WPE01 Assessing the value added of NEON for using machine learning to quantify vegetation mosaics and woody plant encroachment at Konza Prairie

Summary

Short name: 

WPE01

Abstract: 

Woody encroachment, or invasion of woody plants, is rapidly shifting tallgrass prairie into shrub and evergreen dominated ecosystems, mainly due to exclusion of fire. Tracking the pace and extent of woody encroachment is difficult because shrubs and small trees are much smaller than the coarse resolution (>10m2) of common remote sensed images. However, the US government has been investing in finer resolution (<2m2) remote sensing through USDA NAIP and the National Ecological Observatory Network (NEON), both of which cost multi-million dollars each year and contain different remote sensed products. We compared two methods of classification (random forests and support vector machines) with these two freely available remotely sensed aerial images to determine if and how much NEON adds to classification accuracy and determine which method of machine learning was more accurate. All models have very high overall classification accuracy (>91%), with the NEON image a few percent more accurate than NAIP. The NEON image significantly relies on canopy height (LiDAR) to make classifications, but the importance of bands is more evenly distributed during NAIP classification. Lastly, accuracy for Eastern Red Cedar specifically is high with NEON (78-84%), compared to the relatively low classification accuracy using NAIP imagery (55-61%).

Published on EDI/LTER Data Portal

Citation Suggestion: 

Noble, B., Ratajczak, Z. . 2022. WPE01 Assessing the value added of NEON for using machine learning to quantify vegetation mosaics and woody plant encroachment at Konza Prairie . Environmental Data Initiative. http://dx.doi.org/10.6073/pasta/a7b40e41080460bb1123dcc7b6d4d942.

Data set ID: 

167

EML revision ID: 

1

Data sources: 

Variables: 

DataCode
  • Label:
  • Definition: Dataset code
  • Type: Nominal
  • Missing values: None specified
RecType
  • Label:
  • Definition: dataset record type
  • Type: Nominal
  • Missing values: None specified
Species_classification
  • Label:
  • Definition: classification of species
  • Type: Nominal
  • Missing values: None specified
X2019_NAIP_hc_konza1
  • Label:
  • Definition: Value of near-infrared pixel in rs_stack.1
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
X2019_NAIP_hn_konza1
  • Label:
  • Definition: Redness of each pixel in rs_stack.2
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
X2019_NAIP_hn_konza2
  • Label:
  • Definition: Greenness of each pixel in rs_stack.3
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
X2019_NAIP_hn_konza3
  • Label:
  • Definition: Blueness of each pixel in rs_stack.4
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
evi_2020_neon_mosaic
  • Label:
  • Definition: Similar to NDVI, estimates vegetation greenness and biomass in rs_stack.5
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
ndni_mosaic
  • Label:
  • Definition: Relative nitrogen concentration in canopy in rs_stack.6
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
ndli_2020_neon_mosaic
  • Label:
  • Definition: Uses shortwave IR to estimate water content in canopy in rs_stack.7
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
savi_mosaic
  • Label:
  • Definition: Reduces soil brightness in areas where vegetation cover is low - rs_stack.8
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
arvi_2020_neon_mosaic
  • Label:
  • Definition: Reduces atmospheric noise from dust, smoke, rain, etc. in rs_stack.9
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
ndvi_2020_neon_mosaic
  • Label:
  • Definition: Calculated from NIR and red bands, indicates live green vegetation density in rs_stack.10
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
can_height_2020
  • Label:
  • Definition: Height of canopy above bare earth in rs_stack.11
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
layer1
  • Label:
  • Definition: Avg. values of near-infrared of surrounding pixels - rs_stack.12
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
layer2
  • Label:
  • Definition: Avg. redness of surrounding pixels in rs_stack.13
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
layer3
  • Label:
  • Definition: Avg. greenness of surrounding pixels in rs_stack.14
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
layer4
  • Label:
  • Definition: Avg. blueness of surrounding pixels in rs_stack.15
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
layer5
  • Label:
  • Definition: Avg. NDVI values of surrounding pixels in rs_stack.16
  • Type: Physical quantity
  • Unit: dimensionless
  • Maximum: Not specified
  • Minimum: Not specified
  • Precision: Not specified
  • Missing values: None specified
point
  • Label:
  • Definition: Training points vs evaluation points
  • Type: Nominal
  • Missing values: None specified
Comments
  • Label:
  • Definition: Comments
  • Type: Nominal
  • Missing values: None specified

Categories
Dates

Date Range: 

Tuesday, June 1, 2021 to Friday, December 30, 2022

Publication Date: 

Friday, December 2, 2022
People

Owner/Creator: 

Contact: 

Downloads