Aprilianto, Teguh Dharmawan and Prof. Dr. Ir. Sri Rahayu Utami,, M.Sc. and Ir. Aditya Nugraha Putra,, S.P., M.P. (2022) Penyusunan Model Pendugaan Kadar Nitrogen Tanah Dan Tanaman Di Lahan Sawah Pada Berbagai Bentuk Lahan Menggunakan Citra Multispektral. Sarjana thesis, Universitas Brawijaya.
Abstract
Pemanfaatan penginderaan jauh berbasis citra multispektral untuk menduga kadar nitrogen (N) tanah dan tanaman telah berkembang pesat. Pada penelitian sebelumnya, transformasi citra ke indeks vegetasi dan tanah mampu menduga kadar N dengan baik. Namun, kelemahan muncul ketika formula yang dihasilkan pada teknologi ini tidak bisa memenuhi syarat uji multi lokasi. Penelitian ini bertujuan untuk menyusun formula pendugaan kadar N tanah dan tanaman terbaik di lahan sawah pada berbagai bentuk lahan (landform) di Kabupaten Malang, Jawa Timur, Indonesia. Lahan sawah di Kabupaten Malang tersebar ke dalam bentuk lahan aluvial, vulkanik dan karst. Pola ini selanjutnya digunakan untuk menyusun satuan peta lahan (SPL) dan penentuan titik pengamatan. Citra Sentinel-2A ditransformasi dalam beberapa indeks, antara lain NDVI, GNDVI, TVI, EVI2, EVI2-2, ARVI 2, NDSI, MSAVI, MSAVI Hyper, SARVI3 dan OSAVI. Titik observasi berjumlah 130 (terdiri dari 100 untuk penyusunan model dan 30 untuk validasi) disebarkan di masing-masing SPL dengan metode stratified random sampling dengan memperhatikan keterwakilan seluruh nilai indeks. Kadar N tanah dan tanaman berkorelasi dengan nilai indeks GNDVI (vegetasi) dan MSAVI (tanah) dan memiliki koefisien determinasi sebesar 0,50 dan 0,54. Melalui persamaan berbentuk polymonial, distribusi N tanaman dapat diduga melalui persamaan 18,79(GNDVI)2 – 10,28(GNDVI) + 3,05, sedangkan N tanah diduga melalui persamaan -1,39(MSAVI)2 + 1,31(MSAVI) – 0,11. Formula pendugaan N tanah dan tanaman di berbagai bentuk lahan tersebut valid dan dapat diaplikasikan ke bentuk lahan aluvial, vulkanik dan karst sekaligus.
English Abstract
The use of remote sensing based on multispectral imagery to estimate the nitrogen (N) content of soil and plants has developed rapidly. In previous studies, image transformation to vegetation and soil indices was able to predict N levels well. However, weaknesses arise when the formulas generated in this technology cannot meet the requirements of the multi-site test. This study aims to develop a formula for estimating the best soil and plant N levels in rice fields in various landforms in Malang District, East Java, Indonesia Rice fields in Malang Regency are spread into alluvial, volcanic and karst landforms. This pattern is then used to compile Land Map Units (LPU) and determine observation points. Sentinel-2A images were transformed to NDVI, GNDVI, TVI, EVI2, EVI2-2, ARVI 2, NDSI, MSAVI, MSAVI Hyper, SARVI3 and OSAVI. There are 130 observation points (consisting of 100 for modeling and 30 for validation) distributed in each LPU using stratified random sampling method by taking into account the representation of all index values. Soil and plant N content correlated with index values GNDVI (vegetation) and MSAVI (soil) and had coefficients of determination of 0,50 and 0,54, respectively. Through a polymonial equation, the distribution of plant N can be estimated using the equation 18,79(GNDVI)2 – 10,28(GNDVI) + 3,05, while the soil N is estimated through the equation -1,39(MSAVI)2 + 1,31(MSAVI) – 0,11. The formula for estimating soil and plant N in various landforms is valid and can be applied to alluvial, volcanic and karst landforms at the same time
Other obstract
-
| Item Type: | Thesis (Sarjana) |
|---|---|
| Identification Number: | 0522040625 |
| Subjects: | 300 Social sciences > 338 Production > 338.1 Agriculture > 338.16 Production efficiency |
| Divisions: | Fakultas Pertanian > Agroekoteknologi |
| Depositing User: | PKN 01 UB |
| Date Deposited: | 13 Jun 2023 06:29 |
| Last Modified: | 13 Jun 2023 06:29 |
| URI: | http://repository.ub.ac.id/id/eprint/201532 |
![[thumbnail of DALAM MASA EMBARGO]](http://repository.ub.ac.id/style/images/fileicons/text.png) |
Text (DALAM MASA EMBARGO)
Teguh Dharmawan Aprilianto.pdf Restricted to Registered users only until 31 December 2024. Download (6MB) |
Actions (login required)
 |
View Item |