EWRI2013

Sergio Morera-Julca, Abel Mejia-Marcacuzco, Jean Loup Guyot; Carlos Glvez, Francisco Salinas, Manuel Collas, and Eusebio
Ingol-Blanco, Uncertainty in Suspended Sediment Load Estimates for Mountain Rivers. Case of Study of Central Andes in Peru, World Environmental & Water Resources Congress 2013, EWRI-ASCE, CINCINNATI, OHIO, May 19-23, 2013
Uncertainty in Suspended Sediment Load Estimates for Mountain Rivers. Case of Study of Central Andes in Peru S. B. Morera1, Abel Mejia-Marcacuzco1, Jean Loup Guyot2; Carlos Glvez3, Francisco Salinas4, Manuel Collas5, and Eusebio Ingol-Blanco6
1
Water Resources Program, Universidad Nacional Agraria La Molina, Lima, Peru; PH (511) 6147800; email: sergiobaymorera@gmail.com, jabel@lamolina.edu.pe. 2 Institut de Recherche pour le Dveloppement, Peru. 3 Proyecto especial Chavimochic, Trujillo, Peru. 4 Proyecto especial Zaa-Jequetepeque, Lambayeque, Peru. 5 Proyecto especial Tacna, Tacna, Peru 6 Universidad San Ignacio de Loyola, Lima, Peru; PH (511) 317-1000, annex. 3262; email: eingol@usil.edu.pe. Abstract This study evaluates the uncertainty in the estimation of annual and monthly sediment flows in four mountain basins of the Pacific Ocean in Peru. For this end, samples at different time scales (hourly, daily, and monthly) were recorded. The database was broken down into a numerical base to simulate several sampling frequencies. From these timing frequencies, the annual and monthly flow was computed. Likewise, we compare the observed and simulated data, calculating the degree of uncertainty using statistical tests. The results for the 25 and 75 percentiles show that there is a high temporal variability in these Andean watersheds, whose strategy of measurement of solid flows can vary from 6 to 12 days underestimating ~ 20% annually. While for estimating suspended sediment yield (SY) to a monthly time step in the rainy season (December to May), a sampling frequency between 1-5 days which would lead to a maximum error of 40% is needed. However, for the dry season (June to September), the sampling can be done one time per month for an error of 45% which is less than 2 % in an annual balance. Finally, the results show that the sampling frequency values mentioned above should be reduced to half, for exceptional events (rainy season) where the annual error estimate is around 300%. Key words: Uncertainty, Suspended Sediment Load, Solid Flow, Mountain Rivers, Peru. 1. Introduction Uncertainty is an inherent characteristic of water resources systems (e.g.; Raje & Mujumdar, 2010). In a general context, the existence of the uncertainty appears in the whole monitoring process, data processing and management; these include: the bias of the estimated flows during measurement (Cerdan et al., 2010), laboratory analysis (e.g.; Harmel et al., 2006), sampling strategy (Meybeck et al., 1992) and the calculation procedures method (Phillips et al., 1999); the latter is important in the period and algorithm used (eg.; Duvert et al., 2011). From the above, Birgand et al., (2010) mentions that during the evaluation of the sediment load (SSL), the uncertainty generally has two major sources: one is the sampling frequency and the other one is the method of calculation. In fact, Morera et al., (2011) estimated the SY in the watershed under study from hourly records; they determined the resulting accuracy are given by the methods of calculation of Colby, (1956) and Walling, (1977); however, understanding sediment dynamics goes through the quantification of uncertainties in the liquid flow and suspended concentration (SCC), and its becomes even more complex without a proper strategy sampling with low temporal frequency (annual, bimonthly, monthly, weekly, etc.). One main characteristic of the study area, is the short-
Sergio Morera-Julca, Abel Mejia-Marcacuzco, Jean Loup Guyot; Carlos Glvez, Francisco Salinas, Manuel Collas, and Eusebio Ingol-Blanco, Uncertainty in Suspended Sediment Load Estimates for Mountain Rivers. Case of Study of Central Andes in Peru, World Environmental & Water Resources Congress 2013, EWRI-ASCE, CINCINNATI, OHIO, May 19-23, 2013
and long time scale variability, which is founded on sub-seasonal periods (eg., rainfall-DEF) by taking into account the seasonal and interannual variability (Reckhow, 1994); where concentration loads vary widely during liquid flow events (e.g.; Meybeck et al., 2003). Moreover, mountain watersheds typically produce large amounts of fine sediment (e.g.; Dietrich & Dunne, 1978), which appear in random pulse mode, and those are influenced by distinctly landslides or other natural events (rain, wind) (Cui et al., 2003). However, the SSL frequency and magnitude are affected by human factors such as mining (eg.; Morera et al., Summited), agriculture, roads, etc (e.g.; Wohl, 2006; Montgomery, 2007). On top of that not only is well known that most SY is transported in important hydraulic events but also some studies show that 90% of the SY occurs during 25-10% of the hydrologic cycle, as 50% in 1% of the time, and even 63 and 97% of the SY at 2% of the year (e.g.; Guyot et al., 1994; Molina et al., 2008; Duvert et al., 2011). As consequence, an accurate estimate of specific sediment yield (SSY) requires an effective strategy for monitoring the SSC (e.g.; Duvert et al., 2011). Hence, the requirement of evaluating uncertainty as a result of low sampling frequency is widely recognized (e.g.; Webb et al., 1997). So, the time scale selection that reflects the solid flow behavior plays a fundamental role to solve this problem. This paper aims at evaluating the uncertainty introduced by infrequent sampling of the SSC (hourly, daily, weekly, and monthly) in terms of SY, in four mountain basins distributed along the Peruvian coast (Figure 1). It also seeks to optimize the sampling strategy based on uncertainty. Study area The study area is localized between 6 to 18 latitude south and 69 to 79 longitude west , including watersheds from the western central Andes (Peru), which drain to the Pacific Ocean. Among the studied watersheds, it includes the largest watershed of the Peruvian coast, and one of the most arid basins in southern of Peru; the areas range from 1 757 a 10 411 km2.
Yonan station
200
Water discharge (m3.s-1)
Tablachaca station
100 80 60 40 20 0
150 100 50
Yonan station
0 Jab Feb
Tablachaca station
Jab Feb Mar 100 Apr May
Yonan station
100 80 60 40 20 0
Tablachaca station
150
200 Mar Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Yonan station Jun Jul Aug Sep Oct Nov Dec

100 80 60 40 20 0
Tablachaca station
200
80 60
100 50 0 Jab
Condorcerro station
150 100
10 8
Sambalay station
40 20 0
500
400 300 200 100
50
Dec
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
6 4
Jab
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
ay
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jab
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jab
2 0
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jab
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Condorcerro station
500
Jab Feb Mar Apr 400 May Jun Jul Aug Sep Oct Nov Dec
Sambalay station
Jab Feb Mar
10 Apr May Jun 8 6 4 2 0 Jab Feb Mar
Jul
Aug Sep
Oct
Nov Dec
ondorcerro station
Sambalay station
500
Condorcerro station
10
Sambalay station
10 300
8 6 4 2 0
400 300 200

Dec
200 100
8 6 4 2 0
Jab
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Apr May Jun
Jul
Aug Sep
Oct
Nov Dec
100
ay
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jab Feb
Mar
Apr May Jun
Jul
Aug Sep
Oct
Nov Dec
Jab
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Nov
Dec
Jab Feb
Mar
Apr May Jun
Jul
Aug Sep
Oct
Nov Dec
Figure 1. Shaded-relief and elevation map from see level to the highest point in the Central Andes (Peru). Runoff stations are represented by black triangles. Box plots show median water discharge (m3.s-1) for the period 2004-2011(red line in the centre of box), the little red cross represent atypical events, the extremes of the dotted line represent the 5th and 95th percentile, 25th and 75th percentiles
(bottom and top of box, respectively). From the graphics we realise seven rainfall months (NovApr), and five dried months (May-Oct). Hidro-sedimentologies stations are located in the lower basin; this feature gives them clearly a special configuration of piedmont, delineating watersheds highly influenced by rainfall. Some features of the physical parameters for each watershed is shown in the Table 1. Table 1. Description and interpretation of sedimentary and lithology
Code 1 2 3 4 Station Catchments Yonan Tablachaca Condorcerro Sambalay Altitude (masl) Min
441 525 470 924
Average
2,571.3 3,269.6 3,654.7 3,540.7
Max
4,195 4,969 6,685 5,719
Area (km2)
3,290.4 3,131.3 10,403.0 1,701.0
Perimeter (km)
342.0 339.4 789.9 313.7
Main river
96.6 100.9 260.3 81.1
Drainage density
0.315 0.555 0.598 0.387
Form factor (Kf)

0.353 0.308 0.154 0.259
Gravelius Index (kc)

1.681 1.710 2.184 2.145
2. Data and Methodology Water discharge and SSC Flow measurements and SSC are recorded by the special projects: Jequetepeque- Zaa at Yonan station (3293 km2); Chavimochic project at Tablachaca (3191 km2) and Condorcerro stations (10411 km2), and lastly the Tacna special project at Sambalay station (1757 km2, Figure 1). The common links to the four stations are: the monitoring period from 01/09/2004 to 31/08/201, continuous monitoring of the flow rates (water level recorder) and discreet monitoring a high frequency of SSC (4, 6 and 12 hours).
Figure 2. Hydrological time series of instantaneous water discharge (m3.s-1) and suspended sediment concentration SSC (mg.l-1). a) Yonan station, b) Tablachca station and c) Condorcerro station axis at the same scale. Finally d) Sambalay station axis at different scale axis.
The protocol of measurement is based on the Standard Test Methods for Determining Sediment Concentration (ASTM, 2002), which consists in the extraction of the SSC sample by the vacuum method, using suction pumps or gravity pumps. Samples of SSC are taken every 12 hours throughout the year at Condorcerro and Tablachaca station, because of the permanent regimen of the rivers. For station Yonan, four samples are taken per day during the flood season; also one sample per day is taken during the drought season. Finally, six samples are taken per day during the rainy season at Sambalay station; SSC samples are not taken in low water levels, because the river reaches the dry condition completely (Figure 2). Methodology The method is divided into three parts: 1) Under the assumption that the higher the sampling frequency, the uncertainty is reduced (e.g.; Coynel et al., 2004), we calculate sediment flow and its uncertainty from discrete water discharge and SSC samples (higher frequencies-observed data), 2) we break down the SSC time series into a numerical dataset to simulate a low time frequency sampling (Birgand et al., 2010), and 3) we simulate the sediment flux with the new datasets; finally the uncertainty was estimated comparing the observed against the simulate sediment flows. Estimating the suspended sediment loads a. Getting continuous series Different calculation methods differ on how to integrate the equation (1). A traditional method was used to calculate the sediments load in the studied watersheds (Verhoff et al., 1980). The method is based on calculating the SSL from the instantaneous reading of SSC and liquid flow, both simultaneously taken. n (1) 3
SSLid SSLt dt with
i 1
SSLt Qi . SSCi . 10
dS
Where SSLid is the instantaneous suspended sediment flux in a discrete monitoring at high sampling frequency t=12 hours (t.s-1). Qi is the instantaneous water discharge (m3.s-1) and SSCi is the instantaneous suspended sediment concentration (g.l-1). (S) is a measured at a point of a cross section. After the calculation of solid flows, some gaps were observed which were completed with the method of Colby, (1956) (equation (2).
SSLi a (Qi )b
(2)
SSLi is the instantaneous suspended sediment flux generated from the Qi , a y b are regression parameters from the entire available record.
b. Observed and simulated load estimation This method was developed by Phillips et al., (1999), who coded as method 15; now in the equation (3). This equation was applied to the new continuous time series in reference to the time step required: daily, weekly, and monthly.
SSLt t * (SSLid or SSLi )
(3)
where SSLt the instantaneous suspended sediment flux transported during the time t (t.s-1); t time interval (from daily to monthly).
Sergio Morera-Julca, Abel Mejia-Marcacuzco, Jean Loup Guyot; Carlos Glvez, Francisco Salinas, Manuel Collas, and Eusebio Ingol-Blanco, Uncertainty in Suspended Sediment Load Estimates for Mountain Rivers. Case of Study of Central Andes in Peru, World Environmental & Water Resources Congress 2013, EWRIASCE, CINCINNATI, OHIO, May 19-23, 2013
3. Results Uncertainty estimation for annual SSL Relative Error Figure 3 shows the development of the uncertainty against a sampling strategy. Example hourly, daily and monthly frequency, in general, the mean relative error tends to increase or decrease with a decrease of the sampling rate for the four watersheds. Yonan station shows a strong trend to overestimate for sampling frequency from 1 to 6 days (red line, Figure 3). From the seventh day, the trends tend to be randomized (see 25 and 75 percentile in the upper and lower limits). The degree of variability increases with decreasing sample rate in a range from 10% to 46%. Finally, there is a critical point when the sampling frequency is greater than 12 days, which shows a variation of ~120%. Also, ; also, it is found that during extreme events this rate will reach 300% of variation based on an instantaneous water discharge. At Tablachaca station, the SSC sampling frequency, less than nine days tend to underestimate the SY. Frequencies more than 9 days tend to be randomly; however, it could be meaningful in an annual estimate of the SY (Figure 3). Condorcerro station shows a tendency to sub-estimate at the seventh day. The Sambalay station (Figure 3) uncertainty is much larger than the previous three stations, there is a clear trend until the second day, the third day there is a high trend to underestimate ~ 50% during the rainy season.
Yonan station
100
Tablachaca station
100
Error distribution (%)

12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d13d 14d 15d 31d
50
50
-50
-50
-100
-100
Condorcerro station
100
100
Sambalay station
Sampling strategy
Sampling strategy
50
50
-50
-50
-100
-100
Figure 3. Box plot of statistical distribution of the relative error uncertainties during the hydrological period 2004-2011, to estimate annual sediment flow from hourly, Sampling strategy daily, etc. Sampling strategy strategy simulation sampling. The medians are showed by the horizontal red line inside the boxes, the little red cross represent the atypical events, the extremes of the dotted line represent the 5th and 95th percentile, the bottoms and tops of the boxes show 25th and 75th percentiles.
Root Mean Square Error (RMSE) Yonan Station shows the highest accumulated error, with an upward trend until the sixth day. Tablachaca station showed a positive trend until the 8th day, followed Condorcerro station that showed an upward trend until the 6th day. Finally Sambalay station showed a positive trend until the 3rd day (Figure 4).
Yonan station
2500
Tablachaca station
2500
RMSE (t.km2.year-1)
12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d 13d 14d 15d 31d Sampling strategy
RMSE (t.km2.year-1)
2000
2000
1500
1500
1000
1000
500
500
12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d 13d 14d 15d 31d
Sampling strategy
Condorcerro station
2500
Sambalay station
2500
RMSE (t.km2.year-1)
2000
RMSE (t.km2.year-1)
2000
1500
1500
1000
1000
500
500
Sampling strategy
Sampling strategy
Figure 4. Box plot of statistical distribution of the Root Mean Square Error (RMSE) to estimate specific sediment yield by simulation hourly and daily strategy of sampling during a the hydrological period 2004 -2011. The medians are showed by the black circles inside the boxes, the little blue circles represent atypical events, the extremes of the dotted line represent the 5th and 95th percentile, the bottoms and tops of the boxes show 25th and 75th percentile. Estimation of uncertainty for a monthly balance of SSL Seasonal variability Generally, the study regions, during the year, show a dominant climate unimodal regime, with a very strong contrast between the rainy and dry season periods (semi-arid). Furthermore, a strong interannual variability is observed (Figure 1). In addition to this, the presence of anomalies, with wetter years than other ones, produce very strong runoff events. As a result, sediment rates are altered. For example, the Sambalay basin in dry years, its seasonality tends to decrease to the point of no distinction between wet and dry periods (Figure 1). In contrast, under the presence of ENSO events, Yonan basin tends to increase the transport of suspended sediment. Monthly relative error From May to October (dry season), the sampling frequency required is about 1 per month or each two months. For example, the Yonan station with strong seasonality, the RMSE shows
high range of variation in peak events (rainy months Nov-Apr); which is reduced markedly along to the hydrologic cycle. The highest variability was founded during April in the studies basin. The monthly CUSUM test The cumulative sum of the specific SSL (CUSUM, Figure 5), explains how uncertainty progresses throw a monthly testing period. The CUSUM and RMSE tests show consistent results during the dry season; whereas, in the rainy season the observed and simulated SSL tend to overestimate or underestimate according to the month and sampling frequency. For example in Figure 5 wet years during the rainy season tends to underestimate flows for the four sedimentary basins; this statement is reflected in the increased range between observed and simulated SSL.
Yonan station
250
Tablachaca station
600
200
500
Cumulative load (t.km2)
400
150
300
100
200
50
100
0 0
10
20
30
40
50
60
70
80
0 0
10
20
30
40
50
60
70
80
Months
Condorcerro station
350 300
Months Sambalay station

35 30

20 30 40 50 60 70 80
250 200 150 100 50 0 0
25 20 15 10 5 0 0
10
10
20
30
40
50
60
70
80
Months
Months
Figure 5. Time series plot of monthly measured and simulated cumulative specific SSL (t.km2.month-1) at station Yonan, Tablachaca, Condorcerro y Sambalay in the testing period 2004-2011. 4. Discussion Uncertainty variability during the annual estimation of the SSL The relative error shows a high variability, increasing the uncertainty in proportion to the decrease of the sampling frequency; at first time it is also related to the watershed size (eg.; Meybeck, 2003). Nevertheless, the SSL is highly influenced by the factors which control the SY. It is observed that weekly sampling strategy underestimates the SSL; this effect is attributed to the lack of representativeness of the activity related to the SY in the watershed. As example, whether the sampling begins sunday this sampling strategy considers all sunday of the year,
avoiding activities that are directly related to the production of SSC as mining, population, industry, agriculture, among others. Uncertainty tends to be more stable at non-intermittent basins such as Condorcerro and Tablacahaca; reducing the sampling rate trends to underestimate the SSL; nonetheless, intermittent watershed show deferent behaviour also no clear trend can underestimate or overestimate the characteristics of the basin. The strong variability in the topography, climate and human activity generates a SY dynamics, where it is not possible to establish a fixed frequency sampling strategy throughout the year, so each basin must be analysed seasonally an individually. Uncertainty in the estimation of monthly SSL Seasonal uncertainty analysis found the sub-seasonal variability (Dec. to Feb.), as a key issues to correct possible error (Diarra et al., 2007); because the occurrence of floods events, where is presumed high concentration of SSC in a short time series tends to underestimate the frequency of atypical events and therefore increase our range error (Cheviron et al., 2011); reflected in the underestimation of sediment flow calculation. The relative error test shows strong variability during the rainy season, whereas for the dry season the pulse of SSL are much more stable; although there are dissolution and concentration events, those are no representative in an annual balance because the very low rates of water discharge and so on low SSL (~ 0 t.day-1). Such as events are appreciated frequently in the Sama basin at Sambalay station, also strong seasonality is appreciated (Figure 2); still, along rainy season the estimation of the SSL is highly sensible to the variation. Analysis of the 5th and 95th shows wide range of uncertainty, one of the factors is attributed to the size of the basin; based on the higher frequency pulses of SSL and water discharge, which are related to the time and the runoff concentration; nevertheless, climate, geology and anthropic factor which control the SSL tend to change this relation (e.g.; Vente et al., 2011; Morera et al., Summited). One example is the influence of the watershed location along the north-south (Lavado et al., 2012), also from east to west gradient in the Pacific coast. The CUSUM test showed that for exceptional years is required a sampling frequency, which must not be exceed one half of the maximum tolerated sampling at each watershed; this strategy would require us 35% error in our monthly SSL estimation. Sampling strategy from hourly to daily The rating curve was built to analyst the relationship between the SSC and water discharge. The relation was highly significant (P<0.0001 and R> 0.9) in the study watersheds. This result helped to explain the behaviour of sediment transport process; because in other cases, the ratio between the water discharge and SSL are usually very complex patterns including hysteresis, and discontinuities (Meybeck & Moatar, 2011). Finally, the SSC sampling strategy was based on Verhoff et al., (1980); they established the average of the instantaneous reading of SSC and liquid; where SSL estimation depends directly on the frequency of sampling. 5. Summary and conclusion There is a knowledge gap on the uncertainty produced by the infrequent sampling of the SSC and water discharge along the Pacific coast at the Andean region; consequently, the failure
to adopt a sampling strategy. At local and regional scale, numerous studies have been carried out considering reports from other continents; however, those basins report different situations in the SY. In this context, we analyse four basins (1757-10411 km2), distributed along foothills of western Peruvian Andes. The used databases belong to the same period and the SSC was obtained under the same technique. Finally, the database was decomposed to simulate numerically different sample. This study is based on hourly frequency data (SSC and water discharge), which was recorded manually in a section of the river (hydro-sedimentology station). Results showed high temporal variability in these Andean watersheds (25th and 75th percentile); where the SSL monitoring strategy can range from 6 to 12 days to underestimate ~20% annually. Nevertheless, monthly analysis suggests that water regime plays an important role to establish better sampling strategy. Estimation in the rainy season (dec-may) needs to be sampled between 1 to 5 days frequency which involve 40% of error. However in the dry season (Jun-Sep), sampling can be performed once a month, whose error could reach 45%, the last one would not achieve 2% of error in an annual balance. Frequency time samples should be reduced during exceptional events because this could influence at ~300% of error on the annual balance. Acknowledgements This work was initiated as part of a Ph.D. dissertation at the La Molina Agrarian National University; it is financially supported by a doctoral research scholarship from the FINCyT (www.fincyt.gob.pe). The authors are grateful to the French Institut de Recherche pour le Dveloppement (IRD, www.ird.fr) to the funding through the HYBAM observatory. References
ASTM. (2002). Standard Test Methods for Determining Sediment Concentration in Water Samples. D3977, from http://www.astm.org Birgand, F.; Faucheux, C.; Gruau, G.; Augeard, B.; Moatar, F., & Bordenave, P. (2010). Uncertainties in assessing annual nitrate loads and concentration indicators: Part 1. Impact of sampling frequency and load estimation algorithms. Transactions of ASABE. 53 (2): 437-446. Cerdan, O.; Govers, G.; Le Bissonnais, Y.; Van Oost, K.; Poesen, J.; Saby, N.; Gobin, A.; Vacca, A.; Quinton, J.; Auerswald, K.; Klik, A.; Kwaad, F.J.P.M.; Raclot, D.; Ionita, I.; Rejman, J.; Rousseva, S.; Muxart, T.; Roxo, M. J., & Dostal, T. (2010). Rates and spatial variations of soil erosion in Europe: A study based on erosion plot data. Geomorphology. 122 (1-2): 167-177. doi: 10.1016/j.geomorph.2010.06.011 Cheviron, B.; Delmas, M.; Cerdan, O., & Mouchel, M. (2011). Parameter uncertainty and sensitivity analysis in sediment flux calculation. Hydrol. Earth Syst. Sci. doi: doi:10.5194/hessd-8-1469-201 Colby, B. R. (1956). The relationship of sediment discharge to streamflow Open File Report (Vol. R29, pp. 170). Washington, D.C.: U. S. Geological. Cui, Y.; Parker, G.; Lisle, T. E.; Gott, J.; Hansler-Ball, M. E.; Pizzuto, J. E.; Allmendinger, N. E., & Reed, J. M. (2003). Sediment pulses in mountain rivers: 1. Experiments. Water Resour. Res. 9 (39): 1239. doi: 10.1029/2002WR001803, 2003 Diarra, L.; Fofana, A.; Givone, P.; Morand, P.; Orange, D., & Traor, P. (2007). Connaissance du fleuve, volution et indicateurs. In IRD (Ed.), Avenir du fleuve Niger (pp. 193-308). Dietrich, W. E., & Dunne, T. (1978). Sediment budget for a small catchment in mountainous terrain, Zeits. Geomorphology. 29: 191-206.
Duvert, C.; Gratiot, N.; Nmery, J.; Burgos, A., & Navratil, O. (2011). Sub-daily variability of suspended sediment fluxes in small mountainous catchments - implications for community-based river monitoring. Hydrol. Earth Syst. Sci. 15: 703-713. doi: 10.5194/hess-15-703-2011 Guyot, J. L.; Bourges, J., & Cortez, J. (1994). Sediment transport in the Rio Grande, an Andean river of the Bolivian Amazon drainage basin. Int. Assos. Hydrol. Sci. 224: 223-231. Harmel, R. D.; Cooper, R.; Slade, R.; Haney, R., & Arnold, J. (2006). Cumulative uncertainty in measured streamflow and water quality data for small watersheds. Transactions of ASABE: 49(43): 689-701. Lavado, W.; Ronchail, J.; Labat, D.; Espinoza, J. C., & Guyot, J. L. (2012). Basin-scale analysis of rainfall and runoff in Peru (19692004): Pacific, Titicaca and Amazonas drainages. Hydrological Sciences. doi: DOI:10.1080/02626667.2012.672985 Meybeck, M. (2003). Global analysis of river systems: from Earth system controls to Anthropocene syndromes, The Royal Society, 20. Meybeck, M.; Laroche, L.; Durr, H. H., & Syvistski, J.P.M. (2003). Global variability of daily total suspended solids and their fluxes in rivers. Global and Planetary Change. 39: 65-93. Meybeck, M., & Moatar, F. (2011). Daily variability of river concentrations and fluxes: indicators based on the segmentation of the rating curve. Hydrological Processes. Meybeck, M.; R., Kimstach, & R., Helmer. (1992). Assessment of the Quality of the Aquatic Environnement Through Water, Biota and Sediment Strategies for water quality assessment (Chapman & Hall ed., pp. 19-50): London. Molina, A.; Govers, G.; Poesen, J.; Van Hemelryck, H.; De Bievre, B., & Vanacker, V. . (2008). Environmental factors controlling spatial variation in sediment yield in central america mountain area. Geomorphology 98: 176-186. Montgomery, D. R. (2007). Soil erosion and agricultural sustainability. PNAS, 104, 13268-13272. Retrieved from doi:10.1073/pnas.0611508104 Morera, S. B.; Condom, T.; Crave, A.; Vauchel, P., & Galvez, C. (2011). Erosion rates and dynamics flow of sediments over the Santa River Basin in Peru. Revista Peruana geo-atmosfrica. GeoArmosfrica. 3: 25-37. Morera, S. B.; Condom, T.; Vauchel, P.; Guyot, J. L.; Galvez, C., & Crave, A. (Summited). What is the pertinent spatio-temporal scale of observation to understand sediment yield control factors in Andean Region? The case of the Santa River (Peru). Hydrol. Earth Syst. Sci. Phillips, J. M.; Webb, B. W.; Walling, D. E., & Leeks, G.J.L. (1999). Estimating the suspended sediment loads of rivers in the LOIS study area using infrequent samples. Hydrological Processes. 13 (7): 1035-1050. doi: 10.1002/(SICI)1099-1085(199905)13:7<1035::AID-HYP788>3.0.CO;2-K Raje, D., & Mujumdar, P. P. (2010). Reservoir performance under uncertainty in hydrologic impacts of climate change. Advances in Water Resources. 33: 312-326. doi: 10.1016/j.advwatres.2009.12.008 Reckhow, K. (1994). Water quality simulation modeling and uncertainty analysis for risk assessment and decision making. Ecological Modelling. 72: 1-20. Vente, J.; Verduyn, R.; Verstraeten, G.; Vanmaercke, M., & Poesen, J. (2011). Factors controlling sediment yield at the catchment scale in NW Mediterranean geoecosystems. Journal Soils Sediments. 11: 690-707. doi: 10.1007/s11368-011-0346-3 Verhoff, F. H.; Yaksich, S. M., & Melfi, D. A. (1980). River nutrient and chemical transport estimates. J. Environ. Engng. Div. ASCE. 10: 591-608. Walling, D. E. (1977). Assessing the Accuracy of Suspended Sediment Rating Curves for a Small Basin. Water Resour. Res. 13 (3): 531-538. Webb, B.; Phillips, J.; Walling, D., & Littlewood, C. (1997). Load estimation methodologies for British rivers and their relevance to the LOIS RACS(R) programme. Sci. Total. Environ: 194-195: 379389. Wohl, E. (2006). Human impacts to mountain streams. Geomorphology 79: 217-248. Geomorphology. 79 (3-4): 217-248. doi: 10.1016/j.geomorph.2006.06.020

EWRI2013

Загружено:

Сведения о документе

Исходное описание:

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

EWRI2013

Загружено:

Авторское право:

Доступные форматы

Sergio Morera-Julca, Abel Mejia-Marcacuzco, Jean Loup Guyot; Carlos Glvez, Francisco Salinas, Manuel Collas, and Eusebio

Water discharge (m3.s-1)

Water discharge (m3.s-1)

Water discharge (m3.s-1)

200 Mar Apr

Yonan station Jun Jul Aug Sep Oct Nov Dec

Water discharge (m3.s-1)

Water discharge (m3.s-1)

400 300 200 100

Water discharge (m3.s-1)

Water discharge (m3.s-1)

Water discharge (m3.s-1)

Water discharge (m3.s-1)

10 Apr May Jun 8 6 4 2 0 Jab Feb Mar

Water discharge (m3.s-1)

400 300 200

Water discharge (m3.s-1)

Water discharge (m3.s-1)

Apr May Jun

Apr May Jun

Apr May Jun

Form factor (Kf)

Gravelius Index (kc)

Error distribution (%)

Error distribution (%)

12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d13d 14d 15d 31d

Error distribution (%)

12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d13d 14d 15d 31d

12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d13d 14d 15d 31d

12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d 13d 14d 15d 31d

12h 24h 2d 3d 4d 5d 6d 7d 8d 9d 10d 11d 12d 13d 14d 15d 31d

Cumulative load (t.km2)

Cumulative load (t.km2)

Months Sambalay station

Cumulative load (t.km2)

Cumulative load (t.km2)

250 200 150 100 50 0 0

Вам также может понравиться