SPDSS_Task89-2_CropLandUseClassificationProcedures_20060929

• Title: SPDSS Task 89.2 - Crop and Land Use Classification Procedures for Year 2001
• Description: This memorandum describes the activities conducted under Task 89: ‘Mapping of Irrigated Land Use and Irrigated Parcel Boundaries’ for year 2001 and complements the SPDSS Memoranda for Task 89.1 and Task 90.2. This memorandum also provides details on the methods used to conduct a number of Task 89 activities, including Task 89.3: Determine Irrigated Vs. Non-irrigated Lands, Task 89.4: Identify Crop Types In Each Polygon, Task 89.5: Review, Revision and Final Classification, and Task 89.6: Conduct Accuracy Assessment, as well as the results obtained from these activities for year 2001.
• Decision Support - Doc Type: Task Memorandum
• Date: 9/29/2006
• DSS Category: GIS
• DSS: South Platte
• Basin: South Platte
• Contract/PO #: C153960
• Grant Type: Non-Reimbursable
• Bill Number: SB01-157, HB02-1152, SB03-110, HB04-1221, SB05-084, HB06-1313, SB07-122
• Prepared By: Riverside Technology inc.

MEMO 89.2
Figure 16 shows the development of overall accuracy and Kappa Index of Agreement (KIA) of three
iterations of the classification refinement process. The Initial Classification (1) and the Refined
Classification (2) were conducted using the initial and optimized thresholds shown in Table 4 of Section
2.5.1. The `Final Classification' corresponds to the last iteration of the classification refinement process,
after manual delineation of irrigated parcel boundaries and edits from water user inputs. The overall
accuracy achieved in this classification was 96 percent. Notable is not only the progressive improvement
of overall accuracy and KIA with each refinement iteration but the diminishing gap between the KIA and
overall accuracy, indicating an improvement in the reliability of the results6. Error matrices and their
corresponding derived accuracy assessment parameters for all three classifications are presented in
Appendix D.
100% 1

90% 0
9
.
80% 0
8
.
sv
70% 0
7 ~
.
~ ~
60% 0
6 Q
3
~ .
.
~
x
~
Q 50% 0.5 0,

40% D
0
4
.
O
30% fD
0
3 ~
.
rt
20% - 0.2
10% 0.1
0% 0
(1) Initial Classification (2) Refined ~3~ Final Classification
Classification
(Overall Accuracy 68.42°k 76.54°k 96.16°k
~ KIA 0.37 0.53 0.92
Figure 16. Overall Accuracy and Kappa Index of Agreement (KIA) for the Iterations of Irrigated Lands
Classification
The error matrices in Appendix D show an initial overestimation of irrigated lands in the first two
classifications. This overestimation was reduced in the classification refinement process as a result of the
improved calibration of the thresholds, manual delineation of parcels, and later on, with edits from
information obtained from water users. The overestimation can be explained in part by a decision made
by the analysts to be inclusive in the initial classification attempts and then proceed to eliminate the
identified errors gradually in the classification refinement process. Table 5 shows an evaluation of edits
made to the irrigated parcels data based on comments from water users. The purpose of this evaluation
was to corroborate results obtained with the accuracy assessment described above using a different
evaluation approach. From Table 5 it is evident that the amount of parcels eliminated was approximately
6 The Kappa Index of Agreement (KIA) is based on the difference between the actual agreement in the error matrix (i.e., the
agreement between the classification and the reference data as indicated by the major diagonal) and the chance agreement (i.e.,
indicated by the marginals) (Congalton and Green 1999). The higher the KIA, the farther apart the classification results from a
classification resulting from chance.
Page 24 of 45 ~R~versfde FecAnotogy, fnc.
4'JaYer Resources Errgi~ecr:np an~i CansuFlrnp