Controlling agricultural production areas

1. An irrigation system for an agricultural production area comprising:

an irrigation actuator;

a receiver for receiving wide-area meteorological gridded prediction data;

a sensor network comprising a sensor deployed within the agricultural production area to collect local-area sensor data at a respective sensor location;

a processor configured to

store data from the receiver as historical wide-area meteorological gridded prediction data, the prediction data comprising a prediction for multiple meteorological variables made at a first point in time for a second point in time and for an area covering the sensor location;

at the second point in time store data from the sensor network captured at the second point in time as historical local-area sensor data at a sensor location, the sensor data comprising multiple sensor data variables;

for each sensor data variable determine a relationship between

(i) the historical wide-area meteorological gridded prediction data for the multiple meteorological variables for the second point in time and

(ii) the historical local-area sensor data for that sensor data variable captured at the second point in time at the sensor location;

receive at a present point in time wide-area meteorological gridded prediction data from the receiver, the prediction data comprising a prediction for the multiple meteorological variables for a point in time that is in the future from the present point in time;

calculate a prediction on water deficit at the sensor location within the agricultural production area for the point in time that is in the future from the present point in time based on

(i) the present wide-area meteorological gridded prediction data for the point in time that is in the future from the present point in time, and

(ii) the relationship for each sensor data variable between the historical wide-area meteorological gridded prediction data and the historical local-area sensor data for that sensor data variable at the sensor location; and

Repeatedly update the relationship over time based on further wide-area meteorological gridded prediction data and further local-rea sensor data at the sensor location to thereby improve the relationship and calculate improved predictions;

wherein the irrigation actuator is controlled, based on the prediction on water deficit, to define an amount of water to be used for irrigating the agricultural production area.

2. A method for controlling an agricultural production area, the method comprising:

determining a relationship between historical wide-area meteorological gridded prediction data and historical local-area sensor data at a sensor location based on historical wide-area meteorological gridded prediction data for multiple meteorological variables made at a first point in time for a second point in time and historical local-area sensor data captured at the second point in time at the sensor location, the sensor data comprising multiple sensor data variables and the relationship being determined for each of the multiple sensor data variables;

calculating at a present point in time a prediction on a local-area agricultural parameter for a future point in time based on

present wide-area meteorological gridded prediction data for the future point in time, and

the relationship for each sensor data variable between the historical wide-area meteorological gridded prediction data and the historical local-area sensor data for that sensor data variable at the sensor location; and

controlling the agricultural production area based on the prediction on the agricultural parameter; and

repeatedly updating the relationship over time based on further wide-area meteorological gridded prediction data and further local-area sensor data at the sensor location to thereby improve the relationship.

3. The method of claim 2 , wherein prediction relates to at least 24 hours into the future.

4. The method of claim 2 , wherein the historical wide-area meteorological gridded prediction data and the historical local-area sensor data at a point relate to at least 5 days in the past.

5. The method of claim 2 , wherein calculating the prediction on the local-area agricultural parameter is based on an agricultural model, the agricultural model is based on plant growth, the agricultural model comprises a value indicative of evapotranspiration of plants, and the value indicative of evapotranspiration of plants is variable over time.

6. The method of claim 2 , wherein the historical wide-area meteorological gridded prediction data and the present wide-area meteorological gridded prediction data comprises wind data and determining the relationship and calculating the prediction is based on the wind data.

7. The method of claim 2 , wherein

the agricultural production area comprises multiple sub-areas,

there is at least one local-area sensor in each of the multiple sub-areas, and

determining the relationship and calculating the prediction is performed for each of the sub-areas.

8. The method of claim 2 , wherein calculating a prediction on a local-area agricultural parameter comprises calculating a prediction of a plant state and controlling the agricultural production area is based on the plant state.

9. The method of claim 8 , further comprising calculating a prediction on future local-area sensor data at a point, wherein controlling the agricultural production area is based on the predicted plant state and the future local-area sensor data at the point.

10. The method of claim 2 , wherein controlling the agricultural production area comprises one or more of:

plant;

irrigate;

harvest;

protect; and

feed.

11. The method of claim 2 , further comprising creating a graphical user interface to present the prediction on the local-area agricultural parameter to a user, wherein the method comprises repeating the step of calculating the prediction for multiple future times and creating the graphical user interface to present a time series of the prediction for the multiple future times.

12. The method of claim 11 , wherein the graphical user interface comprises input elements to allow the user to input planned controlling actions.

13. The method of claim 2 , further comprising determining a suggestion for controlling the agricultural production area based on the prediction on the agricultural parameter.

14. The method of claim 13 , further comprising determining a prediction on the local area sensor data based on the current wide-area meteorological gridded prediction data and the relationship between the historical wide-area meteorological gridded prediction data and the historical local-area sensor data at a point, wherein determining the suggestion is based on a predefined risk associated with local area sensor data where that risk is likely to occur and the suggestion is determined based on the prediction on the local area sensor data to reduce the risk.

15. The method of claim 13 , further comprising creating a user interface to display the suggestion.

16. The method of claim 2 , wherein the local-area agricultural parameter is a water deficit or water surplus.

17. The method of claim 2 wherein the prediction on the local-area agricultural parameter comprises a quality parameter indicative of a predicted quality of a produce from the agricultural production area and controlling the agricultural production area comprises optimising the quality parameter.

18. The method of claim 17 , further comprising repeating the step of calculating the prediction on the quality parameter for multiple future times and creating a graphical user interface to present a time series of the prediction on the quality parameter for the multiple future times.

19. The method of claim 17 , wherein the quality parameter comprises an expected shelf life.

20. A computer system for controlling an agricultural production area comprising:

a receiver for receiving wide-area meteorological gridded prediction data for multiple meteorological variables and local area sensor data comprising multiple sensor data variables;

a processor to

determine for each sensor data variable a relationship between historical wide-area meteorological gridded prediction data and historical local-area sensor data at a sensor location based on the historical wide-area meteorological gridded prediction data made at a first point in time for a second point in time and the historical local-area sensor data for that sensor data variable captured at the second point in time at the sensor location;

calculate at a present point in time a prediction on a local-area agricultural parameter for a future point in time based on

present wide-area meteorological gridded prediction data for the future point in time, and

the relationship for each sensor data variable between the historical wide-area meteorological prediction data and the historical local-area sensor data for that sensor data variable at the sensor location;

repeatedly update the relationship over time based on further wide-area meteorological gridded prediction data and further local-area sensor data at the sensor location to thereby improve the relationship; and

an output port to control the agricultural production area based on the prediction on the agricultural parameter.