Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

141 

 

The Dynamics of Flow Discharge and Suspension Flow 

Discharge in Volcano Watershed with Agroforestry Land 

Cover 

La Ode Hadini1, Junun Sartohadi2, M. Anggri Setiawan3, Djati Mardiatno3 

1 Faculty of Earth Sciences and Technology, Universitas Halu Oleo, Kendari 93132 
2Faculty of Agriculture, Universitas Gadjah Mada, Yogyakarta 55281 
3Faculty of Geography, Universitas Gadjah Mada, Yogyakarta 55281 

hadini74@gmail.com1 

Received 12-01-2021; accepted 27-06-2021 

Abstract. The suspension flows the upper part of a volcano watershed, which has a very thick 

soil condition, is sensitive to land use. Agroforestry is the dominant land use in the volcanic 

landscape of Indonesia. This research, performed in the agroforestry area, covered the 

characteristics of the correspondence between flow discharge and suspension discharge during 

the flow. The suspension flow was measured at the outlet of key watershed areas, which yielded 

436 suspension data. The measurement analysis was conducted at every rain event in the field 

and the laboratory. The crop characteristics in the catchment area were recorded in detail during 

the field survey. The characteristics of the channels converging toward the gully system were 

observed during the field survey. There were the relationship patterns between the peak flow 

discharge and the suspension discharge with the average time interval between the rain events, 

and the occurrence of suspension flow was 17.7 minutes, and the peak suspension content varied 

with an average of 1.03 g/L; then the grain size of the suspension was dominated by clay fraction 

with an average of 73% at the rising stage and average of 69% the falling stage.  

Keywords: Agroforestry, Discharge, Suspension, Volcano, Watershed. 

1.  Introduction 

The land use pattern in volcanic lands in Indonesia is distinctive. Indonesia has more than 400 

volcanoes, 127 of which are included in the active category [1]. The order of a volcanic landscape starts 

from the cone, the upper slope – the foot slope. This landscape's typical land use pattern is the non-

intensive utilization of cone and upper slope due to high-intensity volcanic hazards [2]. Production 

activities in the form of agroforestry have recently appeared in the middle slope [3]. Agricultural land 

usually occupies areas from the foot slope [4]. 

The suspension flow from the upper part of a volcanic watershed with very thick soil conditions is 

sensitive to land utilization. The suspension flow dynamics illustrate the flow response to the dynamics 

 
1 Cite this as: Hadini, L.O., Sartohadi, J. &Setiawan, M.A., Mardiatno, D. (2021). The Dynamics of Flow 

Discharge and Suspension Flow Discharge in Volcano Watershed with Agroforestry Land Cover. Civil and 

Environmental Science Journal (Civense), 4(2), 141-153. doi: https://doi.org/10.21776/ub.civense.2021.00402.4 

mailto:hadini74@gmail.com


 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

142 

 

of the watershed characteristics, one of which is the land use pattern. The suspension flow dynamics are 

essential to identify the watershed criticality qualitatively. This approach, however, requires further 

development to achieve time and cost-effectiveness, and efficiency [5][6][7]. The suspension flow is 

strongly related to soil loss rate and soil fertility deterioration process. It also triggers deposition and 

sedimentation, which lead to siltation [8][9][10][11]. 

The dynamics of the suspension flow characteristics and the change in watershed conditions are 

observable during certain rainfall events. The relationship between these two dynamics is explainable 

through suspension hydrograph analysis. In a hydrograph, flow discharge and suspension discharge 

parameters depict the aforementioned dynamics [12][13][14][15][16][17][18][19]. These dynamics are 

associated with the balance system between rainfall as the input, infiltration, and soil water storage in a 

watershed area [18][20][21][22][23]. The suspension flow dynamics presented in the suspension 

hydrograph’s response demonstrate any changes in rain input, infiltration, and soil water storage.  

The studies of flow suspension dynamics under the land use of agroforestry have been reported to be 

optimal enough to control the formation of suspension flow [18][24][25][26][27][28]. However, they 

do not deal with the physical characteristics of watersheds and landscapes covered with homogeneous 

agroforestry as crucial watershed areas. Furthermore, the studies of flow and suspension flow dynamics 

in volcanic watersheds have so far taken place in watershed areas with varied land use patterns along 

with other geophysical conditions that involve many assumptions or generalizations. Therefore, the 

results of studies based on this assumption generalization pose a potential bias toward the real states in 

the field.  

This research aims to involve a key area approach particularly for a watershed with small area 

coverage and homogeneous agroforestry and geophysical conditions. This approach enables planning 

the watershed’s physical characteristics in a more detailed and uniform manner to create a study that 

approximates the actual conditions in the field. Also, it can make a study applicable to volcanic 

watersheds with similar characteristics. The study of flow suspension dynamics in volcanic watersheds 

with agroforestry covers the following problems: the corresponding responses of flow and suspension, 

the time lag between rain events and the initial formation of suspension flow, and the grain size of the 

suspension during the flow. 

2.  Material and Methods 

This study used a key area method. With a large area of ±300 ha (0.03 km2), Bompon Watershed was 

designated as the key area because its land utilization was agroforestry, mainly in the volcanic foot slope 

area. It is a volcanic watershed on the borders of Magelang Regency, Purworejo Regency, and 

Wonosobo Regency, Central Java (Figure 1). It lies between 9163200 mN - 916400 mN and 396300 

mE - 397800 mE with an elevation between 377 and 539 m above sea level. Unevenly distributed rainfall 

with an annual average of 2,214.5 mm typifies its climatic characteristics. Bompon Watershed is located 

in the transition zone between the material deposition zones of the Tertiary and Quaternary Volcanoes 

on the foot slope of Sumbing Volcano. It experiences a volcanic intrusion that causes an intensive 

alteration process on the bedrocks. The intensive alteration and weathering processes result in a soil 

layer with a thickness of over 10 meters, which is categorized as super thick soil 

[24][25][26][27][28][29]. The vegetation cover is in the form of agroforestry, i.e., land use with diverse 

plant types like durian, coconut, green cottonwood, mahogany, Albizia chinensis, rosewood, Gnetum 

gnemon, Lansium dookoo, Lansium domesticum cv. Kokossan, jackfruit, bamboo, banana, Salacca 

zalacca, turmeric, Javanese turmeric, and cardamom. At the base of the tree stands, there are plants 

attached to the ground surface, namely grass and aromatic ginger. The vegetation cover is dense (478.78 

trees/m2), with a wide canopy spanning between 1-12 meters. The height variation of the vegetation 

stands forms a plant layering structure (multilayer canopy). 

The study used a key area method, which was carried out by measuring the suspension flow at the 

gully outlet; this measurement produced 436 suspension data. The suspension flow was measured at 

every rain event in the field and in the laboratory. The characteristics of the crop in the rain catchment 

area were recorded in detail during the field survey. In addition, the characteristics of the channels 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

143 

 

converging toward the gully system were observed in the field survey. These data were presented in 

tables and graphs (suspension hydrograph) to describe the causal relationship between rain phenomena 

and suspension flows.  

The suspension flow analysis was built based on rainfall and water level data. The rainfall data 

included the dynamics of rain depth, intensity, and duration at the initial formation of the suspension 

flow. The suspension was analyzed using the filtration method, which produced suspension weight and 

concentration data. The suspension discharge was obtained by multiplying the concentration of the 

suspensions by flow discharge, as Strand (1982) proposed in [30]. The flow discharge obtained for each 

water level observed at the outlet of a stream gauge with broad-crested weir was calculated using the 

Weir discharge equation [31]. 

 

 

Figure 1. The Location of Key Areas in the Study Site and the 

Geomorphological Unit Conditions in Bompon Watershed. 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

144 

 

3.  Result and Discussion 

3.1.  The correspondence pattern between flow discharge and suspension discharge 

During the rain events, the dynamics of the suspension flow showed correspondence patterns 

between flow discharge and suspension discharge in the rising and falling phases. At the rising stage, 

an increase in flow discharge was accompanied by a rise in the suspension discharge and vice versa, a 

decrease in flow discharge was followed by a decline in the suspension discharge (Figure 2). At the peak 

condition, three correspondence patterns between the flow discharge and the suspension discharge were 

identified. Namely, (1) the peak flow discharge corresponded to the peak suspension discharge, (2) the 

peak flow discharge preceded the peak suspension discharge, and (2) the peak flow discharge occurred 

after the peak suspension discharge.  

 

Table 1. The suspension conditions during the flows in each rain event 

No Rain events Number of data 
Rain intensity 

(mm/hour) 

Peak 
suspension 

concentration 

Cp 
(g/L) 

Peak runoff 

discharge Qp 

(L/s) 

Peak 

suspension 
discharge Qsp 

(g/s) 

Type 

tQp and 

tQsp 

1 2 3  5 6 7 11 

1 18 February 2017 4 8.94 0.0711 13.2745 0.9432 tQp=tQsp 
2 20 February 2017  10 8.88 0.2989 14.3273 4.2820 tQp>tQsp 

3 21 February 2017-1 7 5.14 0.0016 8.4806 0.0136 tQp<tQsp 

4 21 February 2017-2 9 4.44 0.0259 8.4806 0.2197 tQp=tQsp 
5 22 February 2017  6 6.86 0.0252 8.0463 0.2025 tQp<tQsp 

6 23 February 2017  2 1.80 0.0081 7.2035 0.0586 tQp=tQsp 

7 25 February 2017  28 10.71 0.9894 28.4727 28.1720 tQp=tQsp 
8 27 February 2017  2 4.40 0.0147 7.2035 0.1057 tQp=tQsp 

9 28 February 2017  25 19.55 2.3453 43.5071 102.0355 tQp=tQsp 

10 01 March 2017  2 6.60 0.1560 8.4806 1.3230 tQp=tQsp 
11 01 March 2017-2  31 29.25 3.0703 129.9952 399.1225 tQp=tQsp 

12 02 March 2017  34 22.34 1.9138 106.7442 204.2839 tQp=tQsp 

13 05 March 2017  13 12.74 0.5655 39.5293 1.9314 tQp>tQsp 
14 07 March 2017  11 27.16 0.3209 25.0950 8.0522 tQp>tQsp 

15 18 March 2017  15 25.34 0.6623 18.8337 12.4739 tQp=tQsp 

16 25 March 2017  10 16.80 0.9541 18.8337 17.9699 tQp=tQsp 
17 26 March 2017  28 36.39 2.8932 142.2171 411.4681 tQp=tQsp 

18 5 April 2017  39 75.08 2.7803 316.0675 878.7592 tQp<tQsp 

19 6 April 2017  23 26.45 0.8329 118.1667 98.4224 tQp<tQsp 
20 18 April 2017  13 29.18 0.9152 32.0076 29.2937 tQp=tQsp 

21 19 January 2018 3 10.58 0.1120 10.7772 1.2066 tQp<tQsp 

22 20 January 2018 7 3.45 0.8137 11.7526 9.5627 tQp=tQsp 
23 21 January 2018  9 5.64 0.1742 8.4806 1.4773 tQp>tQsp 

24 23 January 2018 2 4.60 0.2566 6.3958 1.6410 tQp=tQsp 

25 24 January 2018-1 5 1.80 0.1519 10.7772 1.6368 tQp>tQsp 
26 24 January 2018-2 21 9.60 2.1257 65.4142 139.0481 tQp=tQsp 

27 24 January 2018-3 4 2.52 0.2945 10.7772 3.1741 tQp=tQsp 

28 24 January 2018-4 4 3.84 0.2765 10.7772 2.9798 tQp<tQsp 
29 26 January 2018 3 3.75 0.0553 14.3273 0.7918 tQp>tQsp 

30 29 January 2018 3 7.80 0.1513 9.3745 1.4187 tQp=tQsp 
31 31 January 2018-1 5 21.42 0.6379 8.4806 5.4094 tQp=tQsp 

32 31 January 2018-2 6 16.80 0.1665 8.4806 1.4118 tQp<tQsp 

33 1 February 2018 7 22.14 1.2545 14.8651 18.6486 tQp=tQsp 
34 4 February 2018-1 5 20.88 0.7136 11.2610 8.0355 tQp=tQsp 

35 4 February 2018-2 14 17.28 1.1217 51.8768 58.1903 tQp<tQsp 

36 08 February 2018 5 12.48 4.7086 21.2571 100.0904 tQp=tQsp 
37 13 Februri 2018 21 45.12 2.7227 115.8493 315.4219 tQp=tQsp 

38 23 February 2018 12 38.40 0.9552 47.6241 45.4924 tQp>tQsp 

39 24 February 2018 5 16.56 0.8113 13.2745 10.7700 tQp=tQsp 
40 07 March 2018 11 15.10 2.2446 32.0076 71.8442 tQp=tQsp 

41 08 March 2018 16 49.69 3.7600 106.7442 401.3581 tQp=tQsp 

 Total 480      

  Min. 2 1.80 0.0016 6.3958 0.0136 0.4500 

  Mean 12 17.26 1.0330 41.1108 82.8962 4.31 

  Max. 39 75.08 4.7086 316.0675 878.7592 18.7692 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

145 

 

In general, the peak conditions showed a corresponding pattern between the peak flow discharge and 

the peak suspension discharge. Out of the 41 rain events observed in the field, (1) the peak flow 

discharge corresponded to the peak suspension discharge in 26 events, (2) the peak flow discharge 

preceded the peak suspension discharge in 7 events, and (3) the peak flow discharge occurred after the 

peak suspension discharge in 8 events (Table 1). 

The flow discharge condition substantially affects the suspension discharge; therefore, it determines 

the correspondence shape patterns generated by the peak flow discharge and the peak suspension 

discharge. The influence of flow discharge on suspension discharge is as reported in [32] [33] and [34]. 

The dynamics of runoff discharge are part of the balance system between the dynamics of rain input, 

infiltration capacity, and soil water storage [13]. The rain input triggers the formation of suspension 

flow that follows the dynamics of flow (runoff) formation when infiltration capacity and soil water 

storage are exceeded [12][14][15][35][36]. 

 

 

Figure 2. The types in the correspondence patterns of the peak suspension discharge and the 

peak flow discharge during the flows in each and several rain events (Rain event No. 11 in of 

the tQp=tQsp; Rain event No. 2 in of the tQp>tQsp; and Rain even No 19, in of the tQp<tQsp 

types) 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

146 

 

The response of runoff and suspension flow formation dynamics is identifiable by suspension 

hydrograph analysis. Hydrograph analysis illustrates the fluctuating movement of suspension flow 

following the dynamics of the rainfall characteristics [16][17][18][19]. Figure 2 shows that at the 

beginning of a rain event, the rain intensity and duration are still low, and, therefore, the flow discharge 

tends to be small. As the rain intensity and duration increase, the flow discharge and the peak suspension 

flow rise. Increased suspension flow discharge occurs because sediment production by raindrops and 

scouring (erosion) by water flows on the ground surface also increases. Increased suspension discharge 

at the same time as high flow discharge is caused by adequate suspension transport energy. At the 

recession stage, the rain intensity decreases and, therefore, the flow discharge declines. Decreased flow 

discharge reduces suspension production and sediment transport energy; accordingly, the suspension 

discharge becomes lower. 

3.2.  The time lag of the suspension flow formation 

The time lag from rain events to the initial formation of suspension flow at the outlet varies widely 

(Table 2). It ranges between 4 and 55 minutes with an average of 17.7 minutes and a standard deviation 

of ± 13 minutes. A high standard deviation indicates that the time lag from the beginning of the 

suspension flow formation and the rain events to the time at which the flows reach the outlet is highly 

diverse. In this research, the time lag of the initial formation of suspension flow is controlled by the 

response of the highly complex watershed condition, which includes the dynamics of the intensity and 

duration of the previous rain events, the channel’s base flow, the intensity dynamics of the occurring 

rain event. The statistical analysis results from the correlation test showed that the causal factors with 

significant correlation coefficients were rain intensity (0.381) and runoff discharge (0.443), whose 

significance levels were 0.05 and 0.001, respectively. The other factors, namely the dynamics of the 

time lag, the maximum intensity and duration of the previous rain events, and the base flow condition 

of the channel, had weak influence with insignificant correlation coefficients.  

This study generally revealed different time lags from rain events to the initial formation of the 

suspension flow as a combination of the dynamics of the intensity and duration of previous rain events, 

the state of the channel’s base flow, and the intensity dynamics of the occurring rainfall. At the beginning 

of the suspension flow formation, the rain intensity ranged from 1.2 to 97.2 mm/hour with an average 

of 15.6 mm/hour. The lowest rain intensity to initiate the formation of the suspension flow was 1.2 

mm/hour with a time lag of 8-10 minutes. At the beginning of the suspension flow formation, the highest 

rain intensity was 97.2 mm/hour, with a time lag of 24 minutes. Low rain intensity with a shorter time 

lag is possible when the previous rainfall has high intensity and long duration. The base flow is present 

at the channel during the occurring rain event (events no. 15 and 17). In rain events no. 15 and 17, the 

lowest rain intensity (1.2 mm/hour) resulted in shorter time lags, i.e., 8 minutes and 10 minutes, 

respectively, when the previous rain events had high intensities, i.e., 39.6 mm/hour and 32.4 mm/hour, 

and long durations, i.e., 90 minutes and 120 minutes, respectively. 

With high rainfall intensity of 22.14 mm/hours, the time lag was wide (slow) because the previous 

rainfall had low intensity and short duration and the base flow in the occurring rain event was low (event 

no. 33). In the event no. 33, the rain intensity was 97.2 mm/hour, but it had a long time lag, i.e., 24 

minutes, because of the low intensity (7.2 mm/hour) and short duration (15 minutes) of the previous 

rainfall. The time lag of the initial formation of suspension flow also varied even when the rain fell with 

the same intensity. In several observations (events no. 9, 16, 34), the rain intensities were equal (2.4 

mm/hour). However, the previous rain events had different intensities, i.e., 6 mm/hour, 4.8 mm/hour, 

and 97.2 mm/hour, and different durations, i.e., 256 minutes, 75 minutes, and 83 minutes, respectively, 

indicating different time lags, i.e., 10 minutes, 5 minutes, and 27 minutes, respectively. Events no. 15 

and 17 also had equal rain intensities, i.e., 1.2 mm/hour. However, the rain events preceding them had 

different intensities, i.e., 39.6 mm/hour and 32.4 mm/hour, and different durations, i.e., 90 minutes and 

120 minutes, which also resulted in different time lags, i.e., 8 minutes and 5 minutes.  



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

147 

 

Table 2. The dynamics of the aspects of previous rain events, occurring rain events, and the formation of suspension runoff 

No. Rain events Previous rain events Occurring rain events   Suspension flow events 

No. Rain events 

Time lag to 

previous 

rain (hour) 

Max. rain 

intensity 

(mm/hour) 

Rain 

depth 

(mm) 

Rain 

duration 

(minute) 

Rain 

intensity 

at the 

beginning 

of 

suspension 

flow 

(mm/hour) 

Rain 

depth at 

the 

initial 

flow 

The time lag 

between the 

initial 

formation of 

suspension flow 

and flow 

reaching outlet 

(minutes) 

Base 

flow 

condition 

(mm) 

Suspension flow 

events 

Runoff 

discharge 

(L/s) 

Suspension 

discharge 

(gram/s) 

1 2 3 4 5 6 7 8 9 10 11 12 13 

1 16/02/2017 21 1.2 2.1 15 2.6 4.2 - 35 Not formed  - - 

2 17/02/2017 24 3.6 5.1 15 1.2 0.3 - 35 Not formed  - - 

3 18/02/2017 18 1.2 0.3 15 15.6 9.6 10 32 Formed 11.75 0.08 

4 19/02/2017 27 4.8 15.6 45 2.4 1.5 - 5 Not formed - - 

5 20/02/2017 23 2.4 1.5 15 20.4 24.6 55 25 Formed 7.2 0.33 

6 21/02/2017 (1) 14 20.4 24.6 95 10.8 6 10 22 Formed 7.2 0.45 

7 21/02/2017 (2) 10 10.8 6 30 16.6 18.3 20 25 Formed 7.2 0.04 

8 22/02/2017 24 16.6 18.3 120 6 14.7 37 25 Formed 7.2 0.19 

9 23/02/2017 11 6 14.7 256 2.4 1.5 10 25 Formed 7.2 0.06 

10 24/02/2017 33 2.4 1.5 30 4.8 4.5 - 19 Not formed  - - 

11 25/02/2017 16 9.6 8.7 135 14.4 24.9 28 22 Formed 7.2 0.04 

12 26/02/2017 18 14.4 24.9 225 4.8 4.5 12 05 Not formed  - - 

13 27/02/2017 20 4.8 4.5 90 7.2 4.8 8 25 Formed 7.2 0.11 

14 28/02/2017 13 7.2 4.8 60 6 1.5 14 25 Formed 8.48 0.52 

15 01/03/2017 (1) 15 39.6 25.5 90 1.2 0.3 10 25 Formed 8.48 1.32 

16 01/03/2017 (2) 2 4.8 6 75 2.4 0.6 5 25 Formed 8.48 1.87 

17 02/03/2017 20 32.4 51.3 120 1.2 1.2 8 30 Formed 8.48 0.99 

18 04/03/2017 40 33.6 36 105 2.4 1.8 - 25 Not formed  - - 

19 05/03/2017 28 2.4 2.4 45 2.7 10.8 4 35 Formed 13.27 2.08 

20 06/03/2017 24 21.6 9.9 60 28.4 1.4 - 01 Not formed  - - 

21 07/03/2017 23 26.4 6.9 15 49.2 12.3 14 30 Formed 10.78 1.27 

22 14/03/2017 (1) 72 12 4.8 15 6 1.5 - 5 Not formed  - - 

Continued to the next page 

 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

148 

 

Table 2 (continued) 
1 2 3 4 5 6 7 8 9 10 11 12 13 

23 14/03/2017 (2) 2 6 2.1 15 2.4 1.5 - 5 Not formed  - - 

24 14/03/2017 (3) 1 9.6 12.9 60 2.4 2.4 - 5 Not formed  - - 

25 17/03/2017 48 28 21.3 75 2.4 2.2 - 32 Not formed  - - 

26 18/03/2017 17 4.8 2.4 90 28.8 7.2 39 - Formed 1.58 1.41 

27 20/03/2017 21 40.8 57 150 4.8 1.8 - - Not formed  - - 

28 24/03/2017 72 18 6.9 45 2.4 1.5 - 22 Not formed  - - 

29 25/03/2017 16 2.4 2.4 30 5.1 20.4 10 15 Formed 4.54 0.54 

30 26/03/2017 11 30 42 120 9.6 38.4 14 15 Formed 4.54 0.94 

31 04/04/2017 (1) 15 2.4 19.8 75 8.4 2.4 - - Not formed  - - 

32 04/04/2017 (2) 8 8.4 3.1 15 7.2 1.8 - - Not formed  - - 

33 05/04/2017 19 7.2 5.7 15 97.2 24.3 24 15 Formed 18.83 11.13 

34 06/04/2017 19 97.2 74.4 83 2.4 0.6 27 15 Formed 4.54 0.38 

35 17/04/2017 120 2.4 1.2 15 16.8 5.4 - - Not formed  - - 

36 18/04/2017 17 16.8 5.4 15 22.8 5.7 12 - Formed 0.55 0.02 
 Mean 24.5 15.3 14.8 68.7 11.7 7.4 17.7 21   7.7 1.2 
 Max. 120 97.2 74.4 256 97.2 38.4 55 35   18.83 11.13 
 Min. 1 1.2 0.3 15 1.2 0.3 4 01   0.55 0.02 
 Sdev. 22.6 18.2 17.6 58.6 17.9 9.1 13.0 10   4.0 2.4 

Source: Field data processing (2017) 

 

 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

149 

 

This study argued that the time lag of the initial formation of the suspension flow was influenced 

dynamically by a combination of factors, including the intensity and duration of previous rain events, 

the intensity of the occurring rainfall, and the base flow of the channel. The combination of these factors 

significantly controls the initial formation of suspension flow and is supported by the concept of flow 

formation mechanism. The flow formation mechanism explains that a flow forms when the soil surface 

receives rain input and absorbs water vertically into the soils by infiltration and percolation. As well as 

when these processes affect the initial humidity status of the soil surface [36], make the soil water-

saturated, and form runoff on the soil surface in the event of excess rainfall [24][22][35]. 

This study provides the time lag between the initial formation of suspension flow and rain event, 

which tends to be longer (17.7 minutes) in a volcanic watershed with super thick soils and land cover in 

the form of agroforestry. The super thick soil in the volcanic watershed has a high clay content (> 50%), 

and it can bind and store more water; therefore, suspension flow takes a longer time to form [40]. The 

plant root density under the agroforestry practices is beneficial in initiating the formation of fractures 

and secondary soil pores to increase infiltration and soil water storage capacity. Soil conditions with 

high infiltration and soil water storage capacity can slow down the process of runoff (flow) formation 

that carries suspension on the surface, as previously reported by [16][37]. According to [38], soil 

permeability reflects soil infiltration capacity. The permeability of the super thick soils in the study area 

ranged from 0.0259 cm/hour to 80.0759 cm/hour with an average of 0.4492 cm/hour (slow). Slow soil 

permeability triggers a faster formation of suspension flow on the soil surface. In line with [39], the 

combination of initial water content and infiltration affects the initial infiltration rate. When the need to 

reach the initial soil moisture content is higher, the lower the initial infiltration rate. Suspension flow 

forms when rain input exceeds infiltration capacity; in other words, the formation of suspension flow is 

highly dependent on the fulfillment of infiltration capacity.  

The time lag between the suspension flow formation and rain event in this research was wide (17.7 

minutes). The wide time lag proves that the vegetation roots in agroforestry land use, which create 

fractures and soil pores, can increase water absorption in infiltration and soil water storage capacity. The 

super thick soil condition with high infiltration capacity and large soil water storage in the volcanic 

watershed can store more water and slow the formation of suspension flow.  

3.3.  The characteristics of the grain size of the suspension  

The results presented in Table 1 show that the suspension level at the peak suspension flow discharge 

has a concentration ranging from 0.0016 g/L to 4.71 g/L with an average of 1.03 g/L. The most negligible 

suspension concentration was 0.0016 g/L, which occurred in the event no. 2 with a peak runoff discharge 

of 8.48 L/s. The highest suspension concentration was 4.71 g/L, which happened in rain event no. 36 

with a peak runoff discharge of 21.26 L/s. In general, the study shows that the peak flow discharge 

influences the suspension level, i.e., low peak flow discharge has a small suspension concentration. In 

contrast, high peak flow discharge has a large suspension concentration. The situation in which peak 

flow discharge affects suspension content is in line with the results of previous studies, including the 

correlation between suspension’s grain size and flow conditions, namely flow discharge and rate 

[41][42]. 

In the study area, the grain size of the suspension in the suspension flow is dominated by clay fraction. 

The dominant clay fraction corresponds to the clay fraction found in the surface soil layer in the volcanic 

watershed area. The grain size of the suspension was grouped according to the percentages of the clay, 

silt, and sand fractions whose granular scales are <0.002 mm, 0.002-0.02 mm, and 0.02-2 mm, 

respectively. The grain size of the suspension at the rising stage had the following mean percentages: 

2% sand, 26% silt, and 73% clay. At the recession stage, the grain size of the suspension was composed 

of the following mean percentages: 3% sand, 28% silt, and 69% clay. The fraction of the surface soil 

layer consisted of the following mean percentages: 2% sand, 26% silt, and 73% clay. During the flow 

events, the clay-sized suspension fraction showed a decrease from 73% at the rising stage to 69% at the 

falling phase. 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

150 

 

Meanwhile, the sand-sized suspension increased from 2% to 3%, and the silt-sized suspension also 

increased from 26% to 29%. The increases in the sand and silt-sized fractions and the decline in clay-

sized grain in the suspension during the flow events not only indicate different levels of transportability 

between the sand, silt, and clay-sized suspension but also demonstrate an increase in the transportation 

of sediment originating in the channel during the flow events. Clay fractions are mostly transported in 

the rising phase when flow discharge is low because clay has a very small and delicate size that is more 

easily suspended and transported. Meanwhile, the silt and sand-sized fractions are mostly carried in the 

recession phase when flow discharge accumulates and becomes larger. Silt and sand are larger and 

coarser so that their disaggregation and transportation processes require stronger flow discharge energy 

and more prolonged time [43][44][45]. The increased silt and sand contents in the suspension during the 

flow events indicate that the increased flow discharge can trigger intensive soil disaggregation and 

sediment transport. This research argues that significant accumulation of flow discharge prompts 

intensive suspension transport and flow in a volcanic watershed area, which occurs not only on soils 

with clay content but also on sand and silt fractions. 

4.  Conclusions 

During the rain events, the dynamics of the suspension flow showed correspondence patterns 

between flow discharge and suspension discharge in the rising and falling phases. At the peak condition, 

there are three relationship patterns between the peak flow discharge and the peak suspension discharge, 

namely (1) the peak flow discharge corresponds to the peak suspension discharge, (2) the peak flow 

discharge precedes the peak suspension discharge, and (3) the peak flow discharge occurs after the peak 

suspension discharge.  

The time lag from rain events to the formation of suspension flow ranges from 4 minutes to 55 

minutes, with an average of 17.7 minutes. The wide time lag proves that the vegetation roots in 

agroforestry land use, which create fractures and soil pores, can increase water absorption in infiltration 

and soil water storage capacity. The peak suspension content varies between 0.0016 g/L and 4.71 g/L 

with an average of 1.03 g/L. The grain size of the suspension is mainly from clay fraction with a range 

of 71% to 76%, and the average is 73% in the rising phase. Furthermore, the recession phase ranges 

between 68% and 71%, with an average of 69%. Silt and sand fractions in the suspension are averagely 

26% and 2% in the rising phase and 28% and 3% at the recession phase.  

Acknowledgments 

The authors would like to express their gratitude to all parties, especially the Transbulent Team for 

the togetherness during data collection in the field until the realization of this publication material. The 

authors would also like to thank LPDP (Lembaga Pengelola Dana Pendidikan―Indonesia Endowment 

Fund for Education) for the financial support during the course of this Education Program. 

References 

[1] Badan Geologi Indonesia, 2011, Data Dasar Gunung Api Indonesia, Edisi ke-2, Kementrian 

Energi dan Sumber Daya Mineral, Bandung. 

[2] Asriningrum, W., Noviar, H., & Suwarsono. 2004. Pengembangan Metode Zonasi Daerah Bahaya 

Letusan Gunungapi Studi Kasus Gunung Merapi. Jurnal Penginderaan Jauh dan Pengolahan 

Data Citra Digital, 1(1), pp.66–75.  

[3] Nandini, R & Narendra, B.H. 2012. Karakteristik Lahan Kritis Bekas Letusan Gunung Batur di 

Kabupaten Bangli, Bali. Penelitian Hutan dan Konservasi Alam, 9(3), pp.199–211. 

[4] Bachri, S., Utaya, S., Nurdiansyah, F.D., Nurjanah, A.E., Tyas, LWN., Purnama, D.S., & Adillah, 

A.A. 2017. Analisis dan Optimalisasi Potensi Lahan Pertanian sebagai Kajian Dampak Positif 

Erupsi Gunungapi Kelud 2014. Majalah Geografi Indonesia, 1790.  

[5] Kimmins, J.P., Rempel, R.S., Welham, C.V.J., Seely, B., & Van Rees, K.C.J. 2007. Biophysical 

sustainability, process-based monitoring and forest ecosystem management decision support 

systems. The Forestry Chronicle, 83(4), pp.502–514. 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

151 

 

[6] Verstraeten, G., Prosser, I.P., & Fogarty, P., 2007. Predicting the spatial patterns of hillslope 

sediment delivery to river channels in the Murrumbidgee catchment, Australia. Journal of 

Hydrology, 334(3-4), pp.440–454. 

[7] Kironoto, B.A., 2008. Konsentrasi Sedimen Suspensi Rata-Rata Kedalaman Berdasarkan 

Pengukuran 1, 2, dan 3 Titik pada Aliran Seragam Saluran Terbuka. Dinamika Teknik Sipil, 

8(1), pp.59–71. 

[8] Panagos, P, Borrelli, P, Poesen, J, Ballabio, Lugato, E., Meusburger, K., Montanarella, L.,  & 

Allewl, C. 2015. The new assessment of soil loss by water erosion in Europe. Environmental 

Science & Policy, 54, 438–447. http://doi.org/10.1016/j.envsci.2015.08.012. 

[9] Suripin, 2002. Pelestarian Sumber Daya Tanah dan Air. Andi Offset. Yogyakarta. 

[10] Merritt, W.S., Lecther, R.A., & Jakeman, AJ. 2003. A review of erosion and sediment transport 

model. Environment Model Software, 18: 761-799. 

[11] Ma’wa, J., & Andawayanti, U. 2009. Studi Pendugaan Sisa Usia Guna Waduk Sengguruh 

Dengan Pendekatan Erosi dan Sedimentasi. 

[12] Parsons, A.J., & Wainwright, J. 2000. Modeling Surface Runoff. In Schmidt, J. (ed). Soil Erosion, 

Application of Physically Based Models. Germany: Springer. 

[13] Handayani, Y.L., Jayadi, R., & Triatmojo, B., 2005. Optimasi Tata Guna Lahan dan Penerapan 

Rekayasa Teknik dalam Analisa Banjir di Daerah Aliran Sungai: Studi Kasus Daerah Aliran 

Sungai Ciliwung Hulu Di Bendung Katulampa. Manusia dan Lingkungan, 12(2), pp.53–61. 

[14] Oktarina, N.R. 2005. Analisis Hidrograf Limpasan Akibat Variasi Intensitas Hujan dan 

Kemiringan Lahan (Kajian Laboratorium dengan Simulator Hujan). Jurnal Teknik Sipil dan 

Lingkungan. Vol. 3, No. 1, Maret 2015. 

[15] Walker, S., & Mostaghimi, S. 2009. Watershed-Based Systems. In Moore, K.M. (ed). The 

Sciences and Art of Adaptive Management Innovating for Sustainable Agriculture and Natural 

Resource Management.  Ankeny, Iowa: Soil and Water Conservation Society. 

[16] Handayani, W., & Indrajaya, Y. 2011. Analisis Hubungan Curah Hujan dan Debit Sub Sub DAS 

Ngatabaru, Sulawesi Tengah. Jurnal Penelitian Hutan dan Koservasi Alam, Vol. 8, No.2, 

pp.143–153. 

[17] Bisantino, T., Bingner, R., Chouaib, W., Gentile, F., & Liuzzi, G.T. 2013. Estimation of runoff, 

peak discharge and sediment load at the event scale in a medium-size Mediterranean watershed 

using the annAGNPS model. Land Degradation & Development. Available at: 

10.1002/ldr.2213. 

[18] Miller, J.R. Mackin, G., & Miller, S.M.O. 2015. Application of Geochemical Tracers to Fluvial 

Sediment, London: Springer. 

[19] Gao, P., Deng, J.,Chai, X., Mu, X., Zhao, G., Shao, H., & Sun, W. 2017. Science of the Total 

Environment Dynamic sediment discharge in the Hekou – Longmen region of Yellow River 

and soil and water conservation implications. Science of the Total Environment, The, 578, 

pp.56–66. Available at: http://dx.doi.org/10.1016/j.scitotenv.2016.06.128. 

[20] Hergarten, St., Paul, G., & Neugebauer, H.J. 2000. Modeling Surface Runoff. In Schmidt, J. (ed). 

Soil Erosion, Application of Physically Based Models. Germany: Springer. 

[21] Poesen, J., Nachtergaele, J., Verstraeten, G., & Valentina, C. 2003. Gully erosion and 

environmental change: importance and research needs. Catena, 50, 91-133. 

http://dx.doi.org/10.1016/S0341-8162(02)00143-1. 

[22] Arsyad, S. 2006. Konservasi Tanah dan Air. Bandung: IPB Press. 

[23] Fryirs, K.A. & Brierley, G.J. 2013. Geomorphic Analysis of River Systems 1st ed., A John Wiley 

& Sons, Ltd., Publication. 

[24] Soemarto, C.D. 1999. Hidrologi Teknik. Pusat Pendidikan Manajemen dan Teknologi Terapan. 

Malang. 

[25] Dariah, A., Subagyo, H., Tafakresnanto, S., & Marwanto, S. 2003. Kepekaan Tanah terhadap 

Erosi. Jurnal Akta Agrosia Vol. 8, No.2. 

[26] Morgan, R.P.C. 2005. Soil Erosion and Conservation: Third Edition. USA; Blackwell. 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

152 

 

[27] Nicótina, L. et al. 2011. Hydrologic controls on equilibrium soil depths. Water Resources 

Research, 47(4), pp.1–11. 

[28] Rusdi, Alibasyah, M. R., & Abubakar, K. 2013. Evaluasi Degradasi Lahan diakibatkan Erosi pada 

Areal Pertanian di Kecamatan Lembah Seulawah Kabupaten Aceh Besar. Jurnal Konservasi 

Sumber Daya Lahan. Pascasarjana Universitas Syiah Kuala, Vol. 1, No. 1, Mei 2013, 1(1), 

24–39. 

[29] Sartohadi, J. 2013. Genesis Tanah Supertebal dan Kaitannya dengan Longsor dalam di Hulu DAS 

Bogowonto Jawa Tengah. Hibah Penelitian Dosen. LPPM UGM Yogyakarta. 

[30] Wulandari, D.A., Suripin, and Syafrudin. 2014. Evaluasi Penggunaan Lengkung Laju Debit-

Sedimen (Sediment-Discharge Rating Curve) Untuk Memprediksi Sedimen Layang. 

http://eprints.undip.ac.id/4670/DYA. 

[31] Herschy, R.W., 2009. Streamflow Measurement Third edit. T. & Francis, ed., 2 Park Square, 

Milton Park, Abingdon, Oxon OX14 4RN. 

[32] Soewarno. 1991. Hidrologi Pengukuran dan Pengukuran Daerah Aliran Sungai. Bandung: Nova. 

[33] Arianti, F.D., Suratman, Martono, E., & Suprayogi, S. 2012. Dampak Pengelolaan Lahan 

Pertanian terhadap Hasil Sedimen di Daerah Aliran Sungai Galeh Kabupaten Semarang. 

Jurnal Manusia dan Lingkungan. Vol. 19, No. 3. Available at: http://i-

lib.ugm.ac.id/jurnal/detail.php?dataId=12641 [Accessed September 19, 2015]. 

[34] Maulana, R.A., Lubis, K.S., & Marbun, P. 2014. Uji Korelasi antara Debit Aliran Sungai dan 

Konsentrasi Sedimen Melayang pada Muara Sub DAS Padang di Kota Tebing Tinggi. Jurnal 

Online Agroekoteknologi, 2(2337), pp.1518–1528. 

[35] Triatmodjo, B. 2013. Hidrologi Terapan. Cetakan ke-3, Beta Offset, Yogyakarta. 

[36] Neno, A.K. et al. 2016. Hubungan Debit Air dan Tinggi Muka Air di Sungai Lambagu Kecamatan 

Tawaeli Kota Palu. Warta Rimba,Vol. 4 Nomor 2 Desember 2016, 4, pp.1–8. 

[36] Wang, J., Huang, J., Wu, P., Zhao, X., Gao, X., Dumlao, M., Si, B.C. 2015. Effects of soil 

managements on surface runoff and soil water content in jujube orchard under simulated 

rainfalls. Catena, 135, pp.193–201. Available at: 

http://www.sciencedirect.com/science/article/pii/S0341816215300795 [Accessed September 

16, 2015]. 

[37] Gumiere, S.J., Bailly, J.S., Cheviron, B., Raclot, D., Bissonnais, Y.L., & Rousseau, A.N. 2015. 

Evaluating the Impact of the Spatial Distribution of Land Management Practices on Water 

Erosion: Case Study of a Mediterranean Catchment. J. Hydrol. Eng., 2015, 20(6): C5014004, 

20(2004), pp.1–10. 

[38] Haridjadja O, Murtilaksono, K., Sudarmo, Rahman, L.M. 1990. Hidrologi Pertanian. Jurusan 

Tanah. Fakultas Pertanian. Bogor (ID): Institut Pertanian Bogor. 

[39] Asdak, C. 2002. Hidrologi dan Pengelolaan Daerah Aliran Sungai. Yogyakarta: Gadjah Mada 

University Press. 

[40] Mbaya, L.A., Ayuba, H.K., & Abdullahi, J. 2012. An assessment of gully erosion in Gombe 

Town, Gombe State. Journal of Geography and Geology, 4(3), pp.110–122. 

[41] Steegen, A., Govers, G., Nachtergaele, J., Takken, I., & Poesen, J. 2000. Sediment export by 

water from an agricultural catchment in the Loam Belt of central Belgium. Geomorphology, 

33, pp.25–36. 

[42] Tillinghast, E.D., Hunt, W.F., & Jennings, G.D., 2011. Stormwater control measure (SCM) design 

standards to limit stream erosion for Piedmont North Carolina. Journal of Hydrology, 411(3-

4), pp.185–196. Available at: http://dx.doi.org/10.1016/j.jhydrol.2011.09.027. 

 [43] Castillo, V.M., Mosch, W.M., García, C.C., Barberá, G.G., Cano, J.A.N., & Bermúdez, F.L. 2007. 

Effectiveness and geomorphological impacts of check dams for soil erosion control in a 

semiarid Mediterranean catchment: El Cárcavo (Murcia, Spain). Catena, 70(3), pp.416–427. 

Available at: http://www.sciencedirect.com/science/article/pii/S0341816206002438 

[Accessed on September 16, 2015]. 



 

 

 

 
Civil and Environmental Science Journal 

Vol. 4, No. 2, pp. 141-153, 2021 

 

 

 

153 

 

[44]  Haregeweyn, N., Melesse, B., Tsunekawa, A., Tsubo, M., Meshesha, D., & Balana, B.B. 2012. 

Reservoir sedimentation and its mitigating strategies: A case study of Angereb reservoir (NW 

Ethiopia). Journal of Soils and Sediments, 12(2), pp.291–305. 

[45] Nugroho, S.H. dan Basit, A., 2014. Sebaran Sedimen Berdasarkan Analisis Ukuran Butir di Teluk 

Weda, Maluku Utara. Jurnal Ilmu dan Teknologi Kelautan Tropis, 6(1), pp.229–240.