Applied Introduction of Modeling of REDD Projects: A Strategy for Reduce the Impacts of Climate Change

Reducing Emissions from Deforestation and Forest Degradation (REDD) is a climate change
mitigation strategy employed to reduce the intensity of deforestation and GHGS emissions in
developing countries. In recent decades, drastic land-use changes in the Mazandaran province caused
a substantial reduction in the amount of Hyrcanian forest. The present research has been conducted
with the knowledge of the objectives of REDD projects. Accordingly, forest cover changes in a range
of Nowshahr and Noor in the Mazandaran province using Landsat satellite images (1984, 2000, and
2014) were examined. Then, transition potential modeling using the multi-layer perceptron artificial
neural network (MLP) was implemented and continuing the accuracy assessment of the model was
performed. In the end, according to the Voluntary Carbon Standard (VCS) methodology and the
calibration period 1984-2014, forest cover change was predicted for the 30 next years (2044) and CO2
emissions up to the 2044 was computed. The results showed that forest cover decreased by about
3413 ha during 1984-2014. Also, accuracy assessment indicates the good accuracy of the model with
ROC equal to 0/95. Finally, REDD baseline modeling results showed that 827591.5 tCO2e will be
released during 2014-2044 (the next 30 years), but REDD project implementation will prevent the
release of 584056.38 tCO2e. The findings of this research shows that using the methodology
presented, the rate of land-use changes, GHGS emissions and the impact of REDD projects to reduce
emissions can be predicted and can be used for the Project Design Document (PDD) of Clean
Development Mechanism (CDM) in the Iran.