نوع مقاله : مقالات پژوهشی
نویسندگان
1 گروه مهندسی عمران، دانشکده مهندسی، دانشگاه بوعلی سینا، همدان، ایران.
2 دانشگاه بوعلی سینا
3 گروه گیاه‌پزشکی، دانشکده کشاورزی، دانشگاه بوعلی سینا، همدان، ایران
چکیده
خسارت سرمازدگی بهاره، از جمله مواردی است که مشمول پرداخت غرامت از سوی صندوق بیمه کشاورزی هستند. سرمازدگی ناگهانی از اثرات منفی در میزان باز‌دهی محصولات در باغ‌ها است. به دلیل ارزش اقتصادی نسبتاً بالای محصولات باغی، باید از روش‌های دقیقی در برآورد میزان خسارت استفاده کرد. بر خلاف محصولات زراعی که در سطح وسیع و به‌صورت یکنواخت کشت می‌شوند، تعمیم روش‌های نمونه‌برداری در این مطالعه به کل منطقه بسیار دشوار بوده و موجب کاهش دقت مدل‌‌های برآورد خسارت می‌‌گردد. هدف اصلی این تحقیق، ارائه یک شاخص کمّی برای تخمین میزان خسارت وارده به درختان گردو بر مبنای سری زمانی تصاویر ماهواره‌های Landsat و Sentinel-2 است؛ لذا بر اساس سری زمانی شاخص NDVI، شاخص خسارت سرمازدگی بهاره SFDIبرای باغات گردوی شهرستان تویسرکان محاسبه و در یک دوره زمانی نه‌ساله (2013 تا 2021) مورد ارزیابی قرار گرفت. سری‌های زمانی NDVI مورداستفاده توسط فیلتر SG هموار شده‌اند تا اثرات عواملی مانند ابر، گردوغبار و نویز کاهش یابند. جهت برآورد شاخص SFDI، نیاز به تعیین است، با مقایسه سری زمانی هر سال با سال مرجع و محاسبه سطح بین دو منحنی، شاخص SFDI محاسبه شد که بیانگر میزان خسارت‌های وارده به درختان گردو است. ارزیابی عملکرد شاخص SFDI نسبت به برآوردهای صندوق بیمه کشاورزی نشان می‌دهد که شاخص SFDI برای کلیه روستاها از طریق دو مدل چندجمله‌ای درجه 2 و رگرسیون خطی با ضریب‌های همبستگی به ترتیب 927/0 =r و 823/0 = r قادر به تخمین خسارت است. دراین‌بین، نقشه‌های توزیع مکانی و خسارت سرمازدگی بهاره، تناسب خوبی با خسارت‌های گزارش‌شده از طرف صندوق بیمه کشاورزی دارند. در نهایت، روش پیشنهادی می‌تواند جایگزین مقرون به صرفه‌ای جهت برداشت داده میدانی خسارت برای درختان گردو با استفاده از تصاویر ماهواره‌ای باشد.
کلیدواژه‌ها
موضوعات
عنوان مقاله [English]
Estimation of frost damage in walnut trees using remote sensing
نویسندگان [English]
- Elham Gohari 1
- Hossein Torabzadeh 2
- Saeed Alvandy 3
- Morteza Heidarimozaffar 1
1 Department of Civil Engineering, Faculty of Engineering, Bu-Ali Sina University, Hamedan, Iran
2 Bu Ali Sina University
3 Department of Plant Protection, Faculty of Agriculture, Bu-Ali Sina University, Hamedan, Iran
چکیده [English]
Introduction
Walnut is one of the most important economic products of Iran and the world. Among the climatic factors effective in choosing the right place for planting walnut trees, we can mention the spring cold. Most of the walnut planting areas in Iran are mostly in the areas that have damaging colds in spring. The largest area under walnut cultivation in Iran is concentrated in the mountain hills and heights of the Zagros and Alborz mountain ranges. Every year, walnut production is limited by spring frosts, which, in addition to reducing production, can also damage sensitive plant tissues. Frost damage is one of the cases that are subject to compensation from the Agricultural Insurance Fund. Due to the high economic value of this product, an accurate, fast and low-cost estimation is necessary to evaluate the damages caused by this phenomenon on the walnut product. In this regard, in order to plan for the accurate calculation of the amount of compensation to the insurer, the insurer needs comprehensive information, so that it can estimate the performance losses caused by various factors. Remote sensing methods for frost monitoring using vegetation indicators are very efficient and effective. Therefore, this research examines spring frost damage by providing a quantitative index.
Materials and methods
In this research, frost damage of the Agricultural Insurance Fund was used for walnut trees. The damage assessment was presented over a period of nine years.Vegetation index values extracted from OLI, ETM+, MSI sensors have been used to overcome adverse weather conditions. Each of the vegetation indicators has different noises such as light changes during the day, wind speed, temperature, sun angle, etc. . In order to smooth the indices, SG filter was used. This filter was calculated with window size 11 and degree 2. In this filter, choosing the window size and degree is very important, because the small window size and degree cause noise to remain in the data, and if the window size and degree are chosen large, they lead to the loss of information related to vegetation indices. Meanwhile, choosing the reference year for comparison with other years was of particular importance. According to the consistency of the filtered indicators and the damages reported by the Agricultural Insurance Fund, the reference year was determined. The SFDI index was presented in each year using the filtered values of the relevant vegetation indices, between the reference year and other years. In general, the enclosed area of the two curves between the intersection points at the beginning and the end has indicated the values of the SFDI index in each year.
Results and Discussion
The relationship between SFDI index and damages was done through 2nd degree polynomial model and linear regression. The status of the SFDI index compared to the estimates of the agricultural insurance fund for all villages was investigated through two polynomial models of the 2nd degree and linear regression. The accuracy of the regression model was evaluated with criteria such as: r, RMSE, P-value. The value of r indicates how well a model can predict the value of the dependent variable in percentage terms. The higher the value of r, the better the model. RMSE values indicate how well a regression model can predict the value of the variable. The quantity P-value shows the significance of two independent and dependent variables, and values less than 5% indicate the existence of a relationship between two variables at a significant level of 95%. It is useful to calculate all three quantities for a given model, as each measure provides useful information. Spring frost damage index based on NDVI vegetation index with correlation coefficients of r = 0.927 and r = 0.823 were able to estimate the damage. Finally, the maps of spatial distribution and spring frost damage showed a good fit with the damage reported by the Agricultural Insurance Fund.
Conclusion
The SFDI index has provided a positive effect in the assessment of frost damage. Also, the results of the spatial distribution of the SFDI index have shown the effectiveness of this index, in such a way that with the increase of this index, the frost damage percentage of the agricultural insurance fund has also increased. The results showed that this index evaluated the effects of spring frost effectively and quickly and showed a good correlation with the decrease in crop yield.
کلیدواژه‌ها [English]
- Remote sensing
- SFDI index
- Spring frost damage
- Walnut tree
)
ارسال نظر در مورد این مقاله