Laserfiche WebLink
MEMO 89.2 <br />Temporal Mean NDVI Trend <br />190 <br />180 <br />170 <br />~ 160 fSmall_Grains <br />D Corn <br />~ 150 ~ NI Small Grains <br />~ NI Corn <br />~ 140 <br />130 <br />120 <br />110 <br />q, ~p ti~ ~ tip` °' ti~ ~ ~~ '1 ~q; 1 q`l% <br />Figure 8. Example of Multitemporal Signatures of Selected Irrigated and Non-irrigated Crop Types using <br />Reference Data Collected from FSA <br />As a result of this signature analysis and comprehensive image interpretation, threshold values for <br />irrigated lands were determined for the multitemporal NDVI image composites (Table ~. The thresholds <br />were optimized iteratively using accuracy assessment procedures (Section 2.7 Accuracy Assessment) <br />until the best results were obtained. <br />Table 4. Initial and Revised NDVI Thresholds for Landsat frame 033/032 <br />Landsat Acquisition Date Initial NDVI Threshold c~> Optimized Threshold ~~> <br />Apri126 >153 >155 <br />May 12 >163 >165 <br />Junes >175 >175 <br />July 7 >148 >150 <br />August 24 >136 >150 <br />September 25 >133 >145 <br />November 4 ~ >131 - <br />n Usine re-scaled NDVI values described in Section 2.3.1 <br />i <br /> <br /> ~ <br /> <br /> ~ <br />i ~ <br /> <br />Consequently, pixels with NDVI values above the predetermined NDVI thresholds for all Landsat <br />acquisition dates were classified as irrigated. The outcome of the thresholding procedure was a <br />preliminary classification of irrigated lands. This preliminary classification was recoded into a binary <br />mask with a value of 1 for irrigated areas and a value of 0 for non-irrigated areas (Figure 9) <br />Page 13 of 45 ~Rlversfde FecAnotogy, fnc. <br />4'JaYer Resources Errgi~ecr:np an~i CansuFlrnp <br />