ArkDSS Task 1, Spatial System Integration Subtask 1.1.c, Irrigation Well Information Layer Prepared for: Colorado Water Conservation Board and Colorado Division of Water Resources Prepared by: HRS Water Consultants, Inc. and ElephantFish, LLC FINAL September 21, 2020 Contract No. CT2018-00578, ArkDSS Spatial System Integration Component HRS Job No. 17-02
14
Embed
ArkDSS Task 1, Spatial System Integration Subtask 1.1.c ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ArkDSS Task 1, Spatial System Integration
Subtask 1.1.c, Irrigation Well Information Layer
Prepared for:
Colorado Water Conservation Board
and
Colorado Division of Water Resources
Prepared by:
HRS Water Consultants, Inc.
and
ElephantFish, LLC
FINAL
September 21, 2020
Contract No. CT2018-00578, ArkDSS Spatial System Integration Component
HRS Job No. 17-02
1
Table of Contents Executive Summary ....................................................................................................................................... 2
SEO Compiled Historic Well Datasets ........................................................................................................... 4
Split Historic Well Datasets ........................................................................................................................... 5
Well to Parcel Assignment ............................................................................................................................ 6
Assign Well Attribute Data to Irrigated Parcels .......................................................................................... 11
Comments on Deliverable Limitations ........................................................................................................ 12
Final Deliverables ........................................................................................................................................ 12
State Refinements ....................................................................................................................................... 13
List of Tables
Table 1 Historic Well Datasets Table 2 Key attributes in the historic wells’ datasets Table 3 Snapshot year historic well datasets Table 4 Input and output datasets used with the ET GeoWizard’s Tool “Global Snap Points” Table 5 Excerpt of the attribute table for the irrigated parcel dataset, including joined attributes
from the historic well datasets.
List of Figures
Figure 1 General historic well information layer workflow Figure 2 Location and extent of historic wells datasets Figure 3 Historic wells inside and outside the HI Model Boundary for each snapshot year Figure 4 Original well location and snapped well location Figure 5 Flow chart displaying steps followed during the inspection phase. Figure 6 “ET_Status” of historic wells per snapshot year – snapped, manual assign, and NA
Figure 7 Example of a well that was not snapped to a parcel.
2
Executive Summary This technical memorandum documents the approach, deliverables, and limitations for Subtask 1.1.c,
Irrigation Well Information Layer, prepared under Task 1, Spatial System Integration, of the Arkansas
River Decision Support System (ArkDSS).
This memorandum and the related GIS deliverables were prepared by HRS Water Consultants, Inc. (HRS)
and ElephantFish, LLC (EF). Staff from the State Engineers Office (SEO), the Colorado Water
Conservation Board (CWCB), and Division 2 staff at Colorado Division of Water Resources
(CDWR/Division 2), also assisted in completing this task.
The original objective of Subtask 1.1.c was to use Hydrobase well information to create a Geographic
Information System (GIS) coverage of irrigation well locations. The well coverage was to list irrigation
well attributes including well “start” and (if applicable) “end” dates, to indicate the period of use of the
irrigation wells in Division 2. However, per discussions with the SEO and Division 2 GIS staff, the Subtask
1.1.c scope was revised to entail assigning “historic wells” as a groundwater source to irrigated parcels
for each irrigated parcel snapshot year (1954, 1975, 1988, and 1998; see ArkDSS Task 1.3) based on a
semi-automated process in GIS. A “historic well” is a well that was not associated with an irrigated
parcel in the SEO’s 2010 irrigated parcel dataset. The historic well information1 was provided by the SEO
and Division 2 as two GIS coverages.
The general approach, displayed in Figure 1, consisted of:
The SEO and Division 2 GIS staff compiling historic well data which was delivered to HRS.
HRS reviewing and aggregating these datasets and filtering the results based each of the wells’
status (i.e., in service/not in service or does not exist) during each irrigated acreage snapshot
year (1954, 1975, 1988, and 1998).
HRS assigning the historic wells as a groundwater source to the Task 1.3 irrigated parcels (i.e.
adding the wells’ WDIDs2 to the irrigated parcels’ attribute table) for each snapshot year based
on geographic location using a semi-automated process in GIS.
The SEO and Division 2 each provided HRS with a point GIS coverage of historic wells. Division 2
provided HRS with historic wells inside the HI model area. The SEO provided HRS with wells mostly
outside of the HI model area. HRS reviewed and aggregated the two GIS coverages using spatial and
temporal filters. The spatial aggregation created two datasets – wells inside the HI model area and wells
outside the HI model area. The temporal filtering created historic well datasets for each snapshot year.
This was needed because not all wells were in service for the full period between 1954 and 1998. This
resulted in eight total datasets (inside and outside HI model datasets for the four snapshot years).
1 The historic well coverage only includes wells in service between 1954 and 1998; this does not include all wells in Division 2. Thus, the wells shown on the maps, graphs, and tables in this memorandum only encompass the historic well information layer provided by the SEO and Division 2. 2 The groundwater diversion ID (WDID) is a unique identifier for well structures in Hydrobase.
3
To assign each well to an irrigated parcel, the “Global Snap Points” tool, an ArcGIS plugin by ET
Geowizards3 was used. This tool “snapped” or moved each well (GIS point) to the nearest irrigated
parcel within ¼ mile, allowing the well to be designated as a water source for the parcel. Some wells
were not within a ¼ mile of a parcel; these wells were visually inspected in GIS. A portion of these wells
were manually attributed to an existing parcel. For the remaining wells, some were linked with newly
digitized parcels and a very small number were not attributed to any parcel. While this approach has
some inherent assumptions and limitations, which are discussed below, it allows for historic wells to be
efficiently linked to nearby irrigated parcels.
The Task 1.1.c GIS deliverables were provided to the State in May 2019 and included:
Two file geodatabases containing wells inside the HI model area (HistoricWells_InsideHI.gdb)
and outside the HI model area ( HistoricWells_OutsideHI.gdb)
Each file geodatabase contained a point file for each snapshot year – 1954, 1975, 1988, and
1998 (8 total point files)
A related deliverable was the Task 1.3 file geodatabase containing the irrigated acreage for the snapshot
years (HistoricIrrigatedAcreage.gdb) with the groundwater sources linked during Task 1.1.c.
Figure 2. Task 1.1.c Memorandum layout and general historic well information layer workflow
In summary, the method used to assign historic wells to the snapshot year irrigated parcels was
primarily based on the proximity of the well to a parcel and the completion date and abandonment date
of the well. In cases where there was no proximal irrigated parcel, HRS visually inspected the area
around the well and either digitized new irrigated acreage or associated the well with a slightly more
distant parcel that could reasonably be served by the well. While this method can be inexact at the
3 ET GeoWizards is a collection of geospatial processing tools that has a seamless integration with ArcGIS Desktop.
More detailed information regarding the plugin can be found here: https://www.ian-ko.com/
changes had occurred; additional irrigated areas that were observed as part of this process were
digitized and added to the dataset. However, irrigated areas that were not near any already digitized
parcel(s) would not have been observed and thus would not have been digitized. This explains why a
small number of parcels were added as part of the well to parcel assignment process.
Figure 4. Original well location and snapped well location
9
Figure 5. Flow chart displaying steps followed during the inspection phase
For all eight of the historic well datasets most of the wells were automatically snapped to irrigated
parcels and most of the remaining wells were manually assigned to an irrigated parcel (Figure 6). Wells
that were not snapped or manually assigned were in areas without reasonably nearby irrigated parcels
or irrigated areas. Most of these un-snapped wells were located outside the HI Model Boundary (see
Figure 7).
10
Figure 6. “ET_Status” of historic wells per snapshot year – snapped, manual assign, and NA
Figure 7. Example of a well that was not snapped to a parcel
11
Assign Well Attribute Data to Irrigated Parcels HRS applied the historic wells as irrigation sources to their associated irrigated parcels. This involved
adding the well’s WDID and ET_Status to the GIS attribute table of the irrigated parcel coverage. Two
main steps were followed to ensure that the attributes were applied appropriately:
Spatial Join – ArcMap’s ‘Spatial Join’ tool was used to join the well datasets’ key attribute fields
to the irrigated parcel dataset attribute table:
o WDID(s) – The unique well structure identifier from Hydrobase. When added to a parcel
in the irrigated parcel dataset, it identifies the well as a groundwater source for that
parcel.
o ET_Status – Indication of how the well was attached, or snapped, to the irrigated parcel.
Clean Up Attribute Table – The spatial join was performed as a ‘one to many’ relationship
between tables, which resulted in a row for each parcel and unique associated well. Thus,
parcels with more than one associated well contained an attribute row for each well (i.e. a
duplicate parcel). To remove the duplicate records, while retaining all relevant WDIDs, HRS
added the fields “WDID and WDID_01-WDID_056” to the attribute table (Table 5), where:
o WDID = Historic well WDID that was either snapped or manually assigned to the parcel
o WDID_01 – WDID_05 = Additional historic well WDIDs if more than one well was
snapped or manually assigned to a parcel.
Table 5. Excerpt of the attribute table for the irrigated parcel dataset, including joined attributes from the historic well datasets.
6 These fields were included in the datasets that HRS submitted to the SEO and Division 2 in May 2019. In the version of the data on the CDSS website, these field have been removed and the WDIDs have been added to the fields GW_ID1 to GW_ID20.
12
Comments on Deliverable Limitations The process developed to assign historic wells to irrigated parcels introduced some limitations into the
final deliverables. These include:
The distance snapping approach did not account for physical barriers such as property lines,
roads, bodies of water, etc. For example, if a parcel across the street from a well was closer
than a parcel on the same side of the street, the well will have been snapped to the parcel
across the street.
A well may have been snapped to the neighbor’s irrigated parcel if that parcel was closer than
the well owner’s irrigated parcel.
A well may have been snapped to a certain parcel in one snapshot year, then snapped to a
different parcel in another snapshot year. This could occur due to the changing extent of the
irrigated parcels over the snapshot years, causing one parcel to be closer to the well in one
snapshot year, then further from the well in another snapshot year.
In summary, the method used to assign historic wells to the snapshot year irrigated parcels was
primarily based on the proximity of the well to a parcel and the completion date and abandonment date
of the well. In cases where there was no proximal irrigated parcel, HRS visually inspected the area
around the well and either digitized new irrigated acreage or associated the well with a slightly more
distant parcel that could reasonably be served by the well. While this method can be inexact at the
parcel level, we believe it provides a reasonable generalized representation of the location and history
of historic groundwater in Division 2.
Final Deliverables As per the Subtask 1.1.c scope of work and revisions thereto, HRS submitted the following data
deliverables:
Two file geodatabases containing wells inside the HI model area (HistoricWells_InsideHI.gdb)
and outside the HI model area (HistoricWells_OutsideHI.gdb)
o Each file geodatabase contained a point file for each snapshot year – 1954, 1975, 1988,
and 1998 (8 total point files):
HistoricWells_InsideHI.gdb
Wells_InsideHI_1954
Wells_InsideHI_1975
Wells_InsideHI_1988
Wells_InsideHI_1998
HistoricWells_OutsideHI.gdb
Wells_OutsideHI_1954
Wells_OutsideHI_1975
Wells_OutsideHI_1988
Wells_OutsideHI_1998
13
As part of ArkDSS Task 1.3 HRS submitted the following deliverable, which included work done as part of
Subtask 1.1.c:
One file geodatabase containing the irrigated acreage for the snapshot years
(HistoricIrrigatedAcreage.gdb) with the newly applied groundwater sources
State Refinements Following HRS’s submission of deliverables, the state refined the assignments of the historical wells to
parcels within the final irrigated parcel datasets by examining available documents for each well
including permits, decrees, and DWR field inspection forms. Although many permits and decrees are
silent on place of use, many do indicate the permitted/decreed place of use while inspection forms
often provided some indication of the parcels on which wells were being used at the time of inspection.
Using HRS’s spatially based assignments as a starting point, a number of well assignments were altered