Academia.eduAcademia.edu

Figure 3 – uploaded by Caspian Journal of Environmental Sciences

See full PDFdownloadDownload figure
Table 3. Results of network training for estimating the annual rangelands production. Table 4. Comparison between the simulated rangelands production and the measured rangelands production in the test stage.

Table 3 Results of network training for estimating the annual rangelands production. Table 4. Comparison between the simulated rangelands production and the measured rangelands production in the test stage.

Related Figures (9)

Rangelands production modeling using an artificial neural network (ANN) and geographic information system (GIS) in Baladeh rangelands, North Iran
Rangelands production modeling using an artificial neural network (ANN) and geographic information system (GIS) in Baladeh rangelands, North Iran
Fig. 1. Location of the studied rangelands in the north of Iran. The study area is located at 51° 29' to 51 ° 45' E longitudes and 36° 05' to 36° 14' N latitude in North Iran (Fig. the southern Caspian coasts in the Alborz highlands. The area of the studied rangelands is about 200 km?. T most area of Baladeh watershed was included rangelands. The climate of the studied area is cold semi-hum: The mean annual precipitation and temperature are approximately 580 mm and 5.3 C® respectively. Precipitati mainly consists of snow. The main part of precipitation falls in the cold season (Gholami et al. 2008). Frost a low temperature are the most important limiting factors in the vegetation growth during the cold seasotr Moreover, climatic data are available from the Baladeh meteorological station. The area consists of rangelan with different types of vegetation included Artemisia fragrans, Amygdalus lycioides, Astragalus spp., Achill spp., Onobrychis cornuta, Psatyrostachys fragillis-Leucopoa scrophylla, Artemisia fragrans-Onobrycl cornuta, Astragalus sp.-Cousinia multiloba, Astragalus sp., Astragalus sp.- Bromus tomentellus and Amygdal sp.-Astragalus sp. Different growth forms of rangelands vegetation are observed such as shrub and gra species. Elevation alterations from 2000 to 4200 m and slope changes from flat parts to 70 degrees in the are The mean slope degree of the area is 24.5. There are nine villages in the studied area and the people jobs a mostly animal husbandry or farming. Human activities are of the main factors resulting in vegetati degradation in North Iran. degradation in North Iran.
Fig. 1. Location of the studied rangelands in the north of Iran. The study area is located at 51° 29' to 51 ° 45' E longitudes and 36° 05' to 36° 14' N latitude in North Iran (Fig. the southern Caspian coasts in the Alborz highlands. The area of the studied rangelands is about 200 km?. T most area of Baladeh watershed was included rangelands. The climate of the studied area is cold semi-hum: The mean annual precipitation and temperature are approximately 580 mm and 5.3 C® respectively. Precipitati mainly consists of snow. The main part of precipitation falls in the cold season (Gholami et al. 2008). Frost a low temperature are the most important limiting factors in the vegetation growth during the cold seasotr Moreover, climatic data are available from the Baladeh meteorological station. The area consists of rangelan with different types of vegetation included Artemisia fragrans, Amygdalus lycioides, Astragalus spp., Achill spp., Onobrychis cornuta, Psatyrostachys fragillis-Leucopoa scrophylla, Artemisia fragrans-Onobrycl cornuta, Astragalus sp.-Cousinia multiloba, Astragalus sp., Astragalus sp.- Bromus tomentellus and Amygdal sp.-Astragalus sp. Different growth forms of rangelands vegetation are observed such as shrub and gra species. Elevation alterations from 2000 to 4200 m and slope changes from flat parts to 70 degrees in the are The mean slope degree of the area is 24.5. There are nine villages in the studied area and the people jobs a mostly animal husbandry or farming. Human activities are of the main factors resulting in vegetati degradation in North Iran. degradation in North Iran.
1 sites, B) Picturing method, and estimating vegetation canopy was evaluated as rangelands production. Therefore, the method of cutting and weighting was used for estimating rangelands production for grass species. For shrub species, the annual vegetation growth was exactly separated by drying the samples (Fig. 3). The samples were dried in the air. Thereafter, samples were weighed using a digital scale with 0.1 g accuracy. Finally, rangelands production was estimated based on the mean of the twenty plots production for each of the sites and consequently, the measured production values in the sites were defined as ANN output in the modeling process.
1 sites, B) Picturing method, and estimating vegetation canopy was evaluated as rangelands production. Therefore, the method of cutting and weighting was used for estimating rangelands production for grass species. For shrub species, the annual vegetation growth was exactly separated by drying the samples (Fig. 3). The samples were dried in the air. Thereafter, samples were weighed using a digital scale with 0.1 g accuracy. Finally, rangelands production was estimated based on the mean of the twenty plots production for each of the sites and consequently, the measured production values in the sites were defined as ANN output in the modeling process.
Fig. 3. A) Measurement of rangelands production by evaluating the annual vegetation growth, B) Estimation of rangelands production by cut and weighting method for grasses and shrub species, C) Separation of annual growth by drying shrub species.
Fig. 3. A) Measurement of rangelands production by evaluating the annual vegetation growth, B) Estimation of rangelands production by cut and weighting method for grasses and shrub species, C) Separation of annual growth by drying shrub species.
Fig. 4. Evaluation of ANN performance for rangelands production estimation during the test stage throughout the comparison between the estimated and measured production values.
Fig. 4. Evaluation of ANN performance for rangelands production estimation during the test stage throughout the comparison between the estimated and measured production values.
Table 1. The network optimal inputs (influencing factors) and outputs (the measured rangelands production) for estimating rangelands production in some of the sampling sites. Received: Sep. 25. 2019 Accepted: Feb. 29. 2020 Article type: Research Caspian J. Environ. Sci. Vol. 18 No. 3 pp. 277~290 DOI ©Copyright by University of Guilan, Printed in LR. Ira
Table 1. The network optimal inputs (influencing factors) and outputs (the measured rangelands production) for estimating rangelands production in some of the sampling sites. Received: Sep. 25. 2019 Accepted: Feb. 29. 2020 Article type: Research Caspian J. Environ. Sci. Vol. 18 No. 3 pp. 277~290 DOI ©Copyright by University of Guilan, Printed in LR. Ira
Table 2. The Pearson correlation between rangelands production and the factors influencing rangelands production.
Table 2. The Pearson correlation between rangelands production and the factors influencing rangelands production.
Fig. 6. The map of rangelands production has resulted from the integrating ANN inference system and GIS capabilities. In this map, the accuracy of the results was evaluated using a comparison between the estimated and measured production values.
Fig. 6. The map of rangelands production has resulted from the integrating ANN inference system and GIS capabilities. In this map, the accuracy of the results was evaluated using a comparison between the estimated and measured production values.
Related topics:
MLP network
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved