PEMANFAATAN CITRA SENTINEL-2A UNTUK ESTIMASI PRODUKTIVITAS TEH DI PERKEBUNAN DAYEUHMANGGUNG KABUPATEN GARUT

Yanti Mega Nurzihan, - and Jupri, - and Hendro Murtianto, - (2025) PEMANFAATAN CITRA SENTINEL-2A UNTUK ESTIMASI PRODUKTIVITAS TEH DI PERKEBUNAN DAYEUHMANGGUNG KABUPATEN GARUT. S1 thesis, Universitas Pendidikan Indonesia.

Abstract

Produktivitas teh di Indonesia menurut data dari Badan Pusat Statistik mengalami penurunan dalam beberapa tahun terakhir akibat perubahan iklim, teknik budidaya yang belum presisi, dan sistem pemantauan hasil panen yang masih dilakukan secara manual dan konvensional. Pemantauan secara manual cenderung menghabiskan banyak tenaga, waktu, uang dan rentan human error. Seiring perkembangan teknologi, penginderaan jauh menjadi alternatif yang efektif dan efisien dalam memantau kondisi vegetasi serta mengestimasi hasil tanaman secara spasial. Tujuan dari penelitian ini yaitu 1) mengetahui produktivitas eksisting teh, 2) mengestimasi produktivitas teh menggunakan indeks vegetasi SAVI, ARVI dan EVI dari Citra Sentinel-2A perekaman Agustus 2024, dan 3) menentukan tingkat akurasi estimasi berdasarkan perbandingan nilai indeks vegetasi dengan data produktivitas eksisting. Data produktivitas eksisting diperoleh melalui wawancara pengelola perkebunan, sedangkan estimasi dilakukan dengan analisis regresi linear, dan uji akurasi menggunakan nilai koefisien determinasi R² dan Root Mean Square Error (RMSE). Faktor ketinggian, tahun pangkas, dan umur tanaman teh turut dianalisis sebagai variabel pendukung. Hasil penelitian menunjukkan bahwa produktivitas eksisting di Perkebunan Dayeuhmanggung sebesar 139 kg/ha. Estimasi dengan indeks SAVI menghasilkan 138,59 kg/ha (R² = 25,7%; akurasi = 75,31%), ARVI sebesar 134,87 kg/ha (R² = 30,9%; akurasi = 62,07%), dan EVI sebesar 134,543 kg/ha (R² = 33,4%; akurasi = 61,29%). Berdasarkan pengolahan tersebut, indeks EVI menjadi indeks yang paling akurat untuk perhitungan estimasi produktivitas teh di Perkebunan Dayeuhmanggung Kabupaten Garut. Penelitian ini diharapkan dapat membantu mengidentifikasi area perkebunan dengan produktivitas rendah, sehingga pengelola atau lembaga terkait dapat merencanakan tindakan manajemen seperti pemupukan atau penyesuaian jadwal pangkas secara tepat. Tea productivity in Indonesia, according to data from the Central Statistics Agency, has declined in recent years due to climate change, imprecise cultivation techniques, and a harvest monitoring system that is still carried out manually and conventionally. Manual monitoring tends to consume a lot of energy, time, money, and is prone to human error. With the advancement of technology, remote sensing has emerged as an effective and efficient alternative for monitoring vegetation conditions and estimating crop yields spatially. The objectives of this study are 1) to determine existing tea productivity, 2) to estimate tea productivity using the SAVI, ARVI, and EVI vegetation indices from Sentinel-2A imagery recorded in August 2024, and 3) to determine the accuracy of the estimates by comparing the vegetation index values with existing productivity data. Existing productivity data were obtained through interviews with plantation managers, while estimations were performed using linear regression analysis, and accuracy tests were conducted using the coefficient of determination R² and Root Mean Square Error (RMSE). Factors such as elevation, pruning year, and tea plant age were also analyzed as supporting variables. The results of the study indicate that existing productivity at the Dayeuhmanggung Plantation is 139 kg/ha. Estimates using the SAVI index yielded 138.59 kg/ha (R² = 25.7%; accuracy = 75.31%), ARVI of 134.87 kg/ha (R² = 30.9%; accuracy = 62.07%), and EVI of 134.543 kg/ha (R² = 33.4%; accuracy = 61.29%). Based on this processing, the EVI index is the most accurate index for estimating tea productivity in the Dayeuhmanggung Plantation in Garut Regency. This study is expected to help identify plantation areas with low productivity, so that managers or related institutions can plan management actions such as fertilization or pruning schedule adjustments appropriately.

	Text S_SIG_2107840_Title.pdf Download (762kB)
	Text S_SIG_2107840_Chapter1.pdf Download (272kB)
	Text S_SIG_2107840_Chapter2.pdf Restricted to Staf Perpustakaan Download (357kB)
	Text S_SIG_2107840_Chapter3.pdf Download (749kB)
	Text S_SIG_2107840_Chapter4.pdf Restricted to Staf Perpustakaan Download (2MB)
	Text S_SIG_2107840_Chapter5.pdf Download (84kB)
	Text S_SIG_2107840_Appendix.pdf Restricted to Staf Perpustakaan Download (767kB)

Item Type:	Thesis (S1)
Additional Information:	https://scholar.google.com/citations?user=dOE0-GwAAAAJ&hl=id ID SINTA Dosen Pembimbing: Jupri: 6681759 Hendro Murtianto: 6115607
Uncontrolled Keywords:	Citra Sentinel-2A, Estimasi Produktivitas Teh, Indeks Vegetasi, Penginderaan Jauh, Regresi. Sentinel-2A Image, Tea Productivity Estimation, Vegetation Index, Remote Sensing, Regression.
Subjects:	G Geography. Anthropology. Recreation > G Geography (General) L Education > L Education (General) S Agriculture > S Agriculture (General)
Divisions:	Fakultas Pendidikan Ilmu Pengetahuan Sosial > Sains Informasi Geografi S1
Depositing User:	Yanti Mega Nurzihan
Date Deposited:	08 Sep 2025 02:42
Last Modified:	08 Sep 2025 02:42
URI:	http://repository.upi.edu/id/eprint/138060

Actions (login required)

View Item