CN108563733B - A kind of land use data assimilation method based on Bayesian inference - Google Patents
A kind of land use data assimilation method based on Bayesian inference Download PDFInfo
- Publication number
- CN108563733B CN108563733B CN201810311341.9A CN201810311341A CN108563733B CN 108563733 B CN108563733 B CN 108563733B CN 201810311341 A CN201810311341 A CN 201810311341A CN 108563733 B CN108563733 B CN 108563733B
- Authority
- CN
- China
- Prior art keywords
- land use
- assimilation
- use pattern
- model
- observation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 27
- 238000004088 simulation Methods 0.000 claims abstract description 17
- 238000011160 research Methods 0.000 claims abstract description 14
- 230000001413 cellular effect Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000007704 transition Effects 0.000 claims description 2
- 238000013316 zoning Methods 0.000 claims description 2
- 230000004927 fusion Effects 0.000 abstract description 6
- 239000002689 soil Substances 0.000 description 3
- 238000013480 data collection Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000007500 overflow downdraw method Methods 0.000 description 2
- 238000013439 planning Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N5/00—Computing arrangements using knowledge-based models
- G06N5/04—Inference or reasoning models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/16—Real estate
- G06Q50/165—Land development
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- Software Systems (AREA)
- Economics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention relates to a kind of land use data assimilation method based on Bayesian inference, the core of this method is that discrete type distribution variable is corresponded to continuous type distribution density, and using continuous type distribution density as intermediate variable, and then calculate the Posterior distrbutionp probability of categorical variable.Bayesian inference thought based on conjugate prior is introduced into land use conflict model by the present invention, is observed data by fusion, is assimilated to the land use pattern of discrete, multidimensional modeling, improves the simulation and forecast precision of land use conflict model.The present invention carries out assimilation research to a variety of land use patterns using multidimensional land use/cover variation dynamic model for the first time in terms of model.The present invention is firstly introduced a kind of assimilation strategy, is updated to discrete type distribution variable in method.
Description
Abstract
The present invention has developed a kind of land use data assimilation method based on Bayesian inference, the core of this method be by
Discrete type distribution variable corresponds to continuous type distribution density, and becomes continuous type distribution density as centre
Amount, and then calculate the Posterior distrbutionp probability of categorical variable.Bayesian inference thought based on conjugate prior is introduced by the present invention
In land use conflict model, data are observed by fusion, the land use pattern of discrete, multidimensional modeling is carried out
Assimilation improves the simulation and forecast precision of land use conflict model.
Technical field
The present invention relates to Geographical Information Sciences technical field, a kind of specifically land use based on Bayesian inference
Data assimilation method.
Background technique
Land use change survey is that mankind's activity influences most direct manifestation mode to earth surface system, by changing the earth
The underlying surface of surface knowledge system affects the cyclic process of the major ring layer of earth surface system.Therefore, it accurate simulation and predicted
Go, now, the true Land Use Change Pattern of the following epigeosphere and process, can be mentioned for land use planning, making policies
For scientific basis, Region Sustainable Development can also be served.
Model and observation are two kinds of basic means of land use change survey research.Model helps to understand land use in depth
Causes rebuild past, prediction future developing trend, Evaluation Environment influence and support land use planning Analysis of Policy Making.
Currently, cellular automata (Cellular Automata, CA) has become one of most common tool of simulation of land use changes.
But error caused by cellular Automation Model structure, static parameter and boundary condition, initial fields and uncertainty can be with moulds
The operation of type is constantly transmitted and is accumulated；And the earth's surface that the moment is observed on various spatial and temporal scales can rapidly and accurately be obtained by observing
True land utilization.But the situation without the observation moment can not be understood, land use change survey continuous in time cannot be provided
Situation.How observation information and model to be organically blended, it is various not true in land use simulation and prediction to quantify and reduce
It is qualitative, the Land-use of space and time continuous is obtained, is the significant challenge faced in current land use change survey research.
Data assimilation (Data Assimilation), also referred to as " model-data fusion " (mode-data fusion) are
A kind of dynamics frame based on model is considering mould by merging the directly or indirectly observation of separate sources, different resolution
It constantly relies on and observes and self-optimizing model track, Optimized model state on the basis of type and various observation errors, and can be same
It walks Optimal Parameters, quantitative assessment and reduces a kind of probabilistic method.Data assimilation is fusion observation and land use model
Theoretical frame is provided, is filled with fresh blood for the research method in the field.
However, no matter international or domestic, land use data assimilation research is at the early-stage, and there is also not for theory and method
Perfect place.Different from the hydrology, meteorology and land process model, the output of land use conflict model is discrete type
Variable (land use pattern) lacks the state variable that can be directly used for assimilation.It makes a general survey of Land Use and utilizes data assimilation
Correlative study does not carry out the soil to multidimensional it is found that be two-dimensional land use conflict model used in (1) current research
Ground utilizes the assimilation of dynamic model.(2) most researchs are based on particle filter and set Kalman filtering algorithm to continuous
Model parameter optimize, lack the update to discrete, multidimensional type distribution variable (land use pattern).
In consideration of it, it is necessary to aim at model in land-use study in the world-measurement fusion method.
Summary of the invention
For two above-mentioned main problems, the present invention proposes a kind of based on Bayesian inference from the angle of methodology
Land use data assimilation strategy, to improve the simulation and forecast precision of land use conflict model and the number of discrete variable
Preferable solution is provided according to assimilation method research.
The output of land use conflict model is a discrete land use pattern.Based on Bayesian inference method to soil
Ground is discrete type distribution variable to be corresponded to continuous type distribution density, and will connect using the core concept assimilated
Continuous type distribution density is as intermediate variable, to calculate the Posterior distrbutionp probability of categorical variable.
Specific land use data assimilation implementation steps are as follows:
S1: operation land use conflict model obtains the analogue value α in assimilation time, while extracting the sight in assimilation time
Measured value c.The analogue value and observation herein refers both to the corresponding number of each land use pattern.
S2: the Bayesian inference formula based on conjugate prior is used, obtains assimilation number, i.e., each land use in each observation point
The corresponding distributed density values of type.Wherein, the Bayesian inference formula based on conjugate prior is as follows:
In formula,It is the land use pattern after assimilationDistributed density values；It is land use patternObservation number
Mesh；The land use pattern simulated in representative modelNumber；It is the observation number of all land use patterns；It is
All land use pattern numbers of simulation.
S3: according to assimilation number, to current analog result, that is, the land use pattern simulated is modified, and obtains the year
Land use pattern after part assimilation.
The beneficial effects of the present invention are:
This brand-new developing direction of model-measurement fusion method in research on utilization of the present invention is visited from the angle of methodology
The new way of rope land-use study, on the basis of multidimensional land use conflict model, fusion observation data, optimization multidimensional from
Scattered type distribution variable, the room and time distribution characteristics of land use is more accurately provided with it.
The present invention is in terms of model, for the first time using multidimensional land use/cover variation dynamic model to a variety of land uses
Type carries out assimilation research.
The present invention is firstly introduced a kind of assimilation strategy, is updated to discrete type distribution variable in method.
Detailed description of the invention
Fig. 1 is the flow chart that the present invention implements land use data assimilation method.
Fig. 2 is the flow chart being modified to the analog result of land use conflict model.
Fig. 3 is that assimilation result and analog result and 2001 and 2009 true land utilization spaces of the invention are distributed
Contrast schematic diagram
Specific embodiment
Explanation is further elaborated to the present invention in the following with reference to the drawings and specific embodiments.
Embodiment 1
Research object in the present invention is positioned at the Zhangye And Its Surrounding Oasis Ganzhou District in the Middle Reaches of Heihe River area.Ganzhou District is Zhangye
The administrative center in oasis, about 3665 square kilometres of the gross area.Ganzhou District is the wet land protection policy of the starting of Zhangye And Its Surrounding Oasis in 2008
Emphasis implement region, and the Urban Land Expansion in the region and expansion plough it is more typical in Zhangye And Its Surrounding Oasis.It is adopted in this research
It is that multiple dimension logic returns Markov cellular Automation Model (MLRMCA) with land use conflict model, the test data of use
Are as follows: calibration data set of the Ganzhou District land use historical variations data set of 1992-1999 as MLRMCA model, with determination
Model parameter；Observation and result verification data of -2009 years 1999 Ganzhou District land use data collection as assimilation, every 2
Year is once assimilated.The above land use data collection is formed by TM/ETM+ image interpretation, using identical categorizing system.Institute
There are the data in region to be all made of same coordinate system, the raster data file with scale.
Fig. 1 is flow chart of the method for the present invention, mainly including the following steps:
Step 1: land use conflict model is calibrated using historical variations data set, obtains the parameter value of model.
Step 2: research zoning is divided into several square nets, each grid includes k × k pixel (cellular).Choosing
Several square nets therein are selected as observation point.
Step 3: it is general to obtain land use pattern conversion when simulating to when observing the time for operation land use conflict model
Rate and analogue value α, while extracting the observation c in assimilation time.The analogue value and observation herein refers both to each land use
The corresponding number of type.
Step 4: Bayesian inference formula of the operation based on conjugate prior obtains assimilation number.Wherein, based on conjugate prior
Bayesian inference formula is as follows:
In formula,It is the land use pattern after assimilationDistributed density values；It is land use patternObservation number
Mesh；The land use pattern simulated in representative modelNumber；It is the observation number of all land use patterns；It is
All land use pattern numbers of simulation.
Step 5: according to assimilation number, to current analog result, that is, the land use pattern simulated is modified, and is obtained
Land use pattern after obtaining observation time assimilation.
The process of amendment analog result is shown in Fig. 2, and specific amendment step is as follows:
S1: in observation point j, it is based on assimilation numberWith the area of square grid, each land use is calculated
The varied number of type, calculation formula are as follows:
S2: the land use pattern of varied number is successively found.
S3: by land use pattern transition probability from small to large by pixel reduce
A cellular (pixel), while the land use pattern of these pixels is assigned a value of 0.
S4: all land use pattern is redistributed according to step S3, finally obtaining needs again in observation point j
All cellulars of distribution.
S5: in cellular to be allocated, by remaining land use pattern, i.e., land use pattern is successively increased
Cellular.
S6: repeating step S1-S5, until completing the amendment of all observation points.
Step 6: being again input to revised result in model, as initial value, carries out the simulation assimilation in time.
When next observation moment is arrived in simulation, repetition step 3~step 5 terminates until the simulation assimilation time.
Fig. 3 is the Comparative result for assimilating result and modeling.The chart is bright based on the updated soil of method of the invention
Ground is closer to truth using type space distribution, can be improved the simulation and forecast precision of model.
Specific assimilation is as follows with simulation precision result:
(OA is overall accuracy, and Kappa is Kappa coefficient, and FM is performance index).
Claims (2)
1. a kind of land use data assimilation method based on Bayesian inference, it is characterised in that: the assimilation method include with
Lower step:
Step 1: land use conflict model is calibrated using historical variations data set, obtains the parameter value of model；
Step 2: research zoning is divided into several square nets, each grid includes k × k pixel；If selecting therein
Dry square net is as observation point；
Step 3: operation land use conflict model obtains land use pattern transition probability when simulating to when observing the time, with
And analogue value α, while extracting the observation c in assimilation time；
Step 4: Bayesian inference formula of the operation based on conjugate prior obtains assimilation number；Wherein, based on the pattra leaves of conjugate prior
This rational formula is as follows:
In formula,It is the land use pattern after assimilationDistributed density values；It is the observation number of land use pattern；
The land use pattern number simulated in representative model；N is the observation number of all land use patterns；It is simulation
All land use pattern numbers；
Step 5: according to assimilation number, current analog result is modified；
Specific amendment step is as follows:
S1: in observation point, it is based on assimilation numberWith the area of square grid, each land use class is calculated
The varied number of type, calculation formula is as follows:
S2: varied number is successively foundLand use pattern；
S3: willLand use patternTransition probability from small to large by pixel reduce
A cellular, i.e. pixel；The land use pattern of these pixels is assigned a value of 0 simultaneously；
S4: will ownLand use pattern redistributed according to step S3, finally obtain observation point
The middle all cellulars that need to be redistributed；
S5: in cellular to be allocated, by remaining land use pattern, i.e.,Land use pattern, successively
IncreaseA cellular；
S6: repeating step S1-S5, until completing the amendment of all observation points；
Step 6: being again input to revised result in model, as initial value, carries out the simulation assimilation in time；
When next observation moment is arrived in simulation, repetition step 3~step 5 terminates until the simulation assimilation time.
2. a kind of land use data assimilation method based on Bayesian inference as described in claim 1, it is characterised in that:
The analogue value and observation in step 3 refer both to the corresponding number of each land use pattern.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810311341.9A CN108563733B (en) | 2018-04-09 | 2018-04-09 | A kind of land use data assimilation method based on Bayesian inference |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810311341.9A CN108563733B (en) | 2018-04-09 | 2018-04-09 | A kind of land use data assimilation method based on Bayesian inference |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108563733A CN108563733A (en) | 2018-09-21 |
| CN108563733B true CN108563733B (en) | 2019-07-26 |
Family
ID=63534350
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810311341.9A Expired - Fee Related CN108563733B (en) | 2018-04-09 | 2018-04-09 | A kind of land use data assimilation method based on Bayesian inference |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108563733B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113505948B (en) * | 2021-09-13 | 2021-11-19 | 四川师范大学 | Future ecological security pattern prediction and optimization method based on Bayesian network |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737155A (en) * | 2011-04-12 | 2012-10-17 | 中国科学院寒区旱区环境与工程研究所 | Bayesian fitering-based general data assimilation method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101814117A (en) * | 2010-04-14 | 2010-08-25 | 北京师范大学 | Multi-source environment ecological information data assimilation method |
| CN102034027A (en) * | 2010-12-16 | 2011-04-27 | 南京大学 | Method for assimilating remote sensing data of soil humidity in watershed scale |
-
2018
- 2018-04-09 CN CN201810311341.9A patent/CN108563733B/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737155A (en) * | 2011-04-12 | 2012-10-17 | 中国科学院寒区旱区环境与工程研究所 | Bayesian fitering-based general data assimilation method |
Non-Patent Citations (2)
| Title |
|---|
| Identifying a land use change cellular automaton by Bayesian data assimilation;Judith A.Verstegen et.al;《Environmental Modelling & Software》;20131220;第121-136页 * |
| 数据同化算法研究现状综述;马建文 等;《地球科学进展》;20120710;第27卷(第7期);第747-754页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108563733A (en) | 2018-09-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wijesekara et al. | Assessing the impact of future land-use changes on hydrological processes in the Elbow River watershed in southern Alberta, Canada | |
| Fan et al. | Temporal and spatial change detecting (1998–2003) and predicting of land use and land cover in Core corridor of Pearl River Delta (China) by using TM and ETM+ images | |
| CN113902580B (en) | Historical farmland distribution reconstruction method based on random forest model | |
| Liu et al. | Rural land use spatial allocation in the semiarid loess hilly area in China: Using a Particle Swarm Optimization model equipped with multi-objective optimization techniques | |
| Li et al. | Pattern of spatial evolution of rural settlements in the Jizhou District of China during 1962–2030 | |
| Long et al. | Reconstruction of historical arable land use patterns using constrained cellular automata: A case study of Jiangsu, China | |
| CN112884226A (en) | Multi-agent algorithm-based territorial spatial pattern simulation planning method and system | |
| CN109271683A (en) | A kind of groups of building automatic placement algorithm towards sunshine constraint | |
| CN103886220B (en) | Water data discretization method for setting weight based on BP network and Gini coefficient | |
| CN105389469A (en) | Automatic calibration method of storm water management model parameters | |
| Chen et al. | Simulating the optimal land-use pattern in the farming-pastoral transitional zone of Northern China | |
| CN109858180B (en) | Regional crop water consumption space pattern optimization design method | |
| CN108733952B (en) | Three-dimensional characterization method for spatial variability of soil water content based on sequential simulation | |
| Xu et al. | Land-use change modeling with cellular automata using land natural evolution unit | |
| CN112528516A (en) | Watershed water environment management method coupling land utilization type and climate change | |
| CN113743659B (en) | Urban layout prediction method based on component method and Markov cellular automaton and application | |
| Wen et al. | Application of a loose coupling model for assessing the impact of land-cover changes on groundwater recharge in the Jinan spring area, China | |
| CN115952421A (en) | High-precision time-space simulation method for coupling ecological process model and machine learning algorithm | |
| CN108563733B (en) | A kind of land use data assimilation method based on Bayesian inference | |
| CN111859303B (en) | Soil humidity fusion method and system based on dynamic Bayesian average | |
| CN106546958B (en) | A kind of radar data assimilation method of optimization | |
| Schaldach et al. | Conceptual design and implementation of a model for the integrated simulation of large-scale land-use systems | |
| CN111914465A (en) | Data-free regional hydrological parameter calibration method based on clustering and particle swarm optimization | |
| Gadgay et al. | Novel ensemble neural network models for better prediction using variable input approach | |
| Shi et al. | Determination of the effective utilization coefficient of irrigation water based on geographically weighted regression |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200714 Address after: 730000 No.318, Donggang West Road, Chengguan District, Lanzhou City, Gansu Province Patentee after: NORTHWEST INSTITUTE OF ECO-ENVIRONMENT AND RESOURCES, CHINESE ACADEMY OF SCIENCES Address before: 730000 Gansu city of Lanzhou province Donggang West Road No. 320 Patentee before: COLD AND ARID REGIONS ENVIRONMENTAL AND ENGINEERING Research Institute,CHINESE ACADEMY OF SCIENCES |
|
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190726 |