<![CDATA[Abstrak. Potensi bahaya erosi pada lahan berlereng seperti lahan perkebunan perlu kiranya dilakukan pengukuran dan perkiraan erosi. Metode perkiraan erosi salah satunya adalah model MUSLE (Modified Universal Soil Loss Equation) yang awalnya dikembangkan pada skala petak pertanian di Amerika Serikat, sehingga untuk mengetahui sejauhmana MUSLE mampu memperkirakan erosi maka dilakukan akurasi tingkat erosi dengan membandingkan erosi pengukuran dengan erosi model MUSLE pada dua plot percobaan tanaman kacang tanah dan tanaman bayam. Erosi pengukuran pada penelitian ini yaitu menggunakan simulasi hujan buatan berupa alat rainfall simulator. Sampel tanah pada rainfall simulator diambil dari lahan Perkebunan Kurma dengan kelerengan yang curam 25% dan jarak tanam 8 x 8 m2 tanpa adanya penerapan tanaman penutup tanah. Pengukuran erosi dilakukan per kejadian pemberian hujan dengan rainfall simulator dengan terlebih dahulu mengukur limpasan menggunakan literan dan kemudian menimbang tanah yang tertampung pada bagian bawah plot (reservoir). Hasil penelitian menujukkan adanya perbedaan erosi model MUSLE dan erosi pengukuran rainfall simulator pada plot percobaan tanaman kacang tanah berkisar antara9 x 10-6 ton/ha - 6,62 ton/ha, dan plot percobaan tanaman bayam berkisar 1,92 x 10-5 ton/ha - 6,37 ton/ha. Tingkat akurasi erosi model MUSLE (Modified Universal Soil Loss Equation) dengan erosi pengukuran rainfall simulator pada plot tanaman kacang tanah memiliki persamaan y = 0,000086x-0,000028 dengan nilai korelasi (r) = 0,82 (sangat kuat), (R2) = 0,67 (kuat) dan RMSE = 1,6. Sedangkan pada plot tanaman bayam memiliki persamaan y = 0,00095x-0,0001 dengan nilai korelasi (r) = 0,86 (sangat kuat), (R2) = 0,74 (kuat) dan RMSE = 1,44. Nilai RMSE 1 dikarenakan variasi nilai yang dihasilkan oleh suatu model MUSLE memiliki nilai yang jauh dari variasi nilai pengukurannya.Comparison of MUSLE Erosion Models with Rainfall Simulator Erosion MeasurementAbstract. The potential for erosion hazards on sloping land such as plantation land needs to be measured and estimated for erosion. There are several methods of estimating erosion, one of which is the MUSLE (Modified Universal Soil Loss Equation) model. The MUSLE model is an erosion estimation method developed on a farm plot scale in the United States, so to find out to what extent MUSLE is able to estimate erosion, erosion rate accuracy is carried out by comparing erosion measurements with the erosion of the MUSLE model on two experimental plots of peanut and spinach plants. Measurement erosion in this study uses artificial rain simulations in the form of a rainfall simulator. Soil samples in the rainfall simulator were taken from the Barbatee Dates Plantation with a steep slope of 25% and a spacing of 8 x 8 m2 without the application of ground cover crops. Erosion measurements are carried out per rainfall incidence using a rainfall simulator by first measuring runoff using liters and then weighing the land that is stored at the bottom of the plot (reservoir). The results showed that there were differences in erosion of the MUSLE model and the erosion of the rainfall simulator measurement on the experimental plot of peanut plants ranging from 9 x 10-6 tons/ha - 6.62 tons/ha, and the experimental plot of spinach plants ranging from 1.92 x 10-5 tons. / ha - 6.37 tonnes / ha. The accuracy rate of erosion of the MUSLE model (Modified Universal Soil Loss Equation) with the erosion of the rainfall simulator measurement on the peanut plant plot has the equation y = 0.000086x-0.000028 with a correlation value (r) = 0.82 (very strong), (R2 ) = 0.67 (strong) and RMSE = 1.6. Whereas the spinach plant plot has the equation y = 0.00095x-0.0001 with a correlation value (r) = 0.86 (very strong), (R2) = 0.74 (strong) and RMSE = 1.44. The value of RMSE 1 is because the variation in the value generated by a MUSLE model has a value that is far from the variation in the measurement value.]]>
