Chapter Three: Incorporating Agrodiversity Data into the PLEC Database
Introduction
After the PLEC Biodiversity database is up and running it can be expanded to include related data. Related data will primarily fall into the category of agrodiversity data, but can also include socioeconomic surveys (see China Agrodiversity Database), GPS coordinates (see chapter six), etc. An Access table must be created to hold data before they can be linked to the database. Next, at least one of the fields within the new table must be linked to the database through a relationship. For instance, if the new table contains information on management techniques in plots, it can be linked to the plot number field in the Plot Description table. Once the relationship is created, the new table becomes part of the database and can be sorted, summarized, and analyzed in queries, forms, and reports. This chapter will explain how agrodiversity data can be organized into a table and linked to the main database. Converting qualitative data into categorical or quantitative data is also discussed. The Amazonia Management Database will be used as an example. This chapter, however, should act as a guide for the reader to create unique agrodiversity tables that reflect the data available to a specific cluster.
Relationship, Scale, and Categorization
In the PLEC Amazonia Management Database, management information is collected on more than one scale. General data on the management of an entire sample area, including data on seed broadcasting and vine removal, are linked by sample area number to the Sample Area table. Specific management data on species in a particular plot are linked to the database via the plot ID field in the Species Data table. Data should be organized according to scale before creating tables. If data are about a single species and vary from plot to plot, they should be linked to the Species Data table. Data on management techniques that encompass an entire sample area, for instance plowing or irrigation, should be linked to the Sample Area table. Data regarding a specific species that does not vary from plot to plot or farmer to farmer can be linked to the Species List table. Organizing data according to scale will make for clearer and less redundant relationships (see box 3.1).
Databases can store just about any type of data. However, for quantified or discrete data numerous summary and analysis options are available, while qualitative data offer fewer statistical options. Incorporating qualitative data, especially management data, into a database requires categorization or quantification. In Amazonia, for example, Fernando Rabelo has recorded extensive descriptions of seed dispersal and seedling transplanting by farmers in their fallows. While these descriptions are an important part of the Amazonia cluster’s work, they cannot be entirely incorporated into the database because paragraphs or even sentences cannot be summarized or analyzed in a database. For this reason, qualitative or descriptive data are categorized as yes/no, multiple choice, or quantified data. In the case of transplanting and broadcasting, there are two Yes/No columns in the Fallow/Forest Management table of the Amazonia Management Database. The transplant and broadcast columns distinguish between fallows where farmers do or do not broadcast and/or transplant. With these distinct groupings, analysis can be performed to compare these two management techniques in relation to any other data in the database. It is important to note that these techniques can be broken down and divided into categories of specific types of broadcasting (throwing or selective placement) or transplanting (bare root or root ball). Some qualitative data can be quantified. For instance, weeding and watering cycles can be recorded as ‘weeding per year’ or ‘watering per month’ and entered into the database as numbers. Depending on the quantity and quality of data, management tables can encompass a wide range of specific management techniques on multiple scales. See the Amazonia Management and Yunnan Agrodiversity Databases on the CD for examples of categorizing management and socioeconomic data.
Designing a Management Table
Once the data are organized by scope and structured into discrete or quantitative categories, a management table can be designed. To give an overview of the process the Amazonia Fallow/Forest Management table will be used as an example. It may be helpful to open the Amazonia Management Database’s Fallow/Forest Management table in design view (click once on the table and then select the design icon on the right side of the window) in order to follow the example.
The necessary components of a new table are (a) the relationship field and (b) the primary key. The relationship field is the column that will link the table to the database. The Fallow/Field Management table contains management information for each sample area. Therefore, each entry into the table is a description of a specific sample area. The obvious link between this new table and the rest of the database is the Sample Area ID in the Sample Areas table. By adding a Sample Area ID column to the new management table, each entry can reference the information in the Sample Area table and the rest of the database through the Sample Area ID of the sample area being described in the new table. The relationship field can be connected to the database from the relationship window (see chapter one). The primary key is the field that uniquely identifies each record stored in the table. The most important point to remember when selecting a primary key field is that no two values can be the same. In the case of the Amazonia Management Database, the Sample Area ID was selected as a primary key. Since each sample area will be only one row in the table and will never be repeated, the Sample Area ID is the best choice for the primary key. Frequently, as in this example, the relationship and primary key fields are the same, but this is not always the case.
Once all the fields are categorized and the primary key and relationship fields are selected, a new table can be constructed (See chapter one for instructions for designing new tables). The table on the next page provides a brief overview of the process for converting management information into a database. It should be noted that any of fields described in the table could be subdivided to include more specific data. For instance, the timber field can be changed from a Yes/No field to a Number field to incorporate the quantity of timber extracted. Furthermore, if data are available, the timber field can be removed and a new table can be created with multiple fields describing the timber species, quantities, as well as extraction and management methods. Adding management tables to the database allows one to accurately represent the significant or unique management practices in the sample area or plot. When constructing these tables, important questions to be addressed are: What are the most significant management practices in this sample area or plot? What are the unique practices in the sample area or plot? How can these practices be quantified or categorized? Can any of the categories be further divided to include relevant sub-categories? Finally, the nature of management and socioeconomic data allows for infinite quantities of tables and fields. While an extremely large set of tables and fields may be tempting, it is important to recognize the limits of available data and resources.
Observation / Field Name(as in database) / Data Type
(as in table design window) / Description
(as in table design window)
Forests and fallows have three basic purposes. / Purpose / Text / Will this be used as an agricultural field, enriched forest, or is it a forest (FO)?
Some farmers thin the vegetation in their fallows or forests. / Thinning/Year / Number
Single
/ How many times will the vegetation be thinned per year?Some farmers remove vines from their fallows or forests. / Vines/Year / Number
Single
/ How many times will the farmer remove vines per year?Some farmers broadcast economically valuable seeds in their fallows or forests. / Broadcasting / Yes/No / Does the farmer broadcast seeds in the area?
Some farmers transplant economically valuable seedlings in their fallows or forests. / Transplanting / Yes/No / Does the farmer transplant seedlings in the area?
Some farmers create gaps in their fallows or forests to facilitate natural regeneration. / Gaps / Yes/No / Does the farmer create gaps in the area?
Timber is extracted from some fallows and forests. / Timber / Yes/No / Does the farmer extract timber from the area?
Fruits are harvested from some fallows or forests. / Fruits / Yes/No / Does the farmer harvest fruits from the area?
Some farmers hunt in their fallows or forests. / Hunting / Yes/No / Does the farmer hunt in the area?
Some farmers plant crops in their fallows or forests. / Crop / Yes/No / Does the farmer plant crops in the area?
To complete the table add the relationship and primary key field
The management data must be linked to the database through a relationship. / SampleAreaID / Number
Long Integer
/ The SampleAreaID of the sample area in the Sample Area table15