CN105699624A - Soil organic carbon storage amount estimation method based on soil genetic horizon thickness prediction - Google Patents

Abstract

The invention relates to a soil organic carbon storage amount estimation method based on soil genetic horizon thickness prediction. Better technological ideas are provided for continuous prediction of soil mass distributed in the horizontal dimension in consideration of soil attributes through unstructured information encapsulation of the genetic horizon; with the adoption of the technologies of genetic horizon merging, prediction and recalculation, the limitation of a conventional prediction method in continuous description of soil mass is corrected while character information of the soil genetic horizon is not missed, a universal soil organic carbon storage amount estimation technology for standard description and accurate prediction is realized, and the method has wide industrial application prospect in engineering survey of agricultural application, environmental protection, territorial resources and other relevant departments.

Description

A kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction

Technical field

The present invention relates to a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction, belong to metering soil technology field。

Background technology

Soil carbon mainly includes soil organic matter and soil inorganic carbon, and soil inorganic carbon storehouse (carbonate carbon) is relatively stable。Soil organic matter is mainly distributed in 1 meter of degree of depth soil body, and the physics of soil, chemical property are had material impact, and directly affect soil quality。One of important indicator as fertility evaluation, the continuous reduction of soil organic matter can directly result in the barren problem of agricultural cultivating soil。Soil Carbon Stock is not only one of social problem that the investigation of ecological public good pays close attention to, and is also global basic problem。The global climate that existing numerous application both at home and abroad produce, environmental protection department starts that survival and development of mankind is faced increases warm problem and gives extensive concern。National Carbon budget computing technique is also proposed different demand by relevant Edge Actual, and Soil Carbon Stock estimation has become as the great ecology and the production of environment technical problem that affect national economy, diplomacy。

Soil Organic Carbon Density refers to the organic C storage of certain area fixing soil degree of depth, and unit is kg/m2, and the Soil Carbon Stock of certain area is the product of Organic Carbon Density and region area, and unit is kg。Conventional organic C storage estimating techniques mainly include soil types method, vegetation pattern method, life zone typological method, modelling, dependency relation statistical method, GIS (GIS-Geographic Information System) spatial prediction method。It is the widely used a kind of spatial prediction mode of modern digital soil cartography based on the space predicting method of GIS。It is different from traditional soil investigation and draughtsmanship, the method adopts the quantization Soil-landscape model that domestic and international soil scientist is extensively affirmed, Remote Sensing Image Processing Technology, digital Terrain Analysis technology, GIS Spatial Data Analysis and soil investigation technology can be effectively integrated, by the analysis of view information being predicted the spatial distribution of Soil Organic Carbon Density。The key operation flow process of GIS space predicting method is by the field sampling of diverse geographic location, lab analysis soil physical chemistry attribute, sets up organic C storage appraising model and then the carbon reserves of estimation target area based on environmental variable and known sampling point Soil attribute data。

At present, conventional carbon reserves estimating techniques are based primarily upon following two production calculation pattern: first calculate and predict (CTM) again, first predict and calculate (MTC) again, here " calculating " is the Soil Organic Carbon Density (based on soil organic carbon, the soil weight, soil gravel concentration, thickness of soil) calculating pedon, and " prediction " is the spatial distribution of spatial interpolation prediction target soil attribute。Prior art application is main adopts a kind of computation schema, and contrast application model still rarely has data see。

In recent years, along with computer, remote sensing, soil investigation, surveying and mapping technology development, the particularly fast development of high-definition remote sensing, Digital Mapping technology, the data acquisition that region class other environmental variable information comprises tested more heterogeneous details is possibly realized, by the soil-like point data of spatial spreading, quantitatively, objective, in real time, exactly Simulation of Complex Landscape Region region soil attribute space be distributed to concrete production of soil information relevant departments and have higher requirement, this also develops into the important development direction of international high precision soil carbon reserves estimation。

Genetic horizon is the important evidence differentiating soil types, and has a series of quantitative explanation in nature。As the fundamentals of forecasting of organic C storage estimation, the actual distribution situation (including level dimension and vertical dimension) of Soil-landscape model hypothesis soil attribute is closely-related with view attribute。Particularly in the vertical dimension distribution aspect of soil attribute, genetic horizon often has the soil attribute feature of homogenizing。But, existing Soil Carbon Stock appraising model both faces towards the soil attribute prediction of constant depth。Global soil digital mapping plan (GSM) joint agreement specifies that the production soil layer thickness of special topic pedological map all adopts the fixed form of 0-5,5-15,15-30,30-60,60-100,100-200cm。

At present, Soil Carbon Stock estimation have some limitations on Complex Landscape region and production technology, be specifically summed up following some:

(1) computation schema of fixing layer thicknesses is from have ignored the theoretical model that soil occurs to a certain degree, has certain limitation in production and processing。The complexity made a variation due to the reason soil vertical dimension space of landform, weather, artificial disturbance is far beyond the analog capability of existing computing technique。If the attribute information disappearance of soil sampling point, when especially less and sampling point the spatial representative of sampling point total amount is poor, practical application is difficult to the Soil Carbon Stock of accurate estimation area。It is true that this is also cause the drawing of existing digital soil to express the one of the main reasons that there is critical regions property indeterminacy phenomenon。Nanjing Soil Inst., Chinese Academy of Sciences and soil information reference center, the ISRIC world (Holland) all have been pointed out, and the application framework of modern digital soil cartography technology should not ignore the soil information feature of genetic horizon。

(2) in actual soil information production application process, existing production technology is difficult to consider soil genetic horizon thickness。Existing soil investigation engineering still adopts traditional field investigation pattern, does not simply fail to obtain the soil physical chemistry attribute data of large-scale area and descriptive information in real time, also tends to the technical capability being limited to investigation business models with the investigation operation of personnel。Soil taxonomy relates to many soil diagnosis characteristics, and different diagnostic features has quantitatively theoretical with soil generation model qualitatively, brings very big difficulty to the actual production of technical staff。Therefore, existing basic Soil Carbon Stock estimation engineering often lacks the merging techniques considering soil genetic horizon feature。

(3) prior art focuses mostly in the fixing computation schema of one, lacks strengthening comparative study。Widely apply case it has been shown that different production models (CTM, MTC) often shows totally different regional precision problem in different model hypothesis, view Sudden change region both at home and abroad。Therefore, the different computation schema of strengthening contrast has important directive significance for concrete production link。Meanwhile, the non-structured data structure of soil data is cause the incomplete major reason of existing production technology with the metering soil characteristic of genetic horizon。

The deficiency of the above existing Soil Carbon Stock estimating techniques, different application department production and processing soil information product and engineer applied bring bigger difficulty, estimate compared with the organic C storage in the regions such as complex region, highly variation forest zone in such as remote mountain areas, man's activity and engineer applied can bring potential erroneous decision support, and then directly contribute loss to national economy planning。

Summary of the invention

The technical problem to be solved is to provide the brand-new architecture design of a kind of employing, it is possible to be effectively improved the Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction of Soil Carbon Stock estimation precision and estimation work efficiency。

The present invention is to solve above-mentioned technical problem by the following technical solutions: the present invention devises a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction, comprises the steps:

Step 001., for target soil region, arranges each sampling point position, and the quantity adding up sampling point position is I, subsequently into step 002；

Step 002. is in target soil region, it is respectively directed to each sampling point position, obtain the environmental information of sampling point position, if the degree of depth of this sampling point position soil region straight down is be more than or equal to predetermined depth L simultaneously, then obtain the hatch region of this sampling point position predetermined depth L straight down；If the degree of depth of this sampling point position soil region straight down is less than predetermined depth L, then obtain the hatch region of this sampling point position soil straight down；And the job requirements according to Soil Taxonomy, divide soil genetic horizon for this hatch region, it is thus achieved that each soil genetic horizon corresponding to this sampling point position；Obtain thickness information and the soil morphology information of this sampling point position each soil genetic horizon corresponding more respectively；And then obtain the environmental information of each sampling point position in target soil region respectively, and thickness information and the soil morphology information of each soil genetic horizon corresponding is distinguished in each sampling point position, subsequently enters step 003；

Step 003. is respectively directed to each sampling point position in target soil region, in each soil genetic horizon corresponding to sampling point position, gather the pedotheque of preset quality respectively, and measure the soil attribute information obtaining each soil genetic horizon respectively, and then obtain the soil attribute information of each sampling point position each soil genetic horizon corresponding respectively in target soil region, subsequently into step 004；Wherein, soil attribute information include soil organic carbon, more than the gravel concentration of preset diameters and the soil weight；

Step 004. is respectively directed to each sampling point position in target soil region, the soil attribute information of each soil genetic horizon, thickness information corresponding to sampling point position, it is thus achieved that the Soil Organic Carbon Density measured value SOCD of each sampling point position_i, i={1 ..., I}, subsequently into step 005；

Step 005. is respectively directed to each sampling point position in target soil region, each soil genetic horizon corresponding to sampling point position carries out merger process, obtain each merger layer corresponding to this sampling point position, then each soil genetic horizon of correspondence is distinguished according to each merger layer of this sampling point position, soil attribute information for this each soil genetic horizon of sampling point position is weighted, obtain the soil attribute information of this sampling point position each merger layer corresponding, and the thickness information for this each soil genetic horizon of sampling point position carries out read group total, obtain the thickness of this sampling point position each merger layer corresponding；And then obtain each merger layer that each sampling point position is corresponding respectively in target soil region respectively, and the soil attribute information of each merger layer and thickness information, enter back into step 006；

Step 006. is respectively directed to each sampling point position in target soil region, judge that whether the thickness sum of sampling point position each merger layer corresponding is less than predetermined depth L, it is then under each merger layer corresponding to this sampling point position, with horizon d, merger layer is set, make the thickness sum of this sampling point position each merger layer corresponding equal to predetermined depth L, and enter step 007；Otherwise it is directly entered step 007；

Step 007. obtains the kind of merger layer corresponding to all sampling point positions in target soil region, constitutes target soil region merger layer kind set；According to target soil region merger layer kind set, merger layer corresponding to each sampling point position in target soil region carries out unified operation, make the kind of merger layer corresponding to each sampling point position in target soil region mutually the same, and enter step 008；

Step 008. adopts linear congruent algorithm, and the sampling point position in target soil region is divided into prediction sampling point location sets and checking sampling point location sets, and according to the Soil Organic Carbon Density measured value SOCD of each sampling point position in target soil region_i, it is thus achieved that the Soil Organic Carbon Density measured value of each checking sampling point position in checking sampling point location setsConstitute the Soil Organic Carbon Density measured value set V of each checking sampling point position in checking sampling point location sets；And enter step 009, and wherein, i₂=1 ..., I₂, wherein, I₂For the quantity of checking sampling point position in checking sampling point location sets, it was predicted that in sampling point location sets, the quantity of prediction sampling point position is I₁, I₁>I₂；

Step 009. is according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the thickness information of each prediction sampling point position each merger layer corresponding respectively, adopt random forest method, each forecast model that it is target with each merger layer thickness information respectively that training obtains, constitutes the first forecast model set；

Simultaneously, according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the soil attribute information of each prediction sampling point position each merger layer corresponding respectively, adopt random forest method, each forecast model that it is target with each merger layer soil attribute information respectively that training obtains, constitutes the second forecast model set；Subsequently into step 010；

Step 010. is respectively directed to each checking sampling point position in checking sampling point location sets, environmental information according to checking sampling point position, by each forecast model in the first forecast model set, obtain the first thickness prediction information of this checking sampling point position each merger layer corresponding respectively respectively；Simultaneously, it is respectively directed to each checking sampling point position in checking sampling point location sets, environmental information according to checking sampling point position, by each forecast model in the second forecast model set, obtains this checking sampling point position respectively and is distinguished the first prediction soil attribute information of each merger layer corresponding；And then obtain the first thickness prediction information and the first prediction soil attribute information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets；Subsequently into step 011；Wherein, the first prediction soil attribute information includes the first organic carbon content information of forecasting, the first soil weight information of forecasting and the first gravel prediction content more than preset diameters；

Step 011. is respectively directed to each checking sampling point position in checking sampling point location sets, it is thus achieved that L verifies the drawing coefficient Rc of sampling point position each merger layer the first thickness prediction information sum corresponding relatively；Then according to the first prediction soil attribute information, the first thickness prediction information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets, it is thus achieved that the first Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the first Soil Organic Carbon Density predictive value set PMTC-D of each checking sampling point position in checking sampling point location sets；Subsequently into step 012；

Step 012. is according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the Soil Organic Carbon Density measured value of each prediction sampling point position, adopt random forest method, the 3rd forecast model that it is target with Soil Organic Carbon Density measured value that training obtains；Then each checking sampling point position in checking sampling point location sets it is respectively directed to, environmental information according to checking sampling point position, by the 3rd forecast model, obtain the second Soil Organic Carbon Density predictive value of this checking sampling point position, and then obtain the second Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location sets, constitute the second Soil Organic Carbon Density predictive value set PCTM of each checking sampling point position in checking sampling point location sets；Subsequently into step 013；

Step 013. is for sampling point positions all in target soil region, unified by presetting division rule, based on L, each merger layer corresponding to sampling point position is divided into each matching layer, the quantity of each matching layer corresponding to sampling point position and matching layer is identical, and obtains the thickness information of each matching layer；Then each prediction sampling point position in prediction sampling point location sets it is respectively directed to, each merger layer of correspondence is distinguished according to each matching layer of prediction sampling point position, soil attribute information for this prediction sampling point position each merger layer sampling genetic horizon is fitted, it is thus achieved that the soil attribute information of each matching layer in prediction sampling point position；And then obtain the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, and enter step 014；

Step 014. is according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the soil attribute information of each prediction sampling point position each matching layer corresponding respectively, adopt random forest method, each forecast model that it is target with each matching layer soil attribute information that training obtains, constitutes the 4th forecast model set；Then each checking sampling point position in checking sampling point location sets it is respectively directed to, environmental information according to checking sampling point position, by each forecast model in the 4th forecast model set, obtain the 3rd soil attribute information of forecasting of this checking sampling point position each matching layer corresponding respectively, and then obtain the 3rd soil attribute information of forecasting of each checking sampling point position each matching layer corresponding respectively in checking sampling point location sets, subsequently into step 015；Wherein, the 3rd prediction soil attribute information includes the 3rd organic carbon content information of forecasting, the 3rd soil weight information of forecasting and the 3rd gravel prediction content more than preset diameters；

Step 015. is the 3rd soil attribute information of forecasting of each matching layer corresponding to each checking sampling point position in checking sampling point location sets, and thickness information, it is thus achieved that the 3rd Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the 3rd Soil Organic Carbon Density predictive value set PMTC-F of each checking sampling point position in checking sampling point location sets；Subsequently into step 016；

Step 016. carries out accuracy test according to Soil Organic Carbon Density measured value set V, obtain the optimum Soil Organic Carbon Density predictive value set in the first Soil Organic Carbon Density predictive value set PMTC-D, the second Soil Organic Carbon Density predictive value set PCTM and the three Soil Organic Carbon Density predictive value set PMTC-F, and enter step 017；

Step 017. is by target soil discrete region space lattice data, using the sampled data of the respectively corresponding genetic horizon of all sampling point positions in target soil region as prediction data set, if optimum Soil Organic Carbon Density predictive value set is the first Soil Organic Carbon Density predictive value set PMTC-D, then step 009 to the method for step 011 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；If optimum Soil Organic Carbon Density predictive value set is the second Soil Organic Carbon Density predictive value set PCTM, then the method for step 012 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；If optimum Soil Organic Carbon Density predictive value set is the 3rd Soil Organic Carbon Density predictive value set PMTC-F, then step 013 to the method for step 015 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；Subsequently into step 018；

The step 018. Soil Organic Carbon Density spatial distribution raster data according to target soil region, it is thus achieved that the Soil Carbon Stock in target soil region。

As a preferred technical solution of the present invention, described step 001 specifically includes following steps:

Step 00101. obtains the soil types scattergram in target soil region, Soil Utilization scattergram and soil-geological scattergram, and soil types scattergram, Soil Utilization scattergram and soil-geological scattergram are carried out space overlapping, obtain target soil district multiple-layer stacked figure, subsequently into step 00102；

Step 00102. obtains the area of each figure spot region in target soil district multiple-layer stacked figure respectively, and adds up and obtain area ratio and exceed each figure spot region of figure spot area threshold value；Then being respectively directed to this each figure spot region, according to accessible detecting requirement on the spot, arrange sampling point position, the quantity of statistics sampling point position is I, enters back into step 002。

As a preferred technical solution of the present invention: described environmental information includes vegetation coverage, rock exposure area ratio, landform, important symbol thing, earth's surface coarse fragment size, ground crack situation, earth's surface salting stain information。

As a preferred technical solution of the present invention: described soil morphology information represents soil dry-wet situation, soil color, root system information, hole information, sample structure, speckle component, warty tuberculosis material, degree of consolidation, lime reaction information。

As a preferred technical solution of the present invention, described step 004 specifically includes following operation:

It is respectively directed to each sampling point position in target soil region, the soil attribute information of each soil genetic horizon, thickness information corresponding to sampling point position, as follows:

${\mathrm{SOCD}}_i=\frac{\underset{n_i=1}{\overset{N_i}{\Σ}}({\mathrm{SOC}}_{n_i}\×{\mathrm{BD}}_{n_i}\×(1-{\mathrm{Gr}}_{n_i})\×T_{n_i})}{L}$

Obtain the Soil Organic Carbon Density measured value SOCD of each sampling point position_i, wherein, i={1 ..., I}, n_i=1 ... N_i, n_iRepresent the n-th soil genetic horizon corresponding to i-th sampling point position, N_iRepresent the sum of soil genetic horizon corresponding to i-th sampling point position；Represent the organic carbon content of the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the soil weight of the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the gravel concentration more than preset diameters in the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the thickness information of the n-th soil genetic horizon corresponding to i-th sampling point position, subsequently into step 005。

As a preferred technical solution of the present invention, in described step 005, it is respectively directed to each sampling point position in target soil region, each soil genetic horizon corresponding to sampling point position carries out merger process, obtain the process of each merger layer corresponding to this sampling point position, specifically include following steps:

Step 00501. is respectively directed to each sampling point position in target soil region, the soil morphology information of each soil genetic horizon corresponding to sampling point position, and Soil Taxonomy, it is thus achieved that the characteristic expression symbol that each soil genetic horizon of this sampling point position is corresponding respectively；And then obtain the characteristic expression symbol that in target soil region, each soil genetic horizon of each sampling point position is corresponding respectively, subsequently into step 00502；

Step 00502., according to Soil Taxonomy, sets up genetic horizon characteristic merger table, as shown in table 1 below:

Table 1

Wherein, it is identical that Arabic numerals are that soil characteristic expresses symbol, and there is each soil genetic horizon of Further Division from the soil profile arrangement by order from top to bottom；

Then each sampling point position in target soil region it is respectively directed to, each soil genetic horizon corresponding to sampling point position, divide by humus top layer, illuvium, chorizon, and press in genetic horizon characteristic merger table the soil genetic horizon characteristic expression symbol corresponding to each merger level, from this sampling point position profile region from top to bottom, divide each soil genetic horizon corresponding to this sampling point position successively, it is thus achieved that each merger layer corresponding to this sampling point position；

Meanwhile, in being respectively directed to target soil region, each sampling point position carries out in the process of above-mentioned merger operation, if occur identical soil genetic horizon characteristic expression symbol in two merger levels corresponding to sampling point position, then enters step 00503；

Step 00503. judges in each soil genetic horizon characteristic expression symbol corresponding to this sampling point position, whether existence is arranged in the soil genetic horizon characteristic expression symbol in genetic horizon characteristic merger table first merger level from top to bottom, it is that the soil genetic horizon that this identical soil genetic horizon characteristic expression symbol is corresponding respectively is merged, then perform step 00502 again and carry out merger process, it is thus achieved that each merger layer of correspondence is distinguished in this sampling point position；Otherwise enter step 00504；

Step 00504. is by this identical soil genetic horizon characteristic expression symbol, it is arranged in the soil genetic horizon characteristic expression symbol of lower floor's merger layer, it is divided in the merger layer at same soil genetic horizon characteristic expression symbol place, and by each soil genetic horizon characteristic expression symbol corresponding to this sampling point position, move up a merger level by the merger hierarchical sequence in genetic horizon characteristic merger table, it is thus achieved that each merger layer of correspondence is distinguished in this sampling point position。

As a preferred technical solution of the present invention, in described step 007, according to target soil region merger layer kind set, merger layer corresponding to each sampling point position in target soil region carries out unified operation, make the kind of merger layer corresponding to each sampling point position in target soil region mutually the same, specifically include following operation:

It is respectively directed to each sampling point position in target soil region, it is judged that merger layer corresponding to sampling point position, whether equal to target soil region merger layer kind set, is do not do any operation；Otherwise arrange the merger layer that its relative target soil region merger layer kind set lacks to this sampling point position, the thickness information simultaneously arranging this merger layer is 0.00001cm, and the soil attribute information arranging this merger layer is 0.00001；And then make the merger layer that in target soil region, each sampling point position is corresponding respectively be equal to target soil region merger layer kind set。

As a preferred technical solution of the present invention, described step 011 specifically includes following operation:

It is respectively directed to each checking sampling point position in checking sampling point location sets, it is thus achieved that L verifies the drawing coefficient Rc of sampling point position each merger layer the first thickness prediction information sum corresponding relatively；Then according to the first prediction soil attribute information, the first thickness prediction information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets, as follows:

${\mathrm{SOCD}}_{i_2}^{\′}=\frac{\underset{n_{i_2}=1}{\overset{N_{i_2}}{\Σ}}({\mathrm{SOC}}_{n_{i_2}}^{\′}\×{\mathrm{BD}}_{n_{i_2}}^{\′}\×(1-{\mathrm{Gr}}_{n_{i_2}}^{\′})\×Rc\×T_{n_{i_2}}^{\′})}{L}$

Obtain the first Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the first Soil Organic Carbon Density predictive value set PMTC-D of each checking sampling point position in checking sampling point location sets；Wherein, Represent checking sampling point location sets in i-th₂The sum of merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂The n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First organic carbon content information of forecasting of the n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First soil weight information of forecasting of the n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂More than the first gravel prediction content of preset diameters in n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First thickness prediction information of the n-th merger layer corresponding to individual checking sampling point position；Subsequently into step 012。

As a preferred technical solution of the present invention, in described step 013, it is thus achieved that the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, specifically include following operation:

It is respectively directed to each prediction sampling point position in prediction sampling point location sets, according to homalographic Spline function, adopts equation below:

$S\left(k_{i_1}\right)=\stackrel{\‾}{f}\left(k_{i_1}\right)+e\left(k_{i_1}\right)$

Obtain the soil attribute information of this each matching layer of prediction sampling point position；And then obtain the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, and wherein, i₁=1 ..., I₁, I₁Represent the quantity of prediction sampling point position in prediction sampling point location sets, Represent prediction sampling point location sets in i-th₁The number of plies of matching layer corresponding to individual prediction sampling point position,Represent prediction sampling point location sets in i-th₁The soil attribute information of kth layer matching layer corresponding to individual prediction sampling point position,RepresentFunction existsLayer withThe meansigma methods of layer fitting result,Represent prediction sampling point location sets in i-th₁The experimental determination analytical error of kth layer matching layer corresponding to individual prediction sampling point position,Function is homalographic Spline function。

As a preferred technical solution of the present invention, described step 015 specifically includes following operation:

The 3rd soil attribute information of forecasting of each matching layer corresponding to each checking sampling point position in checking sampling point location sets, and thickness information, as follows:

${\mathrm{SOCD}}_{i_2}^{\′\′}=\frac{\underset{k_{i_k}=1}{\overset{K_{i_2}}{\Σ}}({\mathrm{SOC}}_{k_{i_2}}^{\′\′}\×{\mathrm{BD}}_{k_{i_2}}^{\′\′}\×(1-G_{k_{i_2}}^{\′\′})\×Rc\×T_{k_{i_2}}^{\′\′})}{L}$

Obtain the 3rd Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the 3rd Soil Organic Carbon Density predictive value set PMTC-F of each checking sampling point position in checking sampling point location sets；Wherein, Represent checking sampling point location sets in i-th₂The sum of matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂Kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂3rd organic carbon content information of forecasting of kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂3rd soil weight information of forecasting of kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂More than the 3rd gravel prediction content of preset diameters in kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂The thickness information of kth matching layer corresponding to individual checking sampling point position；Subsequently into step 016。

A kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction of the present invention adopts above technical scheme compared with prior art, has following technical effect that

(1) a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction designed by the present invention, by the encapsulation for genetic horizon unstructured information, the soil body seriality prediction for considering the distribution of soil attribute horizontal dimension space provides good technical thought；Wherein, adopt " genetic horizon merger; prediction calculates again " technology; while ensureing that soil genetic horizon characteristic information does not lack; have modified the limitation that traditional prediction method describes in soil body seriality; achieve the universal soll carbon reserves estimating techniques of " Description standard, it was predicted that accurate ", in the engineering sounding of the relevant departments such as agricultural application, environmental conservation, land resources, there is wide industrial applications prospect；

(2) a kind of based in the Soil Carbon Stock evaluation method of soil genetic horizon thickness prediction designed by the present invention, proposed genetic horizon thickness prediction and drawing coefficient thereof calculate has certain universality, its technical scheme is not only oriented to Soil Organic Carbon Density prediction, can also combine with genetic horizon type prediction, constitute the techniqueflow of soil types prediction, therefore, the Predicting Technique that the present invention proposes also has good stability, the genetic horizon symbol merger mechanism of dynamic construction is expected to while ensureing precision of prediction, also reduce the error of soil genetic horizon thickness prediction process；

(3) a kind of based in the Soil Carbon Stock evaluation method of soil genetic horizon thickness prediction designed by the present invention, proposed genetic horizon merging techniques can simplify the structuring application of soil information to the full extent, it is expected to the engineer applied for other, as charted for three-dimensional soil, Digital Radio soil cartography and region organic C storage temporal-spatial evolution analysis provide technological guidance, and the advantage that the strengthening contrastive pattern used has taken into account different computation schema, ensure that Soil Organic Carbon Density is carried out spatial prediction by the computation schema that can use optimum, and then can be more quantitative, the carbon reserves of objective evaluation target area。

Accompanying drawing explanation

Fig. 1 is the schematic flow sheet that the present invention designs the Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction；

Fig. 2 a is in institute of the present invention Application Example when depth of soil is be more than or equal to predetermined depth 1m, gathers the schematic diagram in the soil profile region of the 1m degree of depth；

Fig. 2 b be in institute of the present invention Application Example when depth of soil is less than predetermined depth 1m, gather the schematic diagram of the hatch region of soil straight down；

Fig. 3 is the schematic diagram of different soils genetic horizon thickness space change in the embodiment of the present invention；

Fig. 4 divides, for 1m depth section region, the schematic diagram obtaining fixing layer thicknesses merger layer in the embodiment of the present invention；

Fig. 5 a to Fig. 5 i is the spatial distribution map of the thickness prediction of each merger layer in the embodiment of the present invention；

Fig. 6 is the drawing coefficient distribution schematic diagram of the relative thickness prediction in 1m depth section region in the embodiment of the present invention；

Fig. 7 a to Fig. 7 c is the relative analysis schematic diagram that in the embodiment of the present invention, distinct methods predicts the outcome。

Detailed description of the invention

Below in conjunction with Figure of description, the specific embodiment of the present invention is described in further detail。

For above-mentioned technological deficiency, the expertise that the present invention develops with soil, for instructing, devises a set of new Soil Carbon Stock based on soil genetic horizon thickness prediction and estimates engineering scheme。Give the techniqueflow of the prediction of carbon reserves estimation this technological system。Simultaneously, it is considered to the height space variation property of Soil Organic Carbon Density, the present invention innovatively proposes the organic C storage estimation computation schema that contrast use is different。Strengthening Comparing method is expected to the Optimal calculation pattern providing tradition to ignore based on GIS estimating techniques。

The basic thought of the present invention is in soil investigation sampling process, the genetic horizon characteristic of system log (SYSLOG) various soil mass with describe information。After laboratory soil attribute assay, structurized Soil attribute data structure is used to input non-structured soil information。With genetic horizon characteristic expression symbol for operating unit, build and there are the similar genetic horizon standardization soil profile data with equal genetic horizon quantity。Use genetic horizon thickness and the soil attribute of machine learning algorithm prediction impact point position, based on the drawing coefficient of prediction genetic horizon thickness, update the calculating of Soil Organic Carbon Density。By other computation schema of relative analysis, select the computation schema of optimum, the final carbon reserves estimating target area。

Shown in Fig. 1, a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction designed by the present invention, in the middle of actual application, specifically include following steps:

Step 001., for target soil region, arranges each sampling point position, and the quantity adding up sampling point position is I, subsequently into step 002。

Wherein, step 001 specifically includes following steps:

Step 00101. obtains the soil types scattergram in target soil region, Soil Utilization scattergram and soil-geological scattergram, and soil types scattergram, Soil Utilization scattergram and soil-geological scattergram are carried out space overlapping, obtain target soil district multiple-layer stacked figure, subsequently into step 00102。

Step 00102. obtains the area of each figure spot region in target soil district multiple-layer stacked figure respectively, and adds up and obtain area ratio and exceed each figure spot region of figure spot area threshold value；Then being respectively directed to this each figure spot region, according to accessible detecting requirement on the spot, arrange sampling point position, the quantity of statistics sampling point position is I, enters back into step 002。

Step 002. is in target soil region, it is respectively directed to each sampling point position, obtaining the environmental information of sampling point position, environmental information includes vegetation coverage, rock exposure area ratio, landform, important symbol thing, earth's surface coarse fragment size, ground crack situation, earth's surface salting stain information；Meanwhile, as shown in Figure 2 a, if the degree of depth of this sampling point position soil region straight down is be more than or equal to predetermined depth L, then the hatch region of this sampling point position predetermined depth L straight down is obtained；As shown in Figure 2 b, if the degree of depth of this sampling point position soil region straight down is less than predetermined depth L, then the hatch region of this sampling point position soil straight down is obtained；And the job requirements according to Soil Taxonomy, divide soil genetic horizon for this hatch region, it is thus achieved that each soil genetic horizon corresponding to this sampling point position；Obtaining thickness information and the soil morphology information of this sampling point position each soil genetic horizon corresponding more respectively, soil morphology information represents soil dry-wet situation, soil color, root system information, hole information, sample structure, speckle component, warty tuberculosis material, degree of consolidation, lime reaction information；And then obtain the environmental information of each sampling point position in target soil region respectively, and thickness information and the soil morphology information of each soil genetic horizon corresponding is distinguished in each sampling point position, subsequently enters step 003。

Wherein, Soil Taxonomy table is as shown in table 2 below:

Genetic horizon symbol	Genetic horizon refers to information and describes	Genetic horizon symbol	Genetic horizon refers to information and describes
				A	Humus top layer	l	Reticulate pattern
B	Illuvium	m	Strong rubber is tied
				C	Chorizon	n	Sodium is built up
R	Basement rock	o	Root system dish is tied
				a	High de-agglomeration organic substance	p	Cultivation effect
b	Buried horizon	q	Secondary silicon is built up
				c	Skinning	r	Oxidoreduction
d	Freeze thawing laminated structure	s	Ferrimanganic is built up
				e	Partly decompose organic substance	t	Glutinous grain is built up
f	Permafrost haorizon	u	Artificially pile up impact
				g	Incobation speckle	v	Vertic features
h	Humus is built up	w	On the spot air slaking formed colour developing, have structure sheaf
				i	Low decomposition and undecomposed organic substance	x	Hard cementing of solid-state, does not form huge rock
j	Autunezite	y	Gypsum Fibrosum is built up
				k	Carbonate is built up	z	Soluble salt is built up

Table 2

Step 003. is respectively directed to each sampling point position in target soil region, in each soil genetic horizon corresponding to sampling point position, gather the pedotheque of preset quality respectively, and measure the soil attribute information obtaining each soil genetic horizon respectively, and then obtain the soil attribute information of each sampling point position each soil genetic horizon corresponding respectively in target soil region, subsequently into step 004；Wherein, soil attribute information includes soil organic carbon (unit: g/kg), gravel concentration (percent by volume, %) more than preset diameters 0.2mm and the soil weight (unit: g/cm³)。

Step 004. is respectively directed to each sampling point position in target soil region, the soil attribute information of each soil genetic horizon, thickness information corresponding to sampling point position, it is thus achieved that the Soil Organic Carbon Density measured value SOCD of each sampling point position_i, i={1 ..., I}, subsequently into step 005。

Described step 004 specifically includes following operation:

It is respectively directed to each sampling point position in target soil region, the soil attribute information of each soil genetic horizon, thickness information corresponding to sampling point position, as follows:

${\mathrm{SOCD}}_i=\frac{\underset{n_i=1}{\overset{N_i}{\Σ}}({\mathrm{SOC}}_{n_i}\×{\mathrm{BD}}_{n_i}\×(1-{\mathrm{Gr}}_{n_i})\×T_{n_i})}{L}$

Obtain the Soil Organic Carbon Density measured value SOCD of each sampling point position_i, wherein, i={1 ..., I}, n_i=1 ... N_i, n_iRepresent the n-th soil genetic horizon corresponding to i-th sampling point position, N_iRepresent the sum of soil genetic horizon corresponding to i-th sampling point position；Represent the organic carbon content of the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the soil weight of the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the gravel concentration more than preset diameters in the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the thickness information of the n-th soil genetic horizon corresponding to i-th sampling point position, subsequently into step 005。

Step 005. is respectively directed to each sampling point position in target soil region, and each soil genetic horizon corresponding to sampling point position carries out merger process, it is thus achieved that each merger layer corresponding to this sampling point position；Then each soil genetic horizon of correspondence is distinguished according to each merger layer of this sampling point position, soil attribute information for this each soil genetic horizon of sampling point position is weighted, obtain the soil attribute information of this sampling point position each merger layer corresponding, and the thickness information for this each soil genetic horizon of sampling point position carries out read group total, it is thus achieved that the thickness of this sampling point position each merger layer corresponding；And then obtain each merger layer that each sampling point position is corresponding respectively in target soil region respectively, and the soil attribute information of each merger layer and thickness information, enter back into step 006。

In step 005, being respectively directed to each sampling point position in target soil region, each soil genetic horizon corresponding to sampling point position carries out merger process, it is thus achieved that the process of each merger layer corresponding to this sampling point position, specifically includes following steps:

Step 00501. is respectively directed to each sampling point position in target soil region, the soil morphology information of each soil genetic horizon corresponding to sampling point position, and Soil Taxonomy, it is thus achieved that the characteristic expression symbol that each soil genetic horizon of this sampling point position is corresponding respectively；And then obtain the characteristic expression symbol that in target soil region, each soil genetic horizon of each sampling point position is corresponding respectively, subsequently into step 00502。

Step 00502., according to Soil Taxonomy, sets up genetic horizon characteristic merger table, as shown in table 1 below:

Table 1

Wherein, it is identical that Arabic numerals are that soil characteristic expresses symbol, and there is each soil genetic horizon of Further Division from the soil profile arrangement by order from top to bottom；Note: (1) main genetic horizon or characteristic genetic horizon can be further subdivided into some subgrades by the difference on its occurrence degree。All represent side by side with Arabic numerals and capitalization, for instance C1, C2, Bt1, Bt2, Bt3；(2) different unit matrix soil layer represents: represent before being placed in genetic horizon symbol with Arabic numerals, such as 2C。

Then each sampling point position in target soil region it is respectively directed to, each soil genetic horizon corresponding to sampling point position, divide by humus top layer, illuvium, chorizon, and press in genetic horizon characteristic merger table the soil genetic horizon characteristic expression symbol corresponding to each merger level, from this sampling point position profile region from top to bottom, divide each soil genetic horizon corresponding to this sampling point position successively, it is thus achieved that each merger layer corresponding to this sampling point position。

Meanwhile, in being respectively directed to target soil region, each sampling point position carries out in the process of above-mentioned merger operation, if occur identical soil genetic horizon characteristic expression symbol in two merger levels corresponding to sampling point position, then enters step 00503。

Step 00503. judges in each soil genetic horizon characteristic expression symbol corresponding to this sampling point position, whether existence is arranged in the soil genetic horizon characteristic expression symbol in genetic horizon characteristic merger table first merger level from top to bottom, it is that the soil genetic horizon that this identical soil genetic horizon characteristic expression symbol is corresponding respectively is merged, then perform step 00502 again and carry out merger process, it is thus achieved that each merger layer of correspondence is distinguished in this sampling point position；Otherwise enter step 00504。

Step 00504. is by this identical soil genetic horizon characteristic expression symbol, it is arranged in the soil genetic horizon characteristic expression symbol of lower floor's merger layer, it is divided in the merger layer at same soil genetic horizon characteristic expression symbol place, and by each soil genetic horizon characteristic expression symbol corresponding to this sampling point position, move up a merger level by the merger hierarchical sequence in genetic horizon characteristic merger table, it is thus achieved that each merger layer of correspondence is distinguished in this sampling point position。

Step 006. is as shown in Figure 3, it is respectively directed to each sampling point position in target soil region, judge that whether the thickness sum of sampling point position each merger layer corresponding is less than predetermined depth L, it is then under each merger layer corresponding to this sampling point position, with horizon d, merger layer is set, make the thickness sum of this sampling point position each merger layer corresponding equal to predetermined depth L, and enter step 007；Otherwise it is directly entered step 007。

Step 007. obtains the kind of merger layer corresponding to all sampling point positions in target soil region, constitutes target soil region merger layer kind set；According to target soil region merger layer kind set, merger layer corresponding to each sampling point position in target soil region carries out unified operation, make the kind of merger layer corresponding to each sampling point position in target soil region mutually the same, and enter step 008。

In step 007, according to target soil region merger layer kind set, merger layer corresponding to each sampling point position in target soil region carries out unified operation so that in target soil region, corresponding to each sampling point position, the kind of merger layer is mutually the same, specifically includes following operation:

It is respectively directed to each sampling point position in target soil region, it is judged that merger layer corresponding to sampling point position, whether equal to target soil region merger layer kind set, is do not do any operation；Otherwise arrange the merger layer that its relative target soil region merger layer kind set lacks to this sampling point position, the thickness information simultaneously arranging this merger layer is 0.00001cm, and the soil attribute information arranging this merger layer is 0.00001；And then make the merger layer that in target soil region, each sampling point position is corresponding respectively be equal to target soil region merger layer kind set。

Step 008. adopts linear congruent algorithm, and the sampling point position in target soil region is divided into prediction sampling point location sets and checking sampling point location sets, and according to the Soil Organic Carbon Density measured value SOCD of each sampling point position in target soil region_i, it is thus achieved that the Soil Organic Carbon Density measured value of each checking sampling point position in checking sampling point location setsConstitute the Soil Organic Carbon Density measured value set V of each checking sampling point position in checking sampling point location sets；And enter step 009, and wherein, i₂=1 ..., I₂, wherein, I₂For checking sampling point location sets is verified the quantity of sampling point position, equal to 25%I, it was predicted that in sampling point location sets, the quantity of prediction sampling point position is I₁, equal to 75%I。

Step 009. is according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the thickness information of each prediction sampling point position each merger layer corresponding respectively, adopt random forest method, each forecast model that it is target with each merger layer thickness information respectively that training obtains, constitutes the first forecast model set。

Simultaneously, according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the soil attribute information of each prediction sampling point position each merger layer corresponding respectively, adopt random forest method, each forecast model that it is target with each merger layer soil attribute information respectively that training obtains, constitutes the second forecast model set；Subsequently into step 010。

Step 010. adopts MTC-D method, it is respectively directed to each checking sampling point position in checking sampling point location sets, environmental information according to checking sampling point position, by each forecast model in the first forecast model set, obtain the first thickness prediction information of this checking sampling point position each merger layer corresponding respectively respectively, as shown in Fig. 5 a to Fig. 5 i；Simultaneously, it is respectively directed to each checking sampling point position in checking sampling point location sets, environmental information according to checking sampling point position, by each forecast model in the second forecast model set, obtains this checking sampling point position respectively and is distinguished the first prediction soil attribute information of each merger layer corresponding；And then obtain the first thickness prediction information and the first prediction soil attribute information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets；Subsequently into step 011；Wherein, the first prediction soil attribute information includes the first organic carbon content information of forecasting, the first soil weight information of forecasting and the first gravel prediction content more than preset diameters。

Step 011. is respectively directed to each checking sampling point position in checking sampling point location sets, it is thus achieved that L verifies the drawing coefficient Rc of sampling point position each merger layer the first thickness prediction information sum corresponding relatively, as shown in Figure 6；Then according to the first prediction soil attribute information, the first thickness prediction information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets, it is thus achieved that the first Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the first Soil Organic Carbon Density predictive value set PMTC-D of each checking sampling point position in checking sampling point location sets；Subsequently into step 012。

Wherein, step 011 specifically includes following operation:

It is respectively directed to each checking sampling point position in checking sampling point location sets, it is thus achieved that L verifies the drawing coefficient Rc of sampling point position each merger layer the first thickness prediction information sum corresponding relatively；Then according to the first prediction soil attribute information, the first thickness prediction information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets, as follows:

${\mathrm{SOCD}}_{i_2}^{\′}=\frac{\underset{n_{i_2}=1}{\overset{N_{i_2}}{\Σ}}({\mathrm{SOC}}_{n_{i_2}}^{\′}\×{\mathrm{BD}}_{n_{i_2}}^{\′}\×(1-{\mathrm{Gr}}_{n_{i_2}}^{\′})\×Rc\×T_{n_{i_2}}^{\′})}{L}$

Obtain the first Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the first Soil Organic Carbon Density predictive value set PMTC-D of each checking sampling point position in checking sampling point location sets；Wherein, Represent checking sampling point location sets in i-th₂The sum of merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂The n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First organic carbon content information of forecasting of the n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First soil weight information of forecasting of the n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂More than the first gravel prediction content of preset diameters in n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First thickness prediction information of the n-th merger layer corresponding to individual checking sampling point position；Subsequently into step 012。

Step 012. adopts CTM method, according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the Soil Organic Carbon Density measured value of each prediction sampling point position, adopt random forest method, the 3rd forecast model that it is target with Soil Organic Carbon Density measured value that training obtains；Then each checking sampling point position in checking sampling point location sets it is respectively directed to, environmental information according to checking sampling point position, by the 3rd forecast model, obtain the second Soil Organic Carbon Density predictive value of this checking sampling point position, and then obtain the second Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location sets, constitute the second Soil Organic Carbon Density predictive value set PCTM of each checking sampling point position in checking sampling point location sets；Subsequently into step 013。

Step 013. is as shown in Figure 4, for sampling point positions all in target soil region, unified by presetting division rule, based on L, each merger layer corresponding to sampling point position is divided into each matching layer, the quantity of each matching layer corresponding to sampling point position and matching layer is identical, and obtain the thickness information of each matching layer, in the present embodiment example, hatch region for the 1m degree of depth, specifically it is divided into each matching layer of the constant depth of 0-5cm, 5-15cm, 15-30cm, 30-60cm and 60-100cm, is divided into 5 matching layers altogether；Then each prediction sampling point position in prediction sampling point location sets it is respectively directed to, each merger layer of correspondence is distinguished according to each matching layer of prediction sampling point position, soil attribute information for this prediction sampling point position each merger layer sampling genetic horizon is fitted, it is thus achieved that the soil attribute information of each matching layer in prediction sampling point position；And then obtain the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, and enter step 014。

Wherein, in described step 013, it is thus achieved that the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, following operation is specifically included:

It is respectively directed to each prediction sampling point position in prediction sampling point location sets, according to homalographic Spline function, adopts equation below:

$S\left(k_{i_1}\right)=\stackrel{\‾}{f}\left(k_{i_1}\right)+e\left(k_{i_1}\right)$

Obtain the soil attribute information of this each matching layer of prediction sampling point position；And then obtain the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, and wherein, i₁=1 ..., I₁, I₁Represent the quantity of prediction sampling point position in prediction sampling point location sets, Represent prediction sampling point location sets in i-th₁The number of plies of matching layer corresponding to individual prediction sampling point position,Represent prediction sampling point location sets in i-th₁The soil attribute information of kth layer matching layer corresponding to individual prediction sampling point position,RepresentFunction existsLayer withThe meansigma methods of layer fitting result,Represent prediction sampling point location sets in i-th₁The experimental determination analytical error of kth layer matching layer corresponding to individual prediction sampling point position,Function is homalographic Spline function。

Step 014. adopts MTC-F method, according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the soil attribute information of each prediction sampling point position each matching layer corresponding respectively, adopt random forest method, each forecast model that it is target with each matching layer soil attribute information that training obtains, constitutes the 4th forecast model set；Then each checking sampling point position in checking sampling point location sets it is respectively directed to, environmental information according to checking sampling point position, by each forecast model in the 4th forecast model set, obtain the 3rd soil attribute information of forecasting of this checking sampling point position each matching layer corresponding respectively, and then obtain the 3rd soil attribute information of forecasting of each checking sampling point position each matching layer corresponding respectively in checking sampling point location sets, subsequently into step 015；Wherein, the 3rd prediction soil attribute information includes the 3rd organic carbon content information of forecasting, the 3rd soil weight information of forecasting and the 3rd gravel prediction content more than preset diameters。

Step 015. is the 3rd soil attribute information of forecasting of each matching layer corresponding to each checking sampling point position in checking sampling point location sets, and thickness information, it is thus achieved that the 3rd Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the 3rd Soil Organic Carbon Density predictive value set PMTC-F of each checking sampling point position in checking sampling point location sets；Subsequently into step 016。

Described step 015 specifically includes following operation:

The 3rd soil attribute information of forecasting of each matching layer corresponding to each checking sampling point position in checking sampling point location sets, and thickness information, as follows:

${\mathrm{SOCD}}_{i_2}^{\′\′}=\frac{\underset{k_{i_2}=1}{\overset{K_{i_2}}{\Σ}}({\mathrm{SOC}}_{k_{i_2}}^{\′\′}\×{\mathrm{BD}}_{k_{i_2}}^{\′\′}\×(1-{\mathrm{Gr}}_{k_{i_2}}^{\′\′})\×Rc\×T_{k_{i_2}}^{\′\′})}{L}$

Obtain the 3rd Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the 3rd Soil Organic Carbon Density predictive value set PMTC-F of each checking sampling point position in checking sampling point location sets；Wherein, Represent checking sampling point location sets in i-th₂The sum of matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂Kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂3rd organic carbon content information of forecasting of kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂3rd soil weight information of forecasting of kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂More than the 3rd gravel prediction content of preset diameters in kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂The thickness information of kth matching layer corresponding to individual checking sampling point position；Subsequently into step 016。

Step 016. uses Lins ' s concordance correlation coefficient (Lins'sconcordancecorrelationcoefficient) as precision test index, is predicted the precision evaluation of result:

${\mathrm{\ρ}}_c=\frac{2S_{vp}}{S_v^2+S_p^2+{(\stackrel{\‾}{v}-\stackrel{\‾}{p})}^2}$

Wherein, Represent checking sampling point location sets in i-th₂The Soil Organic Carbon Density measured value of individual checking sampling point position, i.e. Soil Organic Carbon Density measured value Represent checking sampling point location sets in i-th₂The Soil Organic Carbon Density predictive value of individual checking sampling point position, namely distinguishes corresponding first Soil Organic Carbon Density predictive value, the second Soil Organic Carbon Density predictive value, the 3rd Soil Organic Carbon Density predictive value；Represent all checking sampling point positions Soil Organic Carbon Density measured value in checking sampling point location setsMeansigma methods；Represent all checking sampling point positions Soil Organic Carbon Density predictive value in checking sampling point location setsMeansigma methods；Lins ' s concordance correlation coefficient is unrelated with the target data magnitude of accuracy test, and codomain is [-1,1], and maximum 1 represents best predicting the outcome, completely the same with observation data。In the middle of actual application, namely the first Soil Organic Carbon Density predictive value set PMTC-D, the second Soil Organic Carbon Density predictive value set PCTM and the three Soil Organic Carbon Density predictive value set PMTC-F it are respectively directed to, it is respectively adopted above-mentioned Accuracy Assessment, wherein, when adopting the first Soil Organic Carbon Density predictive value set PMTC-D, then each value in the first Soil Organic Carbon Density predictive value set PMTC-D is corresponded toWhen adopting the second Soil Organic Carbon Density predictive value set PCTM, then each value in the second Soil Organic Carbon Density predictive value set PCTM is corresponded toWhen adopting the 3rd Soil Organic Carbon Density predictive value set PMTC-F, then each value in the 3rd Soil Organic Carbon Density predictive value set PMTC-F is corresponded to

Accuracy test is carried out according to Soil Organic Carbon Density measured value set V, obtain the optimum Soil Organic Carbon Density predictive value set in the first Soil Organic Carbon Density predictive value set PMTC-D, the second Soil Organic Carbon Density predictive value set PCTM and the three Soil Organic Carbon Density predictive value set PMTC-F, and enter step 017。

Step 017. is by target soil discrete region space lattice data, using the sampled data of the respectively corresponding genetic horizon of all sampling point positions in target soil region as prediction data set, if optimum Soil Organic Carbon Density predictive value set is the first Soil Organic Carbon Density predictive value set PMTC-D, then step 009 to the method for step 011 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；If optimum Soil Organic Carbon Density predictive value set is the second Soil Organic Carbon Density predictive value set PCTM, then the method for step 012 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；If optimum Soil Organic Carbon Density predictive value set is the 3rd Soil Organic Carbon Density predictive value set PMTC-F, then step 013 to the method for step 015 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；Subsequently into step 018。

The step 018. Soil Organic Carbon Density spatial distribution raster data according to target soil region, it is thus achieved that the Soil Carbon Stock in target soil region。

Above-mentioned for present invention design is applied in reality based on the Soil Carbon Stock evaluation method of soil genetic horizon thickness prediction, as being estimated as example with Soils of Liaoning Province carbon reserves。

Liaoning Province is positioned at the south of Northeast Area of China, and northeast and Jilin Province border on, and northwest and Inner Mongolia Autonomous Region are adjacent。Liaoning Province's land resource is not enough, and cultivated land resource is less, and land use pattern is more。With the soil investigation data of 2009 for input, on the basis of the present invention, use different Forecasting Methodologies and genetic horizon merger strategy, it is possible to get the carbon reserves estimated value based on best practice。With in the practical application in Liaoning Province, final as shown in Fig. 7 a to Fig. 7 c, for different Forecasting Methodology PMTC-D, the predicting the outcome of PCTM, PMTC-F, use Lins ' s concordance correlation coefficient (Lins'sconcordancecorrelationcoefficient) as precision test index, be predicted the precision evaluation of result。Precision evaluation result is the highest (ρ of CTM_c=0.21), MTC-D secondly (ρ_c=0.18), the worst (ρ of MTC-F_c=0.13)。Therefore, the carbon reserves in Liaoning Province is estimated, it is recommended to use CTM method predicts the carbon reserves in this region；Then by target soil discrete region space lattice data, using the sampled data of the corresponding genetic horizon respectively of all sampling point positions in target soil region as prediction data set, then the method for step 012 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；Soil Organic Carbon Density spatial distribution raster data finally according to target soil region, it is thus achieved that the Soil Carbon Stock in target soil region。

Being different from the conventional Soil Carbon Stock evaluation method based on GIS, the present invention takes into full account the spatial continuity of soil attribute homogeneity on genetic horizon basis and level dimension。The use of cross validation is that goals research region organic C storage estimation engineering provides precision of prediction and ensures with Optimal calculation mechanism, and a kind of spatial prediction techniques of customer service exists circumscribed technical bottleneck effectively。This technology is formulated effective administrative mechanism for science, is realized the sustainable development of resource and give full play to the ecological benefits of ecosystem have highly important directive significance。

Above in conjunction with accompanying drawing, embodiments of the present invention are explained in detail, but the present invention is not limited to above-mentioned embodiment, in the ken that those of ordinary skill in the art possess, it is also possible to make a variety of changes under the premise without departing from present inventive concept。

Claims (10)

1. the Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction, it is characterised in that comprise the steps:

Step 001., for target soil region, arranges each sampling point position, and the quantity adding up sampling point position is I, subsequently into step 002；

Step 002. is in target soil region, it is respectively directed to each sampling point position, obtain the environmental information of sampling point position, if the degree of depth of this sampling point position soil region straight down is be more than or equal to predetermined depth L simultaneously, then obtain the hatch region of this sampling point position predetermined depth L straight down；If the degree of depth of this sampling point position soil region straight down is less than predetermined depth L, then obtain the hatch region of this sampling point position soil straight down；And the job requirements according to Soil Taxonomy, divide soil genetic horizon for this hatch region, it is thus achieved that each soil genetic horizon corresponding to this sampling point position；Obtain thickness information and the soil morphology information of this sampling point position each soil genetic horizon corresponding more respectively；And then obtain the environmental information of each sampling point position in target soil region respectively, and thickness information and the soil morphology information of each soil genetic horizon corresponding is distinguished in each sampling point position, subsequently enters step 003；

Step 003. is respectively directed to each sampling point position in target soil region, in each soil genetic horizon corresponding to sampling point position, gather the pedotheque of preset quality respectively, and measure the soil attribute information obtaining each soil genetic horizon respectively, and then obtain the soil attribute information of each sampling point position each soil genetic horizon corresponding respectively in target soil region, subsequently into step 004；Wherein, soil attribute information include soil organic carbon, more than the gravel concentration of preset diameters and the soil weight；

Step 004. is respectively directed to each sampling point position in target soil region, the soil attribute information of each soil genetic horizon, thickness information corresponding to sampling point position, it is thus achieved that the Soil Organic Carbon Density measured value SOCD of each sampling point position_i, i={1 ..., I}, subsequently into step 005；

Step 005. is respectively directed to each sampling point position in target soil region, each soil genetic horizon corresponding to sampling point position carries out merger process, obtain each merger layer corresponding to this sampling point position, then each soil genetic horizon of correspondence is distinguished according to each merger layer of this sampling point position, soil attribute information for this each soil genetic horizon of sampling point position is weighted, obtain the soil attribute information of this sampling point position each merger layer corresponding, and the thickness information for this each soil genetic horizon of sampling point position carries out read group total, obtain the thickness of this sampling point position each merger layer corresponding；And then obtain each merger layer that each sampling point position is corresponding respectively in target soil region respectively, and the soil attribute information of each merger layer and thickness information, enter back into step 006；

Step 006. is respectively directed to each sampling point position in target soil region, judge that whether the thickness sum of sampling point position each merger layer corresponding is less than predetermined depth L, it is then under each merger layer corresponding to this sampling point position, with horizon d, merger layer is set, make the thickness sum of this sampling point position each merger layer corresponding equal to predetermined depth L, and enter step 007；Otherwise it is directly entered step 007；

Step 007. obtains the kind of merger layer corresponding to all sampling point positions in target soil region, constitutes target soil region merger layer kind set；According to target soil region merger layer kind set, merger layer corresponding to each sampling point position in target soil region carries out unified operation, make the kind of merger layer corresponding to each sampling point position in target soil region mutually the same, and enter step 008；

Step 008. adopts linear congruent algorithm, and the sampling point position in target soil region is divided into prediction sampling point location sets and checking sampling point location sets, and according to the Soil Organic Carbon Density measured value SOCD of each sampling point position in target soil region_i, it is thus achieved that the Soil Organic Carbon Density measured value of each checking sampling point position in checking sampling point location setsConstitute the Soil Organic Carbon Density measured value set V of each checking sampling point position in checking sampling point location sets；And enter step 009, and wherein, i₂=1 ..., I₂, wherein, I₂For the quantity of checking sampling point position in checking sampling point location sets, it was predicted that in sampling point location sets, the quantity of prediction sampling point position is I₁, I₁>I₂；

Step 009. is according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the thickness information of each prediction sampling point position each merger layer corresponding respectively, adopt random forest method, each forecast model that it is target with each merger layer thickness information respectively that training obtains, constitutes the first forecast model set；

Simultaneously, according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the soil attribute information of each prediction sampling point position each merger layer corresponding respectively, adopt random forest method, each forecast model that it is target with each merger layer soil attribute information respectively that training obtains, constitutes the second forecast model set；Subsequently into step 010；

Step 010. is respectively directed to each checking sampling point position in checking sampling point location sets, environmental information according to checking sampling point position, by each forecast model in the first forecast model set, obtain the first thickness prediction information of this checking sampling point position each merger layer corresponding respectively respectively；Simultaneously, it is respectively directed to each checking sampling point position in checking sampling point location sets, environmental information according to checking sampling point position, by each forecast model in the second forecast model set, obtains this checking sampling point position respectively and is distinguished the first prediction soil attribute information of each merger layer corresponding；And then obtain the first thickness prediction information and the first prediction soil attribute information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets；Subsequently into step 011；Wherein, the first prediction soil attribute information includes the first organic carbon content information of forecasting, the first soil weight information of forecasting and the first gravel prediction content more than preset diameters；

Step 011. is respectively directed to each checking sampling point position in checking sampling point location sets, it is thus achieved that L verifies the drawing coefficient Rc of sampling point position each merger layer the first thickness prediction information sum corresponding relatively；Then according to the first prediction soil attribute information, the first thickness prediction information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets, it is thus achieved that the first Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the first Soil Organic Carbon Density predictive value set PMTC-D of each checking sampling point position in checking sampling point location sets；Subsequently into step 012；

Step 012. is according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the Soil Organic Carbon Density measured value of each prediction sampling point position, adopt random forest method, the 3rd forecast model that it is target with Soil Organic Carbon Density measured value that training obtains；Then each checking sampling point position in checking sampling point location sets it is respectively directed to, environmental information according to checking sampling point position, by the 3rd forecast model, obtain the second Soil Organic Carbon Density predictive value of this checking sampling point position, and then obtain the second Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location sets, constitute the second Soil Organic Carbon Density predictive value set PCTM of each checking sampling point position in checking sampling point location sets；Subsequently into step 013；

Step 013. is for sampling point positions all in target soil region, unified by presetting division rule, based on L, each merger layer corresponding to sampling point position is divided into each matching layer, the quantity of each matching layer corresponding to sampling point position and matching layer is identical, and obtains the thickness information of each matching layer；Then each prediction sampling point position in prediction sampling point location sets it is respectively directed to, each merger layer of correspondence is distinguished according to each matching layer of prediction sampling point position, soil attribute information for this prediction sampling point position each merger layer sampling genetic horizon is fitted, it is thus achieved that the soil attribute information of each matching layer in prediction sampling point position；And then obtain the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, and enter step 014；

Step 014. is according to the environmental information of each prediction sampling point position in prediction sampling point location sets, and the soil attribute information of each prediction sampling point position each matching layer corresponding respectively, adopt random forest method, each forecast model that it is target with each matching layer soil attribute information that training obtains, constitutes the 4th forecast model set；Then each checking sampling point position in checking sampling point location sets it is respectively directed to, environmental information according to checking sampling point position, by each forecast model in the 4th forecast model set, obtain the 3rd soil attribute information of forecasting of this checking sampling point position each matching layer corresponding respectively, and then obtain the 3rd soil attribute information of forecasting of each checking sampling point position each matching layer corresponding respectively in checking sampling point location sets, subsequently into step 015；Wherein, the 3rd prediction soil attribute information includes the 3rd organic carbon content information of forecasting, the 3rd soil weight information of forecasting and the 3rd gravel prediction content more than preset diameters；

Step 015. is the 3rd soil attribute information of forecasting of each matching layer corresponding to each checking sampling point position in checking sampling point location sets, and thickness information, it is thus achieved that the 3rd Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the 3rd Soil Organic Carbon Density predictive value set PMTC-F of each checking sampling point position in checking sampling point location sets；Subsequently into step 016；

Step 016. carries out accuracy test according to Soil Organic Carbon Density measured value set V, obtain the optimum Soil Organic Carbon Density predictive value set in the first Soil Organic Carbon Density predictive value set PMTC-D, the second Soil Organic Carbon Density predictive value set PCTM and the three Soil Organic Carbon Density predictive value set PMTC-F, and enter step 017；

Step 017. is by target soil discrete region space lattice data, using the sampled data of the respectively corresponding genetic horizon of all sampling point positions in target soil region as prediction data set, if optimum Soil Organic Carbon Density predictive value set is the first Soil Organic Carbon Density predictive value set PMTC-D, then step 009 to the method for step 011 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；If optimum Soil Organic Carbon Density predictive value set is the second Soil Organic Carbon Density predictive value set PCTM, then the method for step 012 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；If optimum Soil Organic Carbon Density predictive value set is the 3rd Soil Organic Carbon Density predictive value set PMTC-F, then step 013 to the method for step 015 is adopted to obtain the Soil Organic Carbon Density spatial distribution raster data in target soil region；Subsequently into step 018；

The step 018. Soil Organic Carbon Density spatial distribution raster data according to target soil region, it is thus achieved that the Soil Carbon Stock in target soil region。

2. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterised in that described step 001 specifically includes following steps:

Step 00101. obtains the soil types scattergram in target soil region, Soil Utilization scattergram and soil-geological scattergram, and soil types scattergram, Soil Utilization scattergram and soil-geological scattergram are carried out space overlapping, obtain target soil district multiple-layer stacked figure, subsequently into step 00102；

Step 00102. obtains the area of each figure spot region in target soil district multiple-layer stacked figure respectively, and adds up and obtain area ratio and exceed each figure spot region of figure spot area threshold value；Then being respectively directed to this each figure spot region, according to accessible detecting requirement on the spot, arrange sampling point position, the quantity of statistics sampling point position is I, enters back into step 002。

3. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterised in that: described environmental information includes vegetation coverage, rock exposure area ratio, landform, important symbol thing, earth's surface coarse fragment size, ground crack situation, earth's surface salting stain information。

4. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterised in that: described soil morphology information represents soil dry-wet situation, soil color, root system information, hole information, sample structure, speckle component, warty tuberculosis material, degree of consolidation, lime reaction information。

5. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterised in that described step 004 specifically includes following operation:

It is respectively directed to each sampling point position in target soil region, the soil attribute information of each soil genetic horizon, thickness information corresponding to sampling point position, as follows:

${\mathrm{SOCD}}_i=\frac{\underset{n_i=1}{\overset{N_i}{\Σ}}({\mathrm{SOC}}_{n_i}\×{\mathrm{BD}}_{n_i}\×(1-{\mathrm{Gr}}_{n_i})\×T_{n_i})}{L}$

Obtain the Soil Organic Carbon Density measured value SOCD of each sampling point position_i, wherein, i={1 ..., I}, n_i=1 ... N_i, n_iRepresent the n-th soil genetic horizon corresponding to i-th sampling point position, N_iRepresent the sum of soil genetic horizon corresponding to i-th sampling point position；Represent the organic carbon content of the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the soil weight of the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the gravel concentration more than preset diameters in the n-th soil genetic horizon corresponding to i-th sampling point position,Represent the thickness information of the n-th soil genetic horizon corresponding to i-th sampling point position, subsequently into step 005。

6. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterized in that, in described step 005, it is respectively directed to each sampling point position in target soil region, each soil genetic horizon corresponding to sampling point position carries out merger process, obtain the process of each merger layer corresponding to this sampling point position, specifically include following steps:

Step 00501. is respectively directed to each sampling point position in target soil region, the soil morphology information of each soil genetic horizon corresponding to sampling point position, and Soil Taxonomy, it is thus achieved that the characteristic expression symbol that each soil genetic horizon of this sampling point position is corresponding respectively；And then obtain the characteristic expression symbol that in target soil region, each soil genetic horizon of each sampling point position is corresponding respectively, subsequently into step 00502；

Step 00502., according to Soil Taxonomy, sets up genetic horizon characteristic merger table, as shown in table 1 below:

Table 1

Wherein, it is identical that Arabic numerals are that soil characteristic expresses symbol, and there is each soil genetic horizon of Further Division from the soil profile arrangement by order from top to bottom；

Then each sampling point position in target soil region it is respectively directed to, each soil genetic horizon corresponding to sampling point position, divide by humus top layer, illuvium, chorizon, and press in genetic horizon characteristic merger table the soil genetic horizon characteristic expression symbol corresponding to each merger level, from this sampling point position profile region from top to bottom, divide each soil genetic horizon corresponding to this sampling point position successively, it is thus achieved that each merger layer corresponding to this sampling point position；

Meanwhile, in being respectively directed to target soil region, each sampling point position carries out in the process of above-mentioned merger operation, if occur identical soil genetic horizon characteristic expression symbol in two merger levels corresponding to sampling point position, then enters step 00503；

Step 00503. judges in each soil genetic horizon characteristic expression symbol corresponding to this sampling point position, whether existence is arranged in the soil genetic horizon characteristic expression symbol in genetic horizon characteristic merger table first merger level from top to bottom, it is that the soil genetic horizon that this identical soil genetic horizon characteristic expression symbol is corresponding respectively is merged, then perform step 00502 again and carry out merger process, it is thus achieved that each merger layer of correspondence is distinguished in this sampling point position；Otherwise enter step 00504；

Step 00504. is by this identical soil genetic horizon characteristic expression symbol, it is arranged in the soil genetic horizon characteristic expression symbol of lower floor's merger layer, it is divided in the merger layer at same soil genetic horizon characteristic expression symbol place, and by each soil genetic horizon characteristic expression symbol corresponding to this sampling point position, move up a merger level by the merger hierarchical sequence in genetic horizon characteristic merger table, it is thus achieved that each merger layer of correspondence is distinguished in this sampling point position。

7. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterized in that, in described step 007, according to target soil region merger layer kind set, merger layer corresponding to each sampling point position in target soil region carries out unified operation, make the kind of merger layer corresponding to each sampling point position in target soil region mutually the same, specifically include following operation:

It is respectively directed to each sampling point position in target soil region, it is judged that merger layer corresponding to sampling point position, whether equal to target soil region merger layer kind set, is do not do any operation；Otherwise arrange the merger layer that its relative target soil region merger layer kind set lacks to this sampling point position, the thickness information simultaneously arranging this merger layer is 0.00001cm, and the soil attribute information arranging this merger layer is 0.00001；And then make the merger layer that in target soil region, each sampling point position is corresponding respectively be equal to target soil region merger layer kind set。

8. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterised in that described step 011 specifically includes following operation:

It is respectively directed to each checking sampling point position in checking sampling point location sets, it is thus achieved that L verifies the drawing coefficient Rc of sampling point position each merger layer the first thickness prediction information sum corresponding relatively；Then according to the first prediction soil attribute information, the first thickness prediction information of each checking sampling point position each merger layer corresponding respectively in checking sampling point location sets, as follows:

${\mathrm{SOCD}}_{i_2}^{\′}=\frac{\underset{n_{i_2}=1}{\overset{N_{i_2}}{\Σ}}({\mathrm{SOC}}_{n_{i_2}}^{\′}\×{\mathrm{BD}}_{n_{i_2}}^{\′}\×(1-{\mathrm{Gr}}_{n_{i_2}}^{\′})\×Rc\×T_{n_{i_2}}^{\′})}{L}$

Obtain the first Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the first Soil Organic Carbon Density predictive value set PMTC-D of each checking sampling point position in checking sampling point location sets；Wherein, Represent checking sampling point location sets in i-th₂The sum of merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂The n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First organic carbon content information of forecasting of the n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First soil weight information of forecasting of the n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂More than the first gravel prediction content of preset diameters in n-th merger layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂First thickness prediction information of the n-th merger layer corresponding to individual checking sampling point position；Subsequently into step 012。

9. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterized in that, in described step 013, obtain the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, specifically include following operation:

It is respectively directed to each prediction sampling point position in prediction sampling point location sets, according to homalographic Spline function, adopts equation below:

$S\left(k_{i_1}\right)=\stackrel{\‾}{f}\left(k_{i_1}\right)+e\left(k_{i_1}\right)$

Obtain the soil attribute information of this each matching layer of prediction sampling point position；And then obtain the soil attribute information of each prediction sampling point position each matching layer corresponding respectively in prediction sampling point location sets, and wherein, i₁=1 ..., I₁, I₁Represent the quantity of prediction sampling point position in prediction sampling point location sets, Represent prediction sampling point location sets in i-th₁The number of plies of matching layer corresponding to individual prediction sampling point position,Represent prediction sampling point location sets in i-th₁The soil attribute information of kth layer matching layer corresponding to individual prediction sampling point position,RepresentFunction existsLayer withThe meansigma methods of layer fitting result,Represent prediction sampling point location sets in i-th₁The experimental determination analytical error of kth layer matching layer corresponding to individual prediction sampling point position,Function is homalographic Spline function。

10. a kind of Soil Carbon Stock evaluation method based on soil genetic horizon thickness prediction according to claim 1, it is characterised in that described step 015 specifically includes following operation:

The 3rd soil attribute information of forecasting of each matching layer corresponding to each checking sampling point position in checking sampling point location sets, and thickness information, as follows:

${\mathrm{SOCD}}_{i_2}^{\′\′}=\frac{\underset{k_{i_2}=1}{\overset{N_{i_2}}{\Σ}}({\mathrm{SOC}}_{k_{i_2}}^{\′\′}\×{\mathrm{BD}}_{k_{i_2}}^{\′\′}\×(1-{\mathrm{Gr}}_{k_{i_2}}^{\′\′})\×Rc\×T_{k_{i_2}}^{\′\′})}{L}$

Obtain the 3rd Soil Organic Carbon Density predictive value of each checking sampling point position in checking sampling point location setsConstitute the 3rd Soil Organic Carbon Density predictive value set PMTC-F of each checking sampling point position in checking sampling point location sets；Wherein, Represent checking sampling point location sets in i-th₂The sum of matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂Kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂3rd organic carbon content information of forecasting of kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂3rd soil weight information of forecasting of kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂More than the 3rd gravel prediction content of preset diameters in kth matching layer corresponding to individual checking sampling point position,Represent checking sampling point location sets in i-th₂The thickness information of kth matching layer corresponding to individual checking sampling point position；Subsequently into step 016。