Addressing incomplete soil sample data in soil enrichment protocol projects
In various embodiments, methods, systems, and computer program products are provided for estimating greenhouse gas emissions. In particular, methods, systems, and computer program products are presented for the task of quantifying emissions reduction from changes in stocks of soil organic carbon (SOC). The methods apply more generally to tasks that involve quantification of a total or average across a population of land (and/or possibly any space and any time) using measurements at a sample from that population. The sample may not be representative or unbiased. In some examples, estimating greenhouse gas emissions may apply to quantifying the total stock of SOC in land, the total nitrous oxide emissions in land and a time interval.
1. A method for estimating greenhouse gas emissions, the method comprising:
receiving a plurality of soil sample locations within a land area, soil attributes for at least some of the plurality of soil sample locations, and regional soil data of the land area;
estimating soil organic carbon (SOC) emissions reduction of the land area;
determining a plurality of pre-strata based on the plurality of soil sample locations and the regional soil data, wherein each pre-stratum of the plurality of pre-strata comprises a plurality of soil sample points across the land area;
for each pre-stratum of the plurality of pre-strata:
determining a potential for bias in an estimate of emissions reduction for the pre-stratum;
comparing the potential for bias with a predetermined threshold; and
when the potential for bias is above the predetermined threshold, determining a plurality of post-strata based on the pre-stratum, wherein the plurality of post-strata are disjoint subsets of the pre-stratum; and
determining a total estimated emissions reduction of the land area based on the plurality of post-strata and the estimated SOC emissions reduction of the land area.
2. The method of claim 1 , wherein determining the potential for bias comprises:
determining a first average value of an attribute among observed sample locations;
determining a second average value of the attribute among all locations where samples were planned to be taken;
determining a pooled standard deviation based on variation associated with the attribute; and
labelling the pre-strata as likely including bias when the difference between the first and second average values, divided by the pooled standard deviation, is above a predetermined threshold.
3. The method of claim 1 , wherein estimating SOC emissions reduction comprises generating a plurality of simulations of SOC emissions reduction based on soil attributes of the land area.
4. The method of claim 3 , wherein the plurality of simulations comprises Monte Carlo simulations.
5. The method of claim 1 , wherein the soil attributes include % SOC, soil texture, pH, and bulk density.
6. The method of claim 1 , wherein the regional soil data comprises land management practices.
7. The method of claim 1 , wherein the regional soil data comprises soil characteristics.
8. The method of claim 1 , further comprising combining at least two of the plurality of pre-strata.
9. The method of claim 1 , wherein determining the plurality of pre-strata comprises applying a zone attribute table and a point attribute table.
10. The method of claim 1 , wherein determining the post-stratum further comprises:
partitioning the land area into a plurality of fields, each field comprising one or more soil sample locations of the plurality of soil sample locations; and
binning the plurality of fields into a plurality of bins based on at least one variable.
11. The method of claim 10 , wherein determining the post-stratum further comprises applying one or more rules to the binned plurality of fields.
12. The method of claim 10 , wherein the at least one variable comprises land management practices.
13. The method of claim 10 , wherein the at least one variable comprises soil texture.
14. The method of claim 10 , wherein the at least one variable comprises crop type.
15. The method of claim 10 , wherein binning comprises applying clustering to the plurality of fields.
16. The method of claim 15 , wherein clustering comprises k-means clustering.
17. The method of claim 10 , wherein at least one of the plurality of fields includes at least one observed soil measurement at the one or more soil sample locations.
18. The method of claim 17 , wherein at least one of the plurality of fields includes at least one missing soil measurements at the one or more soil sample locations.
19. The method of claim 18 , wherein determining the total estimated emissions reduction comprises:
for each post-stratum:
determining a mean emissions reduction based on the at least one observed soil measurement in the plurality of fields within the post-stratum;
determining a total emissions reduction based on the mean emission reduction and a total area of the post-stratum; and
determining a variance of the total emissions reduction.
20. The method of claim 19 , wherein determining the total estimated emissions reduction further comprises:
summing each total emissions reduction; and
summing each variance.
21. The method of claim 1 , further comprising determining a potential for bias for each post-stratum in the plurality of post-strata.
22. A system comprising:
a soil attribute database;
a soil sample location database;
a regional soil data database;
a computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising:
receiving a plurality of soil sample locations within a land area from the soil sample location database, soil attributes for at least some of the plurality of soil sample locations from the soil attribute database, and regional soil data of the land area from the regional soil data database;
estimating soil organic carbon (SOC) emissions reduction of the land area;
determining a plurality of pre-strata based on the plurality of soil sample locations and the regional soil data, wherein each pre-stratum of the plurality of pre-strata comprises a plurality of soil sample points across the land area;
for each pre-stratum of the plurality of pre-strata:
determining a potential for bias in an estimate of emissions reduction for the pre-stratum;
comparing the potential for bias with a predetermined threshold; and
when the potential for bias is above the predetermined threshold, determining a plurality of post-strata based on the pre-stratum, wherein the plurality of post-strata are disjoint subsets of the pre-stratum; and
determining a total estimated emissions reduction of the land area based on the plurality of post-strata and the estimated SOC emissions reduction of the land area.
23. A computer program product for estimating greenhouse gas emissions comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising:
receiving a plurality of soil sample locations within a land area, soil attributes for at least some of the plurality of soil sample locations, and regional soil data of the land area;
estimating soil organic carbon (SOC) emissions reduction of the land area;
determining a plurality of pre-strata based on the plurality of soil sample locations and the regional soil data, wherein each pre-stratum of the plurality of pre-strata comprises a plurality of soil sample points across the land area;
for each pre-stratum of the plurality of pre-strata:
determining a potential for bias in an estimate of emissions reduction for the pre-stratum;
comparing the potential for bias with a predetermined threshold; and
when the potential for bias is above the predetermined threshold, determining a plurality of post-strata based on the pre-stratum, wherein the plurality of post-strata are disjoint subsets of the pre-stratum; and
determining a total estimated emissions reduction of the land area based on the plurality of post-strata and the estimated SOC emissions reduction of the land area.