Road sediment production and delivery in khang ninh commune, ba be, bac kan

MINISTRY OF AGRICULTURE AND RURAL DEVELOPMENT VIETNAM FORESTRY UNIVERSITY STUDENT THESIS ROAD SEDIMENT PRODUCTION AND DELIVERY IN KHANG NINH COMMUNE, BA BE, BAC KAN Major: Advanced Curriculum in Natural Resources Management Code: D850101 Faculty: Forest Resources & Environmental Management Student: Hua Huy Luan Class: K55 Natural Resources Management Student ID:1054030811 Course: 2010 - 2014 Advanced Education Program Developed in Collaboration with Colorado State University, USA Supervisor: Assoc Prof Dr Bui The Doi Co-supervisor: Prof Lee H.MacDonald Hanoi, November, 2014 ACKNOWLEDGEMENTS I would like to express my very great appreciation to Prof Lee MacDonald from Colorado State University and Prof Bui The Doi from Vietnam Forestry University, my research supervisors, for their enthusiastic encouragement and useful critiques of this research work Without their thoughtful, patient guidance through every step of my research, their knowledge on how to correct and analyze the data I collected and the financial support from Prof Lee, this thesis could not have been realized Thank to Vietnam Forestry University and Colorado State University cooperation together especially faculty of Forest Resource and Environmental Management and all professors I would like to extend my thanks to Mr Bui Van Quang, the Vice Director of Ba Be National Park for accepting the responsibility of my research in his national park as well as for his useful and constructive recommendations on this project Many other people have contributed to the success of this thesis but I would like to particularly thank Hua Van Bac, Hua Thi Hop and Kim-Sara Wagner for their support Last but not least this thesis could have never been completed without the support of my mother Mach Thi Noi who always encouraged me during the entire process of doing research and who let me borrow several tools and her motorbike for collecting all data in the field ABSTRACT This study used sediment fences to measure sediment production from road segments for over one month in the wet season in Khang Ninh Village, Ba Be, Bac Kan, Vietnam The mean sediment production rate from native surface roads was 0.60 kg/m² Comparisons among segments showed that recently-graded native surface roads produced more sediment than ungraded native surface roads The sediment production depended on the amount of precipitation, segments slope and especially such as sediment production on native surface roads was best predicted by the product of road area times road slope (A*S) Road sediment production can be reduced by rocks and vegetation, increasing the frequency of road drainage structures, avoiding locations that generate more road surface and ditch runoff, and minimizing grading and traffic Sediment delivery was assessed by a survey of road segments along km of an unpaved road 819/m of the surveyed 2000m of this road are directly connected to the stream Gully initiation increased with road segment length, side-slope gradients, road designs that concentrated road runoff and infiltration capacity of the soil Road sediment delivery can be minimized by the construction of a drain-ditch or reducing the amount of runoff processes on the road surface CONTENTS INTRODUCTION GOALS AND OBJECTIVES METHODOLOGIES 3.1 Sediment Production 3.2 Sediment Delivery 3.3 Survey Surface at Road Segments 3.3.1 Diameter of Rocks, Gravel and Sand 3.3.2 Bulk Density 3.4 Comparision to the WEPP Road model 3.5 Site Description RESULTS 4.1 Road segment characteristics 4.2 Precipitation 4.3 Sediment Production 4.4 Sediment Delivery 13 4.5 Survey of Surface at Road Segments 14 4.5.1 Diameter of Rocks, Gravel, Sand or Soil 14 4.5.2 Bulk Density 15 4.6 Comparision to the WEPP Road model 16 DISCUSSION 17 5.1 Experimental Setup Performance 17 5.2 Data Analysis Performance 17 5.3 Comparisons to Previous Studies 18 5.4 Future Research 18 CONCLUSIONS 20 REFERENCES APPENDICES INTRODUCTION Unpaved roads can contribute a lot of sediment to streams in each wet season, and accumulated road surface erosion in large rain events can have catastrophic effects, such as filling in pools and reducing habitat complexity Road erosion can have a major impact on stream health with very big amount of sediment, especially surface erosion from unpaved roads and the adjacent drainage ditches After heavy rain erosion will occur and then at the end almost all sediment goes to streams causing more turbidity and sediment concentrations, and then it does not only change the morphology of the roads but also changes the size of the rivers or lakes making all of them shallower and narrower Of course it will have a bad effect on water quality and plants and animals in the water Previous studies have identified that unpaved roads can contribute 50 to 80% of the sediment that enters streams (Hagans et al., 1986) The amount of sediment delivered from forests with roads can be more than 300 times greater than from undisturbed forest land (Morrison, 1975) Roads located close to forest land and near rice paddies as well as those leading to rural and suburban parcels may also contribute to sediment problems in a watershed Data on road erosion and sediment delivery rates are critical for assessing road impacts on aquatic resources, and a sound understanding of road erosion processes is needed to minimize road sediment production (Coe, 2006.) Bac Kan is located in the northeast of Vietnam, 240 km from Hanoi It is a developing region of Vietnam, mostly located in high mountains, and the unpaved roads and poor drainage create big amount of road erosion each year Especially the steep roads have a lot of sediment production that deposits to the rice fields, farms, and the river, all of it has bad consequences for the farmers To evaluate the consequences of road erosion it is necessary to make some initial research on road erosion at a specific location This thesis study was in Khang Ninh Commune - Ba Be – Bac Kan Province, as there has not been any study on road sediment production and sediment delivery in this region So this thesis will provide information on road surface erosion and sediment delivery GOALS AND OBJECTIVES The goal of this project is to measure road sediment production and delivery rate in Khang Ninh Commune, Ba Be, Bac Kan The objectives were to: 1) Quantify sediment production from three road segments for at least 10 rains on the road 2) Determine sediment delivery from road segments for at least 10 rains 3) Determine the different effects of surface cover (rocked, graveled and noncover) on road sediment production and delivery 4) Compare the measured road sediment production in Khang Ninh Commune, Ba Be, Bac Kan to predictions from the WEPP (Water Erosion Prediction Model): Road model 3.1 METHODOLOGIES Sediment Production Sediment production rates from three road segments were measured with sediment fences Sediment fences constructed at the drainage point of hill slopes or discrete road segments that have clearly defined contributing areas (Lee, 2007) They are made of sacks stitched together and attached to 4-6 poles that are pounded 0.3-0.5 m into the ground The leading edge of the sack is attached to the ground with landscape staples to prevent underflow and also a floor of fabric is put on the ground so it is easy to remove the sediment Cleaning out the fence in each rain sediment production is measured by sediment and weighing it in kilograms Measurements were taken during 10 rains and the time is marked to know how long the rain took Rainfalls were measured by using a Vietnamese rain gage After each rain sediment productions are measured by using a scale (kg) A sample of 0.3 Kg was taken after each rain at each of the sediment fences and spread out in a pan to let it air dry inside a room The collected data is the basis for following calculations: we have wet-soil minus dry-soil and then calculate the percentage of water stored in the wet-soil (formula: (Water /300 g) * 100 = % of water) To measure how many percent of sediment are in the water the soil has to be dried in a dry room which is not affected by wind or other activities until the soil is completely dry and can be split easily Erosion at a segment are calculated by: Length * Width = Area (m2) Mass / Area = erosion (kg) 3.2 Sediment Delivery Sediment delivery will be measured at road segments within 2km2 in Khang Ninh Commune The discrete point will be determined and then sediment delivery will be measured from the road to the end of sediment visibility Tracking runoff and sediment from each drainage point is necessary to see where it goes Sediment delivery equals the total road area connected to the stream multiplied by the amount of soil (kilogram) that eroded in each rain For measuring sediment delivery the collected data was only compared to the data of sediment fence The reason is that only sediment fence is located on the investigated road All along the road the physical characteristics are very similar so a comparison of the date was very easy Sediment fences and lie on smaller roads which still belong to the investigated road, however they show different characteristics (soil texture, bulk densities and width) 3.3 Survey Surface at Road Segments 3.3.1 Diameter of Rocks, Gravel and Sand The diameter of rocks, gravel and sand covering the road were measured at 50 equally- spaced points in each road segment In particular, it means that a measuring tape used to identify 50 points on the road, each of them has to be 50cm apart from the previous one At each point a sample of road cover which can either be a rock, gravel or soil were taken to measure the diameter with a common ruler This procedure were performed at all three road segments with a total of 150 measurements 3.3.2 Bulk Density Bulk density (p) = mass of dried soil/ Total volume: in this thesis a bamboo tube with the height of 10cm and a ratio of 2.5cm will be used For each road segment bulk density will be calculated by the following steps: Clean soil surface where the soil sample will be taken Use the hammer to pound into ground until the bamboo tube is completely filled with soil It is important to put a piece of wood on the bamboo tube before hammering The piece of wood will distribute the pressure equally so that the soil will not be compressed by the hammer Use the shovel to dig out the bamboo tube and put the soil into a plastic bag and then dry the soil in a dry room to measure the weight With the mass of dry soil and total volume, following calculation will be done: π*r2*h = V  Precipitation data will be collected with a Vietnamese rain gage in 10 rains For each rain the time raining started and stopped will be noted to know how long the rain lasted  Use a clinometer to measure the slop segments and measuring tape to measure the length and width of segments 3.4 Comparision to the WEPP Road model In the WEPP Road Erosion Predictor, Modeling 50 years to get a good average Fill slope as little as possible (0.1 ft) and the same slope as the road, and the buffer also be minimal (0.3 foot) All of the information fill into the WEPP road modeling was the same condition in the road segment 1,2,3 (gradient, length, width, surface, traffic, soil texture and climate) The set of tools that use to were developed by researchers at the USDA Forest Service (USFS) These tools are all based on the physically-based WEPP (Water Erosion Prediction Model), which was developed as an alternative to the better known, but more empirical, Revised Universal Soil Loss Equation (RUSLE) Since the WEPP model is extremely complex and requires several hundred parameters to run, the USFS has developed a series of simplified user interfaces to predict erosion from forests, roads, thinning, fires, peak flows, and the effectiveness of post-fire rehabilitation techniques Each interface prompts the user to enter a few key variables, and then these user-specified variables are used to estimate all the other variables in the WEPP model (http://forest.moscowfsl.wsu.edu/fswepp/) After run the WEPP: Road Model got the results then Used the precipitation in place, where I made study devided total precipitation for 50 years and times the road prism erosion, apply to road segments Took the results compare to the road erosion the Ba Be, Bac Kan province (http://forest.moscowfsl.wsu.edu/cgibin/fswepp/wr/wepproadpl) 3.5 Site Description The study area is a forest road located in Na Kieng Village near Ba Be National Park in Vietnam Na Kieng Village belongs to Khang Ninh Commune, which is one of seven communes of Ba Be district surrounding Ba Be lake, the largest natural lake in Vietnam Image 1: Location of Ba Be National Park, Vietnam The study area lies about 14 Km west of the town of Cho Ra, approximately 70 Km northwest of the provincial capital Bac Kan and 240 Km north of Hanoi, the capital of Vietnam The exact coordinates are the following: (22.233°N 105.566°E) and (22.5166°N 105.733°W) From the table and graphs above the different characteristics of each road segment and the relationship between precipitation and the amount of sediment production per rain can be seen We can see that the sediment fence has the largest area: 201 m2 but the total erosion is the smallest: 998 kg because of the lowest slope compared to the other fence 12° from road segments With slope 15°, more sediment production can be noted in sediment fence Total area is 192.5m2 and the total of sediment is 1298 kg At sediment fence had the highest slope of 22°, the smallest area but produced the largest amount of sediment (1046 kg) From those numbers it can be said that sediment production depends on precipitation and is most affected by the slope 10 Table 2: percentage of actual eroded soil and water Rain Water content of sediment Water content of sediment Water content of sediment fence fence fence Wet- Dry- Percent of Wet- Dry- Percent Wet- Dry- Percent of soil soil(g) Water soil(g) soil(g) of soil(g) soil(g) Soil (g) Water Water 300 220 27 300 190 37 300 210 30 300 215 28 300 200 33 300 220 27 300 215 28 300 195 35 300 225 25 300 215 28 300 190 37 300 195 35 300 210 30 300 190 37 300 210 30 300 210 30 300 195 35 300 225 25 300 210 30 300 190 37 300 210 30 300 215 28 300 195 35 300 195 35 300 220 27 300 200 33 300 210 30 10 300 220 27 300 200 33 300 210 30 According to table and formulas: Length * Width = Area (m2) and Mass / Area = erosion (kg) road erosion is calculated per m2 at sediment fences 11 Table Road erosion per m² at sediment fences Road erosion per m2 at sediment fences Total Sediment/ Total Sediment/ Total Sediment/ Total are(kg) Total are (kg) Total are (kg) Sediment fence Sediment fence Sediment fence3 Precipitation No (mm) 6.9 0.07 0.12 0.10 20.5 0.54 0.83 0.73 8.6 0.09 0.14 0.12 9.2 0.13 0.31 0.15 148.5 1.62 1.86 2.06 17 0.49 0.76 0.61 98 0.67 0.93 0.88 60.5 0.63 0.81 0.82 32 0.59 0.67 0.70 10 9.6 0.139 0.32 0.18 Total 410.8 4.969 6.75 6.35 12 From rain intensity compare to the sediment production of three sediment fences (kg/m2) Showing that if the rain was short and heavy rain had much sediment produced, with high rain intensity were have high runoff processes because the water cannot infiltration much into the soil 4.4 Sediment Delivery Within the road segments there are segments which are connected directly to the stream with a total area of 273* = 819m, and road segments with a total area of 147 * = 441m were leading directly to the rice paddy and pond with very bad consequences for the farmers The calculation was done in kilogram of sediment for each rain that was connected by: Road erosion per m² in the sediment fence times 819m (total area connected) in order to have sediment delivery in kilogram, shown in the chart Figure 8: Sediment delivery compare to sediment fence 13 4.5 Survey of Surface at Road Segments 4.5.1 Diameter of Rocks, Gravel, Sand or Soil Figure 9: Surveyed surface cover segment Figure 10: Surveyed surface cover segment Figure 11: Surveyed surface cover segment 14 Road erosion is not only affected by precipitation and slope but it is also affected by the land cover In the road segment, segment has the highest diameter of the particles (according to graphs above) average is 2.1cm If the diameter of rock or gravel is large it should have less sediment production Proved by road segment in the graph with a total sediment production of 4.969 Kg/m2 In road segment is no vegetation cover In road segment the average diameter of rock and gravel is 1.4 cm which can be produce much sediment (6.75kg/m2) In road segment there is about 5% vegetation cover and the average diameter of rocks and gravel is 1.5cm with the total amount of sediment production during 10 rains of 6.35 Kg/m2 4.5.2 Bulk Density Total Volumes 196.2 cm3 Weight of dry soil 310g (road 280g(road 240g(road segment sample in road segment 1) segment 2) 3) 1.58 g/cm3 1.43 g/cm3 1.22 g/cm3 segment Bulk Density (mass/volume) Soils with less porosity and will have less infiltration and more runoff Following formula mass/volume it can be seen that at road segment an elevated bulk density of 1.58 g/cm3 was measured This is affected by the number of vehicles In road segment has 432 motorbikes per day are estimated In road segment bulk density is 1.43 g/cm3.This road is only used by about motorbikes per day since it is only used to reach the rice paddy so not many people use it Finally road segment is very steep, so people just walk on it without any vehicle affecting this segment In general at all road segments bulk density is normal, not too dense 15 4.6 Comparision to the WEPP Road model According to the table and the chart showing the big difference between the WEPP road and road segments measured in reality greater than times It mean that when measured in the field is much more precise than use the predictor in the website Table 4: Compare between WEPP road and road segments: WEPP Road (kg) 115 222 134 Precipitation (mm) 1700 1700 1700 road segments (kg) 998 1298 1046 Precipitation (mm) 410 410 410 Figure 12: Compare WEPP road model with road segments 16 DISCUSSION During the implementation and analysis of the study several issues had to be faced 5.1 Experimental Setup Performance The basis of the research was the collection of sediment to identify how much sediment is produced during rains Normally a sediment fence is built in a U-shape so that as much sediment as possible can be trapped and no sediment can run around the fence In reality this method could not be applied at two of the three sediment fences that were constructed at the unpaved road near Na Kieng Village As the road has a drain-ditch next to it So if built a U-shaped fence at that point would disturb water runoff in the drain-ditch Furthermore Image 3: Drain-ditch behind the sediment fences water coming from the other side of the hill slope would cross the road but then would also be hindered to runoff with severe effects on the research results Building a Ushaped sediment fence behind the drain-ditch was not an option either because most sediment would erode inside the ditch thus could not be caught by the fence Finally, as a compromise, two of the three fences were built in the shape of a litter bit straight line but surely that two sediment fences still trapped all sediment production 5.2 Data Analysis Performance Research was done during 10 rains which occurred approximately over one month The output are numbers that stress the significance of soil erosion management on unpaved roads, however those numbers might not be representative enough for 17 further scientific research: This study was the first of its kind in the area of Ba Be National Park and it only produced data which can be taken as an average for one month The results show that soil erosion depends on several factors but one of the most significant factors is precipitation Unlike the South, the North of Vietnam lies in a climatic zone with four seasons making it very hard to predict the climate and weather conditions Since research has never been done before in Khang Ninh Commune there is also no data on precipitation over long time available With the data collected during one month in a mostly wet season it is hardly to predict yearly precipitation and compared to previous studies 5.3 Comparisons to Previous Studies The mean one month in rainy season sediment production rate for the native surface road segments ranged from 0.07 up to 2.06 kg m2/ month, the average of three road segments are 0.60 kg/m2 (table 3) Annual yearly precipitation from the WEPP road is 1700 mm/year and rainfall for one month is 410mm/month so calculated by 410/1700 = ¼ So it had * 0.60 kg/m2 = 2.4 kg/m2/year Road erosion rates for unpaved roads the present study is 67% of the reported mean erosion rate of 0.48 kg m2 for unpaved roads in the Idaho batholith (Megahan, 1974), the range sediment production was from 0.13 to 0.99 kg/m2 (JOHN, 2001) The similarity in road erosion rates for the Sierra Nevada and the Idaho batholith Looked at the all data above, showing that the biggest range is in Ba Be Bac Kan province it might be because of the climate condition made the large difference 5.4 Future Research A logical resumption of the study would be to extend the period of data collection to one year including all four seasons The data would be better in terms of quality and could be used for comparison with the WEPP: Road Model as well as previous studies on road erosion Sediment production should be measured with more 18 sediment fences (up to 10), each constructed at a different slope to get more differentiated data The same applies to sediment delivery which should be surveyed on more than eight road segments In addition it would be useful to measure the velocity on the road surface during heavy rains when runoff processes occur Velocity can be a driving key factor for the amount of sediment produced during the rain and therefore it should be investigated, however measurements require more time and more specific tools 19 CONCLUSIONS Over a time span of approximately one month in July and early August 2014, sediment production and sediment delivery were measured at an unpaved road in Na Kieng Village, Bac Kan Province, Vietnam The results of the study make clear that sediment production does depend on precipitation The highest amount of sediment was collected at fence (slope 22°) although it had the smallest area compared to the other two fences After long rains with high rainfall over 350 Kg of sediment could be collected at the fences Sediment production depended strongly on road slope and only a little on road area Recalculating the results in Kg per m² of road showed that less than Kg sediment/m² of road area was eroded; however in cases of long rains with heavy rainfall it is possible that up to Kg of sediment erode per m² of road From the big amount of sediment that affects the river, making it shallower and narrower with a bad effect on water quality as well as plants and animals in the water Looking at sediment delivery, most road segments in the surveyed area delivered their sediment to the river, stream and lake However, calculations showed that during strong rainfall with road erosions of 1,62Kg/m² an alerting amount of 1327Kg of sediment were delivered directly into the adjacent stream In general a larger diameter of rocks or sand mean less sediment production Vegetation cover can decrease sediment production Bulk density is affected by the amount of traffic, frequent passage of motorbikes can increase bulk density from 1,22g/cm³ (segment 3) up to 1,58g/cm³ (segment 1) In the WEPP road showed lower sediment than the reality road 20 REFERENCES Hagans, D.K., W.E Weaver and M.A Madej 1986 Long term on-site and off- site effects of logging and erosion in the Redwood Creek basin, Northern California In: Papers presented at the American Geophysical Union meeting on cumulative effects (1985 December); National Council on Air and Streams, Tech.Bull.No 490, pp.38-66 Megahan WF 1974 Erosion over time on severely disturbed granitic soils: a model USDA Forest Service Research Paper INT-156, Ogden, UT; 14 pp Morrison, P.H 1975 Ecological and Geomorphological Consequences of Mass Movements in the Alder Creek Watershed and Implications for Forest Land Management B.A Thesis University of Oregon, Eugene, OR 102 p Coe D 2006 Sediment production and delivery from forest roads in the sierra nevada, california, M.S thesis Colorado State University 13p MacDonald L, 2007 Road Sediment Production and Delivery in the Central Sierra Nevada, California Watershed Science, Colorado State University, OR 10 p ST.JOHN, 2001 Measurement And Prediction Of Sediment Production From Unpaved Roads, U.S Virgin Islands 14p Website: http://www.usyd.edu.au/agric/web04/Bulk%20density%20the%20final.htm http://forest.moscowfsl.wsu.edu/fswepp/ 8.1 APPENDICES Information on the WEPP road models Figure 2: Run WEPP road model with the same characteristics with the road segment 1, fence Figure 3: Run WEPP road model with the same characteristics with the road segment 2, fence Figure 4: Run WEPP road model with the same characteristics with the road segment 3, fence Pictures took during implement of thesis: