Determining The Areas Prone to The Growth of Rhume ribes L. Specie in Razavi Khorasan Province Using Vector Machine Models

Introduction

In recent years, with the development of computer technologies, remote sensing systems, software and various models, it has been possible to predict the ecological niche of various plant and animal species. In the past decades, with changes in people's lifestyles and industrialization and production processes, atmospheric pollutants have increased, which has led to severe climate changes. Global climate change has led to a change in the range of plant growth, an expansion in plants adapted to warm weather and a decrease in plants adapted to cold weather. Such changes cause subsequent changes in the structure and ecosystems of the whole earth, which directly and indirectly affect ecosystem services (functions) for human well-being and economic growth. Therefore, predicting the effect of climate change on plant distribution has become a major field of research for its conservation measures and programs. The effect of climate changes on the growth range of plants; It is mostly predicted by species distribution models. Therefore, reliable predictions of species distribution are needed to plan effective conservation and maintain sustainable forest ecosystem services under climate change. Considering the importance of this issue, this research was conducted with the aim of identifying the most important climatic and environmental factors affecting the distribution of Rhume ribes L. species and determining its current geographical range in Razavi Khorasan province located in the northeast of Iran.

Materials and Methods

For this purpose, 68 bioclimatic variables including soil characteristics (45 cases), topographical factors (4 cases) and climatic factors (19 cases) were first subjected to correlation analysis as predictive variables and variables with high correlation (above 80%) were removed. Due to the large size of the studied area, sampling of presence points was done with field visits during the period of 1400-1401 of the introduced areas, and a total of 232 presence points from 8 regions were registered as presence points using the global positioning system(GPS). Then all the environmental data and presence points in R software using Biomod 2 package models which include GLM, GBM, GAM, CTA, ANN, SRE, FDA, MARS, RF and MaxEnt models in determining the relationship between vegetation and environmental factors in pastures Khorasan Razavi province was predicted in the present time. The accuracy of the models was evaluated using the values of KAPPA, TSS and ROC indices, which are prominent and widely used indices for determining and identifying the potential areas.

Results and Discussion

The results of this research showed that according to the accuracy evaluation index, the best modeling for the current time is the random forest model with an accuracy of 95.5%, which indicates the accuracy of the modeling at an excellent level. Also, the relative importance of the selected models and the variables that have had the greatest impact at the present time include: digital elevation model (DEM), Average monthly (BIO2), This is the sum of all total monthly precipitation values (BIO12), The average temperatures experienced during the wettest quarter (BIO 8) and the amount of sand at a depth of 15-30 cm from the soil surface (Sand 15-30), which indicates the great influence of climatic factors on the distribution of this species, and in the next stage, the height above sea level and finally the soil factors have the greatest influence. The most distribution of Rhume ribes L. species at the present time is in the east of Khorasan province including the cities of Bakharz, Torbat Jam, Taibad, Zaveh, Khaf and Rashtkhwar in the form of a strip on their border and in the west of the province on the border of Koh Sorkh and Neishabur cities and in the north of the province on the border Binaloud, Zabarkhan and Mashhad cities and in the south of the province in Gonabad city has spread in a strip and limited way.

Conclusion

The results of this research can be used for the improvement and protection as well as the economic exploitation and expansion of the Rhume ribes L. species habitat. Destructive human activities such as livestock grazing and unprincipled exploitation of rhubarb along with climate changes have caused the current habitats of this species in Khorasan-Razavi province to be in serious danger. These unprincipled exploitations without considering environmental capabilities in the field of natural resources is one of the problems in Khorasan province and the country, which gradually causes the loss of water, soil and plants in this region. Although in this study, it was enough to examine the climatic and soil factors in the present time in order to show the areas prone to the presence of rhubarb species, but in order to have a deeper understanding and better knowledge to restore the damaged areas and preserve the areas at risk, as well as to improve the ability of ecological models in forecasting. The potential habitats of plant species (Barnes and Harrison, 1982), in addition to these factors, other factors such as human factors, types of exploitation, livestock grazing, wildlife, economic and social status and many other factors that directly and indirectly affect the distribution of such effects should be considered. are being checked. Many studies have been done on different plant species. Despite the difference in scale, scope and method of work, vector machine methods are introduced as efficient methods. In this research, it was tried to evaluate different models of species distribution vector machine and then the most suitable model which was random forest was selected. Species distribution models are useful and cost-effective tools for the use of natural resource managers and increase their awareness and decision-making power regarding the effects of climate change on species. In general, it can be stated that models based on vector machines provide a very suitable ability to determine the prone areas of this species.