Note: this description is a work in progress developed by the collaborating entities in a workshop. If you would like to contribute reach out to office@space4water.org, or your trusted Space4Water point of contact.

Required Software

• Google Earth Engine

• Google Earth Engine Apps - Global Forest Change

   

1. Data collection

• To collect historic and high-resolution up-to-date imagery over the area, UNOOSA contacted the Land and Information New Zealand Data Service, which provided both historical aerial imagery and LIDAR data sources.
• Historic data for the relevant land patch can be accessed via the Retrolens New Zealand Service.
• Up-to-date aerial photos of the area can be accessed here at the New Zealand Data Service. Tile 503 and 603 are the ones of interest.
• Relevant Landsat data are available from 1989. For the study area, Landsat 7 data is available from 2 July 1999, and Landsat 4 from 2 February 1989.
• Global Forest Change data can be retrieved from Google Earth Engine Apps
• HydroSHEDS: The core data products of HydroSHEDS are a series of gridded datasets designed for use in hydro-environmental model development and custom GIS applications. Data layers include the original digital elevation model (DEM) that underpins HydroSHEDS, a hydrologically conditioned version of the DEM, the derived flow direction and flow accumulation grids, as well as land mask and sink grids. These data products form the digital foundation of the derived secondary data products. HydroSHEDS core data products are currently available for HydroSHEDS v1 only, which is mostly based on SRTM elevation data. HydroSHEDS v2, which is derived from TanDEM-X elevation data, is currently under development and is scheduled for release in 2022.
• A digital elevation model (DEM) is available at 30m resolution by Copernicus is available at the Terrascope website.

2. Mapping the historical land use and land cover surrounding the river (in progress)

• Using Google Earth Engine Apps - Global Forest Change data it is possible to identify recent deforestation (2016) upstream and near the Ngutunui region;
• No other change in the land use is observed upstream the Ngutunui region between 1985-1999.
• According to the vegetation cover analysis, there have not been many changes over the past 20+ years. It has been observed that the Manori community and the surrounding area have maintained almost the same vegetation cover, however some patches adjacent to the community boundary downstream have caused some distractions.
• Limitation- In the case of a small land mass and narrow river, limits many satellite-based analyses.

Other methods - Conducting a community survey

• To obtain historical knowledge on the identification of vegetation, tree species, a community survey appears to be the only option available, since the challenge requires data extending back 50 years. While space-based data (aerial photos) are available, the possibility of identifying each species of tree is very limited, because of the canopy layer, understory plant species cannot be seen.
• This approach will enable to gather data on dominant plant species, their abundance, tree diameters, and the boundaries between different vegetation.
• Data obtained from the community survey provide a valuable historical record of vegetation patterns over the decades and help identify any changes or disruptions.
• "i-nature" app- tree species can be identified by taking a simple picture of a leaf. The app then provides a detailed description of the identified tree species, including information about its characteristics and habitat.

Further information on vegetation identification

Using NDVI allows for identifying the type of vegetation but not the specific species. One can see whether the type of vegetation has changed from trees to grassland, but specific plants cannot be seen.

Retrolense provides aerial photographs taken from an aeroplane at which the relevant bands for NDVI calculation (infrared and red) are missing.

We can examine vegetation cover over the last 30+ years using NDVI with Landsat data.

A study called Aerial photography for assessing vegetation change: A Review of applications and the relevance of findings for Australian vegetation history by Fensham and Fairfax published in 2022 in the Australian Journal of Botany and on the CSIRO page is accessible here.