Managing elevation data: Part 3: Workflow steps

This topic applies to ArcGIS for Desktop Standard and ArcGIS for Desktop Advanced only.

This workflow creates several mosaic datasets from some source data, combines them, and creates one (master) mosaic dataset to use in creating multiple products. It then walks you through creating referenced mosaic datasets and mosaic dataset layers to use in sharing and publishing.

The data was described earlier. Here is the information about the data storage and spatial reference.

Data	Spatial reference	Pixel size	Z units	Bit Depth	NoData
GTOPO	GCS_WGS_1984 Decimal degrees WGS 1984	30 arcsec (1 km)	meters	16-bit signed/unsigned Integer	-9999
SRTM	GCS_WGS_1984 Decimal degrees WGS 1984	90 m	meters	16-bit signed Integer	-32768
NED 30	GDC_North_America_1983 Decimal degrees NAD 1983	1 arcsec (30 m)	meters	Float	-3.4x^38
NED 10	GDC_North_America_1983 Decimal degrees NAD 1983	1/3 arcsec (10 m)	meters	Float	-3.4x^38
Lidar (DEM/DSM)	NAD83_HARN_StatePlane_Oregon_North Foot NAD 1983 HARN	3 ft	feet	Float	-3.4x^38

Data used in the workflow

Note:

If you don't have multiple types of source data, you could create a single mosaic dataset, add the data, and build the overviews. Then follow the workflow where it begins to create multiple products from the master mosaic dataset.

Create a template mosaic dataset

A template mosaic dataset is an empty mosaic dataset that will be modified to contain the default customizations you want applied to any of your mosaic datasets.

Steps:

1. If you do not already have a geodatabase, then create a new geodatabase.

   You can do this by right-clicking a folder in the Catalog window and clicking New > File Geodatabase.

2. Create a new mosaic dataset.

   You can do this by right-clicking the geodatabase in the Catalog window and clicking New > Mosaic Dataset, or opening the Create Mosaic Dataset tool.

3. Use the same spatial reference for all mosaic datasets, since you will be bringing them all together into a single mosaic dataset. The spatial reference does not need to be the same as the input data. Presuming this is a global dataset, Esri recommends web Mercator (using meters, not decimal degrees—unless ALL the data is in decimal degrees). Alternatively, if all of the organization's data is in a limited area (such as a state) and you are confident it won't need to expand later, it is appropriate to use a local coordinate system such as state plane or UTM.

   Click the Coordinate System browse button and expand Projected Coordinate Systems > World > WGS 1984 Web Mercator (Auxiliary Sphere).

4. Since you want to use the mosaic dataset as a template for the source and master mosaic dataset, you need to specify the pixel type, since you'll be bringing together different bit depths when you create the master.

   Expand the Pixel Properties, click the Pixel Type drop-down arrow, and click 32_BIT_FLOAT.

Create custom fields

Custom fields are added to the mosaic dataset's attribute table to manage the metadata unique to each dataset or the mosaic dataset that will be used for providing information to the users, used in queries, or used by the mosaic method. You can do this directly from the Catalog window, or open the mosaic dataset in ArcMap and edit the attribute table in the table window.

Steps:

1. From the Catalog window, right-click the TemplateMosaicDataset and click Properties.
2. Click the Fields tab.
3. Scroll down, select an empty row, and enter these new fields:

   Field Name	Data Type	Description
   Horiz_CE90_m	Float	Used to store the Horizontal Accuracy in meters
   Vert_LE90_m	Float	Used to store the Vertical Accuracy in meters
   Source	Text	Used to record the source of each dataset
   Best	Float	Used to sort the data for which data appears at any zoom

   Description of the fields
4. Add other custom metadata fields at this time, such as a copyright note, date, or owner.

The Best field is recommended to control the ordering using the By Attribute mosaic method. This field is arbitrary and you could identify a numeric value using whatever methodology suits your use cases. In this workflow you'll be populating it with the lowest pixel size for each dataset to ensure the highest resolution is displayed at the appropriate scales.

Learn about adding fields

Create the source mosaic datasets

Source mosaic datasets are created for each of the inputs, unless the input is a single raster dataset. The source data for this workflow includes the following (described earlier):

• GTOPO
• SRTM
• NED 10 & 30
• LIDAR DEM
• LIDAR DSM

You will create a copy of TemplateMosaicDataset and rename it according to each input. It is recommended that you prefix S_, for source, to the name of each mosaic dataset to help you manage and identify their contents, such as S_NED and S_SRTM.

Steps:

1. In the Catalog window, right-click the TemplateMosaicDataset and click Copy.
2. Right-click the geodatabase and click Paste.
3. Rename the Target Name.

Alternatively, you can use the Copy tool to create copies of the TempateMosaicDataset.

Add data to the source mosaic dataset

You will add the appropriate data to each source mosaic dataset.

Steps:

1. In the Catalog window, right-click one of the source mosaic datasets, and click Add Rasters.

   This opens the Add Rasters to Mosaic Dataset tool.

2. Verify the Raster Type is Raster Dataset.

   In most cases, you will be using the Raster Dataset raster type, unless you're adding DTED data.

3. Click the Input drop-down arrow and click Workspace and add the workspace folder (navigate to it or drag it onto the dialog box).

   It is assumed that your data is organized into a folder structure for each separate source; however, if not, you may have to choose Dataset (instead of Workspace) and navigate to individual files.

4. Optionally, check option to not add duplicates if you think your data collection may be full of duplicate files and you do not want them added multiple times. However, this will take additional time to add, so don't use unless necessary.
5. Optionally, enter a file name filter if necessary. For example, if you have a low-resolution .gif file stored with the source data (.tif), you will have to specify a *.tif filter or the .gif files will also be added.

Do not build overviews at this time since they're not needed for these source mosaic datasets. You will be building overviews on the master mosaic dataset.

Fill in metadata fields

You will populate the fields you added in the beginning of the workflow for each source mosaic dataset.

Steps:

1. Add the source mosaic dataset to the ArcMap table of contents.
2. Right-click the Footprint and click Open Attribute Table.
3. If the values you will be entering only apply to some rows in the table, then select those rows first. If all your values apply to every row, then make sure no rows are selected.
4. Right-click the column heading for the field you will be editing and click Field Calculator.
5. Enter the appropriate values for your field.
   • For the Horiz_CE90_m and Vert_LE90_m values, you will enter numeric values. For example, SRTM may have a horizontal accuracy = 20 meters and a vertical accuracy of 16 meters. GTOPO has a vertical accuracy ranging from 30 to 500 meters.
   • For the Source value, enter the value within quotes, such as "GTOPO" or "NED30".
   • For the Best value, you can enter [LowPS] to copy the values from the LowPS field into the Best field; otherwise, enter values you have calculated.

Convert lidar from feet to meters

Since the lidar data is measured in feet and all the other data has an elevation value measured in meters, you will be converting the lidar values to meters. For simplicity, you will add the Arithmetic function to the items in the lidar source mosaic datasets to convert the values from feet to meters. Alternatively, you could add the data to the master mosaic dataset, then select the items and then add the function.

Steps:

1. Right-click the Footprint in the table of contents and click Open Attribute Table.
2. Click the Table Options button in the Table window and click Select All.
3. In the table of contents, right-click the Footprint layer and click Selection > Batch Edit Raster Functions.

   The Raster Functions Editor Wizard opens.

4. The operation should be Insert Function. Then click Next.
5. Check Insert above the top-most function and click Next.
6. Right-click Identity Function and click Insert > Arithmetic Function.
7. Set the following:

   • Operation = Multiply
   • Raster = Raster 2
   • Constant = 0.3048
8. Click OK, and continue through the wizard to insert the function.

Quality control

It is not uncommon to find quality issues with the data (especially NoData values), problems with projections, and so on. Examine each source mosaic dataset in ArcMap. You may need to stretch with DRA checked on in the Image Analysis window since the mosaic dataset doesn't have (or need) statistics. You can examine the data values with the Identify tool . You may need to edit the footprints or define the NoData for some source data.

Footprints

Examine the mosaic dataset without the Image layer checked on. Look at the footprints to be sure they are organized correctly. Sometimes the source data doesn't have its projection information specified correctly and you may have footprints appearing away from the others. You will have to remove this file, correct the projection, and re-add it to the mosaic dataset.

The footprints are generated based on the raster type chosen while adding raster data to the mosaic dataset, and in the case of the Raster Dataset raster type, it's generally just the envelope of the raster dataset. Therefore, you may want to modify it (such as shrink it) using the Build Footprints tool.

Learn about mosaic dataset footprintsLearn about removing raster data from a mosaic datasetLearn about defining or modifying a raster coordinate system

NoData

NoData refers to pixels or areas within a data file or a geographic region where no valid data exists (the absence of data). In a raster dataset, the NoData values are represented by a fixed value that (hopefully) isn’t used as a valid value. There are many possible values used to represent NoData, such as 0 (for 8-bit INTEGER data) or -99,999 (for FLOAT data).

Learn about NoData in raster datasets

The desired result is that any NoData pixels will not be displayed or used in any geoprocessing tool calculations. For more information, see NoData and how it affects analysis.

There are three methods for handling NoData, based on the complexity of the data:

• The footprints (extents of valid imagery in each dataset) and boundary (the union of all footprints) may be used as a mask, then the system will consider everything outside the mask to be NoData.
• For NoData limited to a few numeric values, you can configure the mosaic dataset to identify NoData (for example, -99999), and any pixels with that value will be ignored.
• For datasets with a complicated boundary and/or NoData values within the interior of the boundary, a third option is to use both methods above to create a raster mask of all NoData (as a new raster layer) and then configure the system not to display any pixels excluded from the mask.

Modifying the footprints is typically recommended for handling NoData due to speed, but this generally assumes the NoData values are around the edges of the individual datasets. If there are holes of NoData within each dataset, modifying the footprints will not work. However, configuring the mosaic dataset to have one or more NoData values can slow down the system in the case of multiple overlapping images (for example, check each layer for NoData here).

Esri’s recommended approach is based on the characteristics of the data. For each dataset, identify the value used as NoData, and if possible, determine if any regions of NoData exist in the dataset. Then, for the full project area, determine if there will be multiple datasets that overlap (at the same scales due to the same pixel sizes) and have similar pixel size ranges. For example, NED and SRTM data is tiled and doesn’t overlap (at the same scales); therefore, define the NoData value using the Define Mosaic Dataset NoData tool.

If overlapping data is present, then it may be best to modify the footprints. For example, lidar datasets may cover overlapping areas.

If the individual data files in the source datasets have relatively simple boundaries, but all NoData is around the edges of each dataset (for example, there are no holes of NoData within the image), then redefine the footprints using the Build Footprints tool with a specified range of valid data values (for example, for elevation data, MinimumDataValue = minus100, MaximumDataValue = 9000), and use the default (25) for the Number of Vertices. If the source data files have a small region of overlap (1–5%), it may be acceptable to set a Shrink Distance (default = 0).

Tip:

Be sure to set the Always clip the Image to its Footprint to Yes. To access it, right-click the mosaic dataset in the Catalog window, and click Properties > Defaults. Do this for the source mosaic dataset and any other mosaic datasets where it will be used (such as the master and referenced mosaic datasets).

If the individual datasets have irregular footprints, and there are NoData holes in the interior, it may be necessary to apply both methods above, but note that this can slow down system performance.

If the data has a more complex configuration of NoData, such as multiple NoData values in the data interior and a complicated boundary, one final approach is to define a raster mask as an additional image layer, then apply that layer as a NoData mask by creating a custom function. For example, create an image where the NoData is represented by 0 and the valid data is represented by 1, then use the Arithmetic function and multiply the mosaic dataset by this mask image.

Tip:

You can use the Con(IsNull) expression to create a raster where NoData is 0 and the valid data is 1.

Learn about Is NullLearn about Con

Create master mosaic dataset

Copy the TemplateMosaicDataset, as you did earlier, to create a mosaic dataset named MasterMosaicDataset.

Add data to master mosaic dataset

There are two ways to populate the master mosaic dataset with the source mosaic datasets. The recommended way is to use the Table raster type. This will ensure that every item in the source mosaic dataset is added as individual items in the master mosaic dataset. This will also ensure that you can modify the items individually, if necessary, and reduces processing overhead when publishing the dataset. Optionally, you can add the source mosaic datasets to the master mosaic dataset using the Raster Dataset raster type. This will add each source mosaic dataset as a single item. This limits access to each input in the source mosaic dataset, including access to individual metadata information, footprints, querying, and download access. It's not recommended in most cases.

Steps:

1. Right-click the master mosaic dataset in the Catalog window and click Add Rasters.

   This opens the Add Rasters to Mosaic Dataset tool.

2. Click the Raster Type drop-down arrow and click Table.
3. Click the Input drop-down arrow and click Dataset.
4. Add each source mosaic dataset to the dialog box and click OK. (Do not add the S_Lidar_surface mosaic dataset at this time.)

   

Set mosaic dataset's properties

To ensure your mosaic dataset is configured to display your data, you may want to set the mosaic methods and some other properties.

Steps:

1. In the Catalog window, right-click the MasterMosaicDataset and click Properties.

   The Mosaic Dataset Properties dialog box opens.

2. Click the Defaults tab.
3. Click the Allowed Compression Methods button.

   The Configure Allow List dialog box opens.

4. Click the Default Method drop-down arrow and click LERC.

   LERC is an efficient lossy compression method recommended for data with a large pixel depth.

5. Close the dialog box.
6. Click the Allowed Mosaic Methods button.

   The Configure Allow List dialog box opens.

7. Uncheck Seamline (since your mosaic dataset doesn't have any).
8. Click the Default Method drop-down arrow and click By Attribute.
9. Click the Order Field drop-down arrow and click Best.

   The Order Base Value should be 0.

   The Best field will ensure that the highest resolution data available in any geographic location is the highest priority for display. Note that the MinPS and MaxPS values in the Attribute table provide the primary control for which the image is displayed at any scale.

   Learn about the mosaic methods

10. Close the dialog box.
11. If you modified the footprints of any source mosaic dataset, set Always clip the Image to its Footprint to Yes.
12. Optionally, modify the Allowed Fields to only show those fields you added.
13. Since the Web Mercator projection you're using uses the WGS84 datum, but not all the input data does, click the Geographic Coordinate System Transformation button and choose the appropriate transformations.

    In the case of the data used in this workflow, a transformation will be used with the lidar dataset and the appropriate choice is: NAD_1983_HARN_TO_WGS_1984.

Build overviews

Overviews are used to speed up the display of a mosaic dataset at low resolution. In concept, they are similar to pyramids, although overviews are based on mosaics of multiple images, versus pyramids which are associated with individual images.

Learn about mosaic dataset overviews

Not very many overviews will be generated for the master mosaic dataset, since there are so many different resolutions of data. Overviews are not limited to being the lowest resolution dataset, especially in this case. Pyramids were not built for any of the input data, since they were going to be added to a mosaic dataset containing datasets with varying resolutions. Where the factor between one overlapping resolution and another is greater than 3 (between the MaxPS of the smaller resolution and the LowPS of the larger resolution), an overview will be generated.

There are two ways to build the overviews: one way involves modifying the defaults and the other does not. Overviews are additional files written to disk. By default, they're stored next to the geodatabase where the mosaic dataset is stored. If you want to store them in a different location, use the Define Overviews tool and specify an output location. Alternatively, you can change other properties of the overviews, such as the compression method. This tool will add records to the attribute table representing the overview; however, they are not actually generated. You then need to run the Build Overviews tool to create the overviews you defined.

If you want to accept the defaults, you can just run the Build Overviews tool. The Build Overviews tool will take some time to generate the overviews. You can review the progress in the Results window.

Note:

Unless your application requires low-resolution overviews to be generated from the high-resolution source data, Esri recommends using existing lower-resolution data (for example, GTOPO and SRTM). In this case, you should not build overviews until all datasets have been ingested, and only at that time, consider if additional overviews are required.

Update Best field

Since the Best field was manually edited, and you've just added overviews, the Best field is empty for the new overviews. You need to fill this field in and optionally any other fields you've added.

Steps:

1. Right-click the Footprint layer in the table of contents and click Open Attribute Table.
2. Select all the overview items. You can do this manually or with a query. If using a query, all the overviews should have a Category = 2, and the Name will begin with Ov.
3. Right-click the Best column heading and click Field Calculator.
4. Enter [LowPS] and click OK.

Create referenced mosaic datasets and layers (products)

Now you can focus on creating the products that you will share and publish, such as the hillshade, slope, and aspect items for visualization. The referenced mosaic dataset allows you to generate alternative visualizations without making a copy of the source data.

There are some limitations to referenced mosaic datasets that are important to your workflow and implementation. For example, you cannot edit the attribute table—this ensures that the data is maintained as you designed it in the master mosaic dataset. Also, referenced mosaic datasets maintain a 1:1 relationship between them and the parent mosaic dataset and when the parent (in this case, master mosaic dataset) is updated with new items, they will appear in the referenced mosaic dataset automatically.

Learn about referenced mosaic datasets

Caution:

Be sure there are no records selected in the master mosaic dataset before creating the referenced mosaic datasets; otherwise, only the selected records will be added to the referenced mosaic dataset.

Referenced master mosaic dataset

Creating a referenced master mosaic dataset allows you to provide users access to your master mosaic dataset, without having to share it directly. This will protect it from changes others may unintentionally perform. You can also use this as the source for publishing the image service.

Steps:

1. In the Catalog window, right-click the MasterMosaicDataset and click Create Referenced Mosaic Dataset.

   The Create Referenced Mosaic Dataset tool opens.

2. Name the output however you wish. It is recommended you prepend it with R_ to identify it as a referenced mosaic dataset, such as R_MasterMosaicDataset.
3. Click OK.

Create hillshade

You will create another referenced mosaic dataset, add the Hillshade function, and calculate the statistics.

Create referenced mosaic dataset

Steps:

1. In the Catalog window, right-click the MasterMosaicDataset and click Create Referenced Mosaic Dataset.
2. Enter R_MasterDEM_Hillshade for the Output Mosaic Dataset.
3. Click OK.

Add Hillshade function

The hillshade function will generate an image from the elevation data showing a hypothetical illumination of the surface considering the azimuth and altitude of the light source (sun).

You will insert the Hillshade function and modify some of the mosaic dataset's default properties.

Steps:

1. In the Catalog window, right click R_MasterDEM_Hillshade and click Properties.
2. Click the Functions tab.
3. Right-click the Mosaic Function and click Insert > Hillshade Function.

   You can accept the defaults or modify the azimuth, altitude, or z factor. You may need to experiment with different values to generate the desired hillshade effect.

   Learn about the Hillshade function

4. If you're working with a world-wide dataset (as it's assumed with this workflow), click the Scaling drop-down arrow and click Adjusted.
5. Click OK to close the dialog box.
6. Click the Defaults tab.
7. Open the Allowed Compression Methods and set the default to JPEG.
8. Set the Default Resampling Method to Bilinear Interpolation.
9. Click OK to close the dialog box.

Calculate statistics

The statistics of the mosaic dataset are invalid since the original statistics were calculated on the elevation data (Float) and the hillshade is producing a 8-bit image. There are three instances where statistics can be calculated: for each source raster dataset, for each item in the mosaic dataset, and for the mosaic dataset. For specifics, see Raster dataset statistics. You only need to calculate them for the mosaic dataset.

When calculating the statistics for the mosaic dataset, you do not need to calculate them for every pixel. It is highly recommended you specify a skip factor. A simple equation is the number of columns divided by 1,000. Or if it's a worldwide mosaic dataset, try a very large number, such as 20,000. If you don't specify a skip factor, this process will take a very long time.

Steps:

1. In the Catalog window, right-click the mosaic dataset and click Enhance > Calculate Statistics.
2. Enter the skip values (such as 20000).
3. Optionally, enter ignore values, such as 0 or -9999.

Create slope

You will create another referenced mosaic dataset, add the Slope function, and create a mosaic layer file to define the rendering.

Create referenced mosaic dataset

Steps:

1. In the Catalog window, right-click the MasterMosaicDataset and click Create Referenced Mosaic Dataset.
2. Enter R_MasterDEM_Slope for the Output Mosaic Dataset.
3. Click OK.

Add Slope function

The Slope function will generate an image from the elevation data representing the rate of change of elevation from each cell.

Steps:

1. In the Catalog window, right-click R_MasterDEM_Slope and click Properties.
2. Click the Functions tab.
3. Right-click the Mosaic Function and click Insert > Slope Function.

   You can accept the defaults or modify the z factor.

   Learn about the Slope function

4. If you're working with a world-wide dataset (as it's assumed with this workflow), click the Output Measurements drop-down arrow and click Scaled.
5. Click OK to close both dialog boxes.

The Slope function calculates slope values from 0 to 90. You can calculate statistics; however, you may define some stretch parameters that can't be defined as part of the mosaic properties, so you can forgo this calculation.

Define mosaic layer properties

Steps:

1. Add the mosaic dataset to ArcMap.
2. In the table of contents, right-click the Image layer and click Properties.
3. Click the Symbology tab and scroll to the bottom.
4. Click the Type drop-down arrow and click Minimum-Maximum.
5. Click the Statistics drop-down arrow and click From Custom Settings (below).
6. Enter the following values:
   • Min = 0
   • Max = 90
   • Mean = 45
   • Std Dev =1
7. Optionally, you can scroll up and change the color ramp. For example, you can choose a white-to-blue color ramp. Be sure to check Invert so the color ramp is applied from blue-to-white (low-to-high).

8. Click Apply and OK.
9. Save the mosaic dataset layer. Right-click the mosaic dataset in the table of contents and click Save As Layer File.

Create aspect

You will create another referenced mosaic dataset, add the Aspect function, and create a mosaic layer file to define the rendering.

Create referenced mosaic dataset

Steps:

1. Open the Create Referenced Mosaic Dataset tool.
2. The input is the MasterMosaicDataset.
3. Enter R_MasterDEM_Aspect for the Output Mosaic Dataset.
4. Click OK.

Add Aspect function

Aspect identifies the downslope direction of the maximum rate of change in value from each cell to its neighbors. Aspect can be thought of as the slope direction. The values of the output raster will be the compass direction of the aspect.

Steps:

1. In the Catalog window, right-click R_MasterDEM_Aspect and click Properties.
2. Click the Functions tab.
3. Right-click the Mosaic Function and click Insert > Aspect Function.

   Learn about the Aspect function

4. Click OK to close both dialog boxes.

The Aspect function calculates values from 0 to 360. You can calculate statistics; however, you may define some stretch parameters that can't be defined as part of the mosaic properties, so you can forgo this calculation.

Define mosaic layer properties

Steps:

1. Add the mosaic dataset to ArcMap.
2. In the table of contents, right-click the Image layer and click Properties.
3. Click the Symbology tab and scroll to the bottom.
4. Click the Type drop-down arrow and click Minimum-Maximum.
5. Click the Statistics drop-down arrow and click From Custom Settings (below).
6. Enter the following values:
   • Min = 0
   • Max = 360
   • Mean = 180
   • Std Dev =1
7. Optionally, you can scroll up and change the color ramp.
   1. Right-click the Color Ramp box and click Graphic View (to uncheck it).
   2. Click the Color Ramp drop-down arrow and click Aspect to choose the default aspect color ramp commonly used by ArcGIS users.

8. Click Apply and OK.
9. Save the mosaic dataset layer. Right-click the mosaic dataset in the table of contents and click Save As Layer File.

Create ellipsoidal ground height

Follow the steps in Converting from orthometric to ellipsoidal heights.

Create the DSM

The master mosaic dataset you created in the earlier step was a DEM representing bare earth. You can create another to support the DSM data. It is assumed that the DEM and DSM data will be used together to create the DSM master mosaic dataset, because the DSM data is only available at the highest resolutions. To do this, there are several options (ordered by recommended method):

1. Create a new master mosaic dataset, add the source mosaic dataset as shown earlier, but use the source DSM lidar mosaic dataset instead of the source DEM mosaic dataset. Build overviews, then create any of the referenced mosaic datasets desired.
2. Copy the original master mosaic dataset, remove the lidar DEM data, then add the lidar DSM datasets. Overviews already exist, but may need to be updated. Use the Synchronize Mosaic Dataset tool to update any stale overviews.
3. Alter the original master mosaic dataset by adding the DSM lidar data and adding a new attribute field to identify the DSM versus DEM data. Then modify the mosaic method (if necessary) to display the appropriate data.

Data attribution

Before sharing the mosaic datasets or publishing them as image services, you should add some attribution information. This can include credits, a description, and any information for limiting the use of the data.

Steps:

1. Click Customize > ArcMap Options.
2. Click the Metadata tab.
3. Click the Metadata Style drop-down arrow and select a style to view the metadata.

   The ArcGIS default is Item Description, but you may prefer something with more information, such as the ISO standard.

4. Click OK to close the dialog box.
5. In the Catalog window, right-click the mosaic dataset (or layer file) and click Item Description.
6. On the Description tab, click Edit.
7. It is advisable to enter information for: Title, Tags, Summary, Description, and Credits. You may add more, depending on your requirements.
8. Click Save.

Publishing

You can publish the mosaic dataset or layer file you’ve created as image services. For specific steps, see Serving raster data.

To learn about the different parameters, see Publishing image services.

The mosaic datasets or their layers associated with the data usages are:

• View—Hillshade, Slope (layer), and Aspect (layer).
• Analysis—Slope and Aspect. Optionally create a web application that would process the master mosaic dataset (referenced) and display the result to the user’s display.
• Values—Master mosaic dataset (referenced).

Moving data to a server

In most cases. the desktop machine and server machines. and sometimes their data storage, are not the same. The paths in a mosaic dataset are hardcoded, therefore, if the path names from the server machine are not the same as from the desktop machine, then the paths will need to be updated. It is recommended that you update the paths on the desktop machine before moving. For more information, see Repairing paths in a mosaic dataset.

Make sure you move the overview folder, and the cache folder (if you added lidar data), and update the path to all the items in the mosaic dataset, if the location changes.

Also, be sure the server has access to the locations where the data is stored; otherwise, the image services may be empty.

Capabilities

By default, the image service will be accessible as an image to any ArcGIS application or web application. However, if you will be using it in other applications that don’t support the ArcGIS Image Service, you will want to publish it with either the WMS or WCS capabilities.

Learn about these capabilities

Download

Allowing users to download is another capability that you may or may not want used. Download allows users to download the source elevation data used in the mosaic dataset, such as the SRTM file or NED file. This will impact the load on your server if many users are downloading.

Maintenance

In the event that new data is added to the existing service, you can do one of two things

1. Update an existing source mosaic dataset: Create a copy of the template mosaic dataset and ingest the data to perform QC. Add the new mosaic dataset to the source mosaic dataset using the table raster type. Then run the Synchronize Mosaic Dataset tool on the master mosaic dataset and check Update with new items.
2. Create a new source mosaic dataset: Create a copy of the template mosaic dataset and ingest the data and perform QC, then add the new mosaic dataset to the master mosaic dataset.

If the new data is a higher resolution, such as lidar, and presuming the lower-resolution data already exists and is adequate for overviews, there is no need to generate new overviews.