The Effect of the Forest Stratification Method and its Combined Effect with the PRODAN Six-Tree Sampling Method

Background: Renewable natural resources, such as forests, have played an important role in human life since the past.The principled and sustainable use of forests requires technical planning.Fundamental and technical planning requires data collection and analysis and obtaining useful quantitative and qualitative information.A data collection method in forest populations is 100% statistics, which is rarely implemented in large forest areas due to technical and economic reasons, and it is used in most cases to compare different sampling methods.Sampling methods with a fixed area are mostly used in the preparation of plans and studies of northern forests.Today, different forest sampling methods have been designed and presented, each of which has its own advantages and merits.The use of these methods requires examining and determining the most suitable ones in terms of ease of implementation, time, and accuracy.On the other hand, the industrial forests in the north of Iran have currently become heterogeneous stands due to exploitation in the past.For this type of forest, using the stratification method combined with sampling methods will be useful and effective.To plan to increase efficiency, speed of operation, and reduce costs, it is necessary to examine and use different sampling methods with fixed and variable areas, including multi-tree sampling.This study is aimed at the effect of forest classification and its combined effect with the fixed area sampling method and the PRODAN six tree variable area sampling method and comparing the accuracy of estimators.Methods: This study was conducted in the Safaroud forests with an area of 176 hectares in Mazandaran Province.The selected areas were measured and recorded using the statistical method of 100% of all the trees in the area (counting limit, a diameter at the breast of more than 7.5 cm).To avoid mistakes and make it easier to perform 100% statistics, the studied area was divided into square pieces with dimensions of 50 50 m, and 704 plots were placed in the field.Then, the trees in each of these parts were measured separately.Prodan's six-tree method was carried out in two stages.The first stage was measured in an unclassified area of 176 hectares with a grid size of 100 100 m.The second stage of the investigated forest area using the "Neyman" stratification method into three strata in terms of volume inventory, including less than 200 m 3 , 200-250 m 3 , and more than 250 m 3 per hectare based on Prodan's six-tree method, was performed using new grid dimensions for each stratum.In this study, the effect of segmentation on accuracy and time was investigated using the criterion (E% 2 T).For this purpose, the time required to unload the sample plots for the necessary measurements inside the sample plot and the time required to travel the distance between the adjacent sample plots were also calculated and recorded using a stopwatch.Results: The results of the comparison of the statistical test between the average of the variables calculated with the 100% statistical method (true mean) and six-tree sampling (Prodan) in the first stage without stratification and in the second stage with the implementation of the "Neyman" stratification method showed a significant difference at the level of 0.05%.The results of this study in terms of accuracy and the cost index (E% 2 T) show that the error and statistical error of the measurement characteristics in the standing forest with the six-tree sampling method with stratification was lower than the sampling method without grafting.This amount was calculated for the number of trees per hectare as 8811.81 and 5677.52,respectively.The estimated average of the examined characteristics was 35.56% closer to the real average and the error rate was less than the statistical standard with the forest stratification and Prodan's six-tree sampling implementation.The percentage of error reduction in sampling after seeding was calculated at