MODEL NON LINIER UNTUK MENDUGA TINGGI DAN VOLUME JATI KLON DI KPH SARADAN

DINNY AISYAH ALBA, Dr. Ir. Ronggo Sadono

2017 | Skripsi | S1 KEHUTANAN

Jati klon merupakan jati unggul produk Perum Perhutani yang diperoleh dari progam pemuliaan pohon. Perkembangan tinggi dan volume penting diketahui untuk pemantauan keberhasilan program skala lapangan. Penyediaan informasi perkembangan dilakukan melalui pendugaan dengan diameter setinggi dada sebagai prediktor. Penelitian ini bertujuan untuk memilih model non linier penduga tinggi dan volume jati klon di KPH Saradan. Pengambilan data dilakukan pada petak dengan pertumbuhan baik. Sampel pohon merupakan 30 pohon dominan pada setiap petak terpilih. Data yang dikumpulkan berupa pengukuran tinggi pohon, tinggi batang bebas cabang, diameter tonggak, diameter setinggi dada, diameter pada tinggi dua meter, dan diameter pada tinggi batang bebas cabang. Rumus Smallian digunakan untuk menghitung volume batang bebas cabang. Pohon sampel yang digunakan pada penelitian ini sebanyak 1500 pohon untuk pemodelan dan 360 pohon untuk validasi model yang diperoleh dari pengukuran pada tahun 2015, 2016 dan 2017. Empat model non linier yang diajukan yaitu Schnute, Weibull, Richard dan Gompertz. Model penduga dipilih berdasarkan nilai koefisien determinasi (R2), Standard Error of the Estimate (SEE) dan uji F. Model terpilih diuji dengan data validasi berdasarkan kriteria nilai simpangan agregatif (SA), simpangan rata-rata (SR), dan Root Mean Square Error (RMSE). Hasil penelitian menunjukan bahwa model Richard dipilih untuk menduga tinggi dan model Gompertz untuk menduga volume. Persamaan kedua model tersebut adalah Tinggi = 24,147x(1-(exp(-0,058xDBH)pangkat 1,196)) dan Volume = 8,509x(exp(-8,938x(exp(-0,039xDBH))). Nilai koefisien determinasi (Rkuadrat) masing-masing untuk tinggi dan volume sebesar 0,768 dan 0,837 dengan nilai SEE sebesar 1,251 dan 0,018. Nilai validasi model untuk pendugaan tinggi adalah RMSE = 2,691, SA = -0,024 dan SR = 0,90%. Sedangkan untuk pendugaan volume diperoleh nilai RMSE = 0,031, SA = -0,135, dan SR = 10%. Kedua model tersebut layak digunakan untuk menggambarkan tinggi dan volume jati klon di KPH Saradan.

Clonal Teak is a superior teak of Perum Perhutani obtained from trees breeding program. The development of tree height and volume is important for monitoring the success of field-scale program. Provision of those information is done by model with diameter at breast height as predictor. This research is aimed to apply non linear model for predicting height and volume of clonal teak in Saradan forest district. The data were collected at the plot compartments with a good growth. The sample were 30 dominant trees in each selected compartments. The data were measurement of tree height, height at crown base, stump diameter, diameter at breast height, and diameter at height of two meters, and diameter at height at crown base. The formula of Smallian was used to obtain clear bole volume. Total number of sample trees were 1.500 trees for model fitting, and 360 trees for model validation that obtained from measurement in 2015, 2016, and 2017. Four non linier models were proposed for this research, Schnute, Weibull, Richard and Gompertz. Proposed models were selected based on the value of coefficient of determination (R2), Standard Error of the Estimate (SEE), and F-test. The accepted model was evaluated with validation data based on Agregat Deviation (SA), Mean Deviation (SR), and Root Mean Square Error (RMSE). The results showed that Richard model was selected for predicting height and Gompertz model for predicting volume. The equations of both models were Height = 224,147x(1-(exp(-0,058xDBH)square1,196)) and Volume = 8,509x(exp(-8,938x(exp(-0,039xDBH))). The value of Rsquare were 0.768 and 0.837 with SEE were 1.251 and 0.018 for height and volume repectively. Model validation values for RMSE, agregat deviation and mean deviation for height were 2.691, -0.024 and 0.90%. Model validation values for RMSE, agregat deviation and mean deviation for volume were 0.031, -0.135, and 10%. Both models were feasible to describe the height and volume development of clonal teak in Saradan forest district.

Kata Kunci : jati klon, model non linier, pohon dominan, model Gompertz, model Richard;clonal teak, non linear model, dominant tree, Gompertz model, Richard model