Investigating Methods In Remote Sensing For Invasive Plant Species - - PowerPoint PPT Presentation

Investigating Methods In Remote Sensing For Invasive Plant Species Identification

FRST 443

Alyssia Law, Bryana Ginther, Joyce Chan, Agatha Czekajlo

Invasive Species: Who Cares?

• 1st largest threat to biodiversity: habitat loss
• 2nd largest threat to biodiversity: INVASIVE SPECIES
• Other factors negatively impacted by invasive species:

○ Economy ○ Primary productivity ○ Regional Hydrology ○ ….

Himalayan Blackberry. Retrieved from: Invasive Species Council of BC English holly. Personal photo

Why Use Remote Sensing to Identify Invasive Species?

• Invasive species are incredibly

widespread

• Detailed field surveying = $$$!
• Remote sensing can delineate

invasive species patches

• Areas most at risk can then be

identified and removal teams can be sent in

Outline

• 1. Current technologies and methodologies used to identify

invasive species

• 2. Remote sensing case studies of invasive species present in

British Columbia

• 3. Future direction of research

How are invasive species identified?

• 1. Multispectral Imaging (Landsat)

Advantages

• Accessible, cost effective
• Successful in mapping large dense regions of invasive species

Disadvantages

• Less spectral bands, often lower spatial resolution
• Unsuccessful at mapping less dense regions of invasive species

How are invasive species identified?

• 2. Hyperspectral Imagery (Examples: AVIRIS, SPOT)

Advantages

• High spatial resolution
• Hundreds of spectral bands
• Phenological differences

1. Varying chlorophyll 2. Water absorption 3. Nitrogen levels Disadvantages

• Background knowledge on habitat required

(Asner and Vitousek, 2005)

How are invasive species identified?

• 3. LiDAR

Advantages

• Detects intermediate layers
• Maps canopy structure
• Orographic predictors
• Forest height strata

Disadvantages

• Extensive calibration
• Difficult to distinguish species

(Singh et al., 2015)

Classification by Computer Algorithm Models

• Machine-learning classification algorithms:

○ Random Forest ○ Support Vector Machine (SVM)

• Regression models
• Weed Invasions Susceptibility Prediction

(WISP)

○

Uses GIS to determine areas, and the degree of, invasion susceptibility

Hunt et al. 2007

1. Focus: Detection of leafy spurge (Euphorbia esula) among other vegetation

a.

Yellow-green flower bracts

b.

Flowers in early June to mid-July 2. Methods

a.

20 plots with varying levels of leafy spurge

b.

Sensors: AVIRIS, Landsat 7 ETM+, SPOT 4

Leafy spurge. Retrieved from https://gobotany.newenglandwild.org/species/ euphorbia/esula/

Hunt et al. 2007

1.

Classification

a.

Spectral Angle Mapper classification

2. Results

a. Hyperspectral AVIRIS (48 bands) i. Overall accuracy = 74% b. Multispectral AVIRIS (7 bands in visible + near infrared) i. Overall accuracy = 56% c. Landsat 7 ETM+ (bands 1-5, 7) and SPOT 4 (bands 1-4) i. Overall accuracy insignificant

Hunt et al. 2007

1. Using only visible and near infrared bands

a. Landsat 7 ETM+ (bands 1-4) i. Overall accuracy = 59% b. SPOT 4 (bands 1-3) i. Overall accuracy = 61%

2. Takeaways from the study

a. Higher spectral resolution images = higher accuracy b. Bands in visible/near infrared wavelengths = higher accuracy

Hill et al. 2016

1. Focus: Mapping of Scotch broom (Cytisus scoparius) using Landsat

a. Yellow flowers b. Flowering in June - July

2. Methods a. Sensor: Landsat Thematic Mapper b. 5 images over 4 years c. Aerial observation to find areas of known concentration i. Dense (≥ 75% cover) ii. Moderate (25 - 75% cover) iii. Low (10 - 25% cover)

Hill et al. 2016

1. Results

a. Around 80% (moderate or dense patches) b. Results unreliable for low density c. Often entirely missed patches that were < 0.5 ha

Hill et al. 2016

1. Takeaways from the study a. Landsat is useful for mapping Scotch broom during high flowering 2. Issues a. Low density patches (early detection)

Chance et al., 2016

• Objective: detect invasions of Himalayan blackberry and English ivy in urban

ecosystems (Surrey, BC).

Chance et al., 2016.

RESULTS

• Himalayan Blackberry accuracy: 77.8

to 87.8%

• English Ivy accuracy: 81.9 to 82.1 %
• Presence of species decreased with

distance from roads.

Chance et al., 2016

Chance et al., 2016.

• Increase time scale

○ Multiple growth/flowering seasons ○ Over the year across seasons

• More collaborations between ecologists and RS scientists

○ Ecologists → identify invader-specific plant traits and abiotic tolerances ○ RS Scientists → develop robust RS predictors to identify ecological & environmental markers for detection

Future direction

Future direction

• Co-development of sensors with ecophysiology ground

measurements

○

• Ex. active imaging with LiDAR or RADAR, fine resolution

hyperspectral and hyperspatial tech

• ‘Trait products’ on future satellites with spectrometer–LiDAR systems for

invasive monitoring

○

• Ex. Imaging spectroscopy of chemical and morphological traits across broad regions

Summary of Main Points

• Best RS method of identifying invasive species: combine

LiDAR and Hyperspectral imagery

• Find unique features to detect, such as flower colour
• Ecologists and RS scientists should co-develop RS technology

Literature Cited

Andrew, M. E., & Ustin, S. L. (2009). Habitat suitability modelling of an invasive plant with advanced remote sensing data. Diversity & Distributions, 15(4), 627-640. doi:10.1111/j.1472-4642.2009.00568.x Asner G. P., Knapp D. E., Kennedy-Bowdoin T., Jones M. O., Martin R. E., Boardman J., et al. (2008). Invasive species detection in Hawaiian rainforests using airborne imaging spectroscopy and

• LiDAR. Remote Sens. Environ. 112 1942–1955. 10.1016/j.rse.2007.11.016

Asner G. P., Vitousek P. M. (2005). Remote analysis of biological invasion and biogeochemical change. Proc. Natl. Acad. Sci. U.S.A. 102 4383–4386. 10.1073/pnas.0500823102 Chance, C. M., Coops, N. C., Plowright, A. A., Tooke, T. R., Christen, A., & Aven, N. (2016). Invasive shrub mapping in an urban environment from hyperspectral and LiDAR-derived attributes. Frontiers in Plant Science, 7 doi:10.3389/fpls.2016.01528 Große-Stoltenberg, A., Hellmann, C., Werner, C., Oldeland, J., & Thiele, J. (2016). Evaluation of continuous VNIR-SWIR spectra versus narrowband hyperspectral indices to discriminate the invasive acacia longifolia within a mediterranean dune ecosystem. Remote Sensing, 8(4), 334. doi:10.3390/rs8040334 Großkinsky, D. K., Svensgaard, J., Christensen, S., & Roitsch, T. (2015). Plant phenomics and the need for physiological phenotyping across scales to narrow the genotype-to-phenotype knowledge

• gap. Journal of Experimental Botany, 66(18), 5429-5440. doi:10.1093/jxb/erv345

Hauglin, M., & Ørka, H. (2016). Discriminating between native norway spruce and invasive sitka Spruce—A comparison of multitemporal landsat 8 imagery, aerial images and airborne laser scanner data. Remote Sensing, 8(5), 363. doi:10.3390/rs8050363 Hill, D. A., Prasad, R., & Leckie, D. G. (2016). Mapping of scotch broom (cytisus scoparius) with landsat imagery. Weed Technology, 30(2), 539-558 Homolova, L., Malenovsky, Z., Clevers, J. G. P. W., Garcia-Santos, G., & Schaepman, M. E. (2013). Review of optical-based remote sensing for plant trait mapping. Ecological Complexity, 15, 1-16. doi:10.1016/j.ecocom.2013.06.003 Hunt Jr, E.R., Daughtry, C.S.T., Kim, M.S., Williams, A.E.P. (2007). Using canopy reflectance models and spectral angles to assess potential of remote sensing to detect invasive weeds. Journal of Applied Remote Sensing, Vol. 1, 013506 Hunt Jr, E.R., Gillham, J. H., & Daughtry, C. S. (2010). Improving potential geographic distribution models for invasive plants by remote sensing. Rangeland ecology & management, 63(5), 505-513. Jones, J.P.G. (2011). Monitoring species abundance and distribution at the landscape scale. Journal of Applied Ecology, 48(1), 9-13. doi:10.1111/j.1365-2664.2010.01917.x Niphadkar, M., & Nagendra, H. (2016). Remote sensing of invasive plants: Incorporating functional traits into the picture. International Journal of Remote Sensing, 37(13), 3074-3085. doi:10.1080/01431161.2016.1193795 Simpson, A., Jarnevich, C., Madsen, J., Westbrooks, R., Fournier, C., Mehrhoff, L., . . . Sellers, E. (2009). Invasive species information networks: Collaboration at multiple scales for prevention, early detection, and rapid response to invasive alien species. Biodiversity, 10(2-3), 5-13. doi:10.1080/14888386.2009.9712839 Singh, A., Serbin, S. P., McNeil, B. E., Kingdon, C. C., & Townsend, P. A. (2015). Imaging spectroscopy algorithms for mapping canopy foliar chemical and morphological traits and their uncertainties. Ecological Applications, 25(8), 2180-2197. doi:10.1890/14-2098.1