An Esri White Paper December 2011

Loading Data into a Parcel Fabric

Esri, 380 New York St., Redlands, CA 92373-8100 USA TEL 909-793-2853 FAX 909-793-5953 E-MAIL info@esri.com WEB esri.com

Copyright 2011 Esri All rights reserved. Printed in the United States of America. The information contained in this document is the exclusive property of Esri. This work is protected under United States copyright law and other international copyright treaties and conventions. No part of this work may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage or retrieval system, except as expressly permitted in writing by Esri. All requests should be sent to Attention: Contracts and Legal Services Manager, Esri, 380 New York Street, Redlands, CA 92373-8100 USA. The information contained in this document is subject to change without notice. Esri, the Esri globe logo, ArcGIS, ArcEditor, ArcInfo, ArcMap, ArcCatalog, ArcStorm, SDE, esri.com, and @esri.com are trademarks, registered trademarks, or service marks of Esri in the United States, the European Community, or certain other jurisdictions. Other companies and products mentioned herein may be trademarks or registered trademarks of their respective trademark owners.

J10025

Loading Data into a Parcel Fabric

An Esri White Paper
Contents Introduction........................................................................................... Parcel Fabric Data Model ..................................................................... Tax Parcel Editing Template ................................................................ Migrating to the Parcel Fabric Data Model .......................................... Step 1: Get to Know the Fabric Data Model and Tax Parcel Editing Template................................................... Step 2: Prepare Your New Geodatabase......................................... Step 3: Prepare Your Source Data .................................................. Step 4: Load Your Boundaries and Parcels into a Staging Geodatabase...................................................................... Coverages.................................................................................. Shapefiles.................................................................................. CAD Files ................................................................................. Step 5: Delineate Curves................................................................. Working with Linestrings ......................................................... Working with Data Already Split into Two-Point Lines .......... Step 6: Build and Validate a Topology........................................... Step 7: Model Other Land Types.................................................... Step 8: Maintain Your Attribute Values ......................................... COGO Attributes ...................................................................... Step 9: Load Your Fabric................................................................ Step 10: Annotation Considerations ............................................... Step 11: Load Control Points (optional) ......................................... Appendixes Appendix A: Migration Checklist......................................................... Before Loading the Data................................................................. After Loading the Data ................................................................... Troubleshooting and Q&A ............................................................. Appendix B: Case Study....................................................................... Preparation Work ............................................................................ Environment/Prerequisites........................................................ Data Mapping: Source Data and Target Fabric ........................ Parcel Types..............................................................................
Esri White Paper

Page 1 1 2 2 2 3 5 7 8 8 8 9 10 11 12 12 13 16 16 18 19 21 21 22 23 25 25 25 26 27
i

Loading Data into a Parcel Fabric

J10025

Migration Process ........................................................................... Create a Staging Environment .................................................. Prepare Source (Geometry)....................................................... Add the Fabric Attributes to the Source Data........................... Use Iterator Geoprocessing Tool to Load Data ........................

28 28 28 31 32

December 2011

ii

J10025

Loading Data into a Parcel Fabric

Introduction With the release of ArcGIS 10, Esri is now providing a land records solution as a core part of the ArcGIS platform. This solution will help you efficiently manage your parcel databases while at the same time helping improve the spatial integrity of your database, all through a data model designed specifically for managing parcels. The parcel maintenance solution is part of a bigger land records and local government solution that helps you not only implement efficient editing workflows but also produce great web maps and incorporate best practices from the land records industry at large.
The parcel maintenance solution has three key parts. The first is the parcel data model, also known as the parcel fabric. This dataset manages the spatial and topological relationships inherent in parcel point, line, and polygon data. The second part is the Parcel Editor toolbar, which is included with ArcGIS Desktop at the ArcEditor and ArcInfo license levels. This toolbar contains the tools needed to perform the daily set of splits, combines, and so on, you need to do with your parcel data in a streamlined set of workflows. The third and final part of the solution is the Tax Parcel Editing template. This template supports local government workflows and provides a multiscale editing map that organizes the survey framework (Public Land Survey System [PLSS] and control), subdivisions, lots, tax parcels, and encumbrances in a fabric data model that can be used with the Parcel Editor toolbar. The Tax Parcel Editing template also includes the Local Government Information Model, which is a database schema used with all the applications you can download from Esri's ArcGIS Resource Center (more on that will be discussed later). This harmonized information model connects silos of information in an organization and integrates processes throughout typical government departments. It includes a series of essential foundation layers and operational information that support a range of key maps and apps within a local government. Its design reflects specific application requirements and the cartographic design elements necessary to produce rich, multiscale basemaps and operational layers. You'll find the parcel fabric schema in a feature dataset called ParcelEditing. This schema will show you how to organize many parcel types (e.g., PLSS, subs, lots, tax parcels) in a single parcel fabric. The purpose of this paper is to address one of the first issues in this process, which is how you get your data into a parcel fabric. As with many systems, it's important to have quality data going in. The fabric will then help ensure that you not only maintain the quality of the data but also improve it. This paper also addresses many of the common issues that you may encounter in loading your data but does not attempt to cover every possible situation. The better shape your data is in at the start of this process, the smoother the transition to the fabric and templates will be.

Parcel Fabric Data Model

The parcel fabric is a dataset optimized for the maintenance of parcel data. It stores a continuous surface of connected parcels and ensures that the relationships of polygon features, line features, and point features are maintained. The parcel fabric is made up of three primary components: parcels, lines, and control. Parcels are defined by a series of boundary lines that store measurements as attributes on the lines. When you enter record measurements on parcel lines, they will be preserved irrespective of the changes you make to the parcel geometry to assemble a continuous parcel dataset. The parcel

Esri White Paper

Loading Data into a Parcel Fabric

J10025

fabric also ensures that parcel lines are not broken by adjoining parcels (the classic backlot issue). In doing so, it maintains a continuous set of lines for each parcel and does not intersect those lines when adjacent parcels meet at different parcel corners.

Tax Parcel Editing Template

The maintenance or editing of parcels is one piece in a complete solution for not only the management of parcels but also the analysis and dissemination of information to the public. The ArcGIS for Local Government Resource Center (ArcGIS Resource Center: Land Records) was created to help organizations achieve the goal of a complete solution by providing a series of templates containing best practices for land records management. The resource center contains templates for editing parcels, providing information to the public, and visualizing valuation and revenue trends. All these templates are based on a data model containing a parcel fabric as well as many other components that work toward the goal of a complete solution. One of the templates available from the ArcGIS for Local Government Resource Center is the Tax Parcel Editing template. This template is an ArcGIS 10 editing map authored for local government parcel editors. It organizes the survey framework, subdivisions, lots, tax parcels, and encumbrances in a fabric data model that can be used with the Parcel Editor toolbar. It is an editor that can be used by mapping technicians in an assessor's office, public works agency, registrar of deeds office, or any other agency responsible for managing land records information.

Migrating to the Parcel Fabric Data Model

Recently, Esri has been helping several local governments migrate to the land records solution. When doing so, the first step is generally to migrate existing parcel data to the parcel fabric. This migration provides a platform for the Tax Parcel Editing Map and Parcel Editor tools and opens up opportunities to leverage additional land records applications mentioned earlier. Working through these implementations, Esri has developed a pattern that should be used if you migrate your data to the fabric data model. The migration approach is as follows: Before you start loading data, you'll want to understand the parcel fabric and the Local Government Information Model (the model used with the land records templates) so you can understand where your data will be going. It is suggested that you start by downloading the Tax Parcel Editing template (Template: Tax Parcel Editing Template for ArcGIS 10) from the Land Records Resource Center. The information model within the template supports all the land records maps and apps as well as those in other local government subject areas. After downloading the template, you will want to unzip the contents into a local directory on your machine. The contents of the download include the following:

Step 1: Get to Know the Fabric Data Model and Tax Parcel Editing Template

Application directoryContains a set of add-in tools for extending the Parcel Editor
toolbar for things such as custom parcel naming conventions Information Model, and so on

Documentation directoryContains information about the Local Government MapsandGeodatabase directoryContains map document and sample database Getting Started with the Tax Parcel Editing Template documentPDF file
discussing the contents of the download and how to get started Start by reviewing the Getting Started document in the download. From there you can proceed to open the map document in the MapsandGeodatabase directory and familiarize
December 2011 2

Loading Data into a Parcel Fabric

J10025

yourself with the contents of the template and the parcel fabric data model. The resource center also contains a video (Video: Tax Parcel Editing Template) you can watch to learn how some of the basic parcel editing workflows can be accomplished against the sample data. You will also want to familiarize yourself with the data model and the different components that are modeled within it. Keep in mind that you are looking at the entire Local Government Information Model, and your organization may not need to maintain everything that the model is set up to manage. It is common for organizations to trim down the model to meet their specific needs.

Step 2: Prepare Your New Geodatabase

At ArcGIS 10, geodatabase designs can be shared with a new type of layer package called a schema-only layer package. A schema-only layer package (as the name implies) contains the schema of the data it references and can be used as a template for populating your data and setting up the layer properties (such as symbology and scale) provided in a typical map document. The Local Government Information Model is shared as a schema-only layer package as a way to distribute the geodatabase design and companion map layers. When you download the layer package (Schema: Local Government Information Model for ArcGIS 10), save it in the MapsandGeodatabase folder from your previous download. To build an empty geodatabase schema for your data, start by creating a new geodatabase. Next, browse to the folder where you downloaded the schema layer package. Double-click that file (item.pkinfo) to open the schema through ArcMap.

When ArcMap has started, you will be asked to specify the geodatabase you'd like to import the schema into. The default location is Default.gdb; select the empty geodatabase you created above instead. Note: The sample geodatabase is named Local Government, and all the maps published on the Local Government Resource Center will work with a geodatabase of this name. This is also a good time to specify the spatial reference for the new geodatabase you're creating. A simple way to do this is to select Spatial Reference > Other > Import and pick an existing feature dataset that has the correct spatial reference for your data.

Esri White Paper

Loading Data into a Parcel Fabric

J10025

When the Import Schema Package tool completes, you will see the layers from the package in your map document and an empty schema in the geodatabase you specified. (If for some reason you find yourself in an edit session after this step, save your edits to commit the changes to the database.) From here, you can review the descriptions and domains for each feature and begin thinking about how you can migrate your data to the new parcel fabric information model. Note: Your imagery, your surface models, and any stand-alone tables you'd like to include in this information model must be added manually to this schema. The ParcelEditing feature dataset contains the parcel fabric. The parcel fabric is optimized for editing and should only be used by editors in your organization. The ParcelPublishing feature dataset contains a series of simple geodatabase feature classes that represent the content you'll edit and maintain in the parcel fabric. The simple features should be used to build your parcel basemaps and in your other land records applications. Open the fabric within ArcCatalog or the Catalog window in ArcMap and click the Fabric Classes tab on the Parcel Fabric Properties dialog box. Spend time reviewing the extended fields and domains on the fabric and determine which fields you'll use in your implementation. You can use the layers in the Tax Parcel Editing map to see the attributes Esri has included for each parcel type.

The Parcel Fabric Properties dialog box is where you manage the attribute fields within the fabric.

The Tax Parcel Editing template uses a field called Type to distinguish between the different parcel types (tax parcels, lots, subdivisions, etc.) within the fabric. This is an

December 2011

Loading Data into a Parcel Fabric

J10025

important design decision and a fundamental component of the editing map you'll ultimately author for your users. In ArcGIS 10, users do not directly interact with the parcel fabric in the editing map; instead, they interact with layers derived from the Type field in the parcel fabric. The Type field has a domain called ParcelType assigned to it that should be reviewed before loading polygon data into the fabric. You'll need to decide which parcel types you want to manage in your parcel database. This paper will later describe in more detail how to load the different types and what to do if you want to manage a particular type (such as LotsConveyance Division) but don't currently have that data as polygons. The table below is a list of the ParcelType domain and what can be managed with the data model. When loading data, you'll want to add a Type field to your polygon data and calculate values appropriately based on this table/domain. Code 1 2 3 4 5 6 7 8 9 10 11 Description PLSS Township PLSS Section PLSS Quarter Section Special Survey Simultaneous Conveyance Conveyance Division Tax Parcel Ownership Encumbrance Separated Right Other

Note that the Parcels class has a field called Name. This field is used in the parcel fabric to store the parcel identification number, lot number, subdivision name, etc. You can then alias the field name in ArcMap when you build map layers for each appropriate parcel type and include it in the label expressions for tax parcels, lots, and subdivisions. This paper will further discuss attribute migration in Step 8: Maintain Your Attribute Values.

Step 3: Prepare Your Source Data

As with many systems, it's important to have quality data to start your migration. Once you've migrated, the parcel fabric data structure will then help you maintain the quality of your current data and improve it. This step will vary from organization to organization, but in general, you'll want to start by creating a development or staging geodatabase (but not your production one). This can be a simple file geodatabase on your local machine. It is not the same geodatabase you created in step 2. That database will be the ultimate destination of your data and was set up as a way to review the data model. In your staging geodatabase, you'll also want to create a feature dataset to place the data in. Make sure your feature dataset is in a projected coordinate system identical to what is defined in the database from step 2. Esri's parcel maintenance solution allows you to manage multiple parcel types in a single parcel fabric. The parcel maintenance solution is configured to support the survey framework (PLSS and control), subdivisions, lots, tax parcels, and encumbrances. This configuration supports the National Parcel Data Standard developed by the Federal Geographic Data Committee (FGDC) Cadastral Subcommittee and should be used as a starting point for your implementation of the parcel fabric.

Esri White Paper

Loading Data into a Parcel Fabric

J10025

There are many reasons why organizations choose to manage lots and subdivisions as polygons in the fabric. One of the most common reasons is to maintain a geographic reference and set of attributes that match the legal description for platted lands on the tax roll. In addition, these lot and subdivision boundaries support planning and infrastructure workflows and related data within a local government. Managing polygonal representations of the lots and subdivisions within the parcel fabric also allows you to produce multiscale parcel maps that display appropriate labels at the right scale. The parcel fabric also allows you to store additional metadata about the subdivision in the Plan table. The help system has more information on working with plans and the type of information usually stored there. If your current data does not have lots or subdivisions as polygons, there is a variety of methods you can use to create those polygon features and subsequently load them into your parcel fabric. For instance, the Dissolve geoprocessing tool can be used to create new polygon features from common attribute values on parcel polygons. Note that if you use this approach for subdivision boundaries, the result may not be the legal boundary of the subdivision as identified on the plat. Rights-of-way and other related public uses may affect how well this approach works. For lot boundaries, it is not uncommon for some of the original lines to be stored in a separate line layer from the parcels. To generate a lot polygon layer, you would take the parcel lines, add the lot lines from the separate feature class, and build a new set of polygons using the Feature To Polygon geoprocessing tool. This tool also allows you to specify an optional label feature, which is often a point class made from lot annotation converted to points with the Feature To Point geoprocessing tool.

December 2011

Loading Data into a Parcel Fabric

J10025

Esri has included parcel types in the Tax Parcel Editing template parcel fabric that are typically managed (created or referenced, in the case of PLSS) as part of your land record editing workflows. You may choose to manage one or all of the parcel types in your parcel fabric. Some organizations also maintain other related layers like zoning and land use, but those layers are typically not part of the tax parcel editing workflows. Land use, zoning, and other layers like these should be associated with your parcel fabric so they can be adjusted and kept in sync but not maintained directly with the fabric.

Step 4: Load Your Boundaries and Parcels into a Staging Geodatabase

Regardless of where you start, the recommended approach for loading data into the fabric is to go through a geodatabase and use the Load A Geodatabase Topology Into A Parcel Fabric geoprocessing tool. By going through a geodatabase, you'll get the following benefits:

Ability to locate and fix data inconsistencies through a geodatabase topology Ability to delineate curves so they will be automatically created during the fabric
load

Ability to use geoprocessing tools to help create or delineate other layers to load
(lots, etc.)

Ability to prepopulate staging fields for use with templates

These benefits will be discussed in later sections. There are ways to load data directly into a fabric from coverages and CAD files, but this white paper will only discuss the loading of data from a geodatabase topology as it allows the benefits discussed above. Almost all the organizations Esri has worked with have benefited from one or more of these items. Many different clients requiring parcel management had data that started from a variety of initial sources including coverages, shapefiles, CAD files, and geodatabase feature classes. Naturally, the steps to move into a geodatabase will vary depending on the starting point and other factors. This section will discuss some of the basic steps of moving data into a geodatabase, but it may be necessary for you to consult additional resources (online help, your local regional office, etc.) for your specific situation if it is not covered here. No matter where you are starting from, you begin by creating a geodatabase to hold the data and a feature dataset within that geodatabase to load the parcel data into. When creating the dataset, be sure to create it using the coordinate system you want your fabric to ultimately be in. This is often the same coordinate system as your current data. Once the database and dataset are created, you can move on to loading your specific data. There are two main assumptions being made for the sections below on potential steps to move data into the geodatabase and a parcel fabric from your current data format. The first is that your organization has at least one ArcInfo license available for the conversion from CAD files. Most of the geoprocessing commands that are discussed require an ArcInfo license for use. The second assumption is that the source data is in the proper coordinate system to begin with, and there is no projection required as part of the process. This is particularly important for organizations that might be using CAD files, as it is relatively common to store those files in something other than the coordinate system you'll want to be in.

Esri White Paper

Loading Data into a Parcel Fabric

J10025

Coverages

Coverages are a common initial format for those looking to move into the parcel fabric, as many organizations have existing parcel solutions based on this format. Because the coverage has topology built in, the process of moving to a geodatabase topology is straightforward. The easiest approach is to use the Feature Class To Feature Class geoprocessing tool, which can be accessed from the System toolboxes under Conversion Tools and the To Geodatabase toolbox or by simply right-clicking the feature dataset you create in ArcCatalog or the Catalog window and selecting Import > Feature Class. This geoprocessing tool is preferred because it allows you to remove fields, such as LPOLY# and RPOLY#, that you no longer need as part of the loading process. Use this tool to load both the polygon and arc features within your coverage. If you don't have arc features in your coverage, you'll want to run Build on the coverage with the Line option to create the arc features. If your coverage parcel database is currently within ArcInfo Librarian or ArcStorm, you'll want to move it out into one or more coverages for loading into the fabric. The best method for doing this will vary depending on the number of parcels involved. COGO coverages can be loaded directly into a fabric using the Import Fabric Data wizard, but as mentioned previously, it is recommended that you use the Load A Topology To A Parcel Fabric geoprocessing tool and workflows outlined in this white paper. If you wish to load your COGO coverage directly, review the desktop help topic Migrating parcel data using the Import Fabric Data wizard for more information.

Shapefiles

Moving from shapefiles to a geodatabase is very similar to moving from a coverage to a geodatabase. The Feature Class To Feature Class geoprocessing tool can be used in the same manner as coverages to load your shapefiles. Quite often, organizations using shapefiles have not used both line and polygon shapefiles for managing their parcels. If this is true in your case, you'll need to use other geoprocessing tools to create the missing data. If you just have a polygon shapefile, the Polygon To Line geoprocessing tool (found under Data Management > Features) can be used to generate the necessary line feature class. You'll want to run this tool after loading the polygons into the geodatabase. If you have just a line shapefile, the Feature To Polygon tool can be used to generate polygons. If you have been entering some of your lines via COGO, you may find yourself in a situation where you have lines for some polygons and not others. For this scenario, you will need to create the missing line features. To do this, you can set up a geodatabase topology with one rule: parcel polygon boundary must be covered by parcel lines. After validation, the errors will represent the areas where a polygon doesn't have associated line work. These errors can be used to create the missing line features. Within the topology error inspector, show only these errors, select all of them, right-click, and choose Create Features. This will result in a complete parcel line feature class. These new lines will not have COGO attributes and will be inverted during the fabric loading process. The steps outlined here can also be used in conjunction with creating a topology for loading your data into the fabric (see Step 6: Build and Validate a Topology, below).

CAD Files

There are many different methods for loading data from a CAD file into a geodatabase. The approach you use may vary depending on the schema of your CAD files (whether they have just parcel data or there are many other feature types included in the files that need to be filtered out, etc.), the number of CAD files you have to load, and how these parcels are broken out in the files themselves (whether some parcels split between files, etc.). There's also the question of whether parcels are maintained as just lines or they are stored as polygons with attributes. This discussion assumes that you are trying to bring only line data over from your CAD files. In addition, you will bring over parcel identification number text so you can use that information in

December 2011

Loading Data into a Parcel Fabric

J10025

the polygon-building process. As with other procedures, you will want to try this with a small amount of data initially (for instance, one drawing file) to get the procedures nailed down before trying to process all your data. The first step in the process is to create a new geodatabase, or at least a new dataset, that you can load data into as part of a test procedure while fine-tuning the process. Once you've created a location for your data, the steps for importing from CAD given the assumptions above are as follows: 1. Run the Select geoprocessing tool (in the Extract toolset of the Analysis Tools toolbox), pointing to the polyline features in the CAD file you are importing from. Specify the name of the feature class to create and create an expression to use against the line data. The expression is key because you want to import only those line features that represent the parcel boundaries. Run the Repair Geometry geoprocessing tool to remove features without geometries, duplicate vertices, and extremely short segments. You may find this step unnecessary depending on the quality of your CAD data. Create a geodatabase topology against the new line data and add rules to check for overlapping line segments and dangles. Be sure to use a small tolerance (the default is usually acceptable) when creating the topology. After creating the topology, validate it and fix any errors you find. This step may also be optional depending on the quality of the CAD data. The Planarize command on the Topology toolbar in ArcMap can also be useful if you have a lot of places where lines should be split but are not. When your line work is clean, the Feature To Polygon geoprocessing tool can be used to generate polygons. When running this tool, specify the annotation of your CAD file as the label source if the annotation includes the parcel identification information, as you see in most CAD files with parcel data. The result will be a new set of polygons with the attributes of the CAD annotation as the attributes of the new polygons. Remove all fields but the Text field from the resultant polygon feature class. The Text field should have the annotation string representing the parcel number as its value.

2.

3.

4.

5.

Keep in mind that there are different ways to convert CAD files into feature classes, but this simple approach will work in many cases. Once you have finalized the specific set of steps for the migrating of your CAD files, you will want to put them into a geoprocessing model or Python script so you can easily run them for each of the individual CAD files you are trying to process. You will also need to add a step to combine the individual polygon and line feature classes that are created into a single complete feature classes. This same set of procedures could also be used to pull out lots, subdivisions, or any other polygon types that you want to include with your parcel fabric.

Step 5: Delineate Curves

One of the key aspects in loading your data is making sure curves are maintained if you already have them (the geodatabase and CAD files support true curves), or are properly delineated if you don't. The purpose of delineating the curves is so they can be automatically created during the loading process, since two-point lines and curves are the easiest things to work with when performing updates against a fabric. The parcel fabric also supports linestrings (individual line features with more than two points), but in

Esri White Paper

Loading Data into a Parcel Fabric

J10025

general you should only use these for natural boundary features such as streams and coastlines.

Working with Linestrings

As an example of what you are trying to convert, look at the lines for the parcel depicted in the figure below:

Linestrings often form the boundary of corner parcels.

The boundary lines of this corner parcel include a single line along the street edge (highlighted in the figure, with the vertices displayed). To accurately keep track of the metes and bounds description for this parcel, it would be appropriate for this line to be split into a two-point line, a curve, and another two-point line. Doing this would allow the appropriate bearing information to be maintained on the two straight line segments as well as the curve information on the corner between them. If you can separate the section of the line where the vertices are tightly and evenly spaced (the area of this linestring where the verticesgreen dotsare close together), the loading procedure will be able to create a curve. If the lines are currently linestrings with multiple vertices, use the Split Into 2-Point Curves And Lines custom tool (found at Add-in: Split lines at inflection points and create circular arcs) to delineate the curves. The download is an add-in with a toolbar containing a drop-down menu for defining the line layer to work against and a couple of tools:

The Curves and Lines toolbar is part of the download for the Split Into 2-Point Curves And Lines tool.

To use this tool, load the line layer for your parcel data into ArcMap and open the toolbar. Make sure the drop-down menu on the toolbar is set to the line layer for your parcels, then select the first tool. This tool will select all the lines in the current extent that have more than one vertex in them. Selecting the second tool will display a dialog box for converting the selected lines into two-point lines and potential curves:

December 2011

10

Loading Data into a Parcel Fabric

J10025

The Split into 2-point curves and lines dialog box lets you perform the conversion.

It is important to play with the parameters on this tool before applying it to all your data to see how it works. Use a value of 0.5 for Lateral offset tolerance and a value of 7 for Minimum line length. For the purpose of loading data into a parcel fabric, you will want to be sure the Create circular curves that satisfy the lateral offset option is not checked. After running this tool, be sure to review some of the data (particularly around culs-desac) and make sure the tool is identifying curves as expected. If the parameters are set correctly, you should see the line features in your data separated into features with just two points and features that should be curves (many closely spaced vertices). After checking the data to make sure the lines are being split into individual features correctly, you can run the tool on the remainder of the data in your dataset. Because the tool is just splitting lines at the appropriate locations, you do not have to worry about vertices moving and topology errors being created. However, it will be necessary to validate your data, because the lines are being edited, and dirty areas are being created. The tool for loading data into the parcel fabric will then take care of actually creating the curves within your data.

Working with Data Already Split into Two-Point Lines

If your data has already been split into two-point lines, including the areas that should be curves, you have some other decisions to make. The loading procedures for the parcel fabric will not be able to create curves in areas where there are several two-point line features, so you have a couple of options for how to proceed. Option 1 would be to load the data as is. There are not issues with loading the fabric and working with it when the lines are already split into two-point features. The only difficulty you'll have is when you need to work with (e.g., split, offset from, adjust) one of the boundaries that should be a curve. If you do subsequently have to update a parcel in which the lines were not converted to curves, you may find it necessary to reenter that parcel with the proper curve information before proceeding with the new update. This will depend on the nature of the update you need to perform. For instance, if the new update includes splitting the curve at some distance along it, it will be a little more difficult to accomplish the update when the lines aren't actually curves. Option 2 would be to regenerate the lines from the parcel polygons using the Polygon To Lines geoprocessing tool. You can use this option when you do not need to maintain any of the attributes on your current set of line features. This tool will create a new set of line

Esri White Paper

11

Loading Data into a Parcel Fabric

J10025

features that can then be used with the Split Into 2-Point Curves And Lines tool described in the section above. When using this tool, be sure to enable the Identify and store polygon neighboring information option so that only one set of lines is created.

Step 6: Build and Validate a Topology

As mentioned previously, it's important to ensure that your data is of good quality before loading it into a fabric. By using a geodatabase topology and a specified set of rules, you can be confident that your data is of suitable quality for loading into a fabric. For loading data into the fabric, you will need to create a topology with the lines and polygons defining your parcels and apply the following set of rules:

[Line feature class] Must Be Covered By Boundary Of [Polygon feature class]. [Line feature class] Must Not Self-Overlap. [Line feature class] Must Not Self-Intersect. [Line feature class] Must Be Single Part. [Line feature class] Must Not Intersect Or Touch Interior. [Polygon feature class] Boundary Must Be Covered By [Line feature class].

The default tolerance is normally appropriate when building the topology, but depending on the quality of your data and the accuracy at which it was collected, you may want to change it. Be careful when adjusting the tolerance, though, and be sure to always make a copy of your data before increasing the tolerance number. Once you have created your topology, validate it and verify that there are no errors based on the set of rules above. If there are errors, you will need to start an edit session on the data and fix the errors before moving on to the next step. An alternative to creating a copy while you try out different tolerance numbers would be to validate your topology within ArcMap in an edit session. If you validate within an edit session, you have the option of undoing your edits or not saving them. You can then drop your topology and adjust your tolerance as needed. For more information on using a geodatabase topology for loading data into your fabric, see the help topic About migrating parcel data using geodatabase topology.

Step 7: Model Other Land Types

Now that your parcels are configured and ready to load, look at the other types of data (lots, subdivisions, etc.) you might want to maintain along with your parcels. The local government data model available from the resource center is set up to model lots, subdivisions, and PLSS polygons (through domain values on the parcel Type field), but you may have additional polygon types that you want to manage with your parcel data, meaning land types that you would create or register (in the case of PLSS) as part of your land record editing workflows. As mentioned previously, the best practice for layers that are not part of the normal land records workflows but are related, such as zoning and land use, is to not add them directly to the fabric. There are many reasons why organizations choose to manage things like lots and subdivisions as polygons within the fabric. One of the most common is to maintain attributes for those feature types. Many jurisdictions require lot numbers to be shown as text on the map, so maintaining that information as an attribute on a lot feature makes sense. Storing lots as polygons within the fabric also gives you the advantage of being able to display the original platted boundaries as the dashed lines that are commonly seen in tax maps. For subdivisions, organizations want to maintain information such as subdivision name, as well as other metadata about when the area was originally platted, and so forth. Some of the information for things like subdivisions can also be stored as plan attributes within the fabric. The help system has more information on working with

December 2011

12

Loading Data into a Parcel Fabric

J10025

plans and the type of information usually stored there (Help topic: Plans and the survey record). If your current data does not have lots or subdivisions as polygons, you can use various methods to create those polygon features for loading into the fabric. For instance, the Dissolve geoprocessing tool is often used to take attribute values on parcels (such as subdivision name) to create new polygon layers. For lot boundaries, it is not uncommon for some of the original lines to be stored in a line layer separate from the parcels. To generate a lot polygon coverage, you would then take the parcel lines, add the lot lines from the separate feature class, and build a new set of polygons using the Feature To Polygon geoprocessing tool. This tool also allows you to specify an optional label feature, which is often a point class made from lot annotation converted to points with the Feature To Point geoprocessing tool. Step 2 discussed the use of the Type field by the Local Government Information Model as a way to maintain the different polygon types you may want to keep within your fabric. Creating and setting the Type field correctly within your different polygon types is a key part of the migration of this data to the parcel fabric. Without setting the Type field, it is difficult to distinguish between parcels, lots, and anything else you will want to maintain. Before loading any of the subsequent polygon types, make sure you have added the Type field and set it based on the domain values discussed in step 2. One very important aspect to be aware of when preparing additional polygon types for loading into the fabric is that boundaries for these types should line up precisely with the parcel data that you've already loaded. For instance, if you currently maintain subdivisions as a feature class separate from your parcels, you'll want to make sure those boundaries are coincident before loading the subdivisions. The lines used with the subdivision polygons should be broken at all the same places as the parcel boundaries. This is usually not an issue if you have generated the subdivisions from the parcel boundaries or maintained them together. It is necessary to ensure that the boundaries are coincident so that adjustments done once the data is in the fabric will move all boundaries together. If the boundaries are not coincident (i.e., points do not line up between different polygon types), then data won't move together. Geoprocessing tools, such as Integrate, can be used to make sure feature classes maintained separately are aligned for loading. Once you have the data you want to load as polygons, the procedure to load it into the fabric is the same as it is with the parcels. If you only have polygons, you'll need to use geoprocessing tools such as Polygon To Line to generate the line features. If you only have lines, you'll need to use tools such as Feature To Polygon to create the polygon features. Once you have both, a new topology can be created with the line and polygon data and the same set of rules discussed above for the parcel data. The easiest approach is to create a separate feature dataset for each polygon type you plan to load into the fabric and place the lines; polygons; and, ultimately, the topology for that data in the separate dataset.

Step 8: Maintain Your Attribute Values

After reviewing the local government information model and getting your data ready to load from a geometry standpoint, compare the attribute fields between your data and the template schema for line and polygon features. The fabric for the local government model can be found in the ParcelEditing dataset and is called ParcelFabric. Open the fabric within ArcCatalog or the Catalog window in ArcMap and go to the Fabric Classes tab on the Parcel Fabric Properties dialog box:

Esri White Paper

13

Loading Data into a Parcel Fabric

J10025

The Parcel Fabric Properties dialog box is where you manage the attribute fields within the fabric.

As mentioned previously, you will notice that the Parcels class has a field called Name. This field is used in label expressions for parcels, lots, and subdivisions within the Tax Parcel Editing template, as well as in some of the web applications for land records. When loading your data into the fabric, use the Name field as your parcel identification field, or you will need to update the label expressions and/or web applications to use a field with a name of your choice. The recommended approach would be to add a field called Name (text field with a length of 50) to your polygon data (parcels as well as lots, subdivisions, etc.), then calculate the value of that field based on the naming field you are currently using. You could also use your existing field name and set the alias for that field to Name, but that can become confusing. Other things to consider for polygon attributes include

The template data model and map document both use the Type field to distinguish

between the different parcel types (parcels, lots, subdivisions, etc.) within the fabric. The Type field has a domain called ParcelType assigned to it that should be reviewed before loading polygon data into the fabric. As with the Name field, the recommended approach is to add a field called Type (long integer field) to your polygon data and calculate it appropriately (based on the ParcelType domain: 7 for Tax, 6 for Conveyance Division, 5 for Simultaneous Conveyance, etc.) before loading the data.

December 2011

14

Loading Data into a Parcel Fabric

J10025

The ParcelType domain within the Local Government Information Model maintains different land types you might track within your system.

If there are attribute values on your polygon features that you want to maintain, add
fields with identical names and definitions to the Parcels class in the fabric before loading the data. The attribute values will transfer over during loading if the field definitions are the same.

Things to consider for line attributes include

The template data model and map document use the CartoLineType field for

distinguishing the different line types you might have within the parcel fabric. The CartoLineType field uses a domain (also called CartoLineType) that you will want to review before loading your data. If you already have a field or method you use for maintaining line type, the recommended approach would be to add a field called CartoLineType (text field with a length of 50) to your data then calculate the values based on your current attribute values. If there are additional line types that you want to maintain, you can add those values to the CartoLineType domain. As an example, if you plan on loading both parcels (Tax) and lots (Conveyance Division) into your fabric, add the CartoLineType field to the line feature classes for each type. Using the values defined in the CartoLineType domain, calculate the CartoLineType field for the parcel boundaries to Parcel Boundary and the CartoLineType field for lot boundaries to Conveyance Division before loading the data. with identical names and definitions to the Lines class in the fabric before loading

If there are attribute values on your line features that you want to maintain, add fields

Esri White Paper

15

Loading Data into a Parcel Fabric

J10025

the data. The attribute values will transfer over during loading if the field definitions are the same.

COGO Attributes

Many users that have been managing their parcels with Esri-based solutions over the years have at least some COGO attributes on their line features. COGO attributes are fields that store the measurement information that describes each line, such as bearing (Direction or Angle field) and distance of a straight line. These values are there because at least some of the lines in the parcel database were entered with COGO-based tools. If these attributes exist on the lines when they are loaded into the fabric, they are transferred automatically, as is, to retain the legal record information you entered. If the attributes do not exist, the lines are inverted (values are calculated based on the bearing; distance; and, potentially, radius) when loaded into the fabric. In migrations done to date, there has been some confusion when working with the resultant fabric in trying to understand which values were calculated (inverted) and which values are the original stored COGO attributes. There are different approaches to take while loading your data to keep track of these values. The two most common approaches are

Use the built-in Accuracy field on the Lines feature class. To use this option, you

would want to add the Accuracy field to your Lines class and calculate the values before loading them into the fabric. Give the lines with existing COGO values a different accuracy value than the ones without COGO values so that you can distinguish them once they are loaded into the fabric. The Accuracy field is used as part of the least squares adjustment, though, so some care has to be taken when using this field. Calculated to your Lines data, defined as a short integer. For those lines that do not have existing COGO attributes, set the Calculated field value to 1. For those lines that do have values, set the Calculated field value to 0. After migration, it will be easy to identify the different lines based on this field.

Add an additional field to track the calculated values. For instance, add a field called

In both cases, you will have to decide if you want to continue to maintain the attribute values after the migration process. The recommended approach here would be to use the Calculated field as a means of distinguishing between COGO values and inverted values during the migration. It is also worth mentioning that there are sometimes issues when users have COGO values on lines they are migrating into the fabric that are inconsistent with the actual geometries of the lines. The problems arise when trying to do updates with these features once they are in the fabric. As an example, there has been customer data in which the distance of a line was listed as 200 feet in the COGO attributes, but the actual length of the geometry was closer to 10 feet. To avoid these problems when editing your parcel fabric, it is recommended that you review the COGO attributes in your source data prior to loading your parcels into the fabric. For help reviewing and identifying bad COGO values, you can download an additional set of tools (Add-in: Parcel Fabric Data Migration and Analysis Tools) from the Resource Center. These tools will help identify those bad COGO values as well as help with other migration processes.

Step 9: Load Your Fabric

Now that you have prepared your data and downloaded and prepared the Tax Parcel Editing template, you are ready to load your parcels and other appropriate land data into the fabric. Within the geoprocessing system toolboxes there is one called Parcel Fabric

December 2011

16

Loading Data into a Parcel Fabric

J10025

Tools, with tools for the fabric including one specifically for loading data from a geodatabase topology. The tool is called Load A Topology To A Parcel Fabric and can be found in the Data Migration toolset.

The Load a Topology to a Parcel Fabric dialog box is used to load geodatabase topology classes into a fabric.

If you have not already reviewed the help topic About migrating parcel data using geodatabase topology on loading data into a fabric from a topology, you should review it now. Depending on the size of your data, you will follow one of three general procedures for using the loading tool:

If you have a relatively small number of parcels (the actual size will vary depending
on hardware being used, etc.), you can simply use the tool directly. divide it up into chunks and load the data a piece at a time.

If you have more parcels than can be adequately processed by the tool in one pass, Use the iterator models available on the resource center (Add-in: Iteratively Load to
the Parcel Fabric). The process you use will depend on the number of parcels you are trying to load. There is no exact number for which you use one option over the other, though, so it may be necessary to determine the best method for you through trial and error. If you have fewer than 70,000 parcels, you should try loading all the data at once (assuming it is currently in one feature class). If you have more than 70,000 parcels, you will need to think about a way to split the data up into more manageable chunks for loading.

Esri White Paper

17

Loading Data into a Parcel Fabric

J10025

When you are ready to execute the tool for loading data into the fabric, it will be necessary to run it independently for each polygon type (parcels, lots, subdivisions, etc.) you will be loading. If you end up having to separate your data into chunks, it isn't necessary to have the parcels, lots, and subdivisions all broken out the same way; they will all be loaded separately, and you'll most likely not have the same number of each type. It is recommended that you do some practice runs on the data to make sure everything (curves, attribute values, etc.) looks okay in the fabric before running the conversion on all your data. Because the loading tool is a geoprocessing tool, it will work on a selected set, so you can select a few hundred of your parcels and run the tool to see how things look after conversion for just those parcels.

Step 10: Annotation Considerations

Annotation/Text is not loaded into the fabric, but it is still an integral part of most land records data models. Traditionally, all the text required to produce tax maps was stored as annotation within coverages or the geodatabase. As the software has progressed, though, it has made more sense to model some of the text as labels to minimize the time spent editing annotation. The Tax Parcel Editing template and the Local Government Information Model makes use of labels for all polygon-based text within the map. By using labels, you get the advantage of not having to edit the annotation as well as being able to apply different expressions depending on the scale of the map. For instance, for parcel identification numbers, the label expression will display the entire number when the map is at a scale in which the numbers can be displayed inside the parcel. As the scale of the map increases, only a portion of the parcel identification number is shown. This type of adjustment to the text for better viewing is easily accomplished with labels but would require the storage of multiple pieces of text if you try to do it with annotation (one piece of annotation for each scale). As a result, the recommended approach for modeling polygon-based text is to use labeling, which also saves you the time it would take to migrate any current annotation you have. If you want to follow this recommendation for polygon-based text, review the label expressions that are set up in the Tax Parcel Editing template to ensure that they match the requirements for the tax maps you would like to produce. The other important piece of tax map text is the line-based information. Almost every tax map will have some specification of the distance or length of each parcel boundary line. In addition, some tax maps will also include the bearing or angle of that line. Labeling can also be used for these pieces of text, but in the majority of cases, annotation is required to get the proper positioning of the text (correct side of the line, etc.). Within the Tax Parcel Editing template and Local Government Information Model, annotation is used as the recommended method for maintaining dimension information. As a result, most organizations will find it necessary to move their annotation into the same geodatabase and dataset (though it isn't specifically required to have it in the same dataset) as the parcel fabric. For the purpose of parcel management, you will want the annotation to be feature linked to the Line class in the parcel fabric so that you can, at a minimum, take advantage of annotation automatically being placed for you when new parcels are created. Keep in mind, though, that it isn't necessary for your current annotation to be feature linked. There are advantages if your existing annotation happens to be feature linked, but it isn't a requirement in order to load it into a feature-linked class. As with the parcel features, your path to getting annotation into the geodatabase is going to vary depending on the current source of that information:

GeodatabaseThe easiest path will be for dimension information that is already

stored as annotation within a geodatabase. For this situation, you will most likely not have to do anything more than copy the feature class into the new geodatabase

December 2011

18

Loading Data into a Parcel Fabric

J10025

you've created for the fabric. You may also have created the fabric within your existing geodatabase, and the annotation can stay where it is. As mentioned previously, the recommended approach is to use a feature linked annotation class, so you will want to create a composite relationship between your annotation feature class and the Lines class of the parcel fabric to make the class feature linked. If your annotation is already feature linked to your parcel lines, then you will want to take some additional steps (add an additional field to your Lines feature class and populate that field with the original ObjectID of the line, etc.) to maintain the linkage during the loading process.

CoverageTo import coverage annotation, you should begin with the help topic on

the subject (Help topic: Importing coverage and CAD annotation). This help topic goes over the basics on how to import your coverage annotation using the available geoprocessing tools. The goal is to move the annotation into a feature-linked class, so be sure to use that option when running the tool. Some users have also had success using the Convert Coverage Annotation tool within ArcMap. This tool is not on any toolbar by default, so it needs to be added to a toolbar through the Customize dialog box (look in the Label category). annotation from CAD into a geodatabase.

CADAs with coverage annotation, the above help topic is also helpful in loading
Again, the goal is to put it in a feature-linked annotation class (even though the existing annotation will not be feature linked), so choose the appropriate options.

Step 11: Load Control Points (optional)

Those of you who have existing control points with your data will also want to migrate that information to the parcel fabric. Once the data has been migrated, you can continue to import new data using the same process or directly create and maintain control points using tools provided with the parcel fabric. Control data can be imported from a point feature class, table, or comma-delimited (.csv) text file. If the control points are stored in a point feature class, you can use the shape's geometry, and the coordinates will be calculated. During the import process, the control importer tries to match the control point to the parcel points based on the tolerance you specify. This process also occurs when importing new parcels using the topo loader. Since the control points usually cover the entire data extent, if you have a large dataset, it is recommended that you import the control points first and the parcels second. An alternative is to import the control points in pieces, each piece covering a different extent, such that not all the fabric points are cached into memory. Make sure your control point names are unique. This is a best practice for subsequent imports as well as a requirement for some of the tools that interact with control points. If you wish to import historical points or historical observations, you can add a field called Active (long integer) and distinguish between active (Active = 1) and nonactive points (Active = 0). If you export the control points to a simple feature class and reimport them, make sure to delete the PointID field before reimporting them.

Esri White Paper

19

Loading Data into a Parcel Fabric

J10025

The z-value for control heights might be used during the adjustment process to predict the ground to grid distortion; hence, it is recommended that you assign an estimated z-value to control points, even if such a value is not known, by either using an average or deriving it from a digital elevation model (DEM). Such a value can be rounded to indicate it was approximated. Like any other class, the control points class's attributes can be extended to meet specific requirements. For example, you can add a field to support a hyperlink to the control sheet document. Since ArcGIS Projection Engine can project data, there is no reason to add x,y,z attributes for each projection. For more help on importing control points, use the help topic Importing control points.

December 2011

20

J10025

Appendix A: Migration Checklist

The following set of checklists is meant to help guide you through the process of migrating data to the Local Government Information Model and the parcel fabric. The checklists do not replace what is in the rest of the document but provide an outline of the steps that need to be followed as part of the migration.
This list might help you verify that your parcel data is ready for loading and has loaded correctly. It is generic and might not be applicable to all.

Before Loading the Data

Spatial Reference: Make sure the staging data and the target fabric use the same spatial reference. (Tip: Import the staging data to the feature dataset of the target fabric.) Datasets for Load: Separate your data by parcel type to avoid overlapping parcels in the source data. Divide the data into loadable pieces (e.g., 20,00030,000 parcels per load; this depends on the hardware used). Split Curves: Curves are split in the tangent points. You can use the Lines and Curves Add-in to split the lines at the tangents. Either the loader or the add-in converts segmented lines to true curves. Repair Geometry: There are no zero-length line segments. (Run Repair Geometry on the lines and polygons.) Line Topology: The Lines geometry complies with four of the six topology rules required for loading. No Overlaps: In each dataset for loading, easily divided by parcel type, there are no overlapping parcels in the source data. If you have overlapping parcels (e.g., historical parcels), you can load them multiple times, making sure no loads contain any overlapping parcels. Validated Loading Topology: The loading topology is clean and validated. (Apply the six topology rules required for loading.) Data Integrity: Make sure vertices coincide where they are supposed to. For example, make sure the subdivision boundary coincides with the lots boundaries where it is supposed to. After importing to the fabric, boundaries will share common points and therefore always coincide. Match to Fabric Fields: Ensure that source data attributes match the parcel fabric classes' attributes in name and type, for example, the source polygons have a field called Type as long integer and a calculated value of 7 for Tax Parcels.

Esri White Paper

21

Loading Data into a Parcel Fabric

J10025

Extend Fabric Fields: If you have extended attributes (attributes that do not exist in the parcel fabric classes), add them to the appropriate fabric class using the same name and type. Delete Wrong COGO: Wrong COGO attributes have been identified and made NULL such that the geometry is inverted during the import process. Perform this after the geometry is modified and the topology has been validated. Wrong COGO attributes can cause problems if migrated to the fabric. Bearings: There is only one field for bearing, called Bearing or Direction or Angle. Plan Data: If the parcels are associated with a plan (subdivision number, recorder's number, deed number, etc.), a field called PlanName (text) must be added and populated. Create Inventory: For each parcel type, count and document the number of parcels. Populate Line Category: If possible to distinguish between a parcel boundary line and a frontage line, add a field called Category for lines and calculate them (boundary lines [0], frontage lines [5]).

After Loading the Data

Check Fabric: Right-click the parcel fabric in the Catalog window and run the Check Fabric tool. Spatial Index: Right-click the parcel fabric in the Catalog window and recalculate the spatial index for each fabric class to improve performance. Count: Compare the number of source polygons to the number of parcels for each parcel type. Check Attributes: Make sure all the attributes transferred as expected. Unclosed Parcels: Make sure there are no unclosed parcels for parcels that are polygons in the source data. (This can be done by checking the attribute Unclosed <> 1.) Measurement View: If you had COGO data on your lines, open a parcel with COGO attributes and switch to the measurement view. The shape should be similar and the misclosures acceptable. If you see otherwise, you most likely imported bad COGO data. Parcel Coincide: Make sure parcel types that you expect to coincide to each other share common boundary points. Control Points: Make sure control points are associated with parcel corners when they fall within the import tolerance. Relationships: Turn on points and lines and use the Identify tool to check relationships.

December 2011

22

Loading Data into a Parcel Fabric

J10025

Curves: Examine curve center points and measure the distance between them. Consider changing the center point merge tolerance. Plan Directory: If the PlanName field was populated in the source parcels, verify that the plans exist in the plan directory. Consider populating more plan fields by joining the plan table to a source table (open the parcel attribute table and use the Relate button to view the plan table). Part Connectors: Check for part connection lines that are very longif you have multipart parcels that are spatially away from each other, long part connectors have a bad impact on your geodatabase performance.

Troubleshooting and Q&A

# 1

Scenario/Question The topo loader dialog skips some parcel loading percentages, and the results show that no parcels were loaded. Is it better to import control points first or parcels?

Try/Answer This is most likely a memory issue. Try increasing the memory on the machine used for loading or loading data in smaller pieces.

The fabric checker reports errors.

This doesn't really matter. The association tolerance between parcel points and control can be set when importing parcels or importing control points. Error Types (For detailed information on all error types, see the help topic Checking the topological integrity of the parcel fabric.)

Parcel has dangling boundary links

Line has same from and to pointsThis

Suspected error. Usually not a real error.

Not all the parcels were loaded to the fabric, although the topology was clean and validated.

error indicates a zero-length line, which can potentially cause problems with SDE. Curve is missing a radial lineThis can be easily fixed in version 10.1 and is not a showstopper. Parcel contains lines with errorsSee the help documentation. Identify the parcels that were not loaded and check in the attribute table to see if they were loaded as unclosed parcels. Turning on the lines can also identify these parcels. Try selecting these parcels using 10.0 Service Pack (SP) 3 and checking the geometry and COGO of the parcel. Try deleting vertices that are extremely close to each other.

Esri White Paper

23

Loading Data into a Parcel Fabric

J10025

# 5

Scenario/Question There are duplicate plans in the plan table.

Try/Answer If the data is loaded in batches that contain the same plan name, you might end up with duplicate plans in the plan table. Use the Merge Fabric Plans tool (Add-in: Fabric Plan Tools) to remove duplicates. Do not attempt to remove them manually in the plan table, as you will end up with parcels referencing a plan ID that does not exist.

December 2011

24

J10025

Appendix B: Case Study

The following is the pertinent documentation from an Esri Professional Services project done on a midwestern county in the United States to convert its data to the land records data model. This particular client was moving its parcel data from the coverage format to a geodatabase. The information is provided as an example of some of the steps and thought processes that went into moving data into the Local Government Information Model. This is not meant to be a cookbook or an exact set of procedures for you to follow, though, as your data and situation will differ from what was encountered by this county. Hopefully, you will find the information useful in planning your process for moving to a more modern data model and set of applications through the Local Government templates. Preparation Work Environment/ Prerequisites ArcGIS Desktop 10 SP1 at the ArcInfo license level The Iterate.tbx geoprocessing tool to load data incrementally by tiles (Add-in:
Iteratively Load to the Parcel Fabric)

The topology rule file (see rules in the Step 6: Build and Validate a Topology
section)

The parcel fabric schema-only layer package (Template: Tax Parcel Editing
Template for ArcGIS 10)

The complete set of the most current data for migration (lots, rights-of-way [ROWs],
parcels, control, vacations/dedications)

Lines and Curves add-in installed on ArcGIS 10 (Add-in: Split lines at inflection
points and create circular arcs)

Esri White Paper

25

Loading Data into a Parcel Fabric

J10025

Data Mapping: Source Data and Target Fabric

# Feature Class Name SD.ASMT_BSE_PARCELS_R SD_PWCE_BSE_LOTS_R region.ord Description Tax Parcels Lots Assessment Ordinances Source Format Cov.Region Cov.Region Cov.Region Target GDB Class Fabric.Tax Fabric.conveyance division=6 & simcondivtype=Lot Fabric.Encumbrance??? =9 Comments

Can use the set as historic once migrated plus history type set

anno_blknum_CE

anno_parcelblknum anno_parceldesc anno_parceldim anno_rowdim_assess anno_orddim_assess anno_ordnum_assess SD_PWCE_BSE_BLKCIRC_L SD_ASMT_BSE_PARCELBLKC IRC_L SD_ASMT_BSE_PARCELTIELI NE_L SD_PWCE_BND_SUBDIV_R SD_ASMT_BSE_PARCELNUM LEADER_L SD.ASMT_BSE_PARCELDIM_L SD_ASMT_BSE_PARCELROW DIMLEADER_L SD_ASMT_BSE_ORDDIMLEA DER_L SD_ASMT_BSE_ORDNUMLEA DER_L PWSRV_BSE_ROWDETAIL_A

City Engineers (CE) Block Number (Anno or LabelDiffers from Assess Block Number anno_parcelblknum) Assessment Block Number (Anno or Label?Differs from CE Block Number) Parcel Descriptions (Condo namesCan this be replaced with labels?) Parcel Dimensions ROW Dimensions Ordinance Dimensions (Anno or Label?) City Engineering Block Circles (Replaced by compound label?) Assessment Block Circles (Replaced by compound label?) Parcel Tie Lines SubdivisionsWill harvest from current layer from a TBD date forward Parcel Number Leader Lines Parcel Dimension Leader Lines ROW Dimension Leader Lines Assessment Ordinance Dimension Leader Lines Assessment Ordinance Number Leader Lines Right-of-Way Broken Down by Quarter

Cov.Anno

Anno

Cov.Anno Cov.Anno Cov.Anno Cov.Anno Cov.Anno Cov.Anno Cov.Anno Cov.Anno Cov.Anno GDB.Feat Cov.Anno Cov.Anno Cov.Anno Cov.Anno Cov.Anno GDB.Feat

Anno Anno Anno Anno Anno Anno Anno Anno GDB.Feat Fabric.Sub GDB.Anno GDB.Anno GDB.Anno GDB.Anno GDB.Anno Fabric.conveyance division=6 & simcondivtype=public ROW

December 2011

26

Loading Data into a Parcel Fabric

J10025

Parcel Types

Esri White Paper

27

Loading Data into a Parcel Fabric

J10025

Migration Process Create a Staging Environment

1. Create a new file geodatabase. Import the schema-only layer package to create an empty geodatabase. Make sure the projection information is set to the local state plane coordinate system. For each quadrant, create a feature dataset; use the same coordinate system and resolution that will be used on the final database.

2.

Prepare Source (Geometry)

Prior to loading, the source data must be prepared so that it can be loaded into the parcel fabric structure. 1. Create a feature dataset for the quadrant to load and import the spatial reference from coverage.

2. 3. 4.

Export the _arc class from the coverage to the feature dataset referenced above. Export the _Polygon class (region.Parcels) to the same feature dataset. In ArcMap, select by location all lines that share a segment with the parcels and export them to the same feature dataset as everything else.

December 2011

28

Loading Data into a Parcel Fabric

J10025

5.

Create a topology in the same feature dataset as above using the line and polygon feature classes. Load the topology rules supplied and make sure to keep the default cluster tolerance as specified on the dialog box.

Esri White Paper

29

Loading Data into a Parcel Fabric

J10025

6.

Validate the topology and add it to the map. Fix the errors listed in the dialog box by deleting a line that is missing a polygon and subtracting lines that overlap.

7. 8.

Revalidate the topology, and it should be clean. If it is not, fix the remaining errors and save the edits. This source should be loadable without running steps 914. Install the Curves and Lines add-in, double-click it, go to Customize, and add the Curves and Lines toolbar.

9.

Make sure that the Lines feature class is added to ArcMap and that it appears in the Curves and Lines toolbar as shown above.

10. Open the attribute table of the Lines feature class and note the total number of features. Close the attribute table. 11. Use the first button on the toolbar class. to select all polylines in the Lines feature

12. Once the line selection has been made, start editing and use the second button on the toolbar .

December 2011

30

Loading Data into a Parcel Fabric

J10025

13. Use the following tolerances and run the utility; do not create circular curves:

14. Open the attribute table for the Lines feature class and make sure that there are more lines than before the lines were split. (This will ensure that you don't miss splitting any quadrants.) 15. Revalidate the topology; there should not be any errors introduced. 16. Run the attribute assignment below to source and load to parcel fabric.

Add the Fabric Attributes to the Source Data

Add the fields in your fabric schema to the source data so they can be migrated to the fabric. Parcel Type Lots Attribute PlanName Name SimConDivType BlockDesignator Type ParcelSource Name Type StatedArea ParcelSource Field Type (size) Text (255) Text (50) Text (50) Text (10) Long Integer Text (50) Text (50) Long Integer Text (50) Text (50) Action Required Calc from Subdivision Calc from LotNum Calc "Lot" Calc from BlockNum Calc to "6" Calc from Source Calc from ScheduleNum Calc to "7" Calc from CAMA Data Calc from Source

Tax Parcel

Esri White Paper

31

Loading Data into a Parcel Fabric

J10025

Parcel Type ROW

Attribute Type SimConDivType Category (Lines) Type Historical HistoryType Name Type SimConDivType Historical HistoryType Name Type Name

Dedications

Vacations

Easements

Field Type (size) Action Required Long Integer Calc to "6" Text (50) Calc "Public Right of Way" Long Integer Calc to "5" Long Integer Calc to "6" Long Integer Calc to "1" Text (60) Calc to "Retired Tax Parcel" Text (50) Calc from OrdNum Long Integer Calc to "6" Text (50) Calc "Public Right of Way" Long Integer Calc to "1" Text (60) Calc to "ROW Vacation" Text (50) Calc from OrdNum Long Integer Calc to "9" Text (50) Calc from OrdNum

Note: The field names and types must match your fabric schema exactly or they won't be migrated.

Use Iterator Geoprocessing Tool to Load Data

1.

Create a Fishnet feature class for each quadrant. This feature class will be used to iterate through the data, one cell at a time: a. Search for and find the Create Fishnet (Data Management) geoprocessing tool. b. c. d. e. f. Activate the tool and specify the output feature class name and location. Enter the number of rows and columns to generate a cell size that contains up to 50,000 polygons. Choose the geometry type POLYGON. Run the tool and make sure there are approximately 40,00050,000 staging parcels in each cell. Delete any cells that contain no polygons or merge cells that have a few polygons. Once the fishnet is created, it can be used for every data migration.

2. 3.

Use the Iterator geoprocessing tool to load the source data by parcel type. Repeat step 2 for each parcel type and quadrant.

December 2011

32

Esri inspires and enables people to positively impact their future through a deeper, geographic understanding of the changing world around them.
Governments, industry leaders, academics, and nongovernmental organizations trust us to connect them with the analytic knowledge they need to make the critical decisions that shape the planet. For more than 40 years, Esri has cultivated collaborative relationships with partners who share our commitment to solving earths most pressing challenges with geographic expertise and rational resolve. Today, we believe that geography is at the heart of a more resilient and sustainable future. Creating responsible products and solutions drives our passion for improving quality of life everywhere.

Contact Esri
380 New York Street Redlands, California 92373-8100 1 800 447 9778 t 909 793 2853 f 909 793 5953 info@esri.com esri.com Offices worldwide esri.com/locations

usa

Printed in USA