Canada

Southern Alberta Rangelands

Team leader: Dr Anne M. Smith

Agriculture and Agri-Food Canada (AAFC)

A new article has just been published in the Canadian Journal of Remote Sensing (http://www.tandfonline.com/toc/ujrs20/current#.VZ_bX_P4Abw)

Anne M. Smith, Michael J. Hill & Yongqin Zhang (2015) Estimating Ground Cover in the Mixed Prairie Grassland of Southern Alberta Using Vegetation Indices Related to Physiological Function, Canadian Journal of Remote Sensing: Journal canadien de télédétection, 41:1, 51-66

Project Overview

The native grasslands of western Canada support the multi-billion dollar cattle industry as well as provide habitat and forage for wildlife and a recreational resource for the Canadian public. These vast resources are a source of faunal and floral biodiversity as well as being important in local and regional carbon and water fluxes. Thus, management and productivity of the native grasslands play an important role in Canadian economics and ecology. Management of native grasslands not only requires sustained forage production for the cattle industry but also forage production and habitat for wildlife (animals, birds and insects). Adequate vegetation cover, both photosynthetic and non-photosynthetic, prevents land erosion by minimizing soil exposure and potential erosion by wind and water.

Obtaining spatial information on the productivity and health of these extensive and often inaccessible native grasslands is time consuming and labor intensive. However, the Alberta provincial government has developed a comprehensive Grassland Vegetation Inventory as a digital polygon dataset that identifies native prairie grassland, as well as tame pasture and cropland areas in southern Alberta. This along with important datasets describing grassland composition and grassland biomass from cages and exclosures acquired regularly by provincial agencies makes the southern Alberta rangelands an ideal location for development of tools and methods to support GEOGLAM RAPP goals.

This project aims to:

  1. Evaluate multispectral, hyperspectral and radar technologies and develop methods of quantifying the spatial extent and fragmentation of grassland.
  2. Develop a methodology to update the Grassland Vegetation Inventory such that changes in the native grasslands resource as a result of climatic or anthropogenic activities can be quantified in a reliable and timely fashion.
  3. Develop methods to estimate grassland growth rate and biomass production across southern Alberta using an integration of multiple spatial and temporal scales of remote sensing data with climatic variables and modeling.

Collaboration and Stakeholder Involvement

The project has a very strong collaborative environment within which to operate. The project has been framed with long standing and regular consultation with agencies and producers. An organizational diagram (below) shows the nature of this collaboration. This provides a firm connection to conservation and production needs in the region.

Canada organizational diagram

Organizational diagram of agencies and groups involved in research, outreach and utilization of outputs.

Implementation Plans

The implementation is based on legacy project that explored some of the potential for remote sensing application, and three current projects.

The legacy project provided:

    1. A method of estimating grassland change both “wholesale” and as a result of fragmentation and quantifying losses of native grasslands using Landsat and SAR imagery.
    2. A scheme to derive aspects of grassland productivity that could be used in an operational system to determine effects of climate change and grazing based on some pilot studies.
    3. A remote sensing methods to assess the impact and spread of invasive species, enable directed release of biological control agents and a means of estimating regional scale success of control measures. A follow on project focused on leafy spurge and detection with airborne hyperspectral, radar and multispectral remote sensing with a view to developing an operational method.

The three current projects which commenced on July 1 2013 individually address items A, B and C from the legacy work and greatly expand the scope of items A and B.

Project: Updating the Grassland Vegetation Inventory using Satellite Earth Observation Systems

Deliverable: Methodologies to update the GVI using satellite remote sensing and potentially provide a framework for developing similar inventories in other regions of Canada.

The Grassland Vegetation Inventory developed by Alberta Environment and Sustainable Resource Development (AESRD), from 2006 to 2014, provides a standardized tool for stratifying the landscape and providing a common basis for a variety of activities including best management practices for the livestock industry. A critical business requirement for all users of GVI is updating of the baseline inventory at 5 to 10 year intervals to determine changes in the landscape as a result of surface disturbance, land use practices and climatic factors.

Project: Earth Observation Tools for Evaluating Climatic and Anthropogenic Impacts on Native Grassland Phenology and Production

Deliverable: A comprehensive grassland monitoring system that can be used by land managers, government agencies and NGOs in management and policy decisions and strategies.

This project examines the ability to estimate grassland growth rate and biomass production using an integration of multiple spatial and temporal scales of remote sensing data with climatic variables. Grassland growth rate and annual productivity will be estimated using a simple light use efficiency model that is driven by remote sensing data in conjunction with climatic and soil variables. Historic time series of currently available imagery, from MODIS at moderate resolution (250-1000 m), and from Landsat TM at 30 m resolution, will be used to capture the fraction of photosynthetic radiation absorbed by plants, and to develop empirical biomass relationships. The validity of the predicted growth rate and productivity estimates will be assessed with reference to ground data provided by a number of reference sites by collaborating agencies. As the productivity of grasslands is constrained by the soil type, soil fertility, grassland condition and phenological parameters the influence some key seasonal and management information will be evaluated.

Project: Mapping Invasive Species Impacts on the Preservation of Native Grassland Biodiversity

Deliverable: A cost effective operational monitoring system to evaluate shifts in grassland biodiversity due to invasive plant species.

This project will take results from the legacy project and develop an operational mapping system for leafy spurge. In the prior study, processing of the airborne and satellite optical data using first order derivative analysis highlighted areas of the spectrum in which leafy spurge patches of differential density and vegetation types could be discriminated. With respect to radar technologies, the ability to discriminate vegetation types using Freeman-Durden and Cloude-Pottier decomposition parameters as well as the coherence matrix elements derived from Radarsat-2 imagery was also tested. The results showed coherency elements derived from multi-date, multi-incidence angle images offered the potential Radarsat-2 imagery to delineate leafy spurge. . Additional field data will now be collected. In collaboration with AESB, field sites will be selected in Saskatchewan or Manitoba on community pastures. AESB already has extensive knowledge and documentation of leafy spurge infestations on some of these pastures. Operational satellites systems such the already launched RapidEye, or Sentinel-2 scheduled for launch in 2012 and potentially EnMap scheduled for launch in late 2013 and Radarsat-2 will be targeted for data acquisition. The processing system will be streamlined with the assistance of ATIC/University of Lethbridge who are already developing web-based automated tools involving entire data processing chains and intelligent information production systems that support the ingestion of satellite remote sensing data.

Site Description

The project is focused on the extensive native grassland areas of southern Alberta, a western province of Canada (shown below). The spatial extent is best described by the boundary of the grassland ecoregions in Alberta (1b below). These support a major cattle industry as well as many conservation values.

Alberta grasslands.

Alberta grasslands

In Situ Observations

Spatial databases of native grassland extent

  • Agriculture and Agri-Food Canada grassland mask
  • Agriculture Financial Services Corporation grassland mask
  • Alberta Native Prairie Baseline (~1993)
  • Alberta Grassland Vegetation Inventory (2006-2014)
  • Alberta Primary Land Vegetation Inventory (2014-?)

Field campaigns in 2008-2010

  • Fractional cover measurements (sorted to green grass, green forbs, dead, litter), hemispherical photos, nadir quadrat photos, biomass cuts, biomass and cover scores, spectral libraries (400 – 2500 nm) along five 100m transects at each of four field sites.

Agency Databases

National 10 km Climate grid

Alberta Sustainable Resource Development Rangeland Reference Area Monitoring Data.

EO Data Requirements

  • EO-1 Hyperion
  • Landsat 5–TM
  • Landsat-7 +ETM
  • MODIS MCD43A4 reflectance, MOD13A1 vegetation indices
  • Radarsat 1, Radarsat 2
  • Landsat-8
  • VIIRS reflectance and products when available
  • Sentinel-2
  • Sentinel-3

 Location

Project Reports

Final report from the legacy Canadian Space Agency project.