Investigation of affective habitat factors affecting on abundance of wood macrofungi and sensitivity analysis using the artificial neural network (case study: Kheyrud forest, Noshahr)

One advantage of artificial neural networks is application in the management and planning of natural ecosystems. Considering the high biological diversity of northern forests of Iran, it is necessary to know forests ecosystems. Thus, using artificial neural networks is important for modeling and forecasting of presence and abundance of wood macrofungi in forests. Fungi samples were collected in summer and fall seasons on oak and hornbeam trees from compartments 110, 207 and 311 in educational and research forest of the University of Tehran. Totally 231 samples of macrofungi were collected that 112 samples of them belong to wood decay fungi. Results showed that the designed artificial neural network, has suitable potential for modeling of abundance of wood fungi. Network with two hidden layers and 11 neurons in each layer with the highest coefficient of determination, show the best performance of topology optimization. The number of inputs and outputs equal to 112 samples with 11 variables including density class 4 or class of fungi. Sensitivity analysis showed the decay stage of tree, tree health status and its condition and stand microclimate have the most effect on presence and abundance of wood macrofungi.