Evaluation of neural network and regression models to predict species diversity using some soil and physiographic factors (Case Study: Kharabeh Sanji watershed of Urmia)
Subject Areas : forestBehnam Bahrami 1 , Ardovan Ghorbani 2
1 - دانشجوی دکتری |علوم مرتع، گروه مرتع و آبخیزداری، دانشگاه محقق اردبیلی، اردبیل، ایران
2 - استادیار مرتعداری| گروه مرتع و آبخیزداری، دانشگاه محقق اردبیلی، اردبیل، ایران
Keywords: Rangelands, Environmental factors, Vegetation, modelling,
Abstract :
Direct measurement of species diversity is a time consuming and cost effective and somewhat unreliable because of errors in the sampling. This study was conducted by the aim of determining low cost factors for predicting species diversity using artificial neural network, adaptive- fuzzy neural network and regression models. Sampling was conducted using randomized-systematic method from 60 plots along 6 transects with 100m long and from 0-30cm of soil depth. Vegetation data were recorded to calculate species diversity by Shannon-wiener index. Moreover, for determining the affective factors on species diversity, electrical conductivity, pH, bulk density, percentages of organic matter, clay, silt, wet saturation, coarse and fine aggregates and slope and elevation were measured and determined. Then species diversity was determined using multii-layer perceptron neural network, adaptive-fuzzy neural network and regression models. The results show that criteria such as root mean squire error and efficiency coefficient of the regression model were 0.14 and 0.39, in artificial neural network 0.07 and 0.86 and for adaptive- fuzzy neural network 0.09 and 0.7, respectively. that Shannon wiener index was 1.98 for the study area. The artificial neural network model as a powerful tool in predicting species diversity in comparison with the multiple linear regression analysis and neural network-fuzzy adaptive models showed reliable results.
1-Allison, L.E., 1975. Organic carbon. Methods of Soil Analysis, Chemical and Microbiological Properties. American Society of Agronomy, Madison, p. 1367.
2-Bazartseren, B., & G. Hildebrandt, K., Holz, 2003. Short-term water level prediction using neural networks and neuro-fuzzy approach. Neuro computing. 55: 439-450.
3-Binkley, D., & Fisher, R., 2012. Ecology and management of forest soils. 4th .Wiley-Blackwell.
4-Blake, G.R., & K.H., Hartge, 1986. Bulk density Methods of Soil Analysis, Physical & Mineralogical Methods. Soil Science Society of America,9(1): 361-376.
5-Bouyoucos, G.J., 1962. Hydrometer method improved for making particle size analysis of soils. Agronomy Journal, 56: 464-465.
6-Browman, H.I., M.C., Philippe, H., Ray, J., Simon, K.L., Heike, M.M., Pamela, & M., Steven, 2004. Ecosystem-based Management. Marine Ecology Progress Series, 274: 269-303.
7-Cambardella, C.A., & E.T., Elliott, 1992. Particulate Soil Organic Matter Changes across a Grassland Cultivation Sequence. American Journal of Soil Science, 56: 777-783.
8-Caudill, M., 1987. Neural networks primer: Part I, AI Expert.
9-Fahimipour, E., M.A., Zare Chahouki, & A. Tavili, 2010. Study of some index species – environmental factors relationships in mid Taleghan rangelands. Rangeland, 4(1):23-32.
10-Fajry, A., 2009. Feasibility studies- Executive vegetation survey the wreckage of Urmia. University Department of Natural Resources.
11-Fridley, J.D., 2001. The influence of species diversity on ecosystem productivity: how, where, and why? Oikos, 93: 514-526.
12-Ghahsare ardestani, A., M., Basiri, M., Tarkesh, & M., Borhani, 2010. Models of the distribution and diversity of species diversity and environmental factors in four rangeland Isfahan Hill. Rangeland and Watershed, 63(3): 378-397.
13-Hernandez, R., P., Koohafkan, & J., Antoine, 2004. Assessing Carbon Stocks and modeling win-win Scenarios of carbon sequestration throughland-use change, 166 pp.
14-Holmberg, M., M. Forsius, M., Starr, & M., Huttunen, 2006. An application of artificial neural networks to carbon, nitrogen & phosphorus concentration in three boreal streams & impacts of climate change. International Society for Ecological Information 3rd Conference. Grottaferrata, Roma, 195: 51-60.
15-Ingleby, H.R., & T.G., Crowe, 2001. Neural network models for predicting organic matter content in Saskatchewan soils. Canadian Bios stems Engineering, 43: 71-75.
16-Kaya, Z., & J., Raynal, 2006. Biodiversity and conservation of Turkish forest. Biological conservation, 97: 131-141.
17-Lavorel, S., & E., Garnier, 2002. Predicting changes in community composition and ecosystem functioning from plant traits: revisiting the Holy Grail. Functional Ecology, 16: 545-556.
18-Leij, F., M.G., Schaap, & L.M., Arya, 2002. Water retention and storage: Indirect methods. Methods of Soil Analysis, 4(2): 1009-1045.
19-Magurran, A.E., 1988. Ecological Diversity and its Measurement. Princeton University Press, Princeton, NJ, 179pp.
20-Manly, B.F.J., 1994. Multivariate Statistical Methods: A Primer. London: Chapman & Hall.
21-Menhaj, M.B., 1998. Fundamentals of neuralnetworks. First Edition, Professor Hesabi Publishers, 502 pp.
22-Mesdaghi, M., 2005. Plant Ecology. Publication of Jahade Daneshgahi, 187p.
23-Minasny, B., A.B., McBratney, & K.L., Bristow, 1999. Comparison of different approaches to the development of pedotransfer functions for water retention curves. Geoderma, vol. 93, pp. 225-53.
24-Navabian, M., A.M., Liaghat, & M., Homaee, 2007. Comparison of transfer functions of artificial neural network and regression in estimating the saturated hydraulic conductivity. Proceedings of the Tenth Soil Science Congress of Iran, Karaj, 967-969.
25-Noor Alhamad, M., 2006. Ecological & species diversityof arid Mediterranean grazing land vegetation. Journal of Arid Environments, vol. 66, pp. 698-715.
26-Parasurman, K., A., Elshorbagy, & B., Si, 2006. Estimating saturated hydraulic conductivity in spatially variable fields using neural network in Ensembles. Soil Science Society American Journal, 70: 1851-1859.
27-Parsafar, N.A., & S., Marofi, 2011. Estimated temperatures at depths using network neural networks-Fuzzy (Case Study: Kermanshah region). Journal of Soil and Water Science, 21(3): 21-22.
28-Pietrasiak, N., J.R., Johansen, T., LaDoux, & R.C., Graham, 2011. Comparison of Disturbance Impacts to and Spatial Distribution of Biological Soil Crusts in the Little San Bernardino Mountains of Joshua Tree National Park, California. Western North American Naturalist, 74(4): 539-552.
29-Pilevari, A., Sh., Auobi., & H., Khademi, 2010. Comparison of artificial neural network and multiple linear regression analysis to predict soil organic carbon data to the ground. Journal of Soil and Water, 24(6): 1151-1163.
30-Sabziparvar, A. & M., Beiatorkeshi, 2010. Assess the accuracy of fuzzy artificial neural network, neurotropic solar radiation simulation. Iranian Journal of Physics Research, 4(1): 347-536.
31-Schaap, M.G., 1998. Using neural network to predict soil water retention and soil hydraulic conductivily. Soil Till Research, 47: 37-42.
32-Suding, K.N., &, L.J., Goldstein, 2008. Testing the Holy Grail framework: using functional traits to predict ecosystem change. New Philologist, 180: 559-562.
33-Zarechahouki, M.A., M., Jafari & H. Arzani, 2007. The relationship between environmental factors and species diversity in grasslands Poshtkuh Yazd. Pajouhesh & Sazandegi, 21(1): 192-199.
34-Zhang, C.B., J., Wang, W., Liu, S., Zhu, D., Liu, S., Chang, J., Chang, & Y., Ge, 2010. Effects of plant diversity on nutrient retention and enzyme activities in a full-scale constructed wetland. Bioresource Technology, 101: 1686-1692.
35-Zolfaghari, F., A., Pahlevanravi, A., Fakhireh, & M., Jabari, 2010. Investigation on relationship between environmental factors and distribution of vegetation in Agh Toghe basin. Iranian journal of Range & Desert Research, 17(3): 431-444.