Three steps to success in Satellite Image Classification, Method of Project Manondroala Humid forest mapping

IMG_7228_2

Participants to the intensive training on Manondroala’s forest mapping method at IOGA University of Antananarivo, 02.04th.2016

The training was a veritable success! About forty participants from various governmental and non-governmental organizations attended with enthusiasm the workshop on sharing the Manondroala method of humid forest mapping – organized by the Finnish Association for nature Conservation – at the Institute & Observatory of Geophysics Antananarivo (IOGA) University of Antananarivo, Madagascar- last February 04th, 2016.

It is rare that project shares for free, transparently and in details their methods to the network  of conservationists.

says one of participant.

The number of participants was more than expected by organizers. In fact, a very big interests of the conservationists network on Manondroala method and results were noticed during the workshop.

What is the goal?

The Finnish Association for Nature Conservation (FANC, Finland) has partnered with e.g. Transparent World (Russia), Institute & Observatory of Geophysics Antananarivo (IOGA, Madagascar) and Association Mitsinjo (Andasibe, Madagascar) in order to create a network of conservationists and forest administrators who seek solutions to both local and national conservation challenges.

The final aim of the project is to create a new, locally applicable forest monitoring method that can be used in community forest management as well as in research.

So, after four years of field work and GIS analyses, after presenting the first results of the work during the Seminar held in Antananarivo on 2014 (See seminar presentation here) – Manondroala is sharing the humid forest mapping method – developed together with national and international (Transparent World Russia) GIS experts. (See training method presentation here!)

The Manondroala  method was used to produce with success the recently started map-based web portal of Madagascar.

It is still an early version. However, it already plays an important role – putting together and combining the spatial data from various sources – including our forest maps for a number of areas, mainly for the moist forests of Madagascar.

says Mr Dmitry Aksenov, Director of Transparent World, a Russian NGO

Manondroala team was proud to present this achievement during the training on sharing method to GIS persons and forest administrators of Institute & network of conservationists.

What is the Manondroala method?

First of all, before processing the Satellite Image Classification, it is important to know how to measure the level of forest intactness and degradation. Following is Manondroala’s way to translate – local perception of forest degradation/regeneration classes, using dialect from the ground – to the scientific forest classes observed from space.

from ground to space

This discovery is significant and raises interests from various conservation organizations.

Secondly, based on forest degradation classes, the classification scheme is developed as following!classification scheme

Finally, the Satellite Image Classification method is processed. It is in three steps:

  • Step 1. Data processing using GRASS GIS
  • Step 2. Density of shadow calculation and sample creation using QGIS
  • Step 3. Classification of Image using R

random forest algorithm

The method of classification of satellite image was nicely detailed in the Manual on Humid Forest Method of Project Manondroala, especially created for the training and shared to all participants.

Note that tools are raster data (SPOT Imagery) processed on free software such as Quantum GIS 2.6.0, Grass GIS 7.1 and R 3.2.0.

Manondroala team was satisfy of the training and hopes to contribute on the improvement of networking between conservationists! However, for the sake of permanent improvement, feed-back is always welcome!

Next challenge of Manondroala is to develop and share the mapping method of non-humid forests of Madagascar.

Special thanks to all participants and to IOGA, the host of the training.

***

Misaotra – Kiitos – Spaciba – Thank you!

Text and picture : Angela Tarimy

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: