Data output device and data output method

Abstract

To perform presence confirmation of analysis data used by an analysis device before analysis.SOLUTION: A data output device, which includes a processor for executing a program and a storage device for storing the program, is communicably connected to a data source for storing first time-series data regarding a natural environment and an analysis device for inputting analysis target data from the data output device to execute analysis regarding the natural environment, and includes a recording section capable of recording the analysis target data. The processor executes: input processing of receiving an input of an analysis condition regarding the natural environment; first determination processing of determining whether first analysis target data corresponding to the analysis condition input by the input processing is present in the recording section; and first output processing of outputting the first analysis target data to the analysis device on the basis of a first determination result by the first determination processing.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to a data output device and a data output method for outputting data. [Background technology]

[0002] Simulations that analyze data related to the natural environment have a wide range of applications, including flood forecasting, river flow rate and water quality forecasting, renewable energy power generation forecasting, and agricultural crop yield forecasting. These forecast calculations require highly accurate, long-term meteorological information spanning several decades as input data.

[0003] Generally, meteorological data includes time-series observation data at specific locations and distribution data that records distribution quantities within a region. The former typically includes observed values ​​such as temperature, precipitation, and solar radiation at meteorological observation points, while the latter includes distribution data such as precipitation measured by weather radar, temperature based on the results of meteorological simulations, and wind speed based on the results of meteorological simulations.

[0004] There are the following challenges in utilizing this weather data. Because weather observation points are sparsely located, the time-series observation data is information from locations geographically distant from the target area, raising questions about its accuracy. Distribution data contains a large amount of grid point information, and processing is required to extract only the information related to the point of interest.

[0005] Another common issue for both methods is that they are dependent on external data sources, which limits their data transfer capabilities. For example, it takes several hours to acquire 20 years' worth of meteorological data for a river basin, and this acquisition time is significantly longer than the calculation time of the simulation itself.

[0006] Specifically, the limitations of data transfer capacity are due to the fact that external data sources handle extremely large amounts of data, and infrequently used information must be retrieved from a slow archive area. Also, the limitations of the data storage format are due to the fact that multiple images of distribution data (for example, if one time step is one hour, then 8,760 images per year) must all be retrieved and written as time-series data for points of interest.

[0007] Patent Document 1 below discloses a flood restoration analysis tool method. This flood restoration analysis tool method includes a plurality of processes for accessing weather information for a geographical area divided into a plurality of cells and generating runoff data based on the weather information. The runoff data includes predicted unregulated water flow on the ground surface of each cell in the area. This flood restoration analysis tool method further includes a plurality of processes for generating predictions of water inflow and outflow between cells and calculating predicted water depths within each subcell of each cell within the geographical area based on the predictions of inflow and outflow between cells and a hydraulic model. This flood restoration analysis tool method further includes a plurality of processes for generating a flood inundation map showing predicted water depths for each subcell within the geographical area and presenting the flood inundation map on a user interface of a display device.

[0008] Non-Patent Document 1 below discloses a method for creating and storing a time-series data set called "Data Rods" in advance. [Prior art documents] [Patent documents]

[0009] [Patent Document 1] International Publication No. WO2019 / 204254 [Non-patent literature]

[0010] [Non-Patent Document 1] Teng, William, Rui, Hualan, Strub, Richard, and Vollmer, Bruce, 2016. Optimal Reorganization of NASA Earth Science Data for Enhanced Accessibility and Usability for the Hydrology Community. Journal of the American Water Resources Association (JAWRA) 52( 4): 825- 835. DOI: 10.1111 / 1752-1688.12405 Summary of the Invention [Problem to be solved by the invention]

[0011] Simulations of the natural environment require highly accurate, long-term meteorological information as input data. The distribution data described above must be extracted from mesh data and sorted in chronological order. Furthermore, because observation data and distribution data depend on external data sources and are subject to data transfer capacity limitations, the lead time from when data is requested to when data analysis can begin is long.

[0012] Patent Document 1 discloses a flood monitoring system that predicts and displays flooding based on meteorological data. It shows that part of the system's configuration includes a flood analysis device based on the automatic generation of model data, and that meteorological data is appropriately recorded in a cache and on a server. However, it does not mention the specific configuration method or a method for balancing the processing time and storage capacity required for format conversion of large-scale data.

[0013] In Non-Patent Document 1, the data overlaps with the distribution data, occupying twice the storage capacity. A global-scale dataset spanning several decades would reach a petabyte-scale capacity, making it unrealistic for a single user to maintain and manage it. It would be desirable to be able to pinpoint and extract such data from external data sources in accordance with the analysis task. However, the aforementioned access speed problem arises. This issue is particularly problematic in use cases that require numerous recalculations due to condition changes or optimization.

[0014] The present invention aims to check the existence of analysis data to be used by an analysis device before analysis. [Means for solving the problem]

[0015] A data output device according to one aspect of the invention disclosed in the present application is a data output device having a processor that executes a program and a storage device that stores the program, and further comprising: a data source that stores first time series data relating to the natural environment; an analysis device that inputs analysis target data from the data output device and performs analysis relating to the natural environment; and a recording unit that is communicatively connected and can record the analysis target data, wherein the processor executes an input process that accepts input of analysis conditions relating to the natural environment; a first determination process that determines whether first analysis target data that matches the analysis conditions input by the input process exists in the recording unit; and a first output process that outputs the first analysis target data to the analysis device based on the first determination result by the first determination process. [Effects of the Invention]

[0016] According to a representative embodiment of the present invention, the existence of analysis data to be used by the analysis device can be confirmed before analysis. Problems, configurations, and effects other than those described above will become apparent from the following description of the examples. [Brief explanation of the drawings]

[0017] [Figure 1] FIG. 1 is an explanatory diagram illustrating an example of an analysis system. [Figure 2] FIG. 2 is a block diagram showing an example of the hardware configuration of a computer. [Figure 3] FIG. 3 is a block diagram illustrating an example of the functional configuration of the analysis device. [Figure 4] FIG. 4 is an explanatory diagram showing the progress of execution time in the analysis system. [Figure 5] FIG. 5 is a sequence diagram showing a data request sequence 1 (without reusable data). [Figure 6] FIG. 6 is a sequence diagram showing an initial analysis sequence (without reusable data). [Figure 7] FIG. 7 is a sequence diagram showing the analysis target acquisition process 2 (without reusable data). [Figure 8] FIG. 8 is a sequence diagram showing the sequence of data shaping and calibration (without reused data). [Figure 9] FIG. 9 is a sequence diagram showing the sequence of data shaping and validation (without reused data). [Figure 10] FIG. 10 is a sequence diagram showing a data request sequence 1 (with reusable data). [Figure 11] FIG. 11 is a sequence diagram showing an initial analysis sequence (with reused data). [Figure 12] FIG. 12 is a sequence diagram showing a calibration sequence (with reused data). [Figure 13] FIG. 13 is a sequence diagram showing a validation sequence (without reused data). [Figure 14] FIG. 14 is an explanatory diagram showing a display example of the region setting screen. [Figure 15] FIG. 15 is a sequence diagram showing the sequence for selecting a river to be analyzed. [Figure 16] FIG. 16 is an explanatory diagram showing a display example of the analysis management screen. [Figure 17] FIG. 17 is an explanatory diagram showing a display example of the result display screen. DETAILED DESCRIPTION OF THE INVENTION

[0018] <Figure 1 Analysis system> 1 is an explanatory diagram showing an example of an analysis system. Analysis system 100 includes data output device 101, analysis device 102, and data source 103. Data output device 101, analysis device 102, and data source 103 are communicably connected via network 110, such as the Internet, a LAN (Local Area Network), or a WAN (Wide Area Network).

[0019] The data output device 101 forms input data related to the natural environment to be used in the analysis of the analysis device 102. The analysis device 102 performs analysis using the input data formed by the data output device 101. The data source 103 is a website or public database present on the network 110. The data source 103 stores map data and weather data.

[0020] The map data is the latitude, longitude, and altitude of the observation point. The weather data is observation data related to the weather at the observation point, such as temperature, water volume, nitrogen concentration, and phosphorus concentration. The original data is, for example, map data and weather data for a predetermined period (e.g., one year) within the analysis period.

[0021] <Figure 2: Example of hardware configuration of computer (data output device 101, analysis device 102)> FIG. 2 is a block diagram showing an example of the hardware configuration of a computer. The computer 200 includes a processor 201, a storage device 202, an input device 203, an output device 204, and a communication interface (communication IF) 205. The processor 201, the storage device 202, the input device 203, the output device 204, and the communication IF 205 are connected via a bus 206. The processor 201 controls the computer 200. The storage device 202 serves as a working area for the processor 201. The storage device 202 is a non-transitory or temporary recording medium that stores various programs and data. Examples of the storage device 202 include a read-only memory (ROM), a random access memory (RAM), a hard disk drive (HDD), and a flash memory. The input device 203 inputs data. Examples of the input device 203 include a keyboard, a mouse, a touch panel, a numeric keypad, a scanner, a microphone, and a sensor. The output device 204 outputs data. The output device 204 may be, for example, a display, a printer, or a speaker. The communication IF 205 connects to the network 110 and transmits and receives data.

[0022] <Figure 3: Example of functional configuration of analysis device 102> 3 is a block diagram showing an example of the functional configuration of analysis device 102. Data output device 101 has a data management unit 310, a data acquisition unit 311, a data shaping unit 312, and a dummy data generation unit 313. Specifically, data management unit 310, data acquisition unit 311, data shaping unit 312, and dummy data generation unit 313 are realized, for example, by having processor 201 execute a program stored in storage device 202 shown in FIG.

[0023] The data output device 101 has a data management information recording unit 300, an original data recording unit 301, an intermediate data recording unit 302, a formatted data recording unit 303, and a dummy data recording unit 304. Specifically, the data management information recording unit 300, the original data recording unit 301, the intermediate data recording unit 302, the formatted data recording unit 303, and the dummy data recording unit 304 are realized by, for example, the storage device 202 shown in the figure.

[0024] In addition, the original data recording unit 301, intermediate data recording unit 302, formatted data recording unit 303, and dummy data recording unit 304 may be constructed on an independent server such as a file server or distributed storage, or may be provided through a database system such as a relational database or key-value store.

[0025] The data output device 101 has a region setting screen 314, an analysis management screen 315, and a result display screen 316. The region setting screen 314 is a screen for setting a region to be analyzed. The analysis management screen 315 is a screen for managing the analysis. The result display screen 316 is a screen for displaying the analysis results. Specifically, the region setting screen 314, the analysis management screen 315, and the result display screen 316 are realized by, for example, a display, which is an example of the output device 204 shown in FIG. 2 .

[0026] The analysis device 102 includes an analysis unit 321. Specifically, the analysis unit 321 is realized by, for example, causing the processor 201 to execute a program stored in the storage device 202 shown in FIG.

[0027] Analysis device 102 has analysis result recording unit 322. Specifically, analysis result recording unit 322 is realized by, for example, storage device 202 shown in the drawing.

[0028] Analysis device 102 has analysis setting screen 323. Specifically, analysis setting screen 323 is realized by causing processor 201 to execute a program stored in storage device 202 shown in FIG.

[0029] (Data management information recording unit 300) The data management information recording unit 300 records data management information. The data management information is information used by the data management unit 310. For example, the data management information is information about rivers that are the subject of data management. The information about rivers includes geographical information such as the location, length, and width of the river, as well as water quality information such as flow rate and contained substances.

[0030] (Data management unit 310) The data management unit 310 manages the data acquisition unit 311, the data formatting unit 312, and the dummy data generation unit 313, and generates a watershed model required for the simulation. The data management unit 310 executes, for example, the following processes.

[0031] The data management unit 310 receives input from the user 330 and instructs the data acquisition unit 311 to acquire map data and weather data.

[0032] The data management unit 310 refers to the original data recording unit 301 and the intermediate data recording unit 302 and identifies the range of data that requires access to the external data source 103 for each of the calibration period and the validation period.

[0033] The data management unit 310 instructs the analysis device 102 to perform an initial analysis. The data management unit 310 may display the analysis process and expected completion time on the analysis management screen 315. The data management unit 310 refers to the data management information recording unit 300 to search for a list of rivers.

[0034] (Data acquisition unit 311 and original data recording unit 301) The data acquisition unit 311 acquires data from the data source 103 and records the acquired data in the original data recording unit 301. Specifically, for example, in response to an instruction from the data management unit 310, the data acquisition unit 311 accesses the data source 103 to acquire map data and weather data, and records the map data and weather data in the original data recording unit 301. The map data and weather data recorded in the original data recording unit 301 are referred to as original data.

[0035] The data acquisition unit 311 can also make a request to the data source 103 to divide the data acquisition range appropriately in accordance with specifications that the data source 103 can accept (for example, for distribution data, one physical quantity (distribution amount at one point in time in a specified latitude and longitude range including a river basin); for time series data, one observation point over a one-month period).

[0036] (Data formatting unit 312, formatted data recording unit 303, and intermediate data recording unit 302) The data reforming unit 312 reforms the original data recorded in the original data recording unit 301. Specifically, for example, the data reforming unit 312 appropriately converts the original data into a format that is easy to handle internally, extracts and combines the data according to the required spatial domain and time interval, performs format conversion and unit conversion into a form that can be read by the analysis unit 321, and records the result in the reformed data recording unit 303. The data recorded in the reformed data recording unit 303 is referred to as reformed data.

[0037] The shaping-completed data recording unit 303 stores the shaping-completed data without discarding it for a certain period of time even after the analysis by the analysis unit 321 is completed. The shaping-completed data is reused when an instruction is given to analyze the same spatial region or the same time interval again.

[0038] The data shaping unit 312 records data obtained by combining multiple pieces of original data as intermediate data in the intermediate data recording unit 302. For example, if the analysis target period is 10 years and one piece of original data is time-series data for one year, the time-series data for 10 years obtained by chronologically concatenating 10 pieces of original data becomes the intermediate data. The data shaping unit 312 reads out the intermediate data from the intermediate data recording unit 302. The read-out intermediate data can also be reused (extracted and shaped for a different time interval) in other analyses. The original data, intermediate data, and shaped data are collectively referred to as analysis target data.

[0039] (Dummy data generating unit 313 and dummy data recording unit 304) The dummy data generation unit 313 generates dummy data that differs from actual measurements. Specifically, for example, the dummy data generation unit 313 generates dummy data based on statistical quantities such as the range and average value of each numerical value, instead of actual data. For example, the dummy data generation unit 313 may calculate, as dummy data, statistical quantities such as the average value of observed data in the original data recorded in the original data recording unit 301. This allows the analysis unit 321 to perform analysis based on the dummy data without waiting for data acquisition and data shaping in applications where calculation accuracy is not required.

[0040] (Analysis unit 321 and analysis result recording unit 322) After the preparation of the model data is complete, the analysis unit 321 generates a watershed model based on the model data and input from the user 330. The watershed model is a model that represents a watershed using a collection of grids called Hydrological Response Units (HRUs), which are subdivided based on a combination of soil type, land use, elevation, and its gradient, and is, for example, the Sail & Water Assessment Tool (SWAT) model.

[0041] The analysis unit 321 performs analysis using the generated watershed model. For example, the analysis unit 321 calculates the time-dependent changes in river flow rate and water quality based on the topography, precipitation, and vegetation distribution in the watershed model. At this time, the analysis unit 321 uses model equations to calculate the amount of rainwater reaching the ground surface, the amount of water evapotranspiration from the ground surface, the amount of water infiltrating into the soil and the amount of water runoff into rivers and groundwater, or the amount of vegetation growth and material transfer. The analysis unit 321 sequentially records the analysis results in the analysis result recording unit 322. The analysis results can be displayed on the result display screen 316.

[0042] <Figure 4 Execution time progression in the analysis system 100> 4 is an explanatory diagram showing the passage of execution time in analysis system 100. The horizontal axis is the time axis. When analyzing the time that passes from the analysis instruction from user 330 to the display of the results in the operation of analysis system 100, the time required to obtain data can sometimes be significant compared to the calculation time of the main body of the simulation.

[0043] (A) shows the execution time lapse in the conventional technology, that is, the execution time lapse in the case where there is no data output device 101. In (A), the execution is performed in the following order: planning ⇒ data acquisition and pre-processing 1 ⇒ watershed model construction ⇒ calibration ⇒ data acquisition and pre-processing 2 ⇒ validation ⇒ evaluation and consideration.

[0044] (B) shows the execution time when the data output device 101 is used to reuse already acquired data. (B) is also executed in the same order as (A): planning ⇒ data acquisition and pre-processing 1 ⇒ watershed model construction ⇒ calibration ⇒ data acquisition and pre-processing 2 ⇒ validation ⇒ evaluation and consideration. However, in (B), the data output device 101 reuses already acquired data in data acquisition and pre-processing 1, so the processing time for data acquisition and pre-processing 1 is shorter than the processing time for data acquisition and pre-processing 1 in (A). Similarly, in (B), the data output device 101 reuses already acquired data in data acquisition and pre-processing 2, so the processing time for data acquisition and pre-processing 2 is shorter than the processing time for data acquisition and pre-processing 2 in (A).

[0045] In other words, in (B), acquired data is stored internally and can be reused when a data request for the same section and location occurs. This cache function reduces the number of accesses to the slow external data source 103, shortening the waiting time before analysis can begin.

[0046] (C) shows the execution time when data acquisition and analysis are performed in parallel using the data output device 101. In (C), data acquisition and pre-processing 1 are performed in parallel with the construction of the watershed model. Therefore, data acquisition and pre-processing 1 are performed in parallel with the calibration. Therefore, the execution time is shorter than in (A). In this embodiment, processing such as (B) or (C) above is possible.

[0047] In other words, in (C), data acquisition and analysis processing are performed in parallel, hiding the data waiting time. Generally, analysis begins after the construction of the watershed model is complete, but depending on the use case, calculations may be able to begin earlier before all data for the period to be analyzed is available. In such cases, rather than waiting for all data to arrive as in (A), it is possible to perform analysis using only the data that has already been acquired.

[0048] Execution of a river basin model requires three main steps: initial analysis, calibration, and validation. The initial analysis includes building a river basin model and checking whether the data format conforms to the requirements of the analysis unit 321 (file format, input range of values, unit system, etc.).

[0049] Calibration is the adjustment of the model parameters of a watershed model to match observed data (water volume, water level, and water quality) with the data calculated by the watershed model. Validation is the process of verifying the calculation accuracy of a watershed model after calibration is complete.

[0050] Validation involves calculating the watershed model using data from a period different from the period of data used for calibration, and confirming that the watershed model calculations can continue properly after calibration.

[0051] That is, the time series data applied to validation and the time series data applied to calibration are from different dates and do not overlap, so that the time series data for validation can be obtained in parallel while the calibration calculation is being performed.

[0052] Weather data and water level data are used with a time step of one day and a period of 10 years. For example, five years' worth of data is used for calibration, and 20 years' worth is used for validation. During the calibration calculation, the model parameters of the watershed model are changed in various ways, and the optimal model parameters are determined after more than 100 repeated calculations.

[0053] As a result, since the nature and duration of the input data required for the initial analysis, calibration, and validation calculations are different, the acquisition of each and the calculations using that data can be performed in parallel (within certain ordering constraints).

[0054] 4(B), in parallel with data acquisition by data output device 101 of calibration data and pre-processing 1, analysis device 102 performs initial analysis using dummy data, and performs calibration after the initial analysis is completed and data for calibration is acquired. Furthermore, as shown in (C), data output device 101 acquires data for validation and performs pre-processing 2 while analysis device 102 is performing calibration, and analysis device 102 performs validation using the watershed model after calibration is completed.

[0055] In this way, data acquisition and analysis are performed by the independent data output device 101 and analysis device 102 in cooperation with each other, and it is possible to eliminate factors that affect the dependency between the processes and waiting time.

[0056] [River water quality simulation example] Next, an example of a simulation of river water quality will be described. Figures 5 to 9 show an example of a simulation of river water quality when there is no reused data in the data output device 101, and Figures 10 to 13 show an example of a simulation of river water quality when there is reused data in the data output device 101, that is, after the reused data has been registered in Figures 5 to 9.

[0057] The difference between the two is whether or not input data indicating the area and period to be analyzed has already been obtained in the original data recording unit 301 or intermediate data recording unit 302. This section describes how the original data recording unit 301 or intermediate data recording unit 302 operates as a data cache, and how dummy data is used to speed up the execution of initial analysis. The following explains this in chronological order.

[0058] <Figure 5 Data request sequence 1 (no reused data)> FIG. 5 is a sequence diagram showing a data request sequence 1 (without reusable data).

[0059] (Step S501) The data output device 101 receives, through operation of the input device 203 by the user 330, the selection of the river to be analyzed and the input of the period to be analyzed as analysis conditions related to the natural environment.

[0060] (Step S502) The data output device 101 displays the river to be analyzed and the period to be analyzed, and transmits them to the data management unit 310. In this example, the river to be analyzed is the "Tokachi River," and the period to be analyzed is "2004 to 2014."

[0061] (Step S503) The data management unit 310 executes a data acquisition request process. In the data acquisition request process, the data management unit 310 identifies one or more observation points on the river to be analyzed (hereinafter referred to as the analysis target observation points) and sends a data existence query to the data reforming unit 312. In this example, the raw data recording unit 301, the intermediate data recording unit 302, and the reformed data recording unit 303 do not cache the analysis target data (reused data) corresponding to the analysis target observation points "A, B, C (on the Tokachi River)" and the analysis target period "2004 to 2014." Therefore, the data management unit 310 receives a data non-existence notification from the data reforming unit 312.

[0062] Since there is no reusable data, the data management unit 310 sends a request to the data acquisition unit 311 to acquire map data and meteorological data (original data) corresponding to the analysis target observation points "A, B, C (on the Tokachi River)" and the analysis target period "2004 to 2014" from the data source 103. The data management unit 310 then outputs a notification to the analysis management screen 315 indicating that data acquisition is in progress.

[0063] (Step S504) The data output device 101 displays "Data acquisition in progress" on the analysis management screen 315.

[0064] (Step S505) When the data formatting unit 312 receives the data existence inquiry (step S503), it checks whether the analysis target data is cached in the original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303. Here, the data management unit 310 refers to the original data recording unit 301 and the intermediate data recording unit 302, and specifies the range of analysis target data that requires access to the external data source 103 for each of the calibration period and the validation period.

[0065] In this example, since the data to be analyzed is not cached, the data reforming unit 312 receives a data non-existence notification from each of the original data recording unit 301, the intermediate data recording unit 302, and the reformed data recording unit 303.

[0066] When the data reforming unit 312 receives a data non-existence notification from each of the intermediate data recording unit 302 and the reformed data recording unit 303 , it returns a data non-existence notification to the data management unit 310 .

[0067] (Step S506) The original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303 each check whether or not the analysis target data exists in response to a request from the data formatting unit 312. In this example, the analysis target data is not cached in any of the original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303. Therefore, the original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303 each return a data non-existence notification to the data formatting unit 312.

[0068] (Step S507) The data acquisition unit 311 executes the analysis target acquisition process 1. Specifically, for example, when the data acquisition unit 311 receives from the data management unit 310 an acquisition request for analysis target data to be used for calibration from the data source 103, the data acquisition unit 311 accesses the data source 103 and acquires the analysis target data from the data source 103.

[0069] In this case, the data acquisition unit 311 divides the acquisition range of the original data into K pieces (K is an integer equal to or greater than 1) according to the specifications that the data source 103 can accept, and sends an acquisition request for the K pieces of original data (hereinafter referred to as divided original data) to the data source 103. For example, in the case of distribution data, one physical quantity and a distribution amount at one point in time in a range of predetermined latitude and longitude that includes a river basin constitute one piece of original data. Also, in the case of time-series data, for example, observation data from one observation point for a one-month period constitutes one piece of divided analysis target data.

[0070] (Step S508) The data source 103 executes request processing. Specifically, for example, for each acquisition request for divided analysis target data used for calibration from the data acquisition unit 311, the data source 103 acquires the corresponding divided analysis target data and returns it to the data acquisition unit 311.

[0071] (Step S509) The original data recording unit 301 records the divided original data from the data acquisition unit 311 .

[0072] (Step S510) The data management unit 310 transmits a request for generating dummy data to the dummy data generation unit 313. Specifically, for example, the data management unit 310 may transmit the request for generating dummy data to the dummy data generation unit 313 in response to an operation of the input device 203 by the user 330. Furthermore, the data management unit 310 constructs a watershed model using the split analysis target data, and when it detects that the split analysis target data necessary for constructing the watershed model is insufficient, it may transmit a request for generating dummy data to the dummy data generation unit 313 prior to acquiring the missing split analysis target data.

[0073] (Step S511) When the dummy data generating unit 313 receives a request to generate dummy data from the data management unit 310 , it generates dummy data and records it in the dummy data recording unit 304 .

[0074] (Step S512) The dummy data recording unit 304 records the dummy data generated by the dummy data generating unit 313 .

[0075] <Figure 6 Initial analysis sequence (no reused data)> Figure 6 is a sequence diagram showing the initial analysis sequence (without reused data). Note that the start conditions for the initial analysis are the existence of a watershed model including the analysis target, and the analysis target data or its dummy data for calibration. In this example, we will explain the case where the start conditions are met.

[0076] (Step S601) Data management unit 310 executes initial analysis request processing. In the initial analysis request processing, data management unit 310 transmits an initial analysis execution request to analysis device 102 at any timing after execution of data acquisition request processing starts. When data management unit 310 receives an initial analysis completion notification from analysis device 102, it outputs the initial analysis completion notification to analysis management screen 315.

[0077] (Step S602) Upon receiving a request to perform an initial analysis from data management unit 310, analysis device 102 performs the initial analysis. Alternatively, analysis device 102 may read dummy data from dummy data recording unit 304 and perform the initial analysis using the dummy data. In the initial analysis (step S602), analysis device 102 checks for any defects associated with input that does not match the assumptions of analysis device 102, such as failure of calculation results due to missing data or the inclusion of unnatural values. Using dummy data in the initial analysis (step S602) shortens the waiting time before calibration.

[0078] As shown in FIG. 6, even during the execution of the initial analysis (step S601), data acquisition in the data request sequence 1 (FIG. 5) continues.

[0079] (Step S603) When accessed by the analysis device 102, the dummy data recording unit 304 writes dummy data to the analysis device 102.

[0080] (Step S604) The data output device 101 displays "Initial analysis in progress" on the analysis management screen 315.

[0081] (Step S605) The data output device 101 displays "Initial analysis completed" on the analysis management screen 315.

[0082] <Figure 7 Analysis target acquisition process 2 (no reused data)> 7 is a sequence diagram showing analysis target acquisition process 2 (without reuse data). Analysis target acquisition process 2 (without reuse data) is executed after analysis target acquisition process 1 (step S507). Analysis target acquisition process 1 (step S507) is a process for acquiring analysis target data to be used for calibration, while analysis target acquisition process 2 is a process for acquiring analysis target data to be used for validation.

[0083] (Step S707) The data acquisition unit 311 executes the analysis target acquisition process 2. Specifically, for example, when the analysis target acquisition process 1 is completed, the data acquisition unit 311 accesses the data source 103 and acquires the analysis target data from the data source 103.

[0084] In this case, the data acquisition unit 311 divides the acquisition range of the original data to be used for validation into L pieces (L is an integer greater than or equal to 1) according to the specifications acceptable to the data source 103, and sends an acquisition request for the L pieces of divided analysis target data to the data source 103.

[0085] (Step S708) The data source 103 executes the request processing. Specifically, for example, for each acquisition request for source data to be divided from the data acquisition unit 311, the data source 103 acquires the corresponding source data to be divided and returns it to the data acquisition unit 311.

[0086] (Step S709) The original data recording unit 301 records the divided original data from the data acquisition unit 311 .

[0087] <Figure 8 Data shaping and calibration sequence (no reused data)> Figure 8 is a sequence diagram showing the sequence of data shaping and calibration (without reused data). Note that the calibration start condition is that a watershed model and analysis target data for calibration are available. In this example, it is assumed that the start condition is met.

[0088] (Step S801) When the data management unit 310 receives a notification of completion of the analysis target acquisition process 1 (step S507) from the data acquisition unit 311, it executes a calibration request process. In the calibration request process, the data management unit 310 transmits a data shaping request for calibration to the data shaping unit 312. Next, when the data management unit 310 receives a notification of completion of data shaping for calibration from the data shaping unit 312, it transmits a calibration request to the analysis device 102. In addition, the data management unit 310 outputs a notification indicating that calibration calculation is in progress to the analysis management screen 315.

[0089] (Step S802) The data output device 101 displays "Calibration calculation in progress" on the analysis management screen 315.

[0090] (Step S803) When the data shaping unit 312 receives a data shaping request for calibration from the data management unit 310, it executes data shaping processing for calibration. Specifically, for example, the data shaping unit 312 reads original data from the original data recording unit 301, generates intermediate data, and records the intermediate data in the intermediate data recording unit 302. When the data shaping unit 312 finishes recording the intermediate data, it transmits a notification of completion of data shaping for calibration to the data management unit 310.

[0091] (Step S804) When the original data recording unit 301 receives a request to read the original data from the data reforming unit 312 , it writes the original data to the data reforming unit 312 .

[0092] (Step S805) When the intermediate data recording unit 302 receives a request to write intermediate data from the data reforming unit 312 , it writes the intermediate data from the data reforming unit 312 to the data reforming unit 312 .

[0093] (Step S806) When the formatted data recording unit 303 receives a request to write formatted data from the data formatting unit 312 , the formatted data recording unit 303 writes the formatted data from the data formatting unit 312 to the data formatting unit 312 .

[0094] (Step S807) When analysis device 102 receives a calibration request from data management unit 310, it performs calibration after completing initial analysis. Calibration (step S807) is performed using the shaping-completed data read from shaping-completed data recording unit 303 as the analysis target data for calibration, rather than dummy data.

[0095] Furthermore, in calibration (step S807), unlike the initial analysis and validation, the analysis device 102 executes multiple analysis jobs J1 to Jn (n is an integer greater than or equal to 1) for the same period, varying the model parameters of the watershed model, to determine optimal model parameters (e.g., those that minimize the error with the observed data). This makes it possible to absorb the waiting time for the data to be analyzed for validation. When analysis jobs J1 to Jn are completed, the analysis device 102 sends a calibration completion notification to the data management unit 310.

[0096] (Step S808) When a request to read out the shaping-completed data is received from the analysis device 102, the shaping-completed data recording unit 303 writes the shaping-completed data to the analysis device 102.

[0097] (Step S809) When data management unit 310 receives a calibration completion notification from analysis device 102 , it outputs the calibration completion notification to analysis management screen 315 .

[0098] (Step S810) The data output device 101 displays "Calibration completed" on the analysis management screen 315.

[0099] (Step S811) The data output device 101 receives confirmation of the end of calibration by the user 330 operating the input device 203 .

[0100] (Step S812) Upon receiving confirmation of the completion of calibration, the data output device 101 executes a calibration result visualization process. Specifically, for example, the data output device 101 sends a calibration result visualization request to the data management unit 310 and acquires the calibration results from the data management unit 310. Then, the data output device 101 displays the calibration results on the result display screen 316.

[0101] (Step S813) When the data management unit 310 receives the calibration result visualization request, it transmits a calibration result acquisition request to the analysis device 102. Then, the data management unit 310 acquires the calibration results from the analysis device 102. When the data management unit 310 acquires the calibration results from the analysis device 102, it transmits the calibration results to the data management unit 310.

[0102] (Step S814) When analytical device 102 receives a calibration result acquisition request from data management unit 310, it returns the calibration result to data management unit 310.

[0103] <Figure 9 Data formatting and validation sequence (no reused data)> Figure 9 is a sequence diagram showing the sequence of data shaping and validation (without reused data). Note that the validation start condition is that the watershed model, the analysis target data for calibration, and the analysis target data for validation are available. In this example, it is assumed that the start condition is met.

[0104] (Step S901) When the data management unit 310 receives a notification of completion of the analysis target acquisition process 2 (step S707) from the data acquisition unit 311, it executes validation request processing. In the validation request processing, the data management unit 310 transmits a data shaping request for validation to the data shaping unit 312. Next, when the data management unit 310 receives a notification of completion of data shaping for validation from the data shaping unit 312, it transmits the validation request to the analysis device 102. In addition, the data management unit 310 outputs a notification indicating that validation calculation is in progress to the analysis management screen 315.

[0105] (Step S902) The data output device 101 displays "Validation calculation in progress" on the analysis management screen 315.

[0106] (Step S903) The data reforming unit 312 executes data reforming processing for validation when it receives a data reforming request for validation from the data management unit 310. Specifically, for example, the data reforming unit 312 reads original data from the original data recording unit 301, generates intermediate data, and records the intermediate data in the intermediate data recording unit 302. When the recording of the intermediate data is completed, the data reforming unit 312 transmits a completion notification of data reforming for validation to the data management unit 310.

[0107] (Step S904) When the original data recording unit 301 receives a request to read the original data from the data reforming unit 312 , it writes the original data to the data reforming unit 312 .

[0108] (Step S905) When the intermediate data recording unit 302 receives a request to write intermediate data from the data reforming unit 312 , it writes the intermediate data from the data reforming unit 312 to the data reforming unit 312 .

[0109] (Step S906) When the formatted data recording unit 303 receives a request to write formatted data from the data formatting unit 312 , the formatted data recording unit 303 writes the formatted data from the data formatting unit 312 to the data formatting unit 312 .

[0110] (Step S907) When the analysis device 102 receives a validation request from the data management unit 310, it executes validation after completing the initial analysis. The validation (step S907) is executed using the formatted data read from the formatted data recording unit 303 as the analysis target data for validation, rather than dummy data. When the validation is completed, the analysis device 102 transmits a validation completion notification to the data management unit 310.

[0111] (Step S908) When a request to read out the shaping-completed data is received from the analysis device 102, the shaping-completed data recording unit 303 writes the shaping-completed data to the analysis device 102.

[0112] (Step S909) When the data management unit 310 receives the validation completion notification from the analysis device 102 , it outputs the validation completion notification to the analysis management screen 315 .

[0113] (Step S910) The data output device 101 displays "Validation Completed" on the analysis management screen 315.

[0114] (Step S911) The data output device 101 receives confirmation of the end of validation by the user 330 operating the input device 203 .

[0115] (Step S912) Upon receiving confirmation of the completion of validation, the data output device 101 executes validation result visualization processing. Specifically, for example, the data output device 101 sends a validation result visualization request to the data management unit 310 and obtains the validation results from the data management unit 310. Then, the data output device 101 displays the validation results on the result display screen 316.

[0116] (Step S913) When the data management unit 310 receives the validation result visualization request, it transmits a validation result acquisition request to the analysis device 102. Then, the data management unit 310 acquires the validation result from the analysis device 102. When the data management unit 310 acquires the validation result from the analysis device 102, it transmits the validation result to the data management unit 310.

[0117] (Step S914) When the analysis device 102 receives a validation result acquisition request from the data management unit 310 , it returns the validation result to the data management unit 310 .

[0118] <Figure 10 Data request sequence 1 (with reused data)> FIG. 10 is a sequence diagram showing a data request sequence 1 (with reusable data).

[0119] (Step S1001) The data output device 101 receives, through operation of the input device 203 by the user 330, the selection of the river to be analyzed and the input of the period to be analyzed as analysis conditions related to the natural environment.

[0120] (Step S1002) The data output device 101 displays the river to be analyzed and the period to be analyzed, and transmits them to the data management unit 310. In this example, the river to be analyzed is the "Tokachi River," and the period to be analyzed is "2004 to 2022."

[0121] (Step S1003) The data management unit 310 executes a data acquisition request process. In the data acquisition request process, the data management unit 310 identifies one or more analysis target observation points on the analysis target river and sends a data existence query to the data reforming unit 312. In this example, the analysis target data (reused data) corresponding to the analysis target observation points "A, B, C (on the Tokachi River)" and the analysis target period "2004 to 2014" are cached in the original data recording unit 301, intermediate data recording unit 302, and reformed data recording unit 303. Therefore, the data management unit 310 receives a data existence notification from the data reforming unit 312.

[0122] Because there is reusable data, the data management unit 310 sends a request to the data acquisition unit 311 to acquire map data and meteorological data (original data) corresponding to the analysis target observation points "A, B, C (on the Tokachi River)" and the analysis target period "2004 to 2014" from the data source 103. The data management unit 310 then outputs a notification to the analysis management screen 315 indicating that data acquisition is in progress.

[0123] (Step S1004) The data output device 101 displays "Data acquisition in progress" on the analysis management screen 315.

[0124] (Step S1005) When the data formatting unit 312 receives the data existence inquiry (step S1003), it checks whether the analysis target data is cached in the original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303. Here, the data management unit 310 refers to the original data recording unit 301 and the intermediate data recording unit 302, and specifies the range of analysis target data that requires access to the external data source 103 for each of the calibration period and the validation period.

[0125] In addition, in this example, since the analysis target data is cached, the data reforming unit 312 receives data existence notifications from each of the original data recording unit 301, the intermediate data recording unit 302, and the reformed data recording unit 303.

[0126] When the data reforming unit 312 receives a data presence notification from each of the intermediate data recording unit 302 and the reformed data recording unit 303 , it returns the data presence notification to the data management unit 310 .

[0127] (Step S1006) The original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303 each check whether or not the analysis target data exists in response to a request from the data formatting unit 312. In this example, the analysis target data is cached in the original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303. Therefore, the original data recording unit 301, the intermediate data recording unit 302, and the formatted data recording unit 303 each return a data existence notification to the data formatting unit 312.

[0128] (Step S1007) The data acquisition unit 311 executes the analysis target acquisition process 1. Specifically, for example, when the data acquisition unit 311 receives from the data management unit 310 a request to acquire raw data used for calibration from the data source 103, the data acquisition unit 311 accesses the data source 103 and acquires the raw data from the data source 103.

[0129] In this case, the data acquisition unit 311 divides the acquisition range of the original data into M pieces (M is an integer equal to or greater than 1) according to the specifications that the data source 103 can accept, and sends an acquisition request for the K pieces of divided original data to the data source 103. For example, in the case of distribution data, one physical quantity and a distribution amount at one point in time in a range of predetermined latitude and longitude that includes a river basin constitute one piece of original data. Also, in the case of time-series data, for example, observation data from one observation point and for a period of one month constitutes one piece of divided analysis target data.

[0130] (Step S1008) The data source 103 executes the request processing. Specifically, for example, for each acquisition request for original data to be divided used for calibration from the data acquisition unit 311, the data source 103 acquires the corresponding original data to be divided and returns it to the data acquisition unit 311.

[0131] (Step S1009) The original data recording unit 301 records the divided original data from the data acquisition unit 311 . <Figure 11 Initial analysis sequence (with reused data)> Fig. 11 is a sequence diagram showing the initial analysis sequence (with reused data). Note that the start conditions for the initial analysis are the existence of a watershed model including the analysis target, and the analysis target data or its dummy data for calibration. In this example, we will explain the case where the start conditions are met.

[0132] (Step S1101) Data management unit 310 executes initial analysis request processing. In the initial analysis request processing, data management unit 310 transmits an initial analysis execution request to analysis device 102 at any timing after execution of data acquisition request processing starts. When data management unit 310 receives an initial analysis completion notification from analysis device 102, it outputs the initial analysis completion notification to analysis management screen 315.

[0133] (Step S1102) Analysis device 102 executes the initial analysis when it receives a request to execute the initial analysis from data management unit 310. Analysis device 102 also reads out the formatted data from formatted data recording unit 303 and executes the initial analysis using the formatted data.

[0134] As shown in FIG. 11, even during the execution of the initial analysis (step S1101), data acquisition in the data request sequence 1 (FIG. 10) continues.

[0135] (Step S1103) When accessed by the analysis device 102, the shaping-completed data recording unit 303 writes the shaping-completed data to the analysis device 102.

[0136] (Step S1104) The data output device 101 displays "Initial analysis in progress" on the analysis management screen 315.

[0137] (Step S1105) The data output device 101 displays "Initial analysis completed" on the analysis management screen 315. <Figure 12 Calibration sequence (with reused data)> 12 is a sequence diagram showing the calibration sequence (with reused data). Note that the calibration start condition is that a watershed model and analysis target data for calibration are available. In this example, it is assumed that the start condition is met.

[0138] (Step S1201) When the data management unit 310 receives a notification of completion of the analysis target acquisition process 1 (step S1007) from the data acquisition unit 311, it executes a calibration request process. In the calibration request process, the data management unit 310 transmits a calibration request to the analysis device 102. In addition, the data management unit 310 outputs a notification indicating that calibration calculation is in progress to the analysis management screen 315.

[0139] (Step S1202) The data output device 101 displays "Calibration calculation in progress" on the analysis management screen 315. (Step S1208) When a request to read out the shaping-completed data is received from the analysis device 102, the shaping-completed data recording unit 303 writes the shaping-completed data to the analysis device 102.

[0140] (Step S1209) When data management unit 310 receives a calibration completion notification from analysis device 102 , it outputs the calibration completion notification to analysis management screen 315 .

[0141] (Step S1210) The data output device 101 displays "Calibration completed" on the analysis management screen 315.

[0142] (Step S1211) The data output device 101 receives confirmation of the end of calibration by the user 330 operating the input device 203 .

[0143] (Step S1212) Upon receiving confirmation of the completion of calibration, the data output device 101 executes a calibration result visualization process. Specifically, for example, the data output device 101 sends a calibration result visualization request to the data management unit 310 and acquires the calibration results from the data management unit 310. Then, the data output device 101 displays the calibration results on the result display screen 316.

[0144] (Step S1213) When the data management unit 310 receives the calibration result visualization request, it transmits a calibration result acquisition request to the analysis device 102. Then, the data management unit 310 acquires the calibration results from the analysis device 102. When the data management unit 310 acquires the calibration results from the analysis device 102, it transmits the calibration results to the data management unit 310.

[0145] (Step S1214) When analytical device 102 receives a calibration result acquisition request from data management unit 310, it returns the calibration result to data management unit 310.

[0146] <Figure 13 Validation sequence (with reused data)> Figure 13 is a sequence diagram showing the validation sequence (with reused data). Note that the validation start condition is that the watershed model, the analysis target data for calibration, and the analysis target data for validation are available. In this example, it is assumed that the start condition is met.

[0147] (Step S1301) When the data management unit 310 receives a notification of completion of the analysis target acquisition process 2 (step S707) from the data acquisition unit 311, it executes validation request processing. In the validation request processing, the data management unit 310 transmits a data shaping request for validation to the data shaping unit 312. Next, when the data management unit 310 receives a notification of completion of data shaping for validation from the data shaping unit 312, it transmits the validation request to the analysis device 102. In addition, the data management unit 310 outputs a notification indicating that validation calculation is in progress to the analysis management screen 315.

[0148] (Step S1302) The data output device 101 displays "Validation calculation in progress" on the analysis management screen 315.

[0149] (Step S1303) The data reforming unit 312 executes data reforming processing for validation when it receives a data reforming request for validation from the data management unit 310. Specifically, for example, the data reforming unit 312 reads original data from the original data recording unit 301, generates intermediate data, and records the intermediate data in the intermediate data recording unit 302. When the recording of the intermediate data is completed, the data reforming unit 312 transmits a completion notification of data reforming for validation to the data management unit 310.

[0150] (Step S1304) When the original data recording unit 301 receives a request to read the original data from the data reforming unit 312 , it writes the original data to the data reforming unit 312 .

[0151] (Step S1305) When the intermediate data recording unit 302 receives a request to write intermediate data from the data reforming unit 312 , it writes the intermediate data from the data reforming unit 312 to the data reforming unit 312 .

[0152] (Step S1306) When the formatted data recording unit 303 receives a request to write formatted data from the data formatting unit 312 , the formatted data recording unit 303 writes the formatted data from the data formatting unit 312 to the data formatting unit 312 .

[0153] (Step S1307) When analysis device 102 receives a validation request from data management unit 310, it executes validation after completing initial analysis. Validation (step S1307) is executed using the formatted data read from formatted data recording unit 303 as the analysis target data for validation, rather than dummy data. When validation is completed, analysis device 102 transmits a validation completion notification to data management unit 310.

[0154] (Step S1308) When a request to read out the shaping-completed data is received from the analysis device 102, the shaping-completed data recording unit 303 writes the shaping-completed data to the analysis device 102.

[0155] (Step S1309) When the data management unit 310 receives the validation completion notification from the analysis device 102 , it outputs the validation completion notification to the analysis management screen 315 .

[0156] (Step S1310) The data output device 101 displays "Validation Completed" on the analysis management screen 315.

[0157] (Step S1311) The data output device 101 receives confirmation of the end of validation by the user 330 operating the input device 203 .

[0158] (Step S1312) Upon receiving confirmation of the completion of validation, the data output device 101 executes validation result visualization processing. Specifically, for example, the data output device 101 sends a validation result visualization request to the data management unit 310 and obtains the validation results from the data management unit 310. Then, the data output device 101 displays the validation results on the result display screen 316.

[0159] (Step S1313) When the data management unit 310 receives the validation result visualization request, it transmits a validation result acquisition request to the analysis device 102. Then, the data management unit 310 acquires the validation result from the analysis device 102. When the data management unit 310 acquires the validation result from the analysis device 102, it transmits the validation result to the data management unit 310.

[0160] (Step S1314) When the analysis device 102 receives a validation result acquisition request from the data management unit 310 , it returns the validation result to the data management unit 310 .

[0161] In the examples of Figures 10 to 13, the analysis period is shifted slightly for analysis, so the initial analysis and calibration can reuse the previous shaped data (Figures 5 to 9). Therefore, the initial analysis and calibration can begin without waiting for data acquisition. Furthermore, the analysis period for validation is extended until 2022. Therefore, the data output device 101 can additionally acquire raw data for that period and update the intermediate data and shaped data. Furthermore, if additional acquisition of analysis target data for validation is completed in parallel with calibration, validation can be performed immediately after calibration is completed.

[0162] [Screen display example] Next, examples of the display of the region setting screen 314, analysis management screen 315, and result display screen 316 will be described.

[0163] <Figure 14 Area setting screen 314> 14 is an explanatory diagram showing a display example of the area setting screen 314. The area setting screen 314 has a setting area 1401 and a map display area 1402. The setting area 1401 has an area specification button 1410, an area deletion button 1411, an area confirmation button 1412, a selected river display field 1413, a candidate river name list 1414, a calibration start date input field 1415, a calibration end date input field 1416, a validation start date input field 1417, a validation end date input field 1418, and an analysis start button 1419.

[0164] The area designation button 1410 is a user interface for enabling area designation in the map display area 1402 by the input device 203 .

[0165] The area deletion button 1411 is a user interface for deleting an area (specified area 1425) specified in the map display area 1402 by the input device 203.

[0166] The area confirmation button 1412 is a user interface for confirming the specified area 1425 .

[0167] The selected river display field 1413 is an area for displaying the name of a river (selected river) selected from the candidate river name list 1414 with the cursor 1420 using the input device 203.

[0168] The candidate river name list 1414 is a list that displays the names of the rivers included in the specified area 1425 .

[0169] The calibration start date input field 1415 is an area for accepting input of the calibration start date via the input device 203 .

[0170] The calibration end date input field 1416 is an area for accepting input of the calibration end date via the input device 203 .

[0171] The validation start date input field 1417 is an area for accepting input of the validation start date via the input device 203 .

[0172] The validation end date input field 1418 is an area for accepting input of the validation end date via the input device 203 .

[0173] The analysis start button 1419 is a user interface for starting the execution of analysis (initial analysis, calibration, validation) in the area confirmed by the area confirmation button 1412.

[0174] The cursor 1420 is a user interface for selecting a river name from the candidate river name list 1414 using the input device 203 .

[0175] The map display area 1402 is an area where a map is displayed. In the example of FIG. 14, land 1421 and sea 1422 are displayed. Furthermore, rivers are displayed on the land 1421. A river 1423 indicated by a thick line is a river selected by the cursor 1420, and a river 1424 indicated by a dotted line is a river (excluding the selected river) in the candidate river name list 1414. A specified area 1425 is a rectangular area specified by the input device 203.

[0176] <Figure 15: Sequence for selecting rivers to be analyzed> FIG. 15 is a sequence diagram showing the sequence for selecting a river to be analyzed.

[0177] (Step S1501) When the user 330 accesses the area setting screen 314 using the input device 203, the data output device 101 initially displays a scrollable and zoomable map in the map display area 1402. In response to the user 330 operating the input device 203, the data output device 101 displays the area of ​​interest of the user 330 in the map display area 1402.

[0178] (Step S1502) The data output device 101 draws a rectangle to be analyzed in the map display area 1402 in response to the user 330 operating the input device 203. The drawn rectangle is a specified area 1425. Once the specified area 1425 is drawn, a candidate river name list 1414, which is a list of river names within the specified area 1425, is displayed in the setting area 1401.

[0179] The data output device 101 accepts the selection of a river to be analyzed from the candidate river name list 1414 by the user 330 operating the input device 203. The selected river is highlighted in the map display area 1402. In Figure 14, the selected river 1423 is displayed in bold.

[0180] When the data output device 101 detects that the user 330 has pressed the area confirmation button 1412 through the operation of the input device 203 , it confirms the river to be analyzed as the selected river 1423 and outputs the selected river 1423 to the data management unit 310 .

[0181] When the user 330 operates the input device 203 to input dates into the calibration start date input field 1415, the calibration end date input field 1416, the validation start date input field 1417, and the validation end date input field 1418, the data output device 101 outputs the calibration period and the validation period to the data management unit 310.

[0182] (Step S1503) When the data management unit 310 receives the specified area 1425 from the area setting screen 314, it searches the river data table in the data management information recording unit 300 for a list of rivers included in the specified area 1425, and obtains the candidate river name list 1414 as the search result. The candidate river name list 1414 is displayed in the setting area 1401.

[0183] (Step S1504) When the data management unit 310 receives the river to be analyzed, the calibration period, and the validation period, it sends a data existence query to the data reformatting unit 312, which includes one or more observation points to be analyzed on the river to be analyzed and the period to be analyzed (calibration period and validation period) (steps S503, S1003).

[0184] <Figure 16 Analysis management screen 315> 16 is an explanatory diagram showing a display example of the analysis management screen 315. The analysis management screen 315 has an analysis ID 1601, a river code 1602, a river name 1603, an analysis status 1604, a start time 1605, a scheduled end time 1606, an end time 1607, an acquired data amount 1608, and a cache utilization rate 1609.

[0185] The analysis ID 1601 is identification information that uniquely identifies an analysis (initial analysis, calibration, and validation). The river code 1602 is identification information that uniquely identifies a river. The river name 1603 is the name of the river.

[0186] Analysis status 1604 indicates the status of the analysis. Analysis status 1604 is output from data management unit 310 (steps S504, S604, S605, S802, S810, S902, S910, S1004, S1104, S1105, S1202, S1210, S1302, and S1310).

[0187] The start time 1605 is the date and time when the analysis specified by the analysis state 1604 started. The scheduled end time 1606 is the date and time when the analysis specified by the analysis state 1604 will end. The end time 1607 is the date and time when the analysis specified by the analysis state 1604 ended.

[0188] The acquired data amount 1608 is the amount of data (original data, intermediate data, and formatted data) acquired from the external data source 103 during analysis (initial analysis, calibration, and validation). The cache utilization rate 1609 is the utilization rate of the amount of reused data used from the original data recording unit 301, intermediate data recording unit 302, and formatted data recording unit 303 in the acquired data amount 1608.

[0189] <Figure 17 Result display screen> 17 is an explanatory diagram showing an example of the display of the result display screen 316. The result display screen 316 reads out the calculated river water quality data and visualizes the read out data together with the river flow path and drainage basin range on a map. The result display screen 316 has a selection area 1701, a map display area 1402, and a graph display area 1703.

[0190] The selection area 1701 has a selection button 1711, a candidate river name list 1414, and a display button 1712. The selection button 1711 is a user interface for selecting the type of analysis result (in this example, flow rate, nitrogen concentration, or phosphorus concentration). Figure 17 shows that nitrogen concentration has been selected.

[0191] The graph display area 1703 is an area for displaying a graph of time-series data. The horizontal axis represents time, and the vertical axis represents the quantity of the type of analysis result selected by the selection button 1711.

[0192] The map display area 1402 displays a watershed range 1720, which is a closed area of ​​polygonal lines that encompasses the river 1424. The white circle figures are balloons that represent the quantity of the type of analysis result (nitrogen concentration in this example) selected with the selection button 1711. The larger the circle, the higher the nitrogen concentration contained in the water of the river 1424.

[0193] In the above-described embodiment, the minimum necessary configuration for distinguishing between generating dummy data and acquiring real data is shown. By being able to specify in more detail whether to use dummy data or real data, it is expected that the efficiency of analysis can be improved when conducting analysis across multiple rivers and multiple periods within the target area.

[0194] In the above-described embodiment, only two periods, a calibration period and a validation period, are provided, but the data output device 101 may check in more detail for each year and each location of the observation point whether data is present in the intermediate data recording unit 302. Then, based on the check results, the data output device 101 may apply one of the following four-stage classifications (1) to (4) to the generation of dummy data.

[0195] (1) Generate dummy data using the dummy data generator 313: Inaccurate but requires the shortest analysis time

[0196] (2) Generate semi-dummy data that combines intermediate data and dummy data: This is slightly inaccurate but requires less analysis time. This can be useful when you change the analysis period or area.

[0197] (3) Data shaping using recorded intermediate data: Accurate and quick, provided that the data has been analyzed once before.

[0198] (4) Accessing an external data source 103 to prepare input data: Accurate but takes the longest time to analyze.

[0199] For example, the data output device 101 may determine which of the above (1) to (4) quality analysis target data to generate based on the priority of data quality and analysis time required by the analysis jobs J1 to Jn in the calibration.

[0200] As examples of these two extremes, the data output device 101 may apply (1) or (2) for initial analysis, and (3) or (4) for calibration or validation. By generating intermediate data and dummy data in the intermediate state (2), when analyzing areas adjacent to rivers for which data has already been acquired, it is expected that the initial analysis will be faster when new observation point data is required due to differences in the drainage basin range.

[0201] Furthermore, while waiting for the newly required observation point data to be acquired, the data output device 101 can reuse the data already acquired during calibration or validation, and can begin initial analysis with missing data filled in with dummy data. Since the initial analysis does not require high accuracy in reproducing real-world events, there is no problem even if such weather data input contains a mixture of dummy data and actual measurements.

[0202] In the above-described embodiment, it is assumed that time-series data, typically observation data from meteorological observation points, is input to the analysis device 102. Alternatively, the data output device 101 may input distribution data to the analysis device 102 after performing a predetermined conversion. In the case of a river basin model, the data output device 101 typically inputs three types of distribution data: topographical data, land use data, and soil data. All of this distribution data is generated based on the interpretation of surveyed maps, the discrimination of image data such as satellite images and aerial photographs, and elevation data obtained by radar measurements such as Synthetic Aperture Radar (SAR).

[0203] In the case of these data formats, in addition to using them directly, the data output device 101 may identify the attributes of the subject of each pixel based on characteristics such as spectral distribution and regularity of shape pattern, and output the classification results as, for example, farmland, forest, urban area, etc., by discrimination processing based on a learning device. In this case, the data output device 101 may obtain external data and convert its format, as well as perform discrimination processing based on a learning device to convert unprocessed distribution data into classified distribution data.

[0204] In this way, in an analysis example in which the amount of distribution data processed is relatively large (taking up a large proportion of the processing time), a method can be considered in which the data output device 101 records the discrimination result data by the learning device in the intermediate data recording unit 302 and reuses the data. In this case, in addition to managing unacquired data, acquired data, and dummy data, the data output device 101 may manage the data by dividing it into unclassified and classified data, or may start an analysis job by substituting dummy data for unacquired or unclassified data.

[0205] As explained above, the data output device 101, which is placed between the data source 103 and the analysis device 102, absorbs the difference in data format between the data source 103 and the analysis device 102 by data shaping. The data output device 101 uses the following three approaches in combination. (1) The analysis target data of the region of interest specified by the user 330 is sequentially downloaded and stored in a cache. (2) In parallel with the analysis preparation work, data formatting processing (mesh → time series conversion and file format conversion) is performed on the data to be analyzed. (3) The analysis device 102 uses rough, incomplete data (dummy data) during the initial analysis (test run), and performs the main analysis (calibration and validation) after all data formatting is complete. At this time, it is desirable that the order of the locations for generating the time-series data is not limited to chronological order, but can be random.

[0206] This allows the data output device 101 to reduce the number of accesses to the data source 103 by reusing already obtained data, and the analysis device 102 to start analysis without waiting for data in the analysis processing. Therefore, the data output device 101 can automate data acquisition and pre-processing and start analysis earlier, shortening the period from request to evaluation completion. In particular, by acquiring analysis target data for validation and performing analysis processing using analysis target data for calibration in parallel, it becomes possible to shorten the lead time until validation can start.

[0207] The present invention is not limited to the above-described embodiments, and includes various modifications and equivalent configurations within the spirit and scope of the appended claims. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to configurations including all of the described configurations. Furthermore, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Furthermore, the configuration of another embodiment may be added to the configuration of one embodiment. Furthermore, part of the configuration of each embodiment may be added to, deleted from, or replaced with other configurations.

[0208] Furthermore, each of the aforementioned configurations, functions, units, processing means, etc. may be realized in hardware, for example by designing some or all of them as integrated circuits, or may be realized in software, by a processor interpreting and executing a program that realizes each function.

[0209] Information such as programs, tables, files, etc. that realize each function can be stored in storage devices such as memory, hard disks, SSDs (Solid State Drives), or recording media such as IC (Integrated Circuit) cards, SD cards, and DVDs (Digital Versatile Discs).

[0210] In addition, the control lines and information lines shown are those that are considered necessary for explanation, and do not necessarily represent all the control lines and information lines that are necessary for implementation. In reality, it can be assumed that almost all components are interconnected. [Explanation of symbols]

[0211] 100 Analysis System 101 Data output device 102 Analysis equipment 103 Data Sources 201 processor 202 Storage Devices 300 Data management information recording unit 301 Original Data Recording Unit 302 Intermediate Data Recording Unit 303 Formatting completion data recording section 304 Dummy data recording section 310 Data Management Department 311 Data Acquisition Department 312 Data Formatting Department 313 Dummy Data Generation Unit 314 Area setting screen 315 Analysis management screen 316 Results display screen 321 Analysis Department 322 Analysis result recording section 323 Analysis settings screen

Claims

1. A data output device having a processor that executes a program and a storage device that stores the program, a data source that stores first time-series data related to a natural environment; and an analysis device that receives analysis target data from the data output device and performs analysis related to the natural environment, the data source being communicatively connected; a recording unit capable of recording the analysis target data, The processor: an input process for receiving an input of analysis conditions related to the natural environment; a first determination process for determining whether or not first analysis target data that meets the analysis conditions input by the input process exists in the recording unit; a first output process of outputting the first analysis target data to the analysis device based on a first determination result of the first determination process; A data output device characterized by executing the above.

2. 2. The data output device according to claim 1, The processor: a first acquisition process for acquiring the first analysis target data from the data source when it is determined that the first analysis target data does not exist in the recording unit; a first recording process of recording the first analysis target data acquired by the first acquisition process in the recording unit; In the first output process, the processor outputs the first analysis target data to the analysis device. A data output device characterized by:

3. 3. The data output device according to claim 2, In the first acquisition process, the processor acquires a plurality of the first analysis target data; In the first recording process, the processor records the first analysis target data in the recording unit every time the first analysis target data is acquired; The processor: a generation process for generating dummy data to be used for an initial analysis by the analysis device based on the first analysis target data recorded in the recording unit during the period from the start to the end of acquisition by the first acquisition process, In the first output process, the processor outputs the dummy data to the analysis device. A data output device characterized by:

4. 3. The data output device according to claim 2, The processor: execute a shaping process for shaping the first analysis target data; In the first output process, the processor outputs the first analysis target data that has been shaped by the shaping process to the analysis device. A data output device characterized by:

5. 5. The data output device according to claim 4, The processor: a second determination process for determining whether or not second analysis target data outside the analysis target period included in the analysis conditions exists in the recording unit; a second output process of outputting the second analysis target data to the analysis device based on a second determination result of the second determination process after the first analysis using the first analysis target data by the analysis device is completed; A data output device characterized by executing the above.

6. 6. The data output device according to claim 5, The processor: a second acquisition process for acquiring the second analysis target data from the data source when it is determined that the second analysis target data does not exist in the recording unit; a second recording process for recording the second analysis target data acquired by the second acquisition process in the recording unit; In the second output process, the processor outputs the second analysis target data to the analysis device after the first analysis is completed. A data output device characterized by:

7. 7. The data output device according to claim 6, In the second acquisition process, the processor acquires the second analysis target data from the data source during the execution of the first analysis. A data output device characterized by:

8. A data output method executed by a data output device having a processor that executes a program and a storage device that stores the program, comprising: a data source that stores first time-series data related to a natural environment; and an analysis device that receives analysis target data from the data output device and performs analysis related to the natural environment, the data source being communicatively connected; a recording unit capable of recording the analysis target data, The processor: an input process for receiving an input of analysis conditions related to the natural environment; a first determination process for determining whether or not first analysis target data that meets the analysis conditions input by the input process exists in the recording unit; a first output process of outputting the first analysis target data to the analysis device based on a first determination result of the first determination process; A data output method comprising:

Images