DETERMINING INTRA-FIELD YIELD VARIATION DATA BASED ON SOIL CHARACTERISTICS DATA AND SATELLITE IMAGES

1 . A method comprising:

using instructions programmed in a computer system comprising one or more processors and computer memory:

receiving permanent properties data for a plurality of agricultural sub-fields of an agricultural field;

determining whether at least one data item is missing for any sub-field of the plurality of agricultural sub-fields of the agricultural field in the permanent properties data;

in response to determining that at least one data item is missing for any sub-field of the plurality of agricultural sub-fields of the agricultural field in the permanent properties data, generating, based on, at least in part, the permanent properties data, additional properties data for the plurality of agricultural sub-fields that includes the at least one data item;

wherein a data item, of the at least one data item, is generated by interpolating and aggregating two or more data records in the permanent properties data;

generating preprocessed permanent properties data by merging the permanent properties data with the additional properties data;

based on, at least in part, the preprocessed permanent properties data, generating filtered permanent properties data by removing, from the preprocessed permanent properties data, a set of preprocessed permanent properties records corresponding to a subset of the plurality of agricultural sub-fields in which two or more crops were grown in the same year;

applying a regression operator to the filtered permanent properties data to determine a plurality of intra-field variations values that represent intra-field variations in predicted yield of crop harvested from the plurality of agricultural sub-fields;

storing the intra-field variations values in the computer memory.

2 . The method of claim 1 , further comprising: applying a least absolute shrinkage and selection operator (LASSO) to the filtered permanent properties data to determine the plurality of intra-field variations values that represent the intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields.

3 . The method of claim 1 , further comprising: applying a random forest (RF) operator to the filtered permanent properties data to determine the plurality of intra-field variations values that represent the intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields.

4 . The method of claim 1 , further comprising: based on, at least in part, the plurality of intra-field variations values, determining a plurality of yield patterns of the predicted yield of crop harvested from the plurality of agricultural sub-fields, and storing the plurality of yield patterns in the computer memory.

5 . The method of claim 1 , further comprising: using the plurality of intra-field variations values that represent intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields to automatically control a computer control system to manage one or more of: seeding, irrigation, nitrogen application, or harvesting.

6 . The method of claim 1 , wherein the permanent properties data for the plurality of agricultural sub-fields comprises one or more of: soil property data, soil survey maps, topographical properties data, bare soil maps, or satellite images; wherein the soil property data comprises soil measurement data; wherein the topographical properties data comprises elevation data and elevation associated properties data.

7 . The method of claim 1 , further comprising: identifying a particular type of a subset of the permanent properties data; based on, at least in part, on the particular type of the permanent properties data, determining a second plurality of intra-field variations values that represent second intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields for the particular type of properties data.

8 . The method of claim 1 , further comprising: determining whether the at least one data item is missing for a particular sub-field of the plurality of agricultural sub-fields of the agricultural field due to of one or more of: historical data for the particular sub-field is unavailable, the particular sub-field is irrigated, or no crop was harvested from the particular sub-field.

9 . A data processing system comprising:

a computer memory;

one or more processors coupled to the computer memory and programmed to:

receiving permanent properties data for a plurality of agricultural sub-fields of an agricultural field;

determining whether at least one data item is missing for any sub-field of the plurality of agricultural sub-fields of the agricultural field in the permanent properties data;

in response to determining that at least one data item is missing for any sub-field of the plurality of agricultural sub-fields of the agricultural field in the permanent properties data, generating, based on, at least in part, the permanent properties data, additional properties data for the plurality of agricultural sub-fields that includes the at least one data item;

wherein a data item, of the at least one data item, is generated by interpolating and aggregating two or more data records in the permanent properties data;

generating preprocessed permanent properties data by merging the permanent properties data with the additional properties data;

based on, at least in part, the preprocessed permanent properties data, generating filtered permanent properties data by removing, from the preprocessed permanent properties data, a set of preprocessed permanent properties records corresponding to a subset of the plurality of agricultural sub-fields in which two or more crops were grown in the same year;

applying a regression operator to the filtered permanent properties data to determine a plurality of intra-field variations values that represent intra-field variations in predicted yield of crop harvested from the plurality of agricultural sub-fields;

storing the intra-field variations values in the computer memory.

10 . The data processing system of claim 9 , wherein the one or more processors are further programmed to perform: applying a least absolute shrinkage and selection operator (LASSO) to the filtered permanent properties data to determine the plurality of intra-field variations values that represent the intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields.

11 . The data processing system of claim 9 , wherein the one or more processors are further programmed to perform: applying a random forest (RF) operator to the filtered permanent properties data to determine the plurality of intra-field variations values that represent the intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields.

12 . The data processing system of claim 9 , wherein the one or more processors are further programmed to perform: based on, at least in part, the plurality of intra-field variations values, determining a plurality of yield patterns of the predicted yield of crop harvested from the plurality of agricultural sub-fields, and storing the plurality of yield patterns in the computer memory.

13 . The data processing system of claim 9 , wherein the one or more processors are further programmed to perform: using the plurality of intra-field variations values that represent intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields to automatically control a computer control system to manage one or more of: seeding, irrigation, nitrogen application, or harvesting.

14 . The data processing system of claim 9 , wherein the permanent properties data for the plurality of agricultural sub-fields comprises one or more of: soil property data, soil survey maps, topographical properties data, bare soil maps, or satellite images; wherein the soil property data comprises soil measurement data; wherein the topographical properties data comprises elevation data and elevation associated properties data.

15 . The data processing system of claim 9 , wherein the one or more processors are further programmed to perform: identifying a particular type of a subset of the permanent properties data; based on, at least in part, on the particular type of the permanent properties data, determining a second plurality of intra-field variations values that represent second intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields for the particular type of properties data.

16 . The data processing system of claim 9 , wherein the one or more processors are further programmed to perform: determining whether the at least one data item is missing for a particular sub-field of the plurality of agricultural sub-fields of the agricultural field due to of one or more of: historical data for the particular sub-field is unavailable, the particular sub-field is irrigated, or no crop was harvested from the particular sub-field.

17 . One or more non-transitory computer-readable storage media storing one or more computer instructions which, when executed by one or more processors, cause the processors to perform:

receiving permanent properties data for a plurality of agricultural sub-fields of an agricultural field;

determining whether at least one data item is missing for any sub-field of the plurality of agricultural sub-fields of the agricultural field in the permanent properties data;

in response to determining that at least one data item is missing for any sub-field of the plurality of agricultural sub-fields of the agricultural field in the permanent properties data, generating, based on, at least in part, the permanent properties data, additional properties data for the plurality of agricultural sub-fields that includes the at least one data item;

wherein a data item, of the at least one data item, is generated by interpolating and aggregating two or more data records in the permanent properties data;

generating preprocessed permanent properties data by merging the permanent properties data with the additional properties data;

based on, at least in part, the preprocessed permanent properties data, generating filtered permanent properties data by removing, from the preprocessed permanent properties data, a set of preprocessed permanent properties records corresponding to a subset of the plurality of agricultural sub-fields in which two or more crops were grown in the same year;

applying a regression operator to the filtered permanent properties data to determine a plurality of intra-field variations values that represent intra-field variations in predicted yield of crop harvested from the plurality of agricultural sub-fields;

storing the intra-field variations values in a computer memory.

18 . The one or more non-transitory computer-readable storage media of claim 17 , storing additional instructions for: applying a least absolute shrinkage and selection operator (LASSO) to the filtered permanent properties data to determine the plurality of intra-field variations values that represent the intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields.

19 . The one or more non-transitory computer-readable storage media of claim 17 , storing additional instructions for: applying a random forest (RF) operator to the filtered permanent properties data to determine the plurality of intra-field variations values that represent the intra-field variations in the predicted yield of crop harvested from the plurality of agricultural sub-fields.

20 . The one or more non-transitory computer-readable storage media of claim 17 , storing additional instructions for: based on, at least in part, the plurality of intra-field variations values, determining a plurality of yield patterns of the predicted yield of crop harvested from the plurality of agricultural sub-fields, and storing the plurality of yield patterns in the computer memory.