Adding identifiers for paddy-field operations to the ADM

[Jp-rice]

Thank you for discussing our proposals at the 2024 AgGateway Annual Meeting.
I would like to introduce our proposal for adding identifiers for paddy-field operations to the ADM using two methods.

1. Add five paddy-field operations to OperationTypeCodeEnumeration of operation type code.

1. Levee coating (FIELD_PREPARATION_LEVEE_COATING)
2. Land leveling (FIELD_PREPARATION_LAND_LEVELING)
3. Puddling (FIELD_PREPARATION_PUDDLING)
4. Trench digging (FIELD_PREPARATION_TRENCH_DIGGING)
5. Drainage (DRAINAGE)

2. Add ad-hoc ContextItems to clarify the context of the operation and to refer to the external vocabularies (AAO and AGROVOC)

Please review the attached proposal for more details.
AgGateway_proposal_from_Japan.pdf

I look forward to your feedback.

---

Akane Takezaki from NARO in Japan

[kelly-nelson]

Thanks Akane for submitting the proposal here. Arriving at an agreed list of distinct operation types has been an ongoing process with lots of discussion. We started at #111, and then revised it with #133, and from discussions in Austin, most everyone is open to continued refinement.

Other than harvest operations, we have a pattern whereby there is a _GENERAL item for items not otherwise specified. As such, expanding the types with more specificity is viable. In this case, I think we'd just adjust the definition for FIELD_PREPARATION_GENERAL to something like "Non-tillage activities to prepare a field for cultivation including chopping stover, picking rocks, and other activities not otherwise specified by explicit types herein."

I have a couple comments and questions:

-How is Puddling different from Tillage? From your definition and what I read about it, it sounds like the equivalent of tillage with the only distinction being that the soil is covered by a layer of water. One might say that pulverization, leveling, water retention and weeding are also goals of upload-field tillage.

-Can you define "cod" as you state it in the Puddling definition? I was not able to find a definition for that word that seemed to make sense in this context.

-Rather than introducing DRAINAGE as a top-level concept, the classification of "Water Management in your document caught my eye. APPLICATION_IRRIGATION makes sense to me if we don't have anything else related to water, but I'm wondering if we might now want WATER_MANAGEMENT_GENERAL, WATER_MANAGEMENT_DRAINAGE, WATER_MANAGEMENT_TRENCH_DIGGING, and WATER_MANAGEMENT_IRRIGATION (which we could facilitate by means of superseding the APPLICATION_IRRIGATION code by the attributes on the Data Type Definition objects specified in 1.0 of the Standard).
@strhea Curious on your thoughts here. How to classify Fertigation comes to mind, as does the fact that some may want to model an implicit cost of an amount of water in the same way they would manage other inputs. As with previous discussions on Operation Types, it's hard to model cleanly for all cases, it just strikes me that DRAINAGE is something of an odd man out here.

-Our Context Item concept is at once useful and problematic. The more we use ad hoc context items, the less the data in the ADAPT can be harvested in a plug and play manner. For something like mapping to AGROVOC and AAO, I'm wondering if we should instead extend our Data Type Definition and Enumeration Item objects with a defined attribute for "Vocabulary Mappings" in a new release. We could also make extensive use of such an attribute for mapping our Numeric DTDs to ISO11783-11 DDIs. @aferreyra what are your thoughts?

[dubnemo]

@Jp-rice
As I suggested in Austin, I recommend you 'call it what it really is', and not bend your concepts to the crop notion of FIELD. If you agree, please update your submission to the following:

Levee coating (PADDY_PREPARATION_LEVEE_COATING)
Land leveling (PADDY_PREPARATION_LAND_LEVELING)
Puddling (PADDY_PREPARATION_PUDDLING)
Trench digging (PADDY_PREPARATION_TRENCH_DIGGING)
Drainage (PADDY_DRAINAGE)

[dubnemo]

@knelson-farmbeltnorth a vocabulary mapping is classically called a Lexicon. It was a critical notion of the Business Collaboration Framework (BCF) as used by RosettaNet, ebXML, etc.

[Jp-rice]

@knelson-farmbeltnorth san
Thank you for your comments on our proposal.
I respond to your comments.

Q1.How is Puddling different from Tillage? From your definition and what I read about it, it sounds like the equivalent of tillage with the only distinction being that the soil is covered by a layer of water. One might say that pulverization, leveling, water retention and weeding are also goals of upload-field tillage.
＝＝＞
At the Agriculture Activity Ontology", both "puddling" and "tilling" are aimed at "regulating cultivated soil", but while ‘tilling’ is defined as ‘turning over cultivated soil’, ‘puddling’ is defined as "crushing clods".

Q2.Can you define "cod" as you state it in the Puddling definition? I was not able to find a definition for that word that seemed to make sense in this context.
==>
‘Cod’ is a mistake for ‘clod’. "Puddling" is the process of crushing clods.

Q3. Rather than introducing DRAINAGE as a top-level concept, the classification of "Water Management in your document caught my eye. APPLICATION_IRRIGATION makes sense to me if we don't have anything else related to water, but I'm wondering if we might now want WATER_MANAGEMENT_GENERAL, WATER_MANAGEMENT_DRAINAGE, WATER_MANAGEMENT_TRENCH_DIGGING, and WATER_MANAGEMENT_IRRIGATION (which we could facilitate by means of superseding the APPLICATION_IRRIGATION code by the attributes on the Data Type Definition objects specified in 1.0 of the Standard).
==>
I have no objection to the registration of
WATER_MANAGEMENT_GENERAL
WATER_MANAGEMENT_DRAINAGE
WATER_MANAGEMENT_TRENCH_DIGGING
WATER_MANAGEMENT_IRRIGATION

Q4.@strhea Curious on your thoughts here. How to classify Fertigation comes to mind, as does the fact that some may want to model an implicit cost of an amount of water in the same way they would manage other inputs. As with previous discussions on Operation Types, it's hard to model cleanly for all cases, it just strikes me that DRAINAGE is something of an odd man out here.
==>
Since rice paddies store water, it is necessary to intentionally drain the water to dry the soil. This characteristic of rice paddies differs from fields, which have low water retention. Additionally, drainage in rice paddy cultivation significantly impacts growth control and GHG reduction. For these reasons, we request that ‘Drainage’ be added as a new code.

[Jp-rice]

@dubnemo san
Thank you for your comments on our proposal.
I respond to your comments.

---

As I suggested in Austin, I recommend you 'call it what it really is', and not bend your concepts to the crop notion of FIELD. If you agree, please update your submission to the following:

Levee coating (PADDY_PREPARATION_LEVEE_COATING)
Land leveling (PADDY_PREPARATION_LAND_LEVELING)
Puddling (PADDY_PREPARATION_PUDDLING)
Trench digging (PADDY_PREPARATION_TRENCH_DIGGING)
Drainage (PADDY_DRAINAGE)

==>
Including ‘PADDY’ in the name makes it clear that it refers to operations in paddy fields. However, it is expected that systematic naming will become difficult because it is necessary to describe the differences in operations with different land uses (e.g., PASTURELAND_PREPARATION). For these reasons, I am opposed to including ‘PADDY’ in the name.

Please review it.

Akane

[kelly-nelson]

Thanks @Jp-rice . Re tillage and puddling, we have a mismatch between our two vocabularies. ADAPT would classify as Tillage both using a plow to turn over soil and using a disk harrow, etc. to break up clods. We don't currently have a breakdown between types of tillage, which arguably is hard to classify further as various combination implements have become common.

Regarding separating out puddling from tillage, I would ask if there is an important reason to track if the tillage is under water. If so, perhaps there is need for an attribute elsewhere in the model to indicate that the field is underwater during any given operation. There are many ways to do this, my main concern is that our Operation Type classifications are clean and understandable.

[aferreyra]

My reactions:

1. WATER_MANAGEMENT_IRRIGATION: I don't like this at all. Irrigation water is accounted for like any other kind of crop input; it is totally analogous to applying fertilizer or applying chemicals or seeds. I think Irrigation is a first-class citizen.
2. WATER_MANAGEMENT_DRAINAGE: I'm OK with this. It fits.
3. TILLAGE_PUDDLING: I like this.

[strhea]

@knelson-farmbeltnorth, my thoughts are as follows:

• I have no problem with the addition of (WATER_MANAGEMENT_GENERAL, WATER_MANAGEMENT_DRAINAGE, WATER_MANAGEMENT_TRENCH_DIGGING)
• I need to think a bit more about WATER_MANAGEMENT_IRRIGATION vs APPLICATION_IRRIGATION. I lean strongly toward keeping APPLICATION_IRRIGATION.
• I have no problem with the addition of (FIELD_PREPARATION_LEVEE_COATING and FIELD_PREPARATION_LAND_LEVELING)
• With respect to FIELD_PREPARATION_PUDDLING vs FIELD_PREPARATION_TILLAGE. I believe there is sufficient difference to warrant the addition of the separate operation type, primarily because the effects on soil health ARE different.

[aferreyra]

SAG-SF ISO22006 Processes-SubProcesses.xlsx
Here is the list of processes and sub-processes from ISO 22006. I think this is a useful way of resolving the matter of whether Irrigation is a form of water management or of application.

[dubnemo]

==> Including ‘PADDY’ in the name makes it clear that it refers to operations in paddy fields. However, it is expected that systematic naming will become difficult because it is necessary to describe the differences in operations with different land uses (e.g., PASTURELAND_PREPARATION). For these reasons, I am opposed to including ‘PADDY’ in the name.

Please review it.

@Jp-rice the questions I have is, how do you currently record and classify these operations, and how to you want to consume the data in your analytical processes?

We have had historical 'heartburn' when we had to translate information to a 'canonical' representation, and then when we would need to translate that back again to make sense of the data during data analysis. My 'calling it what it is' approach would then have only ONE layer of data transformation, or best case, depending how you record the data in the paddy, would eliminate all transformations altogether (aka complete pass-through).

[Jp-rice]

@knelson-farmbeltnorth san
Thank you for your comment.

This explains why puddling should be added to the OperationTypeCode, which is used for work records and orders.

Puddling prevents water leakage from the paddy field and helps the planted seedlings take root and grow. Data on puddling work, such as speed, is necessary to analyze its relationship with seedling growth and water levels (degree of water leakage), and to derive optimal conditions for puddling. It is desirable that this data be managed using an individual code for puddling work.

Tillage and puddling are recognized as separate operations in paddy field cultivation. If puddling is registered as ‘TILLAGE’ in the OperationTypeCode, rice growers will not be able to determine whether they are instructed to perform tillage or puddling.

It is necessary to uniquely identify tillage and puddling in paddy field cultivation, so please add ’puddling’ to the OperationTypeCode. This will make the OperationTypeCode classification clear and easy to understand. As Stuart-san has suggested, we would like puddling to be registered separately from tilling as ‘FIELD_PREPARATION_PUDDLING’.

We look forward to your comments.

https://www.kubota.co.jp/kubotatanbo/rice/germination/tilling_02.html
https://www.kubota.co.jp/kubotatanbo/rice/planting/ploughing_01.html
https://www.kubota.co.jp/kubotatanbo/rice/planting/ploughing_02.html