Spatial and temporal characteristics of soil erosion and identification of source contributors of sediments in the tropical rainforest region of Borneo

Figures & data

Figure 1. Study area location map with rain gauges

Table 1. Sources, scale and information of data layers used in the study

Data source	Scale/resolution	Variables generated
Topographic sheets (Land Survey Department, Malaysia)	1:50 000	Study area boundary, drainage networks
Soil series maps (Department of Agriculture, Sarawak, Malaysia)	1:50 000	Soil texture and type
Landsat Images (downloaded from USGS Earth Explorer) Landsat 5 Thematic Mapper (TM) (acquired 1991–1994) Landsat 7 Enhanced Thematic Mapper Plus (ETM+) (acquired 2006–2007) Landsat 8 Operational Land Imager (OLI) (acquired on 2015)	30 m × 30 m	NDVI, C factor, land use/land cover, logging roads
SRTM data (downloaded from USGS Earth Explorer) SRTM Version 3–30 m global elevation data (acquired in 2000)	30 m × 30 m	Elevation and LS factor
Rainfall data (monthly between 1990 and 2014) (Department of Irrigation and Drainage, Malaysia)	mm	Rainfall distribution and R factor

Table 2. Major soil types and proposed soil erodibility values

Soil group	Soil family (local name)	Soil texture	Proposed K factor
Skeletal	Kapit	Residual (saprolite)	0.019
	Meluan	Lithic/paralithic	0.005
	Kapit/Meluan/Tutoh	Residual (clay loam)/lithic/paralithic/colluvium	0.021
	Meluan/Kapit	Lithic/paralithic (residual)	0.005
	Tutoh	Colluvium (clay loam)	0.017
Skeletal – Podzols	Kapit/Silantek	Residual (non-indurated)	0.047
	Bako/Kapit	Sandy loam (residual)	0.010
	Silantek/Kapit	Sandy loam (residual)	0.010
Red-yellow podzolic	Bekenu	Fine loam (residual)	0.007
	Merit	Fine clayey	0.004
	Bekenu/Merit/Nyalau	Fine loamy, fine clayey, coarse loam	0.021
	Merit/Bemang	Fine clay loam	0.017
Andisols	Linau	Clay (residual)	0.009
Peat (oxisols – organic)	Selio/Nibong/Mulu	Residual clay-fine (lateritic with organic content)	0.009
Podzols – arenaceous	Silantek/Tika	Residual (sandy)	0.041
Oxisols	Nibong/Selio	Residual (fine clay)	0.009
Grey-white podzolic – gleysols	Latan/Penara	Clay	0.004
Red-yellow podzolic – arenaceous	Merit/Tika	Fine clayey, sandy	0.038
Organic – gleysols	Bareo/Dalan	Organic, clayey	0.004
Gleysols – alluvial	Bijat/Seduau	Clay (alluvial)	0.009
	Bemang/Dalan	Fine clay loam	0.017
	Julan/Jiwan	Coarse clay loam	0.021
Podzols – organic	Silantek/Umor/Bareo	Sandy loam, residual (organic)	0.010
Organic	Bareo/Umor	Organic (residual)	0.004
Alluvium – red-yellow podzolic	Bemang/Bekenu	Fine loam (alluvial/residual)	0.018
Recent alluvium	Bemang/Seduau	Fine loamy/clayey (alluvium)	0.017

Table 3. Major land use/land cover (LULC) types and corresponding C factor values assigned to each

LULC type	Area (km^2)	Area (%)	C factor
Primary forest (PFR)	4397.42	49.48	0.003
Secondary forest (SFR)	3413.30	38.41	0.03
Montane forest (MF)	30.46	0.34	0.003
Mixed agriculture (MA)	349.68	3.93	0.60
Hill paddy (HP)	72.42	0.81	0.45
Paddy field (wet) (PF)	5.88	0.07	0.25
Bushes/shrubs/grasses (BSG)	286.81	3.23	0.86
Water body (WB)	29.31	0.33	0
Exposed barren land (EBL)	293.39	3.30	1
Pebbles/cobbles in river bed (PC)	2.59	0.03	0.01
Artificial surface (AS)	5.90	0.07	0.01
Total	8887.17	100%

Figure 2. Temporal distribution of: (a–c) rainfall (mm) and (d–f) rainfall erosivity (MJ mm ha⁻¹ h⁻¹ year⁻¹) during the three time frames (T1, T2 and T3). (Places listed are interior villages present in the study area)

Figure 3. Temporal distribution of (a, b) NDVI, (c) land use/land cover and (d–f) corresponding cover management factors. See Table 3 for abbreviations of land-use types

Figure 4. Spatial distribution of K factor in the study area

Figure 5. Spatial distribution of LS factor in the study area

Table 4. Maximum and mean erosion rate in different time frames. T1: 1991–1994; T2: 2006–2007; T3: 2015

			Soil loss (t ha^−1 year^−1)
Time frame	Method of C factor calculation	Dataset^a	Maximum	Mean
T1	NDVI	Landsat 5 TM	483	2
T2	NDVI	Landsat 7 ETM+	671	33
T3	Supervised classification	Landsat 8 OLI	1312	27

^aDataset: 30 m × 30 m pixel size.

Table 5. Areal distribution of soil erosion vulnerability classes in the study area in different time frames

	Soil erosion loss range	Time frame 1 (1991–1994)		Time frame 2 (2006–2007)		Time frame 3 (2015)
Soil erosion vulnerability class	(t ha^−1 year^−1)	Area (km^2)	Area (%)	Area (km^2)	Area (%)	Area (km^2)	Area (%)
Low	0–5	8479.21	95.41	3337.46	37.55	5724.06	64.41
Moderate	5–25	253.94	2.86	3646.24	41.03	2160.1	24.31
High	>25	153.85	1.73	1903.30	21.42	1002.84	11.28

Figure 6. Spatial distribution of soil erosion rate in different time frames

Figure 7. Spatial distribution of soil erosion vulnerability in different time frames

Figure 8. Boxplots explaining the statistical characteristics of soil erosion in the three time frames studied

Figure 9. Areal distribution of erosion vulnerability classes in the study area

Table 6. Mean soil erosion corresponding to the most dynamic land-use activity in the study area

Land use/land cover	Area (km^2)	Area (%)	Mean soil loss (t ha^−1 year^−1)
Exposed barren land	293.39	3.30	228.00
Mixed agriculture	349.68	3.93	88.09
Bushes/shrubs	286.81	3.23	178.76
Hill paddy	72.42	0.81	141.00
Paddy field	5.88	0.07	98.67

Figure 10. Spatial distribution of soil loss characteristics of the study area, identified through differencing of erosion rates calculated for the time frames considered

Figure 11. Spatial distribution of logging roads in the different time frames considered

Table 7. Logging road characteristics in the Upper Baram catchment in different time frames. T1: 1991–1994; T2: 2006–2007; T3: 2015

Characteristics analysed	Time frame
	T1	T2	T3
Total length of road networks (km)	3272	7539	9451
Mean density – whole study area (km km^−^2)	0.36	0.84	1.06
Trend (% change^a)	-	+56%	+25%
Logging road clusters^b	Right bank (North)	Right bank (North, northeast)	Left bank (South, southwest)

^aCompared to the previous time frame.

^bCompared with the Baram River basin.

Figure 12. Spatial distribution of logging road density in the study area

Table A1. Rainfall (mm) and rainfall–runoff erodibility factor (R, MJ mm/(ha/h year)) corresponding to different time frames. Time frame 1: 1991–1994; time frame 2: 2006–2007; time frame 3: 2015

	Time frame 1										Time frame 2						Time frame 3
	1991		1992		1993		1994		Mean 1991–1994		2006		2007		Mean 2006–2007		2015 (using rainfall of 2014)
Station code	Rainfall	R	Rainfall	R	Rainfall	R	Rainfall	R	Rainfall	R	Rainfall	R	Rainfall	R	Rainfall	R	Rainfall	R
Long Akah	4399	5727	4972	4550	4428	5104	4783	3768	4645	4787	5491	4889	4586	3644	5038	4266	3967	5380
Long Lellang	2986	2651	3050	2857	3499	3428	3180	2569	3179	2876	4307	3391	–	–	4307	3391	3253	2822
Long Naha’ah	4267	4610	3490	3389	3756	3653	3757	3294	3817	3736	4591	4299	4589	3708	4590	4003	3629	4518
Long Luteng	–	–	–	–	–	–	–	–	–	–	3620	3563	3925	3539	3772	3551	3378	2353
Long Anap	3822	3333	4861	4507	4253	3841	4564	4856	4375	4134	4287	3994	4536	4403	4411	4199	3846	3798
Long Jee	4659	6196	5269	4844	5343	5901	4877	4313	5037	5314	5779	5926	5846	5949	5812	5937	4396	5908
Long Moh	4407	5368	3925	4649	4158	4632	3929	4350	4105	4750	5282	4358	5387	4980	5335	4669	4470	5269
Lio Matu	3164	2267	2924	2317	3693	3500	3207	3223	3247	2827	3686	2598	3636	2516	3661	2557	- -	–
Bario	2102	1619	1997	1343	2361	1364	1925	1199	2096	1381	2271	1487	2234	1296	2252	1392	1796	1788
Long Seridan	3463	2902	3055	2623	3149	3111	2150	3043	2954	2920	3779	3000	3904	2872	3842	2936	3342	3207
Pa Tik	–	–	–	–	–	–	–	–	–	–	3730	2762	3502	2540	3616	2651	4136	3357
Long Atip	3685	4495	4409	6582	4157	4767	3075	2622	3831	4616	3636	3232	2359	4625	2998	3928	–	–
Long Pilah	3940	3155	3874	3403	3565	3325	3853	5932	3808	3954	3596	6398	4963	3850	4280	5124	–	–
Long Subing	3282	2510	2881	3514	3220	2373	3793	3226	3294	2906	4011	3634	–	–	4011	3634	2043	3639
Long Aton	3721	3484	3400	3649	3641	3519	4098	3202	3715	3463	4209	5013	3533	2500	3871	3756	2650	3423