TY - JOUR

T1 - A Modified KNN Method for Mapping the Leaf Area Index in Arid and Semi-Arid Areas of China

AU - Jiang, Fugen

AU - Smith, Andy

AU - Kutia, Mykola

AU - Wang, Guangxing

AU - Liu, Hua

AU - Sun, Hua

PY - 2020/6/10

Y1 - 2020/6/10

N2 - As an important vegetation canopy parameter, Leaf area index (LAI) plays a critical role in forest growth modeling and vegetation health assessment. Estimating LAI is helpful to understand vegetation growth and global ecological processes. Machine learning methods such as k-nearest neighbors (kNN) and random forest (RF) with remote sensing images have been widely used for mapping LAI. However, the accuracy of mapping LAI in arid and semi-arid areas using the methods is limited due to remote and large areas, high cost for collecting field data and great spatial variability of vegetation canopy. Here, a novel and modified kNN method was presented for mapping LAI in arid and semi-arid areas of China using Sentinel-2 and Landsat 8 images with field data collected in Ganzhou and Kangbao of China. The modified kNN was developed by integrating the tradition kNN estimation and RF classification. The results were compared with those from the kNN and RF regression alone using three sets of input predictors: (i) spectral reflectance bands (input 1); (ii) vegetation indices (input 2); and (iii) a combination of spectral reflectance bands and vegetation indices (input 3). Our analysis showed that in Ganzhou, the red-edge bands of the Sentinel-2 image had a high correlation with LAI. Using the red-edge band derived vegetation indices increased the accuracy of mapping LAI compared with using other spectral variables. Among the three sets of the input predictors, the input 3 resulted in the highest prediction accuracy. Based on the combination, the values of RMSE by the traditional kNN, RF and modified kNN were 0.526, 0.523 and 0.372, respectively, and the modified kNN significantly improved the accuracy of LAI prediction by 29.3% and 28.9% compared with the kNN and RF alone. The similar improvement took place for the input 1 and input 2. In Kangbao, the improvement of prediction accuracy by the modified kNN was 31.4% compared with both the kNN and RF. Thus, this study implied that the modified kNN provided the potential to improve the accuracy of mapping LAI in the arid and semi-arid regions using the images.

AB - As an important vegetation canopy parameter, Leaf area index (LAI) plays a critical role in forest growth modeling and vegetation health assessment. Estimating LAI is helpful to understand vegetation growth and global ecological processes. Machine learning methods such as k-nearest neighbors (kNN) and random forest (RF) with remote sensing images have been widely used for mapping LAI. However, the accuracy of mapping LAI in arid and semi-arid areas using the methods is limited due to remote and large areas, high cost for collecting field data and great spatial variability of vegetation canopy. Here, a novel and modified kNN method was presented for mapping LAI in arid and semi-arid areas of China using Sentinel-2 and Landsat 8 images with field data collected in Ganzhou and Kangbao of China. The modified kNN was developed by integrating the tradition kNN estimation and RF classification. The results were compared with those from the kNN and RF regression alone using three sets of input predictors: (i) spectral reflectance bands (input 1); (ii) vegetation indices (input 2); and (iii) a combination of spectral reflectance bands and vegetation indices (input 3). Our analysis showed that in Ganzhou, the red-edge bands of the Sentinel-2 image had a high correlation with LAI. Using the red-edge band derived vegetation indices increased the accuracy of mapping LAI compared with using other spectral variables. Among the three sets of the input predictors, the input 3 resulted in the highest prediction accuracy. Based on the combination, the values of RMSE by the traditional kNN, RF and modified kNN were 0.526, 0.523 and 0.372, respectively, and the modified kNN significantly improved the accuracy of LAI prediction by 29.3% and 28.9% compared with the kNN and RF alone. The similar improvement took place for the input 1 and input 2. In Kangbao, the improvement of prediction accuracy by the modified kNN was 31.4% compared with both the kNN and RF. Thus, this study implied that the modified kNN provided the potential to improve the accuracy of mapping LAI in the arid and semi-arid regions using the images.

KW - Leaf area index

KW - Medium resolution images

KW - Characteristic variable selection

KW - Modified kNN

KW - Dry regions

U2 - 10.3390/rs12111884

DO - 10.3390/rs12111884

M3 - Article

VL - 12

JO - Remote Sensing

JF - Remote Sensing

SN - 2072-4292

IS - 11

M1 - 1884

ER -