Marine fishery resource quantity prediction device, marine fishery resource quantity prediction method, and marine fishery resource quantity prediction program

Abstract

To predict, with high accuracy, the quantity of marine fishery resources in a wide range of sea area, including the time different from the present point of time such as future.SOLUTION: A marine fishery resource quantity prediction device comprises: a fish school observation information acquisition unit that acquires fish school observation information in a predetermined sea area; an oceanic condition prediction information acquisition unit that acquires oceanic condition prediction information in a prediction target sea area including the predetermined sea area; and a prediction information calculation unit that calculates marine fishery resource quantity prediction information in the prediction target sea area on the basis of the fish school observation information and the oceanic condition prediction information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a technique for predicting information on the amount of marine fishery resources, such as fish stocks. [Background technology]

[0002] Traditionally, fishermen have used locations and ocean conditions that have produced good catches in the past as candidates for good fishing grounds for the current fishery. However, past good fishing grounds do not necessarily become good fishing grounds this time. Specifically, good fishing grounds are considered to be areas where fish migrate around and where the ocean conditions are favorable for fish, but the previous method only took the latter into consideration.

[0003] For this reason, it is possible to present information that has actually been observed (visitor information). However, in this case, there is a problem that only the observation point and the current time are available.

[0004] In order to solve this problem, various systems for predicting marine fishery resources have been devised, as shown in Patent Document 1 and Non-Patent Document 1. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] International Publication No. 2022 / 230333 [Non-patent literature]

[0006] [Non-Patent Document 1] Development of a Pacific saury fishing ground prediction method using AI technology, Takashi Yabuki, JAFIC Technical Review No.1 February 2022 Summary of the Invention [Problem to be solved by the invention]

[0007] However, the above-mentioned conventional methods were not able to predict marine fishery resource abundance with high accuracy for a wide range of ocean areas. Furthermore, while the conventional methods were able to predict marine fishery resource abundance at the present time, they were not able to predict marine fishery resource abundance at times different from the present, such as in the future.

[0008] Therefore, an object of the present invention is to predict with high accuracy the amount of marine fishery resources in a wide range of ocean areas, including times different from the present, such as in the future. [Means for solving the problem]

[0009] A marine fisheries resource quantity prediction device according to one embodiment of the present invention comprises a fish school observation information acquisition unit that acquires fish school observation information for a specified sea area, a sea condition prediction information acquisition unit that acquires sea condition prediction information for a prediction target sea area including the specified sea area, and a prediction information calculation unit that calculates marine fisheries resource quantity prediction information for the prediction target sea area based on the fish school observation information and the sea condition prediction information.

[0010] In this configuration, the amount of marine fishery resources can be calculated with high accuracy based on the fish school observation information obtained through observation and the predicted ocean condition forecast information.

[0011] In the marine fishery resource amount prediction device according to one embodiment of the present invention, the fish school observation information is based on fish school distribution information in a predetermined sea area.

[0012] In this configuration, by specifically using fish school distribution information as fish school observation information, predictions regarding fish school distribution can be made.

[0013] In a marine fisheries resource amount prediction device according to one embodiment of the present invention, fish school observation information is based on fish school distribution information in a specified sea area and fish school observation position information set in at least one location in the specified sea area.

[0014] In this configuration, by specifically using fish school distribution information as fish school observation information, predictions regarding fish school distribution can be made, and by using fish school observation location information, the accuracy of predictions of fish school distribution can be improved.

[0015] In the marine fishery resource amount prediction device according to one embodiment of the present invention, ocean condition prediction information is based on at least one of the water temperature, salinity concentration, and current speed of the prediction target sea area.

[0016] With this configuration, the sea condition prediction information can be set appropriately according to the sea conditions.

[0017] In the marine fishery resource amount prediction device according to one embodiment of the present invention, ocean condition prediction information is based on the water temperature, salinity concentration, and current speed of the prediction target sea area.

[0018] With this configuration, the sea condition prediction information can be set with high accuracy according to the sea conditions.

[0019] In a marine fishery resource amount prediction device according to one embodiment of the present invention, the prediction information calculation unit calculates at least one of the marine fishery resource amount in the prediction target sea area and the change in the marine fishery resource amount over time as marine fishery resource amount prediction information.

[0020] In this configuration, the marine fishery resource amount information, specifically, the marine fishery resource amount and its change over time can be calculated with high accuracy. [Brief explanation of the drawings]

[0021] [Figure 1] FIG. 1 is a functional block diagram showing an example of the configuration of a marine fishery resource amount prediction device according to a first embodiment of the present invention. [Figure 2] FIG. 2 is a diagram showing an example of setting the prediction target region and the individual region. [Figure 3] FIG. 3 is a diagram showing an example of the concept of obtaining the observed fish school quantity. [Figure 4] FIG. 4 is a diagram showing an example of the concept of preference index. [Figure 5] FIG. 5 is a diagram showing an example of a display of predicted fish school distribution. [Figure 6] FIG. 6 is a functional block diagram showing an example of the configuration of a marine fishery resource amount prediction device according to the second embodiment of the present invention. [Figure 7] FIG. 7 is a functional block diagram showing an example of the configuration of a marine fishery resource amount prediction device according to the third embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0022] [First embodiment] A marine fishery resource amount prediction technology according to a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a functional block diagram showing an example of the configuration of a marine fishery resource amount prediction device according to the first embodiment of the present invention. The marine fishery resource amount shown in each embodiment including this embodiment is, for example, the amount of fish. The amount of fish is, for example, the weight of a school of fish.

[0023] 1, the marine fishery resource abundance prediction device 10 includes a fish school observation information acquisition unit 20, an ocean condition prediction information acquisition unit 30, and a prediction information calculation unit 40. The fish school observation information acquisition unit 20 and the ocean condition prediction information acquisition unit 30 are interface function units that acquire various information obtained from the outside. The prediction information calculation unit 40 is composed of, for example, a processing unit such as a computer, and a storage medium that stores a marine fishery resource abundance prediction program executed by the processing unit.

[0024] FIG. 2 is a diagram showing an example of setting the prediction target region and the individual region.

[0025] 2, the prediction target region Rep is set by a plurality of individual regions Re, which correspond to the predetermined region of the present invention.

[0026] More specifically, the prediction target region Rep is set as a two-dimensional region extending in the latitude and longitude directions, and the plurality of individual regions Re are set as a two-dimensional array aligned in the latitude and longitude directions.

[0027] For example, in the case of FIG. 2, the multiple individual regions Re are set in a 5×5 two-dimensional array with five in the latitude direction and five in the longitude direction, with the individual region Re(n, m) at the center.

[0028] Each of the individual regions Re is linked to a position in the absolute coordinate system. For example, the position coordinates of the center of each of the individual regions Re are set by the position coordinates in the absolute coordinate system.

[0029] Note that the example shown in FIG. 2 is just one example, and other examples can also be used as long as a plurality of individual regions Re are set in a predetermined arrangement pattern for the prediction target region Rep.

[0030] The fish school observation information acquisition unit 20 acquires the observed fish school quantity in an individual area Re included in the prediction target area Rep. The observed fish school quantity can be obtained, for example, from observation data from a buoy placed in a specific individual area Re or from a fish finder or sonar placed on a fishing boat in the specific individual area Re.

[0031] Figure 3 is a diagram showing an example of the concept of obtaining the observed fish school quantity. In Figure 3, buoys or ships exist in the individual areas Re(n-2,m+2), Re(n-1,m-1), Re(n+1,m+1), and Re(n+2,m-2) in the prediction target area Rep. The fish school observation information acquisition unit 20 obtains the observed fish school quantity Wo(n-2,m+2) from the individual area Re(n-2,m+2), obtains the observed fish school quantity Wo(n-1,m-1) from the individual area Re(n-1,m-1), obtains the observed fish school quantity Wo(n+1,m+1) from the individual area Re(n+1,m+1), and obtains the observed fish school quantity Wo(n+2,m-2) from the individual area Re(n+2,m-2). At this time, the fish school observation information acquisition unit 20 acquires the observed fish school amount Wo in each individual region Re by linking it to the observation time t.

[0032] The fish school observation information acquisition unit 20 outputs the acquired observed fish school amount Wo to the prediction information calculation unit 40.

[0033] The sea state prediction information acquisition unit 30 acquires externally predicted sea state prediction information for the prediction target area Rep. The sea state prediction information includes time.

[0034] The ocean condition prediction information includes, for example, water temperature distribution, salinity distribution, and current speed. It is preferable that the ocean condition prediction information includes all of water temperature distribution, salinity distribution, and current speed, but it is sufficient if at least one of these can be acquired. Information that cannot be acquired can be supplemented by input from an input unit (not shown), or it can also be supplemented with information from the past (for example, from the immediately preceding or similar period).

[0035] The sea condition prediction information acquisition unit 30 outputs the sea condition prediction information to the prediction information calculation unit 40.

[0036] The prediction information calculation unit 40 calculates fish school distribution prediction information including the fish school amounts Wp of multiple individual areas Re that make up the prediction target area Rep based on the observed fish school amount Wo and sea condition prediction information. More specifically, the prediction information calculation unit 40 calculates the fish school distribution prediction information, for example, as follows.

[0037] As shown in FIG. 1, the prediction information calculation unit 40 includes a fish school distribution correction unit 41, a preference index model setting unit 42, and a fish school distribution prediction unit 43.

[0038] The fish school distribution correction unit 41 receives the predicted fish school abundance Wp of multiple individual regions Re and the observed fish school abundance Wo from the fish school observation information acquisition unit 20. The predicted fish school abundance Wp is, for example, the fish school abundance obtained in the prediction immediately before the current prediction.

[0039] The fish school distribution correction unit 41 corrects the predicted fish school abundance Wp of multiple individual areas Re with the observed fish school abundance Wo. For example, the fish school distribution correction unit 41 corrects the predicted fish school abundance Wp so that the difference between the predicted fish school abundance Wp and the observed fish school abundance Wo in the same individual area Re becomes small (for example, 0).

[0040] The fish school distribution correction unit 41 outputs the fish school amount Wp of the multiple individual areas Re of the prediction target area Rep (the corrected fish school amount Wp for the individual area Re where the fish school amount Wp has been corrected) to the fish school distribution prediction unit 43.

[0041] Ocean condition prediction information is input to the preference index model setting unit 42. The preference index model setting unit 42 sets a preference index P for the fish school to be predicted based on the water temperature distribution and salinity concentration distribution in the ocean condition prediction information.

[0042] The preference index P is an index that indicates the direction in which the fish school to be predicted is likely to move. For example, the preference index P is set to a value between 0 and 1.

[0043] Fig. 4 is a diagram showing an example of the concept of preference index. Fig. 4 shows a preference index P indicating movement from an individual area Re(n,m) to multiple individual areas Re(n-1,m), Re(n,m-1), Re(n,m+1), and Re(n+1,m) adjacent to the individual area Re(n,m).

[0044] Different fish species prefer different water temperatures and salinities. Therefore, if the water temperature and salinity distributions in the prediction target area Rep are known, the direction in which schools of fish of the prediction target fish species are likely to move can be determined. Using this, the preference index model setting unit 42 sets a preference index Pnm(-1,0) from the individual area Re(n,m) to the individual area Re(n-1,m), sets a preference index Pnm(0,-1) from the individual area Re(n,m) to the individual area Re(n,m-1), sets a preference index Pnm(0,+1) from the individual area Re(n,m) to the individual area Re(n,m+1), and sets a preference index Pnm(+1,0) from the individual area Re(n,m) to the individual area Re(n+1,m). At this time, the preference index model setting unit 42 sets the preference index P for the current prediction time.

[0045] The preference index model setting unit 42 sets preference indexes P for all individual regions Re of the prediction target region Rep. Note that the preference index model setting unit 42 can set preference indexes P using at least one of the water temperature distribution and the salinity concentration distribution, but by using both the water temperature distribution and the salinity concentration distribution, the preference index P can be set with higher accuracy.

[0046] The preference index model setting unit 42 sets the maximum swimming speed V for each adjacent individual area based on the current speed in the ocean condition prediction information. The maximum swimming speed V is a value obtained by dividing the maximum movement distance of the target fish school by the time interval.

[0047] For example, the preference index model setting unit 42 sets a maximum swimming speed V for each of the adjacent individual regions.

[0048] The preference index model setting unit 42 sets a preference index model at the target time of prediction by combining the preference index P and the maximum swimming speed V. The preference index model setting unit 42 outputs the set preference index model (preference index P and maximum swimming speed V) to the fish school distribution prediction unit 43.

[0049] The fish school distribution prediction unit 43 predicts the fish school amount Wp of each individual area Re of the prediction target area Rep based on the fish school amount Wp from the fish school distribution correction unit 41 (the corrected fish school amount Wp for the individual area Re in which the fish school amount Wp has been corrected) and the preference index model (preference index P and maximum swimming speed V) from the preference index model setting unit 42.

[0050] More specifically, the fish school distribution prediction unit 43 sets a fish school quantity prediction model based on the preference index model from the preference index model setting unit 42. As the fish school quantity prediction model, for example, a mesh-based fish behavior model can be used.

[0051] The fish school distribution prediction unit 43 predicts the fish school abundance Wp of each individual area Re of the prediction target area Rep using a fish school abundance prediction model that receives the fish school abundance Wp as input from the fish school distribution correction unit 41. The fish school distribution prediction unit 43 outputs the predicted fish school distribution of the prediction target area Rep, using the fish school abundance Wp of multiple individual areas Re as one data set.

[0052] With this configuration and processing, the marine fishery resource abundance prediction device 10 can predict with high accuracy the fish school abundance Wp not only for the present time but also for future times, etc. Furthermore, the marine fishery resource abundance prediction device 10 can predict with high accuracy the marine fishery resource abundance over a wide sea area, not just a small area such as the location of a fishing vessel.

[0053] The marine fishery resource amount prediction device 10 may be provided with a display, for example, to display the predicted fish school distribution in the prediction target area Rep.

[0054] Figure 5 is a diagram showing an example of a predicted fish school distribution display. As shown in Figure 5, the prediction target area Rep is displayed on the display screen of the display. Furthermore, map information showing land, sea, etc. is displayed on the display screen based on the position coordinates of the prediction target area Rep.

[0055] The marine fishery resource quantity prediction device 10 displays a graphic showing the predicted location, predicted fish school quantity (size of the fish school), movement direction, and movement speed of the fish school at multiple times (e.g., t1, t2, t3, t4) based on the predicted fish school distribution in the prediction target area Rep, as shown in Figure 5. For example, in the case of Figure 5, the predicted location is represented by the position of a circle, and the predicted fish school quantity is represented by the size of the circle. The movement direction is represented by the direction of the arrow, and the movement speed is represented by the length of the arrow.

[0056] For example, as shown in Figure 5, it is possible that the center position of the school of fish is shifted from the center of multiple individual regions Re, but the coordinates of the center position of the school of fish at each time can be predicted from the movement direction and movement speed based on the maximum swimming speed V, and the center position of the school of fish can be displayed based on this prediction result.

[0057] It is also possible to calculate the prediction reliability [%] and confidence interval (fish school abundance ±α [tons], etc.) for the predicted fish school abundance and display these, although they are not shown in Figure 5.

[0058] [Second embodiment] A marine fishery resource amount prediction technique according to a second embodiment of the present invention will be described with reference to the drawings. Fig. 6 is a functional block diagram showing an example of the configuration of a marine fishery resource amount prediction device according to the second embodiment of the present invention.

[0059] 6, the marine fishery resource amount prediction device 10A according to the second embodiment has a different correction means than the marine fishery resource amount prediction device 10 according to the first embodiment. Below, only the differences between the marine fishery resource amount prediction device 10A and the marine fishery resource amount prediction device 10 will be described, and a description of similar parts will be omitted.

[0060] The marine fishery resource amount prediction device 10A does not include a fish school distribution correction unit 41 in a prediction information calculation unit 40A, but includes a model correction unit 44.

[0061] The model correction unit 44 sets model correction information for the preference index model based on the predicted fish school amount Wp and the observed fish school amount Wo. The model correction information is information that sets a correction value to correct the preference index model (preference index P and maximum swimming speed V) so that the difference between the predicted fish school amount Wp and the observed fish school amount Wo becomes small (for example, to 0).

[0062] The model correction unit 44 can also select an optimal fish behavior model and include the selected fish behavior model in the model correction information.

[0063] The model correction unit 44 outputs the model correction information to the preference index model setting unit 42A.

[0064] The preference index model setting unit 42A corrects the preference index model (preference index P and swimming speed V) based on the model correction information. The preference index model setting unit 42A outputs the corrected preference index model (preference index P and swimming speed V) to the fish school distribution prediction unit 43A.

[0065] The fish school distribution prediction unit 43A predicts the fish school amount Wp of each individual area Re of the prediction target area Rep based on the observed fish school amount Wo from the fish school observation information acquisition unit 20 and the corrected preference index model from the preference index model setting unit 42A.

[0066] With this configuration and processing, the marine fishery resource abundance prediction device 10A can predict with high accuracy the fish school abundance Wp not only for the present time but also for future times, etc. Furthermore, the marine fishery resource abundance prediction device 10A can predict with high accuracy the marine fishery resource abundance over a wide sea area, not just a small area such as the location of a fishing vessel.

[0067] [Third embodiment] A marine fishery resource amount prediction technique according to a third embodiment of the present invention will be described with reference to the drawings. Fig. 7 is a functional block diagram showing an example of the configuration of a marine fishery resource amount prediction device according to the third embodiment of the present invention.

[0068] As shown in Figure 7, the marine fisheries resource quantity prediction device 10B of the third embodiment has a configuration that combines the configuration of the marine fisheries resource quantity prediction device 10 of the first embodiment and the configuration of the marine fisheries resource quantity prediction device 10A of the second embodiment.

[0069] Specifically, the prediction information calculation unit 40B of the marine fishery resource abundance prediction device 10B includes a fish school distribution correction unit 41, a preference index model setting unit 42B, a fish school distribution prediction unit 43B, and a model correction unit 44. The fish school distribution correction unit 41 is similar to the fish school distribution correction unit 41 of the marine fishery resource abundance prediction device 10 according to the first embodiment. The preference index model setting unit 42B is similar to the preference index model setting unit 42 of the marine fishery resource abundance prediction device 10A according to the second embodiment, and the model correction unit 44 is similar to the model correction unit 44 of the marine fishery resource abundance prediction device 10A according to the second embodiment.

[0070] The fish school distribution prediction unit 43B predicts the fish school amount Wp of each individual area Re of the prediction target area Rep based on the fish school amount Wp corrected by the fish school distribution correction unit 41 and the corrected preference index model from the preference index model setting unit 42B.

[0071] With this configuration and processing, the marine fishery resource abundance prediction device 10B can predict with high accuracy the fish school abundance Wp not only for the present time but also for future times, etc. Furthermore, the marine fishery resource abundance prediction device 10B can predict with high accuracy the marine fishery resource abundance over a wide sea area, not just a small area such as the location of a fishing vessel.

[0072] In the above-described embodiment, the fish school distribution correction, model correction, and fish school distribution prediction are performed by different functional units, but it is also possible to perform these corrections and predictions using a single calculation model.

[0073] <1> a fish school observation information acquisition unit that acquires fish school observation information for a predetermined sea area; a sea condition prediction information acquisition unit that acquires sea condition prediction information for a prediction target sea area including the predetermined sea area; a prediction information calculation unit that calculates marine fishery resource amount prediction information for the prediction target sea area based on the fish school observation information and the ocean condition prediction information; A marine fishery resource amount prediction device comprising:

[0074] <2> <1> A marine fishery resource amount prediction device comprising: The fish school observation information is based on fish school distribution information in the specified sea area. Marine fishery resource forecasting device.

[0075] <3> <1> A marine fishery resource amount prediction device comprising: The fish school observation information is based on fish school distribution information in the predetermined sea area and fish school observation position information set in at least one location in the predetermined sea area. Marine fishery resource forecasting device.

[0076] <4> <1> ~ <3> Any one of the marine fishery resource amount prediction devices described above, The ocean condition prediction information is based on at least one of the water temperature, salinity, and current speed of the prediction target sea area. Marine fishery resource forecasting device.

[0077] <5> <4> A marine fishery resource amount prediction device comprising: The ocean condition prediction information is based on the water temperature, salinity, and current speed of the prediction target sea area. Marine fishery resource forecasting device.

[0078] <6> <1> ~ <5> Any one of the marine fishery resource amount prediction devices described above, The prediction information calculation unit As the marine fishery resource amount prediction information, at least one of the marine fishery resource amount in the prediction target sea area and the time change in the marine fishery resource amount is calculated. Marine fishery resource forecasting device. [Explanation of symbols]

[0079] 10, 10A, 10B: Marine fishery resource forecasting device 20: Fish observation information acquisition unit 30: Ocean condition forecast information acquisition unit 40: Prediction information calculation unit 41:Fish distribution correction section 42, 42A, 42B: Preference index model setting section 43, 43A, 43B: Fish distribution prediction section 44: Model correction section Rep: Prediction target area Re:Individual area Wo: Observed fish population Wp: Fish mass

Claims

1. a fish school observation information acquisition unit that acquires fish school observation information for a predetermined sea area; a sea condition prediction information acquisition unit that acquires sea condition prediction information for a prediction target sea area including the predetermined sea area; a prediction information calculation unit that calculates marine fishery resource amount prediction information for the prediction target sea area based on the fish school observation information and the ocean condition prediction information; A marine fishery resource amount prediction device comprising:

2. The marine fishery resource amount prediction device according to claim 1, The fish school observation information is based on fish school distribution information in the specified sea area. Marine fishery resource forecasting device.

3. The marine fishery resource amount prediction device according to claim 1, The fish school observation information is based on fish school distribution information in the predetermined sea area and fish school observation position information set in at least one location in the predetermined sea area. Marine fishery resource forecasting device.

4. The marine fishery resource amount prediction device according to claim 1, The ocean condition prediction information is based on at least one of water temperature, salinity, and current speed of the prediction target sea area. Marine fishery resource forecasting device.

5. 5. The marine fishery resource amount prediction device according to claim 4, The ocean condition prediction information is based on the water temperature, salinity, and current speed of the prediction target sea area. Marine fishery resource forecasting device.

6. The marine fishery resource amount prediction device according to claim 1, The prediction information calculation unit As the marine fishery resource amount prediction information, at least one of the marine fishery resource amount in the prediction target sea area and the time change in the marine fishery resource amount is calculated. Marine fishery resource forecasting device.

7. Obtaining fish observation information for a specified sea area, Obtaining sea condition forecast information for a forecast target sea area including the specified sea area; calculating marine fishery resource amount prediction information for the prediction target sea area based on the fish school observation information and the ocean condition prediction information; Methods for predicting marine fishery resource abundance.

8. Obtaining fish observation information for a specified sea area, Obtaining sea condition forecast information for a forecast target sea area including the specified sea area; calculating marine fishery resource amount prediction information for the prediction target sea area based on the fish school observation information and the ocean condition prediction information; A marine fishery resource volume prediction program that causes a processing unit to execute processing.

Images