***NOTE ABOUT THE UNBUFFERED VALIDATION ACCURACY TABLES BEGINNING IN 2016: The training and validation data used to create and accuracy assess the CDL has traditionally been based on ground truth data that is buffered inward 30 meters. This was done 1) because satellite imagery (as well as the polygon reference data) in the past was not georeferenced to the same precision as now (i.e. everything "stacked" less perfectly), 2) to eliminate from training spectrally-mixed pixels at land cover boundaries, and 3) to be spatially conservative during the era when coarser 56 meter AWiFS satellite imagery was incorporated. Ultimately, all of these scenarios created "blurry" edge pixels through the seasonal time series which it was found if ignored from training in the classification helped improve the quality of CDL. However, the accuracy assessment portion of the analysis also used buffered data meaning those same edge pixels were not assessed fully with the rest of the classification. This would be inconsequential if those edge pixels were similar in nature to the rest of the scene but they are not- they tend to be more difficult to classify correctly. Thus, the accuracy assessments as have been presented are inflated somewhat. Beginning with the 2016 CDL season we are creating CDL accuracy assessments using unbuffered validation data. These "unbuffered" accuracy metrics will now reflect the accuracy of field edges which have not been represented previously. Beginning with the 2016 CDLs we published both the traditional "buffered" accuracy metrics and the new "unbuffered" accuracy assessments. The purpose of publishing both versions is to provide a benchmark for users interested in comparing the different validation methods. For the 2018 CDL season we are now only publishing the unbuffered accuracy only publishing the unbuffered accuracy assessments within the official metadata files and offer the full "unbuffered" error matrices for download on the FAQs webpage. Both metadata and FAQs are accessible at <https://www.nass.usda.gov/Research_and_Science/Cropland/SARS1a.php>. We plan to continue producing these unbuffered accuracy assessments for future CDLs. However, there are no plans to create these unbuffered accuracy assessments for past years. It should be noted that accuracy assessment is challenging and the CDL group has always strived to provide robust metrics of usability to the land cover community. This admission of modestly inflated accuracy measures does not render past assessments useless. They were all done consistently so comparison across years and/or states is still valid. Yet, by providing both scenarios for 2016 gives guidance on the bias. If the following table does not display properly, then please visit the following website to view the original metadata file <https://www.nass.usda.gov/Research_and_Science/Cropland/metadata/meta.php>.
USDA, National Agricultural Statistics Service, 2018 California Cropland Data Layer
STATEWIDE AGRICULTURAL ACCURACY REPORT
Crop-specific covers only *Correct Accuracy Error Kappa
------------------------- ------- -------- ------ -----
OVERALL ACCURACY** 419,207 84.0% 16.0% 0.828
Cover Attribute *Correct Producer's Omission User's Commission Cond'l
Type Code Pixels Accuracy Error Kappa Accuracy Error Kappa
---- ---- ------ -------- ----- ----- -------- ----- -----
Corn 1 10,430 75.2% 24.8% 0.748 75.4% 24.6% 0.750
Cotton 2 21,776 93.9% 6.1% 0.937 92.4% 7.6% 0.922
Rice 3 61,539 97.4% 2.6% 0.972 98.7% 1.3% 0.986
Sorghum 4 398 40.0% 60.0% 0.400 44.9% 55.1% 0.449
Soybeans 5 - n/a n/a n/a 0.0% 100.0% 0.000
Sunflower 6 4,700 84.2% 15.8% 0.841 75.4% 24.6% 0.753
Pop or Orn Corn 13 1 1.3% 98.8% 0.012 100.0% 0.0% 1.000
Mint 14 1 8.3% 91.7% 0.083 5.6% 94.4% 0.056
Barley 21 2,744 52.1% 47.9% 0.519 84.6% 15.4% 0.846
Durum Wheat 22 17 94.4% 5.6% 0.944 16.5% 83.5% 0.165
Spring Wheat 23 513 47.1% 52.9% 0.471 65.4% 34.6% 0.654
Winter Wheat 24 13,919 60.9% 39.1% 0.602 82.3% 17.7% 0.819
Rye 27 29 8.4% 91.6% 0.084 24.0% 76.0% 0.239
Oats 28 547 12.7% 87.3% 0.126 68.1% 31.9% 0.680
Millet 29 - n/a n/a n/a 0.0% 100.0% 0.000
Safflower 33 1,492 68.1% 31.9% 0.680 69.9% 30.1% 0.698
Mustard 35 - 0.0% 100.0% 0.000 n/a n/a n/a
Alfalfa 36 46,665 84.3% 15.7% 0.834 92.3% 7.7% 0.918
Other Hay/Non Alfalfa 37 9,029 70.1% 29.9% 0.696 58.7% 41.3% 0.581
Sugarbeets 41 2,228 71.7% 28.3% 0.716 68.0% 32.0% 0.679
Dry Beans 42 1,823 67.2% 32.8% 0.671 63.0% 37.0% 0.629
Potatoes 43 454 55.6% 44.4% 0.556 63.0% 37.0% 0.629
Other Crops 44 448 62.7% 37.3% 0.627 79.4% 20.6% 0.794
Sugarcane 45 1 16.7% 83.3% 0.167 10.0% 90.0% 0.100
Sweet Potatoes 46 47 61.0% 39.0% 0.610 79.7% 20.3% 0.797
Misc Vegs & Fruits 47 - 0.0% 100.0% 0.000 0.0% 100.0% 0.000
Watermelons 48 188 23.8% 76.2% 0.237 54.7% 45.3% 0.546
Onions 49 2,614 74.4% 25.6% 0.743 76.7% 23.3% 0.766
Cucumbers 50 267 59.9% 40.1% 0.598 56.0% 44.0% 0.560
Peas 53 67 27.9% 72.1% 0.279 49.3% 50.7% 0.493
Tomatoes 54 16,303 85.5% 14.5% 0.852 89.9% 10.1% 0.897
Herbs 57 229 35.5% 64.5% 0.355 64.9% 35.1% 0.648
Clover/Wildflowers 58 2,215 86.9% 13.1% 0.868 85.1% 14.9% 0.850
Sod/Grass Seed 59 697 42.9% 57.1% 0.429 61.6% 38.4% 0.616
Fallow/Idle Cropland 61 25,776 80.1% 19.9% 0.792 65.1% 34.9% 0.639
Cherries 66 1,596 93.2% 6.8% 0.932 97.3% 2.7% 0.973
Peaches 67 80 46.0% 54.0% 0.460 32.5% 67.5% 0.325
Apples 68 300 92.0% 8.0% 0.920 84.5% 15.5% 0.845
Grapes 69 35,723 98.3% 1.7% 0.983 98.3% 1.7% 0.982
Other Tree Crops 71 954 88.9% 11.1% 0.889 48.2% 51.8% 0.481
Citrus 72 6,236 61.6% 38.4% 0.614 98.6% 1.4% 0.985
Pecans 74 61 45.5% 54.5% 0.455 72.6% 27.4% 0.726
Almonds 75 61,969 93.2% 6.8% 0.927 97.2% 2.8% 0.970
Walnuts 76 18,425 94.2% 5.8% 0.941 96.1% 3.9% 0.960
Pears 77 440 96.7% 3.3% 0.967 98.0% 2.0% 0.980
Pistachios 204 41,096 93.7% 6.3% 0.934 99.1% 0.9% 0.990
Triticale 205 1,125 49.6% 50.4% 0.495 41.7% 58.3% 0.415
Carrots 206 1,053 61.7% 38.3% 0.616 44.1% 55.9% 0.440
Garlic 208 1,720 80.0% 20.0% 0.800 85.0% 15.0% 0.849
Cantaloupes 209 400 37.7% 62.3% 0.377 62.2% 37.8% 0.622
Olives 211 2,132 92.5% 7.5% 0.924 92.6% 7.4% 0.926
Oranges 212 351 80.9% 19.1% 0.808 8.7% 91.3% 0.087
Honeydew Melons 213 159 37.6% 62.4% 0.376 35.9% 64.1% 0.359
Broccoli 214 147 33.5% 66.5% 0.334 22.1% 77.9% 0.220
Avocados 215 2,436 99.9% 0.1% 0.999 96.9% 3.1% 0.969
Peppers 216 92 28.7% 71.3% 0.286 46.9% 53.1% 0.469
Pomegranates 217 1,765 93.1% 6.9% 0.931 95.9% 4.1% 0.959
Nectarines 218 26 65.0% 35.0% 0.650 38.8% 61.2% 0.388
Greens 219 642 51.2% 48.8% 0.512 40.8% 59.2% 0.408
Plums 220 250 64.1% 35.9% 0.641 62.0% 38.0% 0.620
Strawberries 221 4 13.3% 86.7% 0.133 33.3% 66.7% 0.333
Squash 222 1 3.4% 96.6% 0.034 5.9% 94.1% 0.059
Vetch 224 2 3.5% 96.5% 0.035 18.2% 81.8% 0.182
Dbl Crop WinWht/Corn 225 12,596 80.8% 19.2% 0.805 71.5% 28.5% 0.710
Dbl Crop Oats/Corn 226 1,846 66.2% 33.8% 0.661 56.4% 43.6% 0.563
Lettuce 227 304 40.8% 59.2% 0.407 25.4% 74.6% 0.254
Pumpkins 229 - 0.0% 100.0% 0.000 0.0% 100.0% 0.000
Dbl Crop WinWht/Sorghum 236 484 40.1% 59.9% 0.400 48.7% 51.3% 0.487
Dbl Crop Barley/Corn 237 - 0.0% 100.0% 0.000 0.0% 100.0% 0.000
Dbl Crop WinWht/Cotton 238 11 84.6% 15.4% 0.846 61.1% 38.9% 0.611
Blueberries 242 - 0.0% 100.0% 0.000 0.0% 100.0% 0.000
Cabbage 243 2 1.4% 98.6% 0.014 5.7% 94.3% 0.057
Cauliflower 244 44 15.7% 84.3% 0.157 59.5% 40.5% 0.594
Celery 245 - 0.0% 100.0% 0.000 0.0% 100.0% 0.000
Radishes 246 - n/a n/a n/a 0.0% 100.0% 0.000
*Correct Pixels represents the total number of independent validation pixels correctly identified in the error matrix.
**The Overall Accuracy represents only the FSA row crops and annual fruit and vegetables (codes 1-61, 66-80, 92 and 200-255).
FSA-sampled grass and pasture. Non-agricultural and NLCD-sampled categories (codes 62-65, 81-91 and 93-199) are not included in the Overall Accuracy.
The accuracy of the non-agricultural land cover classes within the Cropland Data Layer is entirely dependent upon the USGS, National Land Cover Database (NLCD 2011). Thus, the USDA, NASS recommends that users consider the NLCD for studies involving non-agricultural land cover. For more information on the accuracy of the NLCD please reference <https://www.mrlc.gov/>.
Attribute_Accuracy_Value:
Classification accuracy is generally 85% to 95% correct for the major crop-specific land cover categories. See the 'Attribute Accuracy Report' section of this metadata file for the detailed accuracy report.
Attribute_Accuracy_Explanation:
The strength and emphasis of the CDL is crop-specific land cover categories. The accuracy of the CDL non-agricultural land cover classes is entirely dependent upon the USGS, National Land Cover Database (NLCD 2011). Thus, the USDA, NASS recommends that users consider the NLCD for studies involving non-agricultural land cover.
These definitions of accuracy statistics were derived from the following book: Congalton, Russell G. and Kass Green. Assessing the Accuracy of Remotely Sensed Data: Principles and Practices. Boca Raton, Florida: CRC Press, Inc. 1999. The 'Producer's Accuracy' is calculated for each cover type in the ground truth and indicates the probability that a ground truth pixel will be correctly mapped (across all cover types) and measures 'errors of omission'. An 'Omission Error' occurs when a pixel is excluded from the category to which it belongs in the validation dataset. The 'User's Accuracy' indicates the probability that a pixel from the CDL classification actually matches the ground truth data and measures 'errors of commission'. The 'Commission Error' represent when a pixel is included in an incorrect category according to the validation data. It is important to take into consideration errors of omission and commission. For example, if you classify every pixel in a scene to 'wheat', then you have 100% Producer's Accuracy for the wheat category and 0% Omission Error. However, you would also have a very high error of commission as all other crop types would be included in the incorrect category. The 'Kappa' is a measure of agreement based on the difference between the actual agreement in the error matrix (i.e., the agreement between the remotely sensed classification and the reference data as indicated by the major diagonal) and the chance agreement which is indicated by the row and column totals. The 'Conditional Kappa Coefficient' is the agreement for an individual category within the entire error matrix.