maintaining soil health and sustainable fertility. That is, the indirect assessment data has a higher impact on the sustainability of black soil fertility under this planting scheme, which may increase the second weight of the second fertility assessment result.

[0046] In summary, after evaluating both direct and indirect impacts, the first and second weights are specifically assigned based on the degree of influence of each. The quantification of influence can be achieved using the analytic hierarchy process (AHP).

[0047] Specifically, step S30 in the method includes: Obtain a preset planting plan for the target area, wherein the preset planting plan includes a planting system and crop types; Constrained by the planting system and crop type, and guided by black soil planting, agricultural big data is used to conduct a correlation analysis on the basic physical and chemical properties of the soil and the quality of crop harvest, and the first direct correlation coefficient is output as the first direct influence degree. Constrained by the planting system and crop type, and guided by black soil planting, agricultural big data is used to conduct a correlation analysis on the microbial community monitoring data and crop harvest quality, and output a second direct correlation coefficient as the second direct influence degree. Configure the first weight and the second weight based on the first direct influence degree and the second direct influence degree.

[0048] In this embodiment of the application, the specific content of the preset planting plan is first defined, including the type of crop to be planted, such as corn, soybeans, wheat, etc., and the planting system, such as crop rotation, continuous cropping, intercropping, relay cropping, etc.

[0049] Secondly, taking the elements of the planting plan as the core constraints and the actual planting and production goals of black soil as the guiding direction, the study utilizes historical planting case data, crop growth model data, and soil-crop interaction data stored in the agricultural big data platform. For the assessment of direct impact, basic soil physicochemical properties, such as black soil layer thickness, organic matter content, pH value, cation exchange capacity, available nitrogen, available phosphorus, and available potassium, are used as independent variables, while crop harvest quality, such as yield and quality indicators, are used as dependent variables. Correlation analysis is then conducted using a multiple linear regression mining algorithm.

[0050] Furthermore, a correlation analysis was conducted based on multiple linear regression. A linear regression model was constructed between basic soil physicochemical properties and crop harvest quality. The independent variables were the standardized values ​​of each basic soil physicochemical property, and the dependent variable was the comprehensive score of crop harvest quality, calculated by weighting factors such as yield, protein content, and starch content. The model parameters were estimated using the least squares method to obtain the regression coefficients of each basic soil physicochemical property on crop harvest quality. The absolute value of these regression coefficients reflects the degree of direct influence of the corresponding indicator on crop harvest quality. After normalization, the first direct correlation coefficient was obtained.

[0051] For example, if in a correlation analysis of a corn planting scheme, the regression coefficient of soil organic matter content is 0.4, the regression coefficient of available nitrogen content is 0.3, and the sum of the regression coefficients of other indicators is 0.3, then after normalization, the first direct correlation coefficient of soil organic matter content is 0.4, available nitrogen is 0.3, and so on, the sum of the first direct correlation coefficients of each item is 1. This set of coefficients represents the direct influence of the basic physical and chemical properties of the soil on the crop harvest quality under the preset planting scheme.

[0052] Similarly, for the assessment of indirect impact, key characteristics in the microbial community monitoring data, such as the relative abundance of specific functional microorganisms, α diversity index, and soil respiration rate, are used as independent variables, and the comprehensive crop harvest quality score is used as the dependent variable. A multiple linear regression model is constructed, and the second direct correlation coefficient is obtained by estimating and normalizing the regression coefficients. This is used to quantify the indirect impact of microbial community monitoring data on crop harvest quality.

[0053] Finally, the first and second weights are dynamically configured based on the sum of the first and second direct correlation coefficients. For example, if the sum of the first and second direct correlation coefficients is 0.6 and the sum of the second direct correlation coefficients is 0.4, then the first weight can be configured as 0.6 and the second weight as 0.4, so that the comprehensive fertility assessment results can more accurately reflect the actual contribution of black soil fertility under the preset planting scheme.

[0054] Further, configuring a first weight and a second weight based on the first direct influence degree and the second direct influence degree includes: Based on the microbial community monitoring data, predictive microbial community data for a future time range is obtained, wherein the predictive microbial community data includes predictive microbial taxonomic sequences, predictive microbial community structure sequences, and predictive microbial activity index sequences. Feature extraction is performed on the predicted microbial taxonomic sequence, predicted microbial community structure sequence, and predicted microbial activity index sequence to obtain the predicted taxonomic abundance sequence, predicted community structure feature sequence, and predicted functional activity feature sequence. The community structure features include species evenness, species diversity, community succession rate, community stability index, and fungal / bacterial abundance ratio. The functional activity features include carbon cycle gene abundance, nitrogen cycle gene abundance, phosphorus cycle gene abundance, and microbial metabolic entropy. The predicted taxonomic abundance sequence, predicted community structure characteristic sequence, and predicted functional activity characteristic sequence were analyzed for index volatility, and the microbial growth volatility was calculated by weighting the coefficients of variation of multiple indicators. The first weight and the second weight are configured based on the microbial growth fluctuation, the first direct influence, and the second direct influence.

[0055] In this embodiment, firstly, a pre-constructed microbial community dynamic prediction model is used to predict microbial community data over a future period based on historical microbial community monitoring data and environmental driving factors. This includes predicting the next growing season or the next 3-5 years, generating predicted microbial taxa sequences, predicted microbial community structure sequences, and predicted microbial activity index sequences. The predicted data reflects the potential trends in microbial community changes under the continued implementation of a pre-defined planting plan.

[0056] Secondly, for the predicted microbial taxa sequence, the relative abundance changes of specific functional microbial taxa related to soil nutrient transformation and disease suppression are extracted to form a predicted taxa abundance sequence, such as the abundance prediction values ​​of nitrogen-fixing bacteria, phosphate-solubilizing bacteria, and actinomycetes. For the predicted microbial community structure sequence, species evenness over future time ranges is calculated and extracted, such as species diversity extracted using the Shannon index, community succession rate extracted based on the magnitude of community structure changes at different time points, community stability index and community structure characteristics such as fungal / bacterial abundance ratio are predicted based on the resilience of community structure to disturbances, and a predicted community structure feature sequence is constructed.

[0057] Furthermore, the extraction of predictive functional activity feature sequences focuses on the abundance and metabolic activity of functional genes involved in key soil ecological processes by microorganisms, such as the predicted abundance of cellulase genes related to carbon cycle, nitrification / denitrification genes related to nitrogen cycle, and phosphatase genes related to phosphorus cycle, as well as the predicted value of microbial metabolic entropy, which characterizes the overall metabolic efficiency of microorganisms.

[0058] Subsequently, for each specific indicator in the predicted taxa abundance sequence, predicted community structure characteristic sequence, and predicted functional activity characteristic sequence, the coefficient of variation (i.e., the ratio of standard deviation to mean) is calculated over a future time range to measure the degree of fluctuation of the indicator.

[0059] For example, a large coefficient of variation in the predicted abundance of a certain nitrogen-fixing bacterium indicates that its abundance may fluctuate significantly in the future, having a high impact on the stability of soil nitrogen supply. Based on the importance of each indicator in maintaining soil fertility, different weights are assigned, and the coefficients of variation of each indicator are weighted and summed to obtain a comprehensive microbial growth volatility. Higher microbial growth volatility indicates greater uncertainty in the future microbial community and greater volatility in its potential impact on soil fertility.

[0060] Finally, when configuring the first and second weights, in addition to considering the first and second direct influence degrees, microbial growth volatility is used as an important adjustment factor. If microbial growth volatility is high, it indicates greater uncertainty in the future of the indirect assessment data, which may reduce its reliability for assessing black soil fertility. In this case, the second weight can be appropriately reduced, and the first weight can be increased accordingly. Conversely, if microbial growth volatility is low, the microbial community structure and function are relatively stable, and the reliability of the indirect assessment data is high. In this case, the second weight can be appropriately increased, or the original weight configuration can be maintained.

[0061] Specifically, based on the microbial community monitoring data, predictive microbial community data for a future time range is obtained, including: Based on the aforementioned basic physicochemical properties of the soil, a sample microbial community monitoring dataset was collected using agricultural big data. Historical microbial community monitoring data of different samples within a historical time range was also collected as sample predicted microbial community data to obtain the sample predicted microbial community dataset. Using the sample microbial community monitoring dataset as input and the sample predicted microbial community dataset as supervision, a long short-term memory network is trained until convergence to generate a microbial community prediction plugin. The microbial community monitoring data is input into the microbial community prediction plugin to predict and obtain predicted microbial community data for a future time range.

[0062] In this embodiment of the application, firstly, using basic soil physicochemical properties as constraints, such as black soil layer thickness, organic matter content, pH value, etc., a large number of sample microbial community monitoring data matching the indicators are screened and collected from the agricultural big data platform, including microbial groups, community structure and activity characteristics under different soil physicochemical backgrounds.

[0063] Meanwhile, for each sample of microbial community monitoring data, its continuous monitoring records over a historical time range, such as the past 3 or 5 years, are used as the sample predicted microbial community data for the corresponding sample. This simulates the dynamic change process of the microbial community over time, thereby constructing a sample predicted microbial community dataset with the current microbial community monitoring data as input and the historical predicted microbial community data as output.

[0064] Secondly, a Long Short-Term Memory (LSTM) network was chosen as the basic architecture for the microbial community prediction model. This network can effectively capture long-term dependencies in time-series data and is suitable for predicting microbial community evolution over time. The sample microbial community monitoring dataset was used as the model input, and the sample predicted microbial community dataset was used as the supervision label to construct training sample pairs. During model training, network parameters were set, such as 128 hidden layer neurons, and the time step was determined based on the time interval of historical data (e.g., 4 for quarterly sampling). The Adam optimizer was used, with the root mean square error (RMSE) as the loss function, and the initial learning rate set to 0.001. An early stopping mechanism was implemented to prevent overfitting. The training sample pairs were input into the LSTM network for iterative training until the model's loss on the validation set no longer decreased and tended to stabilize. At this point, the trained microbial community prediction plugin was generated.

[0065] Finally, the currently collected microbial community monitoring data in the target area, including microbial taxonomic sequences, community structure sequences, and activity index sequences, are input into the pre-trained microbial community prediction plugin. The plugin uses a built-in LSTM network to extract temporal features and predict trends from the input data, and automatically outputs predicted microbial community data for a future set time range, such as the next one growing season or two years. This data includes predicted microbial taxonomic sequences, predicted microbial community structure sequences, and predicted microbial activity index sequences, providing dynamic data support for subsequent analysis of microbial growth fluctuations and weight configuration.

[0066] Further, a first weight and a second weight are configured based on the microbial growth fluctuation, the first direct influence, and the second direct influence, including: The ratio of the microbial growth fluctuation to the preset microbial growth fluctuation benchmark value is used as the second weighting influence coefficient; The product of the second weight influence coefficient and the second initial weight is taken as the second weight, wherein the second initial weight is 0.2, and the second weight is greater than or equal to 0.1 and less than or equal to 0.4; The first weight is obtained by subtracting the second weight from 1.

[0067] In this embodiment of the application, firstly, a preset microbial growth fluctuation baseline value is set. This baseline value is determined based on the historical fluctuation data of the microbial community in the long-term fixed-location experiment of black soil. For example, the average value of the microbial community fluctuation under the typical planting system in the black soil area in the past 10 years is selected as the baseline value, which is assumed to be 0.25.

[0068] Secondly, the ratio of the microbial growth fluctuation to the baseline value is calculated to obtain the second weighted influence coefficient. The greater the microbial growth fluctuation, the lower the data reliability and the worse the accuracy of the assessment. For example, if the currently predicted microbial growth fluctuation is 0.3, then the second weighted influence coefficient is 0.3 / 0.25=1.2; if the microbial growth fluctuation is 0.2, then the second weighted influence coefficient is 0.2 / 0.25=0.8.

[0069] Next, the second initial weight is fixed at 0.2. The influence coefficient of the second weight is multiplied by the second initial weight to obtain the preliminary second weight. When the coefficient is 1.2, the preliminary second weight is 0.2 × 1.2 = 0.24; when the coefficient is 0.8, the preliminary second weight is 0.2 × 0.8 = 0.16.

[0070] Then, the initial second weight is subjected to boundary constraints to ensure that it falls within the interval [0.1, 0.4]. If the calculated initial second weight is greater than 0.4, it is adjusted to 0.4; if it is less than 0.1, it is adjusted to 0.1.

[0071] For example, if the microbial growth fluctuation is extremely high, such as 0.5, the influence coefficient of the second weight is 0.5 / 0.25=2, and the initial second weight is 0.4, which does not require adjustment. If the microbial growth fluctuation is extremely low, such as 0.05, the coefficient is 0.2, and the initial second weight is 0.04, which is then adjusted to 0.1. Finally, subtracting the constrained second weight from 1 yields the first weight.

[0072] When the second weight is 0.24, the first weight is 1-0.24=0.76; when the second weight is 0.16, the first weight is 0.84; when the second weight is constrained to 0.4, the first weight is 0.6; when the second weight is constrained to 0.1, the first weight is 0.9.

[0073] The above calculation method enables dynamic adjustment of weights based on the fluctuation of microbial growth, which not only ensures the standardization of weight allocation but also flexibly responds to the impact of changes in the stability of the microbial community on fertility assessment.

[0074] S40: The first fertility assessment result and the second fertility assessment result are weighted and fused according to the first weight and the second weight to output the final black soil fertility assessment result.

[0075] In this embodiment, the first fertility assessment result is multiplied by the first weight to obtain the first assessment component; the second fertility assessment result is multiplied by the second weight to obtain the second assessment component. The first assessment component and the second assessment component are added together to obtain the final black soil fertility assessment result.

[0076] For example, if the full score is 100 points, the first fertility assessment result is 85 points with a first weight of 0.76, the second fertility assessment result is 80 points with a second weight of 0.24, then the first assessment component is 85 × 0.76 = 64.6 points, the second assessment component is 80 × 0.24 = 19.2 points, and the final black soil fertility assessment result is 64.6 + 19.2 = 83.8 points.

[0077] This result integrates the direct influence of basic soil physicochemical properties and the indirect influence of microbial communities. Through dynamic weight adjustment, the contribution of microbial communities to the assessment results is more significant when the stability of the microbial community is high. When the microbial community fluctuates greatly and the uncertainty is high, the weight of basic soil physicochemical properties is increased accordingly, thereby ensuring the accuracy and reliability of the assessment results under different scenarios.

[0078] The final evaluation results are presented in the form of specific scores, grades, or detailed reports that include the contribution of each influencing factor, providing a scientific basis for black soil fertility management, planting program optimization, and sustainable agricultural decision-making.

[0079] In summary, compared to existing technologies, this application achieves a multi-dimensional and hierarchical quantitative assessment of black soil fertility by constructing a direct correlation model between basic soil physicochemical properties and crop harvest quality, and a correlation model where microbial community monitoring data indirectly affects crop harvest quality through its influence on soil physicochemical properties. By introducing a first and a second direct correlation coefficient, and dynamically configuring the first and second weights in conjunction with microbial growth fluctuations, the current state of soil physicochemical properties and the potential influence and future trends of the microbial community are effectively integrated, overcoming the limitations of traditional assessment methods that only focus on a single factor or static assessment.

[0080] In summary, the embodiments of this application have at least the following technical effects: This application provides a method for assessing black soil fertility by combining microbial community detection. First, direct assessment data is collected and a first fertility assessment result is output, directly reflecting the contribution of the current physicochemical properties of the black soil to fertility. Simultaneously, indirect assessment data of the target area is collected and a second fertility assessment result is output, taking into account the role of the microbial community in the black soil ecosystem, thus more accurately assessing its indirect impact on black soil fertility. Next, direct and indirect impacts are evaluated separately according to a preset planting plan for the target area, with first and second weights configured based on the degree of impact, making the assessment method more targeted and practical. Finally, the first and second fertility assessment results are weighted and fused according to the first and second weights to output the final black soil fertility assessment result. This comprehensively considers the current physicochemical state of the soil and the potential fertility driven by microorganisms, outputting an assessment result that reflects the complex characteristics and overall fertility status of the black soil ecosystem, providing a more reliable scientific basis for the refined management and sustainable utilization of black soil. Through the above technical solution, this application realizes a multi-dimensional integrated assessment of basic soil physicochemical properties and microbial community characteristics, breaking through the limitations of traditional methods that rely solely on physicochemical indicators, and significantly improving the comprehensiveness, accuracy and dynamic prediction capability of black soil fertility assessment.

[0081] Example 2, as Figure 2 As shown, based on the same inventive concept as the black soil fertility assessment method combining microbial community detection provided in Embodiment 1, this application also provides a black soil fertility assessment system combining microbial community detection, including: The direct impact assessment module 11 is used to collect the basic physical and chemical properties of black soil in the target area as direct assessment data, and to perform black soil fertility assessment and output the first fertility assessment result. The indirect impact assessment module 12 is used to collect microbial community monitoring data of black soil in the target area within a historical time range as indirect assessment data, and to perform black soil fertility assessment and output the second fertility assessment result. The influence weight configuration module 13 is used to evaluate the direct and indirect influences according to the preset planting plan of the target area, and to configure the first and second weights according to the degree of influence. The evaluation result output module 14 is used to perform weighted fusion of the first fertility evaluation result and the second fertility evaluation result according to the first weight and the second weight, and output the final black soil fertility evaluation result.

[0082] Furthermore, in one embodiment of the application, the basic physicochemical properties of the soil include at least the thickness of the black soil layer, the content of soil organic matter, the soil pH value, the cation exchange capacity, the content of available nitrogen, the content of available phosphorus, and the content of available potassium.

[0083] Furthermore, in one embodiment of the application, the black soil fertility assessment outputs a first fertility assessment result, including: Configure a direct soil fertility assessment system, wherein the direct soil fertility assessment system includes an index weight distribution and an index value-fertility index mapping table; Based on the index value-fertility index mapping table, the first fertility index distribution is output based on the black soil layer thickness, soil organic matter content, soil pH value, cation exchange capacity, alkaline nitrogen content, available phosphorus content and available potassium content. The first fertility index distribution is weighted and fused according to the index weight distribution to obtain the first comprehensive fertility index, which is used as the first fertility assessment result.

[0084] Furthermore, in one embodiment of the application, the microbial community sequence, microbial community structure sequence, and microbial activity index sequence of black soil in the target area are continuously monitored over a historical time range and used as microbial community monitoring data.

[0085] Furthermore, in one embodiment of the application, the black soil fertility assessment outputs a second fertility assessment result, including: Based on historical black soil fertility testing records, sample microbial group sequence sets, sample microbial community structure sequence sets, sample microbial activity index sequence sets, and sample black soil fertility index sets were collected. Using the sample microbial group sequence set, sample microbial community structure sequence set, and sample microbial activity index sequence set as inputs, and using the sample black soil fertility index set as supervision, a long short-term memory network is trained until convergence to generate an indirect black soil fertility assessment model. Using the black soil fertility indirect assessment model, a second comprehensive fertility index is obtained based on the indirect assessment data, which serves as the second fertility assessment result.

[0086] In one embodiment, the influence weight configuration module 13 is specifically used for: Obtain a preset planting plan for the target area, wherein the preset planting plan includes a planting system and crop types; Constrained by the planting system and crop type, and guided by black soil planting, agricultural big data is used to conduct a correlation analysis on the basic physical and chemical properties of the soil and the quality of crop harvest, and the first direct correlation coefficient is output as the first direct influence degree. Constrained by the planting system and crop type, and guided by black soil planting, agricultural big data is used to conduct a correlation analysis on the microbial community monitoring data and crop harvest quality, and output a second direct correlation coefficient as the second direct influence degree. Configure the first weight and the second weight based on the first direct influence degree and the second direct influence degree.

[0087] Further, configuring a first weight and a second weight based on the first direct influence degree and the second direct influence degree includes: Based on the microbial community monitoring data, predictive microbial community data for a future time range is obtained, wherein the predictive microbial community data includes predictive microbial taxonomic sequences, predictive microbial community structure sequences, and predictive microbial activity index sequences. Feature extraction is performed on the predicted microbial taxonomic sequence, predicted microbial community structure sequence, and predicted microbial activity index sequence to obtain the predicted taxonomic abundance sequence, predicted community structure feature sequence, and predicted functional activity feature sequence. The community structure features include species evenness, species diversity, community succession rate, community stability index, and fungal / bacterial abundance ratio. The functional activity features include carbon cycle gene abundance, nitrogen cycle gene abundance, phosphorus cycle gene abundance, and microbial metabolic entropy. The predicted taxonomic abundance sequence, predicted community structure characteristic sequence, and predicted functional activity characteristic sequence were analyzed for index volatility, and the microbial growth volatility was calculated by weighting the coefficients of variation of multiple indicators. The first weight and the second weight are configured based on the microbial growth fluctuation, the first direct influence, and the second direct influence.

[0088] Furthermore, based on the microbial community monitoring data, predictive microbial community data for a future time range is obtained, including: Based on the aforementioned basic physicochemical properties of the soil, a sample microbial community monitoring dataset was collected using agricultural big data. Historical microbial community monitoring data of different samples within a historical time range was also collected as sample predicted microbial community data to obtain the sample predicted microbial community dataset. Using the sample microbial community monitoring dataset as input and the sample predicted microbial community dataset as supervision, a long short-term memory network is trained until convergence to generate a microbial community prediction plugin. The microbial community monitoring data is input into the microbial community prediction plugin to predict and obtain predicted microbial community data for a future time range.

[0089] Further, in one embodiment of the application, configuring a first weight and a second weight based on the microbial growth fluctuation, the first direct influence, and the second direct influence includes: The ratio of the microbial growth fluctuation to the preset microbial growth fluctuation benchmark value is used as the second weighting influence coefficient; The product of the second weight influence coefficient and the second initial weight is taken as the second weight, wherein the second initial weight is 0.2, and the second weight is greater than or equal to 0.1 and less than or equal to 0.4; The first weight is obtained by subtracting the second weight from 1.

[0090] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0091] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0092] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A method for assessing black soil fertility by combining microbial community detection, characterized in that, The methods include: The basic physical and chemical properties of black soil in the target area were collected as direct evaluation data, and the first fertility evaluation result was output after black soil fertility evaluation. Microbial community monitoring data of black soil in the target area over a historical time range were collected as indirect assessment data, and black soil fertility assessment was performed to output a second fertility assessment result. Based on the pre-set planting plan for the target area, the direct and indirect impacts are evaluated, and the first and second weights are assigned according to the degree of impact. The first fertility assessment result and the second fertility assessment result are weighted and fused according to the first weight and the second weight to output the final black soil fertility assessment result.

2. The method for assessing black soil fertility by combining microbial community detection according to claim 1, characterized in that, The basic physicochemical properties of the soil include at least the thickness of the black soil layer, soil organic matter content, soil pH value, cation exchange capacity, alkaline nitrogen content, available phosphorus content, and available potassium content.

3. The method for assessing black soil fertility by combining microbial community detection according to claim 2, characterized in that, The first fertility assessment result for black soil is output, including: Configure a direct soil fertility assessment system, wherein the direct soil fertility assessment system includes an index weight distribution and an index value-fertility index mapping table; Based on the index value-fertility index mapping table, the first fertility index distribution is output based on the black soil layer thickness, soil organic matter content, soil pH value, cation exchange capacity, alkaline nitrogen content, available phosphorus content and available potassium content. The first fertility index distribution is weighted and fused according to the index weight distribution to obtain the first comprehensive fertility index, which is used as the first fertility assessment result.

4. The method for assessing black soil fertility by combining microbial community detection according to claim 1, characterized in that, Microbial community sequences, microbial community structure sequences, and microbial activity index sequences were continuously monitored in the target area over a historical period and collected as microbial community monitoring data.

5. The method for assessing black soil fertility by combining microbial community detection according to claim 4, characterized in that, The black soil fertility assessment outputs a second fertility assessment result, including: Based on historical black soil fertility testing records, sample microbial group sequence sets, sample microbial community structure sequence sets, sample microbial activity index sequence sets, and sample black soil fertility index sets were collected. Using the sample microbial group sequence set, sample microbial community structure sequence set, and sample microbial activity index sequence set as inputs, and using the sample black soil fertility index set as supervision, a long short-term memory network is trained until convergence to generate an indirect black soil fertility assessment model. Using the black soil fertility indirect assessment model, a second comprehensive fertility index is obtained based on the indirect assessment data, which serves as the second fertility assessment result.

6. The method for assessing black soil fertility by combining microbial community detection according to claim 1, characterized in that, Based on the pre-set planting plan for the target area, direct and indirect impacts were evaluated separately. A first and second weight were assigned according to the degree of impact, including: Obtain a preset planting plan for the target area, wherein the preset planting plan includes a planting system and crop types; Constrained by the planting system and crop type, and guided by black soil planting, agricultural big data is used to conduct a correlation analysis on the basic physical and chemical properties of the soil and the quality of crop harvest, and the first direct correlation coefficient is output as the first direct influence degree. Constrained by the planting system and crop type, and guided by black soil planting, agricultural big data is used to conduct a correlation analysis on the microbial community monitoring data and crop harvest quality, and output a second direct correlation coefficient as the second direct influence degree. Configure the first weight and the second weight based on the first direct influence degree and the second direct influence degree.

7. The method for assessing black soil fertility by combining microbial community detection according to claim 6, characterized in that, Configure a first weight and a second weight based on the first direct influence degree and the second direct influence degree, including: Based on the microbial community monitoring data, predictive microbial community data for a future time range is obtained, wherein the predicted microbial community data includes predicted microbial taxonomic sequences, predicted microbial community structure sequences, and predicted microbial activity index sequences. Feature extraction is performed on the predicted microbial taxonomic sequence, predicted microbial community structure sequence, and predicted microbial activity index sequence to obtain the predicted taxonomic abundance sequence, predicted community structure feature sequence, and predicted functional activity feature sequence. The community structure features include species evenness, species diversity, community succession rate, community stability index, and fungal / bacterial abundance ratio. The functional activity features include carbon cycle gene abundance, nitrogen cycle gene abundance, phosphorus cycle gene abundance, and microbial metabolic entropy. The predicted taxonomic abundance sequence, predicted community structure characteristic sequence, and predicted functional activity characteristic sequence were analyzed for index volatility, and the microbial growth volatility was calculated by weighting the coefficients of variation of multiple indicators. The first weight and the second weight are configured based on the microbial growth fluctuation, the first direct influence, and the second direct influence.

8. The method for assessing black soil fertility by combining microbial community detection according to claim 7, characterized in that, Based on the aforementioned microbial community monitoring data, predictive microbial community data for future timeframes are obtained, including: Based on the aforementioned basic physicochemical properties of the soil, a sample microbial community monitoring dataset was collected using agricultural big data. Historical microbial community monitoring data of different samples within a historical time range was also collected as sample predicted microbial community data to obtain the sample predicted microbial community dataset. Using the sample microbial community monitoring dataset as input and the sample predicted microbial community dataset as supervision, a long short-term memory network is trained until convergence to generate a microbial community prediction plugin. The microbial community monitoring data is input into the microbial community prediction plugin to predict and obtain predicted microbial community data for a future time range.

9. The method for assessing black soil fertility by combining microbial community detection according to claim 7, characterized in that, The first weight and the second weight are configured based on the microbial growth fluctuation, the first direct influence, and the second direct influence, including: The ratio of the microbial growth fluctuation to the preset microbial growth fluctuation benchmark value is used as the second weighting influence coefficient; The product of the second weight influence coefficient and the second initial weight is taken as the second weight, wherein the second initial weight is 0.2, and the second weight is greater than or equal to 0.1 and less than or equal to 0.4; The first weight is obtained by subtracting the second weight from 1.

10. A black soil fertility assessment system combining microbial community detection, characterized in that, The method for evaluating black soil fertility in conjunction with microbial community detection as described in any one of claims 1-9 includes: The direct impact assessment module is used to collect basic physical and chemical properties of black soil in the target area as direct assessment data, and to output the first fertility assessment result for black soil fertility assessment. The indirect impact assessment module is used to collect microbial community monitoring data of black soil in the target area over a historical time range as indirect assessment data, and to perform black soil fertility assessment and output the second fertility assessment results. The influence weight configuration module is used to evaluate the direct and indirect influences based on the preset planting plan for the target area, and to configure the first and second weights according to the degree of influence. The evaluation result output module is used to perform weighted fusion of the first fertility evaluation result and the second fertility evaluation result according to the first weight and the second weight, and output the final black soil fertility evaluation result.

Images