METADATA by asafwewe

VIEWS: 78 PAGES: 13

More Info
									                              METADATA
           SOIL PROFILE ANALYTICAL DATABASE OF EUROPE
                         v 2.1.0.0, 29/03/1999


DATA SET IDENTIFICATION
Data set title:
   Soil Profile Analytical Database of Europe version 2.1.0.0, 29/03/1999.
Alternative title:
   Soil Profile Database of Europe.
Alternative title:
   Base de Données Analitique des Profile d'Europe version 2.1.0.0, 29/03/1999.
Alternative title:
   Base de Données des Profiles d'Europe.
Abbreviated title:
   SPADE.




DATA SET OVERVIEW
Abstract, Purpose of production, and Usage:
   The Soil Profile Analytical Database of Europe is an integral part of the European Soil Database. It has been
   compiled through the collaboration of national experts of the 12 EU member countries and is currently being
   extended to include data from the eastern European and Scandinavian countries.

   The driving force for compilation of the database was the need to model the water balance and map the
   available water content in the root zone for the MARS Project (Monitoring Agriculture by Remote Sensing,
   Joint Research Centre, Ispra, Italy). A number of pedotransfer functions or rules for calculating or predicting
   other soil attributes for use in land use and management was also perceived as important for future
   interpretation of soil maps.

   The idea to develop the Soil Profile Analytical Data Base in different stages (levels) was generally regarded
   as the most realistic approach. The number of soil types to be computerized would vary according to the time
   available and the funding provided to establish the database system. It might therefore be necessary to start
   with data for a few soil types and add more later. That would mean making a first approximation (Level 1) to
   a comprehensive soil profile database system and then later following up with a second (Level 2), third (Level
   3) and even a fourth approximation (Level 4).

   Based on discussions in the Soil and GIS Support Group, the MARS Project funded the establishment of a
   Level 1 Soil Profile Analytical Data Base that could later be extended.

   The primary aim was to capture profile data which are representative of the soil types depicted on the EC
   Soil Map. To make the best use of the soil map for modelling purposes, a complete coverage of data are
   needed.

   Two formats (Proformas) were designed to capture the data:
    Proforma I: for profile data that are recognised as truly representative of specific soil types but not geo-
     referenced to any particular locality. Preferably, these data should be obtained by measurement according
     to standard methods of analysis. However, to ensure a complete European data set for modelling
     purposes, experts were requested to provide estimates where no measured data exist for the particular soil
     type or where a non-standard method of analysis has been used.




SPADBE_Metadata                                                                                                 1
    Proforma II: for measured data from specific points in the landscape where the soil has been examined
     and analysed. No attempt is made to harmonise the analytical methods but the methods used are specified
     for each parameter. These data may not be truly representative of soil types shown on the map and some
     data may be missing for some parameters.

   The database includes analytical results for the different soil horizons as follows:
    Texture (& particle size grades)         Electric conductivity
    Organic matter content (C, N)            CEC and exchangeable bases
    Structure                                Soil water retention
    Total nitrogen content                   Bulk density
    pH                                       Root depth
    ESP or SAR                               Groundwater level
    Calcium carbonate content                Parent material
    Calcium sulphate content
Basic concepts:
1. At level 1, the member states are treated as separate regions; in later stages the member states will be divided
   into sub-regions (i.e. level 4).
2. At level 1, typical soil profile descriptions and associated analytical data are identified for each major soil
   type present within each member state. The data are compiled for the dominant soil types on agricultural land
   only (and NOT for all the STUs identified in the SGDBE).
3. At level 2, typical soil profile descriptions and associated analytical data are identified for dominant as well
   as associated soil types on agricultural land (level 2 = level 1 + associated soil types).
4. At level 3, typical soil profile descriptions and associated analytical data are identified for dominant as well
   as associated soil types on agricultural and other land uses (level 3 = level 2 + non-agricultural land uses).
5. Level 4 allows for sub-national subdivision. At level 4 there may be more than one profile for each soil type
   to differentiate some soil types on other criteria such as parent material (pedo-landscapes). Level 4 = level 1
   or 2 or 3 + regional sub-division.
6. Proforma I: soil profiles truly representative of soil types, preferably measured but can be estimated (as a
   „theoretical‟ profile), exhaustive, harmonised, not geo-referenced, intended for spatial modelling at the
   1:1,000,000 scale.
7. Proforma II: soil profiles not necessarily representative of soil types, measured data (from a „real‟ profile),
   not necessarily exhaustive, not harmonised but measurement method specified, geo-referenced, intended as
   the first stage of compilation of a data set of measured data for all Europe.
Spatial sub-schema type:
Spatial reference system type:
   No spatial positioning.
Language:
   English.
Document reference:
   MADSEN, H. Breuning. (1991). The principles for construction of an EC Soil database system. In: Soil
     Survey – a basis for European soil protection. Soil and Groundwater Report 1. J.M. Hodgson (ed.). EUR
     13340 EN, 173-180. Office for Official Publications of the European Communities, Luxembourg.
   MADSEN, H. Breuning and JONES, R.J.A. (1995). Soil profile analytical database for the European Union.
     Danish Journal of Geography, 95, 49-57.
   MADSEN, H. Breuning and JONES, R.J.A. (1996). A Soil Profile Analytical Database for the European
     Union. In: Soil databases to support sustainable development. C. Le Bas and M. Jamagne (eds). European
     Soil Bureau Research Report No.2, EUR 16371 EN, 135-144. Service d‟Etude des Sols et de la Carte
     Pedologique de France de l‟INRA, Orleans (F) and the Joint Research Centre, Institute for Remote
     Sensing Applications, Ispra (I).
   MADSEN, H. Breuning and JONES, R.J.A. (1998). Towards a European Soil Profile Analytical Database.
     In: Land Information Systems: Developments for planning the sustainable use of land resources. H.J.
     Heineke, W. Eckelmann, A.J. Thomasson, R.J.A. Jones, L. Montanarella and B. Buckley (eds). European
     Soil Bureau Research Report No.4, EUR 17729 EN, (1998), 43-50. Office for Official Publications of the
     European Communities, Luxembourg.


SPADBE_Metadata                                                                                                   2
Sample:
   No sample data set provided.
Related data sets:
   Soil Geographical Data Base of Europe at Scale 1:1,000,000.




DATA SET QUALITY INDICATORS
Rationale
   The quality and reliability of the information held in the European Soil Database (ver 1.0) are important
   considerations for users. At present it is not possible to express the accuracy or precision of the data in the
   various data sets according to any quantified standard procedure. The ultimate aim of the Scientific
   Committee of the ESB is to correct this deficiency.

   In practice, the responsibility for the accuracy of the spatial and attribute data rests with the Contributor
   organisations. It has not been possible to make thorough internal checks on the consistency and reliability of
   the data. This will be done in the future as an ongoing process but the ESB Scientific Committee takes the
   view that it is now time to distribute the data and to request user feedback as a part of the validation process.
   In the long-term, this is likely to reveal more errors than a series of „artificial‟ checks.
   The current version of the soil profile analytical database comprises estimated profiles and measured profiles.
   It is the estimated profiles that have been selected to be representative of the STUs. The measured profiles
   are a first stage only in the compilation of a soil profile database of Europe in which all the profiles are geo-
   referenced. This data set is not large enough for the profiles to be linked to SMUs by geo-referencing in any
   meaningful way.

   The medium-term aim is to obtain at least one representative profile for each soil typological unit (STU) –
   currently there are some STUs for which there are no representative profiles. Furthermore, the current linking
   of the estimated soil profile data to STUs needs thoroughly reviewing as many of the current links are not
   explicit or are simply incorrect. There is also a need for internal consistency checks on the various parameter
   values. For example, very acid soils (with pH<5.0) cannot have a very high base saturation (BS > 60%).

   In summary, it is important for users of the soil profile data to understand the following:
   1. The collection of soil profile analytical data began in 1993, before the legend for the version of the Soil
       Geographical Database (current at that time) had been finalised. Therefore the national experts had to use
       their judgement and experience in selecting the profiles that ought to be representative of their nation‟s
       soils. There was no complete list of soil mapping units available at European level to guide the national
       experts at the time.
   2. Since the compilation of the Soil Profile Analytical Database for the EU–12 countries began, the Soil
       Geographical Database has been continuously extended such that it now covers 28 countries. In this
       respect, the process of compiling profile data has lagged behind (and still lags behind) the extension of
       the geographical database.
   3. The soil data in the national archives of the member states are aggregated on the basis of soil units or
       classes belonging to national soil classifications. In most countries, these classifications differ
       significantly from the FAO system (used for the EU Soil Map). This has lead to gaps in the database for
       some soil mapping units that are identified at European level but are not recognised or delineated at
       national level.
   4. Financial resources were made available through the Commission for the extraction and reformatting of
       data from national archives. These funds were limited such that the „target population‟ of soil data was
       reduced to only collated or published information that could be easily extracted. There was no
       opportunity to take new samples or make additional analyses.
   5. The accuracy of the data and the degree to which they are representative of the STUs (and the soil
       mapping units - SMU) in any country, remain the sole responsibility of the national contributor. There
       was no opportunity within the project for further checks on the data supplied.
   6. A brief survey of the data supplied for the EU-12 countries was made and this identified missing data
       items as well as missing profiles.
   7. There are more comprehensive and fully validated profile data in many of the member states. The profile
       data included here must be accepted as the best soil profile data that are available at European level but



SPADBE_Metadata                                                                                                   3
      used in the knowledge that, to date, no internal consistency, compatibility or range checks have been
      made.
   8. Later versions of the European Soil Database (later than 2.1.0.0) will include more data that have been
      checked in the manner described above.
   9. Users are referred to the „Document Reference‟ section of the Metadata file for the soil profile analytical
      database for more information on the compilation process.
Process history:
   1986: The concept of a soil profile database for the EC countries was originally proposed by Madsen,
         following discussion of the subject at a series of annual meetings of the Computerization of Land Data
         Group (EC DGVI) which began in 1982 (at Montpellier [F]). A methodology for compiling a soil
         profile database was agreed by a sub-committee in Brussels.
   1988: The approach was approved at the last full meeting of the Computerization of Land Data Group, in
         Wageningen (Van Lanen and Bregt, 1989). The database would be compiled in phases, firstly at Level
         1, secondly at Level 2 and so on.
   1989: The concept and procedures were endorsed by the Heads of Soil Surveys of the European Community
         countries, at a meeting in Silsoe (Hodgson, 1991). The lack of compatibility between the soil profile
         data collected across Europe would be overcome by defining standard Proformas for recording the
         data.
   1992: Under contract to the MARS Project (JRC), proformas were constructed as spreadsheets in Microsoft
         Excel for the collection of standard data (Proforma I for estimated data where necessary) and measured
         data (Proforma II); guidelines for completing the proformas were also prepared (Madsen and Jones
         1995). The proformas and guidelines were then tested by members of the Soil and GIS Support Group
         of the MARS Project.
   1993: Following receipt of some completed proformas, final versions were produced and distributed to all
         the proposed contributors (EU-12) to the database.
   1995: Completed proformas (I and II) for approximately 350 soil profiles for EU-10 countries were received,
         some on paper, others as digital files.
   1996: The process of extending the profile database to Eastern Europe and Scandinavian countries has
         commenced. Staff at INRA (Orleans) converted the Proforma I data from spreadsheet to relational
         database format. This revealed a number of structural problems with the data that have still to be fully
         resolved.
   1997: The compilation of a Level 1 database for the EU-12 countries was completed (with the inclusion of
         data for Ireland [from the WISE database] and Portugal). Data for only a few eastern European
         countries (e.g. Bulgaria) have been received (to date).
   1998: Version 1.0 of the Soil Profile Analytical Database released to the ESB Scientific Committee.
Overall positional accuracy:
   Not applicable.
Overall thematic accuracy:
   It should be borne in mind that the Soil Profile Analytical Database was compiled BEFORE the list of soil
   map units (SMU) for the SGDBE was finalised. Thus the contributors to the level 1 database could only
   select profile data according to their best judgement that these data would be representative of all important
   soil types on the map of their respective countries. It was fully appreciated at the time that compilation of the
   profile data was not sufficiently coordinated with the updating of the soil geographic database and that this
   would eventually lead to problems of linking the soil profile data to the SGDBE. However, the availability of
   funds from the JRC and the existence of an active expert group (Soil and GIS Support Group) to commence
   the work of compiling the soil profile data were deemed to be of overriding importance.

   The use of expert judgement for estimation of soil profile data (Proforma I) is a pragmatic approach at this
   stage of the development of the European Soil Database. The excavation of soil profiles and the laboratory
   analysis of the samples are both time consuming and expensive processes. Even if sufficient funds were made
   available, it would take a long time to fill in the current gaps in European soil profile data. However, this
   should remain the ultimate aim in the medium to long term.
Overall temporal accuracy:
   The data derive from the analysis of soil samples taken over a long period, mostly 1950-1990, and the expert
   judgement of scientists with at least 25 years field experience of soils.




SPADBE_Metadata                                                                                                   4
Overall logical accuracy:
Overall completeness:
   Estimated 80% for EU-12 country Proforma I at Level 1. For eastern European and Scandinavian countries,
   completeness for Proforma I at mid 1998 is much less (< 35%). For Proforma II data (measured), values are
   missing for many prarameters.




SPATIAL REFERENCE SYSTEM

INDIRECT SPATIAL REFERENCE SYSTEM:
Type of indirect spatial reference system:
   By country.
Reference date:
DIRECT SPATIAL REFERENCE SYSTEM:
Datum:
Ellipsoid:
Map projection:
Height reference system:
   Mean sea level.



EXTENT
CURRENCY OF EXTENT DATA AND COMPLETENESS OF DATA SET:
Extent date:
Extent status:
PLANAR EXTENT:
Bounding XY:
Bounding Area:
Geographic area:
VERTICAL EXTENT:
Min elevation value:
Max elevation value:
TEMPORAL EXTENT:
From date:
To date:



DATA DEFINITION
APPLICATION SCHEMA DESCRIPTION:
Application schema id:
   SOIL.


SPADBE_Metadata                                                                                           5
Application schema text:
   Soil profiles in Europe.


OBJECT TYPE:
Object type name:
Object type definition:
Geometric primitive type:
Structure primitive type:
Object type code:
Occurrences:
Positional accuracy:
Thematic accuracy:
Completeness:


ATTRIBUTE TYPE:
Attribute type name:           See DICTIONARY
Attribute type definition:
Attribute type code:
Attribute type domain:
Thematic accuracy:
Temporal accuracy:


ASSOCIATION TYPE:
Association type name:
Association type definition:
From object type:
To object type:
Cardinality:
Constraints:
Thematic accuracy:
Logical consistency:



CLASSIFICATION
THESAURUS:
Name of thesaurus:


Thesaurus administrator:


SPADBE_Metadata                                 6
THESAURUS ELEMENT:
Term:
Definition:
Synonyms:                                          See Attricod
Related term:
Broader term:
Narrower term:
Picture:



ADMINISTRATIVE META DATA
   See metadata for the Soil Geographical Data Base of Europe.
ORGANISATION AND ORGANISATION ROLE:
Organisation name:
Abbreviated organisation name:
Organisation address:
Organisation role:
Alternative organisation name:
Function of organisation:
POINT OF CONTACT AND POINT OF CONTACT ROLE:
Point of contact name:
Point of contact address:
Point of contact role:

DISTRIBUTION:
   The Soil Profile Analytical Data Base of Europe is distributed only together with the Soil Geographical Data
   Base of Europe. See metadata for the last.
Restrictions on use:
Copyright owners:
Pricing policy:
Unit of distribution:
Media:
Formats:
On-line access:
Order:
Support services:




SPADBE_Metadata                                                                                              7
META DATA REFERENCE
Entry date:
  25/05/1998
Last check date:
  25/05/1998
Last update date:
  20/04/1998
Future review date:
Spatial reference system of metadata:




SPADBE_Metadata                         8
                                                    ANNEX

CONTENT OF ANNEX
Definition of Proforma I
Definition of Proforma II
Definition of the levels
Profile-to-STU linking process
         Typology of profile-to-STU links
         Examples for one country
Data model for the profile database

DEFINITION OF PROFORMA I
 Used to describe estimated profiles.
 Profiles are not georeferenced.
 The key for linkage of the profile to a Soil Typological Unit (STU) is made of the following three attributes:
        Soil name (SOIL),
        Dominant surface textural class (TEXT1),
        Secondary surface textural class (TEXT2).
 The author of the profile may or may not give a list of Soil Mapping Units (SMUs) or STUs to which the
  profile refers.
         If the author does, then the link between that profile and its corresponding STU within that (or
            those) SMU(s) is called “explicit”. This fact will be stored in the data base together with the link.
         For all other links to corresponding STU(s) that can be found using the key for linkage, the link is
            called “implicit” (i.e. not explicitly given by the author of the profile). This fact will also be stored
            in the database together with the link.
 All data in the profile are mandatory with some exceptions (active CaCO3, hydraulic conductivity).

DEFINITION OF PROFORMA II
   Used to describe selected existing measured profiles.
   Profiles may or may not be georeferenced (by X and Y co-ordinates).
   There is no possible linkage of measured profiles to any STU or SMU.
   Most of the data in the profile are not mandatory but some are.

DEFINITION OF LEVELS
 Levels differentiate the number of profiles to be provided by country.
 A "soil type" is identified by the triplet Soil name/Dominant surface textural class/Secondary surface textural
  class (SOIL/TEXT1/TEXT2).
LEVEL 1
   By country.
   For agricultural land only for the soil types where agriculture is at least one of the land uses.
   1 profile for each soil type which appears in 1 or more dominant STU(s).
   This profile does NOT apply to the same soil types that appear in associated (non-dominant) STUs.
LEVEL 2
   By country.
   For agricultural land only.
   1 profile for each soil type, whether dominant or not.
   Level 2 = Level 1 + associated soil types.
LEVEL 3
   By country.
   By land use (agricultural + other(s) (mainly forest, but also scrubs, heather...)).
   1 profile for each soil type, whether dominant or not.
   Levels 1 and 2 provide profiles only for agricultural lands. Level 3 adds 1 profile to each soil type for other
    land use(s) (mainly forest).



SPADBE_Metadata                                                                                                         9
 Level 3 = Level 2 + land use subdivision.
LEVEL 4
 Level 4 allows for sub-national subdivision.
 At Level 4 there may be more than 1 profile for each soil type to differentiate some soil types on other
  criteria such as parent material (pedolandscapes).
 Level 4 = Level 1 or 2 or 3 + regional subdivision.

PROFILE-TO-STU LINKING PROCESS
 The Profile Database currently contains level 1 profiles sometimes with gaps, but it also contains some level
  2, 3 and 4 profiles for some countries.
 It was decided to build a Profile-to-STU link table only for Level 1 profiles only plus Level 4 profiles where
  applicable.
 Therefore the linking process will be run on the list of soil types (SOIL/TEXT1) that appear as dominant
  within each SMU in each country (this corresponds to the level 1 concept and will accept dominant level 4
  profiles).
 Originally, the key for linkage should have been made of the following attributes:

         COUNTRY                               soil type
         SOIL
         TEXT1
         TEXT2
                       }    soil type   }      in
                                               a
                                               country

But because this key is too restrictive, it was decided to work with only the first 3 attributes
(COUNTRY/SOIL/TEXT1).

 When the author of a profile has explicitly given a list of one or more SMUs to which the profile applies,
  then the profile is linked to the dominant soil type (SOIL/TEXT1) in those SMUs for that country under the
  condition that the key for linkage (COUNTRY/SOIL/TEXT1) matches. Such a link is said to be “explicit”.
  This information is stored in the link table together with the link in a LINK_TYPE attribute with code 1
  meaning “explicit link”. Explicit links have a high priority over other link types and are highly reliable.
 When the author of a profile has not indicated any SMU to which the profile applies, then the profile is
  linked to all dominant soil types (SOIL/TEXT1) for that country that hold a matching link key value
  (COUNTRY/SOIL/TEXT1). Such a link is said to be “implicit”. This information is stored in the link table
  together with the link in a LINK_TYPE attribute with code 2 meaning “implicit link”. Implicit links have a
  lower priority than explicit ones and are less reliable.
TYPOLOGY OF PROFILE-TO-STU LINKS
Listed by order of priority and reliability:

1 The link is explicit by author of the profile (i.e. the author has provided 1 or more SMU references for the
  profile).
2 The link is implicit by author of the profile, i.e. profile is a Level 1 profile (i.e. the author has NOT provided
  ANY SMU reference for the profile).
EXAMPLES FOR ONE COUNTRY
The following is a sample of the semantic data set of the Soil Geographical Database of Europe (SGDBE). The
STU table gives, for each STU, its description. We have kept only the attributes that are of interest to our
purpose: STU identifier (STU), soil type (SOIL/TEXT1), and an example attribute (ATT) to exemplify
particular cases. The STU.ORG table holds the composition relationships between SMUs and STUs (e.g. SMU
100 is « made of » 80% of STU 1000 and 20% of STU 1001, meaning that STU 1000 is dominant within SMU
100). These two tables are related through the STU attribute.

From tables STU.ORG and STU above and their related attribute STU, the new table below called
DOMINANT_SOIL_TYPES can be built. It holds each SMU‟s dominant soil type (SOIL/TEXT1).

 STU.ORG                                                        100        1001       20
SMU    STU            PCAREA                                    101        1002       70
100    1000           80                                        101        1003       30



SPADBE_Metadata                                                                                                   10
102        1004      75                                         STU
102        1005      20                                     STU        SOIL      TEXT1        ATT
102        1006      5                                      1000       Jeg       1            A
103        1007      85                                     1001       Lo        2            B
103        1008      15                                     1002       Jeg       1            C
104        1009      80                                     1003       Po        3            A
104        1010      20                                     1004       Be        4            D
105        1011      60                                     1005       Bc        2            D
105        1012      40                                     1006       La        3            A
106        1013      100                                    1007       Be        4            A
107        1014      95                                     1008       Po        2            A
107        1015      5                                      1009       Be        4            E
108        1016      40                                     1010       Ch        5            A
108        1017      30                                     1011       Ec        3            B
108        1018      30                                     1012       Be        4            A
...                                                         1013       Dd        2            E
                                                            1014       Dd        2            B
                                                            1015       Bev       2            C
                                                            1016       Gcf       3            E
                                                            1017       Bx        2            A
                                                            1018       Bx        2            C
                                                            ...


Notice that for SMU 108, the dominant soil type is Bx/2 coming from two STUs which originally are not
dominant because they are different on other attributes such as ATT.

          DOMINANT_SOIL_TYPES
      SMU    PCAREA    SOIL               TEXT1
      100    80        Jeg                1
      101    70        Jeg                1
      102    75        Be                 4
      103    85        Be                 4
      104    80        Be                 4
      105    60        Ec                 3
      106    100       Dd                 2
      107    95        Dd                 2
      108    60        Bx                 2
      …

Each possible case of profile-to-STU linking process will be illustrated using the following sample PROFILES
table. In that table we kept only the attributes that are of interest to our purpose. PROF_NUM holds the profile
identifier. SMU holds the list of SMU identifiers to which the author of the profile wishes to relate that profile.
In this case explicit links have to be built. If the author does not specify any SMU, then this attribute is left blank
and implicit links have to be built. SOIL and TEXT1 hold the soil type which serves as the key for linkage to the
dominant soil type within the specified SMUs (explicit links), or within any SMU if unspecified (implicit links).

          PROFILES
      PROF_NUM     SMU                     SOIL        TEXT1
      1                                    Jeg         1
      2            102,103,105             Be          4
      3                                    Dd          2
      4            107                     Dd          2
      5                                    Bx          2
      ...

From all the four above files, the following PROFILES_TO_STU_LINKS table can be built. In that table
PROF_NUM holds the profile identifier, STU holds the STU identifier to which that profile is linked, and
LINK_TYPE holds the type of that link (1 meaning explicit link, and 2 meaning implicit link). Each example is
explained below.

         PROFILES_TO_STU_LINKS


SPADBE_Metadata                                                                                                     11
     PROF_NUM            STU            LINK_TYPE
     1                   1000           2
     1                   1002           2
     2                   1004           1
     2                   1007           1
     3                   1013           2
     4                   1014           1
     5                   1017           2
     5                   1018           2
     ...


Example 1 illustrated by profile 1: simple implicit link.

In profile 1, the author did not specify any SMU to which that profile should apply. Therefore it implicitly
applies to any SMU whose dominant soil type (SOIL/TEXT1) matches the profile‟s soil type. The link key value
for profile 1 is Jeg/1. Both SMUs 100 and 101 hold this soil type value. Therefore profile 1 is linked to the
corresponding STUs which are STUs 1000 and 1002. That information is stored in the
PROFILES_TO_STU_LINKS table together with the link type, which in this case is 2, meaning that the link is
implicit.

Example 2 illustrated by profile 2: simple explicit link.

In profile 2, the author specified three SMUs to which that profile should apply, namely SMUs 102, 103 and
105. Therefore it explicitly applies to those and only those SMUs under the condition that their dominant soil
type (SOIL/TEXT1) matches the profile‟s soil type. The link key value for profile 2 is Be/4. Both SMUs 102
and 103 hold this soil type value. Therefore profile 2 is linked to the corresponding STUs which are STUs 1004
and 1007. That information is stored in the PROFILES_TO_STU_LINKS table together with the link type,
which in this case is 1, meaning that the link is explicit.

Although SMU 104 also holds the Be/4 matching soil type value, profile 2 is not linked to that SMU because
this was not specified by the author.

The author has also specified SMU 105 in his list of applicable SMUs. But the dominant soil type for SMU 105
is Ec/3, which does not match the profile‟s soil type value Be/4. This is an error from the author of the profile.
No link can be built, but the error will be flagged in a “rejected links” listing. Note that SMU 105 has an
associated, non dominant soil type with the matching value Be/4. But this is not sufficient to build the link,
because at Level 1, only dominant soil types are valid for linkage.

Example 3 illustrated by profiles 3 and 4: explicit link overrides implicit link (level 1 + level 4 links).

Profile 3 illustrates a case of a simple implicit link to STUs 1013 and 1014. Profile 4 illustrates a case of a
simple explicit link to STU 1014. But that STU was already implicitly linked to profile 3. Because explicit links
are more reliable than implicit links, they are considered of a higher priority. Therefore the implicit link of STU
1014 with profile 3 is replaced by an explicit link with profile 4.

This case illustrates situations of “regionalisation” of the profile database. The author has given a profile to
apply generally to a soil type, except in some places (i.e. SMUs) where he has provided another profile, more
specific although for the same soil type. The concept of regionalisation corresponds to Level 4 of the database.
Level 4 may apply to any of the previous levels, i.e. Levels 1 to 3.

Regionalisation can also be managed by the author of the profiles with two or more profiles applying to the same
soil type but where each profile applies to specified different SMUs. In such cases, there may be conflicts such
as two different profiles applying to the same SMU. At Level 1 (but not at Levels 2 and 3), this is an error from
the author of the profile. No link can be built between those two profiles and that SMU, but the error will be
flagged in a “rejected links” listing.

Example 4 illustrated by profile 5: complex implicit link.



SPADBE_Metadata                                                                                                 12
Profile 5 illustrates an implicit link but more complex than those illustrated above. Profile 5 implicitly applies to
any SMU whose dominant soil type (SOIL/TEXT1) matches the profile‟s soil type. The link key value for
profile 5 is Bx/2. SMU 108 holds this soil type value. Therefore profile 5 is linked to all corresponding STUs
within that SMU, which in this case are STUs 1017 and 1018.

Notice that STUs 1017 and 1018 both belong to SMU 108 and are both non-dominant STUs within that SMU.
Their soil types, because they are the same, have become dominant by addition of their respective proportion
within the SMU. Therefore, although the linking process is carried out only at Level 1 of the profile database, it
may happen that links occur from profiles to non-dominant STUs.

The same case may happen with explicit links (complex explicit links) and is not illustrated here because it can
be handled in the same manner.


DATA MODEL FOR THE PROFILE DATABASE
The following data model is adopted to build an Arc/Info data base for integration of the Estimated Profile
Database to the 1/1M Geographical Database:

                    Estimated Profile Attribute Table
             PROF_NUM    SOIL       TEXT1        PM               ...other attributes...




                         Estimated Horizon Attribute Table
         PROF_NUM         HOR_NUM DEPTH_HOR_ DEPTH_HOR_                           HOR_NAME          ...other
                                           START           END                                    attributes...




                    Estimated Profile to STU
                    Link Table
             PROF_NUM     STU          LYNK_TYPE




                         STU Table
             STU        SOIL     TEXT1            MAT1          ...other attributes...

The model does not apply to Measured Profile Database. The measured profiles should not be linked to SMUs
or STUs because they are representative of the soils at particular georeferences and not necessarily
representative of STUs and consequently SMUs. They should be used only for reference and therefore are
supplied as Excel files without any relational structure.




SPADBE_Metadata                                                                                                   13

								
To top