Getirana Et Al, 2011

8/22/2019 Getirana Et Al, 2011

1/19

Assessment of different precipitation datasets and their impacts on the water

balance of the Negro River basin

A.C.V. Getirana a,b,, J.C.V. Espinoza c,d, J. Ronchail e, O.C. Rotunno Filho a

a Universidade Federal do Rio de Janeiro, Programa de Engenharia Civil, Rio de Janeiro, Brazilb Universit de Toulouse, UPS (OMP), IRD, LEGOS, Toulouse, Francec Instituto Geofsico del Peru, Lima, Perud Universidad Agraria La Molina, Lima, Perue Universit Paris 7, LOCEAN/IPSL, Paris, France

a r t i c l e i n f o

Article history:

Received 18 May 2010Received in revised form 24 March 2011Accepted 30 April 2011Available online 11 May 2011This manuscript was handled by K.Georgakakos, Editor-in-Chief, with theassistance of Christa D. Peters-Lidard,Associate Editor

Keywords:

Water balancePrecipitation datasetHydrological modeling

Negro River basinAmazon River basin

s u m m a r y

With the objective of understanding the potential and limitations of available precipitation products forhydrological studies, this paper compares six daily and sub-daily precipitation datasets and their impactson the water balance of the Negro River basin in the Amazon basin. The precipitation datasets containgauge-based data [data derived from the Hybam Observatory Precipitation (HOP) dataset and providedby the Climate Prediction Center (CPC)], satellite-based data [the Global Precipitation Climatology Project(GPCP) one-degree daily and TRMM Multisatellite Precipitation Analysis (TMPA) datasets] and model-based data [the NCEP-DOE AMIP-II re-analysis (NCEP-2) and 40-year ECMWF Re-Analysis (ERA-40) data-sets]. Each dataset has a common set of meteorological forcing data which are used to run the MGB-IPHhydrological model for the period from January 1998 to August 2002.

The average precipitation of all the datasets is 2542 mm for the Negro River basin, with a standarddeviation of 317 mm. TMPA and NCEP-2 have the lowest (2216 mm/year) and the highest (3065 mm/year) precipitation rates, respectively. The HOP and CPC datasets agree best with observed discharge.GPCP gives the best results among the ungauged datasets, followed by ERA-40. TMPA and NCEP-2 arefound to be the least accurate. TMPA can reproduce the water cycles reasonably well, but underestimatesthe precipitation fields and discharges over the basin, while NCEP-2 is unable to represent the rainfallquantity and cycles, and the waterdischarge. Results suggest that gauge-based data are still the most rep-resentative of the actual precipitation in the northern Amazon basin. However, some satellite and model-based can reproduce fairly well the water cycle at the basin scale and monthly time step.

2011 Elsevier B.V. All rights reserved.

1. Introduction

In the past several decades, numerous global and quasi-globalprecipitation datasets have been developed at different time scalesusing input sources such as ground-based observations, satelliteestimates and outputs from general circulation models (Willmott

et al., 1994; Kalnay et al., 1996; Sorooshian et al., 2000; Newet al., 2000; Adam and Lettenmaier, 2003; Rudolf and Schneider,2005; Uppala et al., 2005; Huffman et al., 2007; among others).In view of the many precipitation datasets available, several anal-yses have been carried out to identify their differences at regional(e.g. Gebremichael and Krajewski, 2004; Dinku and Anagnostou,2005), continental (e.g. Syed et al., 2004; Marengo, 2005; Juarezet al., 2009) and global (e.g. Sapiano et al., 2006) scales. Most stud-ies have found that the datasets typically agree in terms of the

main temporal precipitation trends and their global spatial distri-bution but, regionally, they often exhibit marked differences. Forexample, Costa and Foley (1998) highlighted regional differencesfor the Amazon basin, and Adler et al. (2001) showed differencesamong datasets globally. Other comparisons suggest that the larg-est differences among available precipitation datasets occur in the

tropics (Fekete et al., 2004), particularly in the Amazon basin (Raoet al., 2002).

Our main objective is to compare the spatio-temporal heteroge-neities of six precipitation datasets at the daily time step over theNegro River basin, in the northern Amazon basin (Fig. 1). The com-parison of satellite-based and model-based data to gauge-baseddatasets allows us to identify the strengths and weaknesses ofthe different precipitationproducts. Another objective of this paperis to analyze the effects of different precipitation datasets on thesimulated water cycle of the Negro River basin. Indeed, comparingsimulated and observed discharges can be an efficient way to com-plement assessments of precipitation datasets. Several studies ofthis type have been already carried out in recent years (e.g. Yilmaz

0022-1694/$ - see front matter 2011 Elsevier B.V. All rights reserved.doi:10.1016/j.jhydrol.2011.04.037

Corresponding author at: Universidade Federal do Rio de Janeiro, Programa deEngenharia Civil, Rio de Janeiro, Brazil.

E-mail address: [email protected] (A.C.V. Getirana).

Journal of Hydrology 404 (2011) 304322

Contents lists available at ScienceDirect

Journal of Hydrology

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / j h y d r o l
http://dx.doi.org/10.1016/j.jhydrol.2011.04.037mailto:[email protected]://dx.doi.org/10.1016/j.jhydrol.2011.04.037http://www.sciencedirect.com/science/journal/00221694http://www.elsevier.com/locate/jhydrolhttp://www.elsevier.com/locate/jhydrolhttp://www.sciencedirect.com/science/journal/00221694http://dx.doi.org/10.1016/j.jhydrol.2011.04.037mailto:[email protected]://dx.doi.org/10.1016/j.jhydrol.2011.04.037


2/19

et al., 2005; Wilk et al., 2006). Here, the MGB-IPH model (Collis-chonn et al., 2007), forced with different precipitation datasetsand using the same atmospheric forcings (i.e. solar radiation, pres-sure, relative humidity, wind speed and temperature at sea level),has been used to simulate the daily hydrological processes of theNegro River basin for the period from January 1998 to August2002, for which all datasets are available. Then, sets of simulated

runoff, evapotranspiration and soil moisture content have beencompared among themselves and water discharge time series havebeen evaluated with observed data in order to quantify the reliabil-ity of each dataset in terms of providing estimates of the water cy-cle. Evaluating spatio-temporal differences between data sources isparticularly useful for identifying how various components of thewater cycle (such as runoff, evapotranspiration and soil moisture)are affected by changes in the precipitation field used in the hydro-logical models. So, this paper provides important information toboth the hydrological and rainfall-retrieval communities.

This paper is organized as follows: Section 2 gives a briefdescription of the Negro River basin and the datasets used in thisstudy. It also provides information about MGB-IPH model, themodeling setup for the study area and the methodology used to

evaluate the datasets. In Section 3, the results obtained are pre-sented and discussed. Finally, the conclusions are presented inSection 4.

2. Data sets and methodology

2.1. Study area and available data sets

The Negro River basin covers parts of both the northern andsouthern hemispheres with an area of about 712,000 km2, from3140S to 5800N latitude and from 72570W to 58160W longitude,in the northern Amazon basin (Fig. 1). The Negro River is the mostimportant tributary of the Solimes/Amazon River in terms of run-off [4.36 mm/day the mean runoff of the Amazon basin is about

2.9 mm/day (Marengo, 2005)], and the second after the MadeiraRiver in terms of total discharge.

Daily discharge data are available at several gauge stationswithin the Negro River basin and are freely provided by the Brazil-ian Water Agency (ANA Agncia Nacional de guas). These sta-tions drain areas varying from 611 km2 to 291,150 km2 [a list ofmost of gauge stations in service in the Negro River basin can befound in Getirana et al. (2010)]. Five gauge stations (Caracara, Tar-aqua, Cucu, Curicuriari and Serrinha) representing different re-

gions and hydrological regimes of the basin have been selectedto evaluate simulated discharges. The total discharge produced inthe basin cannot be evaluated since daily discharges are not avail-able at the basins outlet. However, other hydrological variablessuch as precipitation, runoff, evapotranspiration and soil moistureare analyzed at the basin scale (represented by the drainage area ofthe Negro River when it passes by Manaus). The main characteris-tics of these stations are given in Table 1.

Indeed, the Negro River basin is one of the rainiest regions of theAmazon basin (about 3000 mm/year). The Guyana Shield, one ofthe three cratons of the South American Plate that underlies Guy-ana, French Guiana, Suriname and parts of Colombia, Venezuelaand Brazil, creates a strong gradient in the rainfall spatial distribu-tion over the basin: humid flow from the Atlantic is hindered by

Fig. 1. Map of the Negro River basin, including the locations of pluviometric stations with data for the period 19802006, and of six fluviometric stations used to evaluatemodel performance. The additional Colombian and Venezuelan stations are found within the dashed lines.

Table 1

Main characteristics of the sub-basins defined by the six gauge stations considered in

this study.

Station Drainagearea (km2)

Mean discharge Meanprecipitation(mm/year)

Upstreamsub-basin

(m3/s) (mm/year)

Caracara 126.085 2903 731 2044 Taraqua 44.255 2755 1972 3557 Cucu 71.132 4940 2207 3192 Curicuriari 191.787 12,.613 2092 3441 Cucu and

TaraquaSerrinha 291.150 18,.082 2211 3241 CuricuriariManaus 712.451 35,943 1591 2667 Caracara

and

Serrinha

A.C.V. Getirana et al. / Journal of Hydrology 404 (2011) 304322 305


3/19

the shield, thereby resulting in less frequent and intense rainevents on the North-eastern side of the Negro River basin (Salatiet al., 1978; Figueroa and Nobre, 1990; Espinoza Villar et al.,2009). Thus, annual rainfall ranges from less than 1200 mm/yearin the driest region (north-eastern basin) to more than 5300 mm/year in the most humid parts (north-western basin). The Roraimahighlands and the Neblina Peak, the highest mountain in Brazil

(Pico da Neblina, 2994 m), are located in this region. The hydrolog-ical cycle of the basin is characterized by a very wet period be-tween May and August, with precipitation peaks occurring inJune and July and no real dry season (Figueroa and Nobre, 1990;Guyot et al., 1993).

An increasing interest in the hydrology of the Amazon basin be-gan near the end of the 1970s. This is reflected by the growingnumber of gauge stations operating in the region (Silva et al.,2007). Nevertheless, the basin still has a very low gauge density($0.12 stations per 1000 km2) (Espinoza Villar et al., 2009), whencompared to other large basins in South America, such as the LaPlata basin, which had a density of 0.63 to 1.06 stations per1000 km2 during the period 19982006 (Su et al., 2008). The NegroRiver basin also has an increasing number of gauge stations over

the last three decades. Fig. 2 shows the growth of the Brazilian pre-cipitation gauge network within the basin between 1980 and 2006.The number of stations providing daily precipitation observationsincreased by about 130% as reported by ANA. In contrast, the dataavailability, defined as the total number of observations averagedfor all stations, remained relatively constant ($85%) over the 27-year period, except for a significant drop in 1991 that coincidedwith a political crisis in the country.

Currently available gauge-based precipitation datasets for theAmazon basin consist mostly in publicly available data, such asthose provided by ANA, with limited a number of gauged data fromthe other countries composing the basin (Rudolf et al., 1994; Newet al., 2000; Silva et al., 2007). To acquire a more comprehensiveunderstanding of the water cycle of the Amazon basin, the Hydrol-

ogy and Geodynamics of the Amazon Basin (Hybam) environmen-tal research observatory (Cochonneau et al., 2006), throughinternational partnerships, created a data base containing hydro-logical data from most of countries that make up the Amazon ba-sin. Espinoza Villar et al. (2009) first presented a monthly HybamObservatory Precipitation (further referred to as HOP) datasetusing data from 1446 gauge stations throughout the basin col-lected between 1964 and 2003 by national agencies from Brazil,Peru, Bolivia, Colombia and Ecuador. This database has been ana-lysed using the regional vector method (RVM) (Hiez, 1977) inorderto assess its quality, and 756 stations have been finally retained.

2.2. Precipitation gridded datasets

This section provides an overview of the major characteristics ofthe precipitation datasets considered in this study (see Table 2).

They are (1) the Hybam Observatory Precipitation (HOP) dataset(Espinoza Villar et al., 2009) for the Negro River basin (consideredas the reference dataset); (2) the Climate Prediction Center (CPC)dataset (Silva et al., 2007); (3) the Global Precipitation ClimatologyProject (GPCP) one-degree daily dataset (Huffman et al., 2001); (4)the Tropical Rainfall Measuring Mission (TRMM) Multisatellite Pre-cipitation Analysis (TMPA) dataset (3B42 version 6 V6) (Huffman

et al., 2007); (5) the 40-yr ECMWF Re-Analysis (ERA-40) (Uppalaet al., 2005) precipitation dataset; and (6) the National Centersfor Environmental Prediction Department of Energy (NCEPDOE) Atmospheric Model Intercomparison Project (AMIP-II)reanalysis dataset (referred to as NCEP-2 hereafter) (Kanamitsuet al., 2002). The datasets are available in both daily or sub-dailytime steps. They were selected since they have been widely usedand evaluated in previous hydrological studies (e.g. Ribeiro Netoet al., 2005; Gonalves et al., 2006; Collischonn et al., 2008; Suand Lettenmaier, 2009; Su et al., 2008; Voisin et al., 2008). It isimportant to highlight that none of the gauge-based datasets usedin this study (e.g. HOP, CPC and CRU the latter one is presented inSection 3) has been corrected for gauge biases, including under-catch errors. The correction of individual records requires detailedlocal meteorological and station meta-information, which are notreadily available (New et al., 2000).

2.2.1. The HOP and CPC gauge-based datasets

The HOP Negro River basin precipitation field was created withdata from 115 gauge stations distributed between Brazil (98),Colombia (14) and Venezuela (3) (Fig. 1) for the 19802006 period.These stations are operated by ANA, MARN (Ministerio de MedioAmbiente y de los Recursos Naturales) and IDEAM (Instituto de Hid-rologa, Meteorologa y Estudios Ambientales), respectively. Braziliandata are freely available at the daily time step. Colombian and Ven-ezuelan data were acquired at the monthly time step and werehomogeneously distributed over the number of days of eachmonth. The well-known inverse squares of distances approach

was used to spatially distribute the precipitation data throughoutthe basin for each computational cell of the hydrological model(more details are found in the Modeling setup section). Onlythose stations located within a radius of two times the closest dis-tance between each computational cell and stations were consid-ered in the interpolation procedure.

HOP has been selected as the reference precipitation dataset be-cause of its relatively dense and more comprehensive network inthe Amazon basin when compared to the other gauge-based data-sets presented in the literature. This does not mean that HOP (orany other precipitation dataset) represents the truth. All datasetsare wrought with error. Gauge-based datasets are affected by both(1) error in gauge observations (systematic bias and random error)and (2) interpolation errors (e.g. station density inhomogeneitiesand non-representative station locations).

The Climate Prediction Center (CPC) of the National Oceanic andAtmospheric Administration (NOAA) provides a South Americagridded precipitation dataset from 1978 to 2007 (Silva et al.,2007). The daily gauged data are gridded at a 1.0 1.0 resolutionover South America, using a modified Cressman scheme. A previ-ous work suggests that CPC provides the most accurate precipita-tion estimates over the Amazon basin at a daily time step (Juarezet al., 2009). This dataset comprises the data available in Brazil.However, the absence of a more comprehensive dataset in theother countries within the Amazon basin increases the inaccuracyof the rainfall in these regions.

2.2.2. The GPCP and TMPA satellite-based datasets

As part of the Global Energy and Water Cycle Experiment(GEWEX), the GPCP dataset was set up to develop monthly

20

40

60

80

100

1980 1993 2006

60

70

80

90

100

Data availability (%)Number of stations

Fig. 2. Number of precipitation gauge stations and precipitation data availabilityprovided by the Brazilian Water Agency for the Negro River basin for the19802006 period.

306 A.C.V. Getirana et al. / Journal of Hydrology 404 (2011) 304322


4/19

precipitation data based on remotely sensed data from geostation-ary and polar-orbiting satellites, plus in situ observations.

Several GPCP precipitation datasets are currently available. TheGlobal Precipitation Climatology Project (GPCP) one-degree daily(1dd) dataset (Huffman et al., 2001) has been used in this study.This dataset is available for the 19972006 period and is basedon multiple passive microwave, infrared satellite observationsand gauge observations. The monthly total GPCP 1dd precipitationdataset matches the monthly values of GPCP version 2, which arebased on satellite and are mostly influenced by gauge observa-tions over land areas (Nijssen et al., 2001). Although its resolutionis coarser (1 1) than those of other satellite datasets, GPCP 1dd(further referred to as GPCP) has been selected for this study be-cause of the extensive applications, analyses and validation stud-ies carried out with its monthly and daily versions (Adler et al.,2003; Gebremichael et al., 2003; Voisin et al., 2008; Juarez

et al., 2009), and because of substantial temporal overlap withthe reanalysis and gauge-based datasets used in our analysis(Fig. 3).

The Tropical Rainfall Measuring Mission (TRMM) MultisatellitePrecipitation Analysis (TMPA) dataset (Huffman et al., 2007) hasalso been used in this study. TRMM is a joint mission, initiated inNovember 1997, between the National Space Development Agency(NASDA) of Japan and the National Aeronautics and Space Admin-istration (NASA) of the United States. The mission was designed to

study tropical rainfall between 35N and 35S. The 3B42 version 6dataset has been used in this analysis. It has a temporal resolutionof 3 h, a spatial resolution of 0.25 0.25 and is available from1998 to the present. This dataset has been largely used for hydro-logical studies (e.g. Arvor et al., 2008; Collischonn et al., 2008; Suet al., 2008).

2.2.3. The ERA-40 and NCEP-2 reanalysis-based datasets

The ERA-40 dataset results from observations from varioussources (land, ship, aircraft, satellite, among others) into a short-term operational forecast model. Data are provided four times aday, on a 1.125 1.125 grid. The numerous analyses of this data-set in the literature (e.g. Betts et al., 2003a,b, 2005; Su et al., 2006;Fernandes et al., 2008; Voisin et al., 2008) motivated its use in thiscomparison in the Negro River basin.

A second reanalysis precipitation dataset is the NCEP-2 (Kanam-

itsu et al., 2002), which is an update of the NCEPNational Centerfor Atmospheric Research (NCAR) reanalysis dataset (Kalnayet al., 1996). NCEP-2 is available from 1979 to the present andincorporates observed data and numerical weather forecast simu-lations. However, precipitation is mostly simulated. The NCEP-2dataset is available on an approximately 1.9 1.9 resolution glo-bal Gaussian grid at 6-h time steps. As in the ERA-40 dataset,NCEP-2 is a model-based precipitation dataset frequently usedfor hydrological and climatologic studies (e.g. Roads, 2003; Fekete

Table 2

Precipitation datasets used in the hydrological modeling.

Dataset Reference Format/SpatialResolution

Spatial coverage Series Span Time-step Data-source

HOP Hybam Observatory Precipitation Espinoza Villaret al. (2009)

Catchmentsa Amazon basin 01/198012/2006

Dailyb Rain-gauge

CPC Climate Prediction Center Silva et al. (2007) Grid (1.0 1.0) South America 01/197812/2007

Daily Rain-gauge

GPCP Global Precipitation ClimatologyProject

Adler et al. (2003) Grid (1.0 1.0) Global 10/1997-present

Daily Satellite + rain-gauge

TMPA Tropical Rainfall Measuring Mission Huffman et al.(2007)

Grid(0.25 0.25)

Quasi-global (180W180E,50N50S)

01/1998-present

3-hourly

Satellite + rain-gauge

ERA-40 European Center for Medium RangeWeather Forecast

Uppala et al. (2005) Grid(1.125 1.125)

Global 01/195708/2002

6-hourly

GCMestimatesc

NCEP-2 NCEP DOE AMIP-II Kanamitsu et al.(2002)

Grid(1.875 1.875)

Global 01/1979-present

6-hourly

GCMestimatesc

a Catchments with areas up to 600 km2.b Espinoza Villar et al. (2009) present a monthly time step dataset while a daily version is used herein.c These models were run by assimilating key variables.

500

1500

2500

3500

4500

1980 1985 1990 1995 2000 2005

HOP CPC GPCP TMPA ERA-40 NCEP-2

TMPAbegins

ERA-40e

nds

P(mm)

Fig. 3. Mean annual precipitation over the Negro River basin between1980 and2006. Thefigurealso shows thetime periods when the six datasets were available. The lowerannual precipitation of ERA-40 in 2002 is due to its end in August 2002.



5/19

et al., 2004; Karam and Bras, 2008). More complete descriptions ofthe datasets can be found in the literature and elsewhere in thispaper.

2.3. Precipitation data intercomparison approach

The precipitation intercomparison has been performed in two

steps: first, an analysis is carried out in terms of statistical coeffi-cients computed at different time scales, and at both the basinand sub-basin scales; and second, a comparison focusing on ananalysis of how the precipitation datasets impact the water bal-ance computed by the MGB-IPH model.

The first step of the analysis is performed by using scatter plotsin which the HOP dataset is considered to be the reference dataset.The root mean squared error normalized by the mean of the ob-served values (Nrmse) and the correlation coefficients (r) are con-sidered in the statistical analysis. They are given by

Nrmse

ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffiffiffiffiffiPntt1 xsimt xobst

2

nt

s x1obs 1

r covxsim;xobs

rxsim rxobs2

respectively, where xobs and xsim stand for the observed and simu-lated time series (in this case, x defines daily precipitation), respec-tively, xobs is the mean value of the observed time series,covxsim;xobs stands for the covariance between observed and sim-ulated variables, and rxobs and rxsim are the standard deviations ofxobs and xsim, respectively. tand ntare, respectively, the time indexand the total number of time steps.

In addition, the intercomparison is carried out by detecting rainevents at different precipitation thresholds over the Negro Riverbasin at a daily time step. It is performed by computing the fre-quency bias index (FBI), the probability of detection (POD), the

false alarm ratio (FAR), and the equitable threat score (ETS). Thesecoefficients, summarized in Table 3, are obtained from a 2 2 con-tingency table made up of four parameters (a, b, cand d), where a isthe number of observed rain events correctly detected, b stands forthe number of observed rain events not detected, c is the numberof false alarms (rainfall events detected but not observed), and dis the sum of cases when neither observed nor detected rain eventsoccurred. FBI indicates whether the dataset tends to underestimateor overestimate rain events, FAR and POD compute the fraction offalse alarms and rain occurrences that were correctly detected,respectively, while ETS provides the fraction of rain events (ob-served and/or detected) which were correctly detected. To quantifythe ability of each dataset in predicting light and heavy rainfall

events, rain event detection coefficients are calculated for precipi-tation thresholds of 0.5, 1, 5, 10, and 20 mm/day.

The second step of the comparison focuses on an analysis ofhow the precipitation datasets impact the computation of thehydrological components (evapotranspiration, discharge and soilmoisture) obtained with the MGB-IPH model. Special attention isgiven to discharge, which enables a quantitative evaluation based

on observed data from five gauge stations within the Negro Riverbasin (Fig. 1).

2.3.1. The hydrological model

The MGB-IPH model has been specially designed for studies oflarge basins with relatively scarce data, using globally availabledata as far as possible. The watershed is divided into elements ofarea (normally square grids or cells) interconnected by channels.In order to reduce the number of model parameters and of requiredinput data, the grouped response unit (GRU) approach (Kouwenet al., 1993) is used to handle sub-grid variability. The MGB-IPHis composed of modules enabling the calculation of soil water bal-ance, evapotranspiration, flow propagation within a cell, and flowrouting through the drainage network. Model concepts were ini-

tially based on the LARSIM (Krysanova et al., 1998) and VIC (Woodet al., 1992; Liang et al., 1994) models, with some changes in theevapotranspiration, percolation and channel-routing modules.

Land use, topography, vegetation cover and soil types are usedin the model as inputs to conduct the selection of parameter val-ues. The model has been applied to several studies of basins inSouth America (Tucci et al., 2005), especially in the Amazon Riverbasin (Ribeiro Neto et al., 2005; Collischonn et al., 2008). Main as-pects of those applications are described by Allasia et al. (2006)and a more detailed description of the model can be found in Col-lischonn et al. (2007) and Getirana et al. (2010). Uncertainty andscale aspects of the MGB-IPH model are not evaluated in this study.Details about these aspects can be found in Nijssen and Lettenma-ier (2004) and Hong et al. (2007).

Parameter values are associated with soil type, land use andvegetation cover and to the characteristics of relief, using theGRU approach. The soil water balance is computed independentlyfor each GRU of each cell, considering only one soil layer (Fig. 4)using

Wki;j Wk1i;j P

ki ET

ki;j Dsup

ki;j Dint

ki;j Dbas

ki;j

Dt 3

where k, i andj are indexes related to the time step, the cell and theGRU, respectively; dtis the model time step (1 day in most applica-tions); Wki;j [mm] is the water storage in the soil layer at the end of

Table 3

Summary of the rain event detection coefficients.

Coefficient Full name Equationa Range Optimalscore

FBI Frequencybias index

FBI a b=a c 01 1

POD Probabilityofdetection

POD a=a c 01 1

FAR Falsealarmratio

FAR c=a c 01 0

ETS Equitablethreatscore

ETS a He =a b c He 1to 1

1

a He a b a c=N where N is the total number of estimates. Fig. 4. Schematic of the MGB-IPH soil water budget applied to each groupedresponse unit (GRU) of a cell (Collischonn et al., 2007).



6/19

the kth time step, of the jth GRU of the ith cell; Wk1i;j [mm] is thesame variable at the end of the previous time step; Pki;j is the precip-itation that reaches the soil; ETki;j [mm dt

1] is the evapotranspira-tion flux from the soil; Dsupki;j [mm dt

1] is the surface runoff, orquick flow; Dintki;j [mm dt

1] is the subsurface flow; and Dbaski;j[mm dt1] is the flow to the groundwater reservoir.

The evapotranspiration is calculated based on soil water storageat the start of the time step, according to an explicit scheme intime. The calculation is performed in two steps. Initially, precipita-tion is assumed to be stored on the surface of the vegetation until amaximum interception storage capacity is reached and only theexcess precipitation (throughfall) passes through the canopy toreach the soil surface. The interception storage capacity is deter-mined for each GRU based on the leaf area index (LAI):

Smax a LAI 4

where Smax is the maximum interception storage capacity for aGRU; and a is a model parameter assumed to be a constant valueof 0.2 mm (Ubarana, 1996). LAI values have been considered con-stant in time for each GRU (Xue et al., 1991). Intercepted water isthen evaporated from the interception storage computed by thePenmanMonteith equation, setting the surface resistance to zero.

The remaining energy is used for evapotranspiration of the veg-etated soil (soil evaporation plus plant transpiration) which is cal-culated according to the PenmanMonteith equation weighted bythe remaining evaporative demand, as proposed by Wigmostaet al. (1994):

ET fDE D A qA cp

Dra

D c 1 rsra

!

M

k qW5

where D [kPa C1] is the gradient of the saturated vapor pressuretemperature function, A [MJ m2 s1] the available energy; qA[kg m3] the specific mass of air; qW [kg m

3] the specific mass ofwater; cp [MJ kg

1C1] the specific heat of moist air; D [kPa] the

vapor pressure deficit; c [kPa C1] the psychrometric constant; rs

[s m1] the surface resistance of the land cover; ra [s m1] the aero-dynamic resistance; k [MJ kg1] the latent heat of vaporization; andMa constant for unit conversion between m s1 and mm dt1. fDE isgiven as follows:

fDE EIP EI

EIP6

where EI and EIP are, respectively, the evaporation and potentialevaporation from the interception storage.

Evaporation flux of open-water bodies is calculated consideringfDE = 1. Available energy and aerodynamic resistance can be calcu-lated following Shuttleworth (1993). Please refer to Collischonnet al. (2007) for more details about the calculation of the water bal-ance and evapotranspiration.

2.3.2. Modeling setup

The Negro River basin has been divided into 1756 catchments(computational cells) having areas of up to 600 km2. Atmosphericforcings (solar radiation, pressure, relative humidity, wind speedand temperature) were provided by the NCEPDOE/AMIP-II reanal-ysis dataset (Kanamitsu et al., 2002) and passed through the in-verse square distance interpolation approach to spatiallydistribute the data throughout the basin. Land cover and soil typeswere derived from the JERS-1 image classification (Martinez and LeToan, 2007) and FAO (FAO, 1995) datasets, respectively. Flowdirections and watershed delineations were acquired from a 200-m SRTM DEM product after a pre-processing phase as suggestedby Getirana et al. (2009a,b).

Using the HOP dataset to force the model, parameters (mostlysoil hydraulic properties that control water storage, infiltration,

subsurface flow and groundwater flow) were manually calibratedand validated using a daily time step from 1997 to 2006 by com-paring model results with observed data at 22 gauge stations.The validation procedure was complemented with water levels de-rived from the ENVISAT radar altimetry satellite from 27 virtualstations (as described in Getirana et al., 2010; 2009c). Anotherstudy compared the model outputs with Gravity Recovery and Cli-

mate Experiment (GRACE)-derived total water storage (TWS) datafrom seven sub-basins defined by gauge stations at Caracara, Tar-aqua, Cucu, Curicuriari, Serrinha, Barcelos and Manaus (Getiranaet al., 2009d). The simulated water level in each computational cellwas derived from a discharge height relationship generalized tomost of the basin, thus permitting a validation of the model against35-day repeated cycles of ENVISAT data. Monthly TWS values werederived from the relation dTWSt

dt Pt Rt ETt where P

stands for the observed monthly precipitation used to force themodel, R and ET stand for monthly valuesof runoff and evapotrans-piration, respectively, provided by the model, and tis the timestep.The simulated discharges and evapotranspiration rates bothshowed satisfactory agreement with observed and satellite-de-rived data. NashSutcliffe coefficients of modeled discharges var-ied from 0.47 to 0.94, and correlation coefficients, r, of waterheights and altimetric data averaged 0.86 for the entire basin. Cor-relation coefficients between simulated and GRACE-derived TWSvaried from 0.64 to 0.84, depending upon the sub-basin involved.These parameters have been used in this study to simulate thehydrology of the Negro River basin from 1998 to 2002 for the dif-ferent precipitation datasets. Since MGB-IPH computes verticalwater and energy balances using a daily time step, the 3-hourly(TMPA) and 6-hourly (ERA-40 and NCEP-2) datasets have beencumulated in a 24-h time step to be in accordance with the modelrequirements.

The model has been evaluated using the correlation coeffi-cient (r), the NashSutcliffe efficiency coefficient (NS) and therelative error (RE) of predicted streamflows. The correlation coef-ficient r is defined by Eq. (2). The NS and RE coefficients are gi-

ven by

NS 1

Pntt1Qobst Qsimt

2Pntt1Qobst Qobs

27

RE% 1

ntXntt1

Qobst Qsimt

Qobst

100 8

where Qobs and Qsim are the observedand simulated daily water dis-charges, respectively and Qobs is the average observed discharge.

3. Results and discussion

3.1. Precipitation analysis

3.1.1. Spatial and temporal distribution of the annual precipitation

As mentioned before, the duration of the study period (19982002) was defined by the available precipitation data. TMPA dataare available since January 1998, while the ERA-40 dataset endsin August 2002, resulting in a nearly 5-year period (Fig. 3). Themean annual precipitation during the study period, averaged overthe Negro River basin, varies significantly among the datasets, from2219 (TMPA) to 3065 mm/year (NCEP-2), while HOP presented2715 mm/year. This means that TMPA and NCEP-2 provide thelowest and highest averaged precipitation estimates, resulting inrelative errors of, respectively, 18% and +13% as compared tothe reference dataset (2715 mm/year). The overestimated precipi-

tation provided by NCEP-2 may be explained by two consecutiveyears (1999 and 2000) during which this dataset gives a



7/19

particularly exaggerated rainfall, especially 1999 (Fig. 3). However,when all previous years (19801998) are considered, NCEP-2 actu-ally shows much lower precipitation rates (2070 mm/year) as com-pared to the reference dataset (2530 mm/year) for the sameperiod.

ERA-40 underestimates precipitation over the Negro River basinby about 18% during the period from January 1998 to August 2002.

This result is similar to those found in previous studies with ERA-40 which showed that precipitation over the Amazon basin had abias of 10% (Fernandes et al., 2008) to 26% (Voisin et al.,2008) compared to gauged-based datasets. Other datasets suchas CPC and GPCP give lower errors (7% and 8%, respectively).

Fig. 5 presents the spatial distribution of the mean annual pre-cipitation amounts derived from the six datasets. Overall, goodagreement exists among the datasets if judged in terms of relativevalues within the basin. All datasets show lower precipitation ratesin the extreme Northeast of the basin than on the western side.However, absolute values vary considerably from one dataset toanother. For instance, the northeastnorthwest rainfall gradientobserved in the Caracara sub-basin (northeastern part of the Ne-gro River basin), as provided by HOP and CPC, is not reproducedby the other datasets. Visual inspection of the results reveals thatsomewhat anomalous distributions are given by the NCEP-2reanalysis-based dataset at the basin-wide scale. The latter onecould not properly represent the precipitation gradient that existsbetween the western and northeastern regions.

To improve the intercomparison of the spatial distribution ofprecipitation fields, four additional rain-gauge datasets have beenincluded in the analysis. Three of these are worldwide datasets:the Climate Research Unit (CRU) dataset (New et al., 2000), theWillmottMatsura (WM) dataset (Willmott and Matsuura, 2009)and the Global Precipitation Climate Center (GPCC) dataset (Rudolfet al., 1994). The fourth one (ANA) is very similar to the HOP data-set, except that it only contains data made available by the Brazil-ian Water Agency ANA. Precipitation fields obtained with the

raingauge-based datasets showthat having less rainfall data gener-ally leads to lower precipitation rates over the western part of theNegro River basin, as shown by the CPC and WM datasets. This isprobably due to the use of data from Colombian and Venezuelangauges that are not located within the Amazon basin but in the lessrainy Andes (as shown in Silva et al., 2007). Consequently, they arenot representative of the hydrological regime of the basin. The pre-

cipitation fields of the Amazon basin presented by Costa and Foley(1998) also appear to have the same limitation. On the other hand,Espinoza Villar et al. (2009) showed that some locations in theNorthwest of the Amazon basin are subject to intense precipitationwith rates up to 6000 mm/year. Underestimated precipitationrates can significantly impact the local water balance as describedin the hydrological modeling section. The role of the gauge densityon the construction of precipitation fields and discharges providedby hydrological models is also addressed by Gebremichael and Kra-jewski (2004).

3.1.2. Daily precipitation analysis

Fig. 6 presents scatter plots between daily Negro River basin-averaged precipitation rates in HOP and in the different datasets.The performance of datasets varies significantly. The best esti-mates are obtained with the CPC gauged-based dataset, giving ahigh coefficient of correlation (r= 0.95) and a low root meansquare (Nrmse = 0.24) and relative (RE = 22%) errors. However,CPC tends to underestimate precipitation, with most points situ-ated to the left of the diagonal.

The satellite-based GPCP and TMPA datasets also perform rela-tively well, albeit less than CPC. GPCP performs somewhat better(r= 0.61, Nrmse = 0.67 and RE = 57%) than TMPA (r= 0.52,Nrmse = 0.79 and RE = 59%). In contrast to the results acquiredfor other large basins in South America, such as the La Plata basin(Su et al., 2008) and the Tapajos River basin (Collischonn et al.,2008), TMPA reveals a slight tendency to underestimate intenserain events (>20 mm/day) in the Negro River basin. ERA-40

Fig. 5. Spatial distribution of mean annual precipitation for the 19982002 period derived from the six datasets considered in this intercomparison (HOP, CPC, GPCP, TMPA,ERA-40 and NCEP-2), and also from ANA, CRU, WM and GPCC datasets. Pmin, Pmax and Pmed represent, respectively, the minimum, maximum and average annual precipitationover the Negro River basin. In order to compensate the early interruption of the ERA-40 time series (August 2002), monthly mean precipitation rates were computed fromJanuary to December and then summed, resulting in the mean annual precipitation rate.



8/19

statistics are comparable to those provided by the satellite-baseddatasets (r= 0.54 Nrmse = 0.70 and RE = 65%), but they show a ten-dency to underestimate high precipitation rates (>15 mm/day).

Similar results were obtained by Fernandes et al. (2008), whonoted that ERA-40 closely reproduces rainfall during the dry andtransitional seasons, but it underestimates the wet season rainfall.The worst results are obtained with NCEP-2 (r= 0.31, Nrmse = 0.91and RE = 115%). The elevated RE value of that dataset is due to thelarge rainfall overestimation in both the wet and the dry seasons.

In order to further elucidate differences between the datasets, arain event detection analysis over the Negro River basin has alsobeen performed. Fig. 7 shows the results for the frequency bias in-dex (FBI), the false alarm ratio (FAR), the probability of detection(POD), and the equitable threat score (ETS) for each of the five pre-cipitation datasets for the 19982006 period. Results vary substan-tially among datasets. The various datasets cannot detect the samefrequency of extreme rain events (20 mm) as the

reference dataset. For a thresholdprecipitation 20 mm/day) is underestimated by CPC and ERA-40, whichhave FBI values of 0.36 and 0.41, respectively, and overestimatedby NCEP-2 (6.0), TMPA (2.32) and GPCP (1.82). However, accordingto the HOP dataset, severe rain events represent less than 1% of thetotal number of events.

The CPC dataset produces optimal FAR values for extreme rainevents (thresholds 20 mm). This means that thedataset does not tend to falsely determine no rain or extremerainfall events. The gauge-based dataset provides higher FAR val-ues for rain events between 0.5 and 20 mm, with the coefficient

varying from 0.11 (threshold 1020 mm) to 0.28 (threshold 0.51.0 mm). The satellite and model-based datasets are unable to givesuch favorable ratios for false alarms. The TMPA, GPCP, ERA-40 andNCEP-2 datasets have a tendency to announce false rain eventswith thresholds up to 1.0 mm, with FAR values varying from 0.77to 1.0. FAR values for these same datasets approximately decrease(and hence their performances increase) for thresholds between1.0 mmand 20 mm, reaching values from 0.41 to 0.58, but then in-crease again for thresholds >20 mm. Among the non-gauge baseddatasets, ERA-40 shows the best FAR values, followed by TMPA,GPCP and NCEP-2. GPCP and NCEP-2 provide quite similar tenden-cies to announce false rain events.

Like FBI, rain occurrence is best detected for rain events be-tween 1 and 10 mm. The CPC again produces the best results, with

POD values being larger than 0.70 for thresholds 15 mm, 510 mm and 1020 mm. Values decrease rapidly (POD < 0.45) for

0

10

20

30

40

50

60

r=0.95

Nrmse=0.24

RE=22%

0

10

20

30

40

50

60

r=0.52

Nrmse=0.79

RE=59%

0

10

20

30

40

50

60

r=0.61

Nrmse=0.67

RE=57%

0

10

20

30

40

50

60

r=0.54

Nrmse=0.70

RE=65%

0

10

20

30

40

50

60

0 10 20 30 40 50 60

0 10 20 30 40 50 60

0 10 20 30 40 50 60

0 10 20 30 40 50 60

0 10 20 30 40 50 60

r=0.31

Nrmse=0.91

RE=115%

TMPA (mm/d) ERA-40 (mm/d)

GPCP (mm/d) NCEP-2 (mm/d)CPC (mm/d)

Gauged(mm/d)

Gauged(mm/d)

Fig. 6. Scatter plots of daily basin wide-averaged precipitation from the five datasets (CPC, TMPA, GPCP, ERA-40 and NCEP-2) versus the gauged estimates (HOP) over theNegro River basin for the January 1998August 2006 period.



9/19

thresholds of 20 mm. Among the non-gauge-baseddatasets, TMPA performs better in detecting extreme rain events(POD = 0.5 for threshold 20 mm), whereas ERA-40 shows the best results overall.

The ability to detect rain events is also evaluated in terms of theequitable threat score (ETS). CPC once again is notably superior indetecting observed rain events, which is easily explained by itsrain-gauge basis. ETS values vary from 0.26 to 0.61. The best resultsfor detection is for thresholds between 1 and 20 mm. Rain eventsof less than 1 mm and above 20 mm show lower ETS values, butare still quite acceptable. Among the non-gauge-based datasets,TMPA shows the best overall ability to detect rainfall events, hav-ing higher ETS values for most thresholds than the other datasets,except for >20 mm for which ERA-40 produces the best results(ETS = 0.10). All datasets provide some skill in detecting rain(ETS > 0) for the different thresholds, except ERA-40 and NCEP-2.Both datasets are unable to detect any observed rain (ETS = 0) withthresholds 20 mm, respectively.

3.1.3. Monthly precipitation analysis

Fig. 8 shows monthly precipitation estimates for the six sub-ba-sins after subtracting a constant value from the whole series, andplots of the correlation coefficients, r, between the different data-sets. The monthly precipitation time series have been shifted usingtheir respective mean annual averages, thus showing the seasonalvariability of the precipitation dynamics. Most datasets properlyrepresent the annual seasonality in all sub-basins. However, theNCEP-2 dataset is unable to satisfactory represent both the annualcycle and the absolute precipitation values.

Correlation coefficients have also been calculated to see howthe various datasets relate to each other. Based on these valuesand considering only the non-gauged based datasets, the best

and worst results are obtained with the GPCP and NCEP-2 datasets,respectively. Values for r vary considerably from one sub-basin to

another. The best agreement among the datasets is found for theCaracara sub-basin (as defined by the Caracara station), for whichr values between two datasets vary from 0.70 (between TMPA andERA-40, and TMPA and NCEP-2) to 0.99 (between HOP and CPC).

TMPA shows the lowest correlation with the HOP dataset(r= 0.78). NCEP-2 and HOP also are not much correlated(r= 0.82), while the other three datasets (produce) display correla-tion coefficients higher than 0.90.

The mean correlation coefficients (r) between a dataset and theother five are listed in Table 4. The HOP and CPC precipitation data-sets show the best correlations for the Caracara sub-basin (bothwith r 0:90) while TMPA (provides) presents the worst correla-tions (r 0:75) for this sub-basin. Correlation coefficients betweendatasets are lower for the other sub-basins. The lowest correlationcoefficients (both for r and r) are obtained for the Taraqua sub-ba-sin (Fig. 8, see the small size of the graphic). Values for r vary be-tween zero (for ERA-40 and NCEP-2), indicating no correlation atall, and 0.82 (HOP and CPC). TMPA provides overall better correla-

tions (e.g. r 0:35 for Taraqua and r 0:66 for Manaus) thanNCEP-2 (0.26 and 0.61, respectively).The contrasting results obtained for the Caracara and Taraqua

sub-basins suggest that most datasets are highly dependent ongauge observations. Datasets are better correlated in basins wherein situ data are available (e.g. Caracara basin), and less correlatedin regions where data are unavailable or inexistent (e.g. Taraquabasin). This means that the quality of satellite-based datasets isintrinsic to the gauge network density.

A low correlation between rainfall datasets found in theTaraqua sub-basin is also noticed in other nested sub-basins (Cur-icuriari and Serrinha), as seen in Table 4. For example, mean corre-lation coefficients vary from 0.46 and 0.42 (both obtained forNCEP-2) to 0.70 and 0.71 for Curicuriari and Serrinha, respectively.

Still, correlations between monthly precipitation time series aver-aged for the entire Negro River basin (represented by the Manaus

Frequency biase index(FBI)

Probability of detection(POD)

Equitable threat score(ETS)

Thresho

ld(mm/day)

False alarm ratio(FAR)

20

Th

reshold(mm/day)

20

0 0.2 0.4 0.6 0.80 0.2 0 .4 0 .6 0.8 10 0.2 0.4 0 .6 0.8 10 0.5 1 1.5 2

0 600 1200 18000 200 400 6000 200 400 6000 200 400 600

ca db

CPC

TMPA

GPCP

NCE

P-2

ERA-40

Fig. 7. Coefficients frequency bias index (FBI), false alarm ratio (FAR), probability of detection (POD) and equitable threat score (ETS), and a, b, c and d for six dailyprecipitation thresholds over the Negro River basin for the CPC, TMPA, GPCP, ERA-40 and NCEP-2 datasets. The HOP dataset is considered to be the reference precipitation. a,b, c and d represent the number of observed rain events correctly detected, the number of observed rain events not detected, the number of false alarms (rainfall eventsdetected but not observed), and the sum of cases when neither observed nor detected rain events occurred, respectively. FBI, FAR, POD and ETS are described in Table 3.



10/19

station) exhibit overall good similarity between the HOP and CPC

datasets (both with r 0:82). GPCP also provides competitive cor-relations, with a mean correlation coefficient ofr 0:80. By com-parison, ERA-40 (r 0:73) gives a better mean correlation valuethan TMPA (r 0:66) and NCEP-2 (r 0:61).

3.2. Model outputs

A model sensitivity analysis was performed considering threehydrological components: evapotranspiration (ET), runoff (R) andsoil moisture content (W). Fig. 9 shows mean daily values of thehydrological components for the 19982002 period averaged forthe six sub-basins within the Negro River basin. The hydrologicalmodel forced by HOP results in moderate ET spatial heterogeneityamong sub-basins, varying from 3.03 mm/day in the Caracara ba-

sin to 4.30 mm/day in the Taraqua basin, averaging 3.28 mm/dayfor the entire basin. Mean ET estimates are in good agreement with

previous studies in the region. For example, Shuttleworth (1988)

found a value of 3.76 mm/day for the Reserva Ducke site, Malhiet al. (2002) obtained 3.08 mm/day at Reserva Biolgica do Cuie-iras, and Tomasella et al. (2007) estimated ET to be 3.86 mm/dayfor the Asu catchment. Runoff values (model forced by HOP) varyfrom 2.52 mm/day (Caracara basin) to 6.64 mm/day (Curicuriaribasin), averaging 4.15 mm/day for the entire basin.

Mean ET values averaged over the entire Negro River basin (asrepresented by the Manaus station) vary from 3.13 (TMPA) to3.44 mm/day (ERA-40), with an average of 3.29 mm/day for alldatasets. Larger differences are found in drier areas (Caracara),where ET varies between 2.83 mm/day (TMPA) and 3.40 mm/day(NCEP-2), and averaging 3.15 mm/day. This corresponds to differ-ences of up to 17% among the datasets, which is less than thedifferences found for precipitation among the same datasets

(33%). More humid locations such as Taraqua present ET values be-tween 4.03 (TMPA) and 4.27 mm/day (HOP), with an average of

0.00

0.25

0.50

0.75

1.00HOP

CPC

GPCP

TMPA

ERA-40

NCEP-2

0.00

0.25

0.50

0.75

1.00HOP

CPC

GPCP

TMPA

ERA-40

NCEP-2

0.00

0.25

0.50

0.75

1.00HOP

CPC

GPCP

TMPA

ERA-40

NCEP-2

-500

0

500

1/98 1/99 1/00 1/01 1/02

-500

0

500

-500

0

500Caracarai

Taraqua

Cucui


Fig. 8. Monthly precipitation estimates (mm/month) for the 19982006 period after subtracting a constant (monthly mean) value from each series (left) and plots of thecorrelation coefficient between each of the datasets andthe other fiveones at themonthlytimestep (right). Thecorrelationof a precipitation dataset with itself (which wouldbe 1.0) is not shown in the correlation plots.



11/19

4.14 mm/day. This represents differences of less than 3% amongdatasets. Small differences in ET were also found by Voisin et al.(2008) for the Amazon basin using the VIC model.

Despite the high spatial heterogeneity of the precipitation

fields, as shown in Fig. 3, mean ET values derived from the differentdatasets rates are very similar. One can notice that a remarkable

gradient of evapotranspiration values is obtained with the six data-sets, as shown in Fig. 10. Mean ET values are lower in the easternpart of the basin, but increases significantly in the western part.Differences between extreme values vary from 73% for CPC (ET-

min = 973 mm/year and ETmax = 1681mm/year) and NCEP-2 (ET-min = 995 mm/year and ETmax = 1654 mm/year) to 123% for HOP(ETmin = 771 mm/year and ETmax = 1718 mm/year). Also note thatin all of the datasets the presence of a lower energy demand inthe central part of the basin. The agreement of ET spatial distribu-tion derived from the six datasets is an indication that ET is notwater limited, which means that water availability (as providedby the precipitation datasets) has a reduced effect on evapotrans-piration. This is demonstrated by the fact that expressive differ-ences between rainfall datasets in the western (and more humid)side of the basin result in small variations in ET (see the monthlytime series of the precipitation averaged for the Taraqua sub-basinin Fig. 7 where ET derived from the six rainfall datasets has insig-nificant differences). The exception is the driest region located in

the north-eastern part of the basin where ET has more accentuateddifferences among datasets. This is probably due to the fact that ET

0.00

0.25

0.50

0.75

1.00HOP

CPC

GPCP

TMPA

ERA-40

NCEP-2

0.00

0.25

0.50

0.75

1.00HOP

CPC

GPCP

TMPA

ERA-40

NCEP-2

0.00

0.25

0.50

0.75

1.00HOP

CPC

GPCP

TMPA

ERA-40

NCEP-2

-500

0

500

1/98 1/99 1/00 1/01 1/02

-500

0

500

-500

0

500Curicuriari

Serrinha

Manaus


Fig. 8 (continued)

Table 4

Mean correlation values (r) of the monthly precipitation time series. Simple

correlations between each monthly rainfall data set and all the others are averaged

in order to obtain the mean correlation between rainfall files, for each dataset and

each sub-basin within the Negro River basin.


Caracara 0.90 0.90 0.88 0.75 0.84 0.79Taraqua 0.54 0.57 0.53 0.35 0.24 0.26Cucu 0.75 0.75 0.73 0.57 0.63 0.60Curicuriari 0.70 0.71 0.65 0.50 0.52 0.46Serrinha 0.69 0.70 0.65 0.49 0.52 0.42

Manaus 0.82 0.82 0.80 0.66 0.73 0.61



12/19

values in drier basins are more sensitive to precipitation than inwetter basins. According to the monthly time series of the evapo-transpiration averaged for the Caracara basin, more significant dif-ferences of ET occur during the dry seasons. Note also that dryseasons are practically not detected by the evapotranspiration de-rived from NCEP-2 due to the overestimated precipitation duringthese periods.

Basin-wide precipitation derived from the HOP dataset is only8% higher than that obtained with CPC. However, as ET valuesare about the same, differences between runoff values derivedfrom the same datasets increase to 26%. The differences are moresignificant in the Taraqua sub-basin where precipitation and runoffrates between HOP and CPC differ by 22% and 125%, respectively. Ingeneral, the averaged basin-wide runoff varies from 2.70 mm/day

(ERA-4) to 4.83 mm/day (NCEP-2). Runoff values derived fromthe satellite-based datasets are underestimated throughout the ba-sin, although GPCP presents overall better results since values arerelatively close to the reference dataset. For example, GPCP andTMPA provide 3.40 and 2.82 mm/day, respectively, for the NegroRiver basin scale.

The spatial distribution (discretized by sub-basins) of the soilmoisture content W shows a high sensitivity to the precipitationdatasets, but the basin-wide values are relatively constant amongthe datasets, with variations of up to 10% between 301 mm(ERA-40) to 334 mm (NCEP-2) averaged over the study period.The simulation with the reference precipitation (HOP dataset) pro-vides W= 333 mm, while the mean W value for all datasets is320 mm. Extreme W values are found for the dry Caracara and

the wet Taraqua sub-basins, where variations among the datasetsare also significant. The mean soil moisture content of the Cara-

cara sub-basin varies between 168 mm (TMPA) and 248 mm(NCEP-2), while HOP produces a value of 210 mm. NCEP-2 alsoprovides the highest mean soil moisture content for the wet Tar-aqua sub-basin (523 mm) while the lowest value is given by CPC(360 mm). HOP has a mean W of 465 mm.

The above results show that disparities among the precipitationdatasets of the western Negro River basin are reflected in similardisparities among the model output variables, notably runoff andsoil moisture content. For this reason, modeled and observed dis-charges are compared at five gauge stations located within thebasin.

Fig. 11 shows the mean annual cycles ofP, ET, R and Waveragedover the Negro River basin. The mean precipitation peaks occur inMay while the ET and W peaks arise 1 month later. Evapotranspi-

ration rates are the highest between July and September, withpeaks occurring in July. The exception is NCEP-2, which shows itsprecipitation peak 1 month sooner. This dataset also overestimatesthe dry period from November to March. Mean R and W values,however, show the same temporal variability as the other vari-ables. Lower precipitation peak values lead to lower amplitude inR and Wpeaks. The ERA-40 dataset shows the lowest precipitationpeak among all datasets, but is able to properly capture the hydro-logical seasonality of the Negro River basin.

3.3. Water discharge analysis

Fig. 12a and b presents the daily hydrographs derived from theMGB-IPH model forced with the six precipitation datasets at five

gauge stations (Caracara, Taraqua, Cucu, Curicuriari and Serrinhastations). Modeled discharges show a general agreement with

Evapotranspirat ion Runoff Soil Moisture

Caracarai

0

3

6

9

12

0

200

400

600

Curicuriari

0

3

6

9

12

0

200

400

600

Taraqua

0

3

6

9

12

0

200

400

600

Serrinha

0

3

6

9

12

0

200

400

600

Cucui

0

3

6

9

12


200

400

600

Manaus

0

3

6

9

12




HOP CPC GPCP TMPA ERA-40 NCEP-2 HOP CPC GPCP TMPA ERA-40 NCEP-20

200

400

600

R,E

T(mm/day)

W(mm)

Fig. 9. Mean daily evapotranspiration (ET), runoff (R) and soil moisture (W) at the selected stations for the 19982002 period.



13/19

gauged data when forced with HOP. CPC leads to good results inthe Caracara basin but not in the western Negro basin becauseof the lack of adequate Colombian and Venezuelan rainfall. Quali-tatively, most datasets (TMPA, GPCP and ERA-40) considerablyunderestimate discharges while NCEP-2 overestimates them. Ex-cept for NCEP-2, all datasets give positive NashSutcliffe (NS) coef-ficients (Table 5) at the Caracara station. Positive NS valuesindicate that the modeled discharges are able to represent gaugeddata more accurately than the mean observed discharge. HOP, CPCand GPCP provide lowrelative errors (up to 17.2% Table 6)atthe

Caracara station, whereas TMPA, ERA-40 and NCEP-2 producehigher relative errors. Correlation values r (Table 7), vary between0.78 (ERA-40) and 0.95 (CPC). The relatively high r values are dueto the same seasonality presented in all of the datasets. CPC showsthe best NS and RE coefficients at the Caracara station, exceedingthose values provided by the HOP dataset. On the other hand, HOPis the only dataset to show a positive NS coefficient (0.69) at theTaraqua station, while the other precipitation datasets have nega-tive values. This means that the observed mean is better than mod-el outputs. The best relative error at Taraqua is given by HOP(4.8%), followed by NCEP-2 (8.6%). The other datasets result inpoorer RE coefficients (less than 44%). Discharges provided bythe MGB-IPH model forced with HOP performed well, with NScoefficients varying from 0.80 to 0.86 and RE values between

2.5% (Serrinha) and 13.7% (Curicuriari). Following HOP, GPCPgenerally produces better performance coefficients for the simu-

lated discharges as reflected by the NS values of 0.65 at Cucuand 0.42 at Curicuriari. The GPCP relative error for streamflowsat the Serrinha station was 27.7%. CPC also provides good results,followed by TMPA and ERA-40. NCEP-2 has the worst results.

Fig. 13 shows the mean monthly observed and simulated dis-charges at five gauge stations. In general, NCEP-2 overestimatesdischarges all over the study period at all stations and gives aweaker annual amplitude than observed. Other datasets providefairly consistent discharges during the wet seasons, but peaks aremostly underestimated. At Caracara, NCEP-2 substantially overes-

timates observations from July to December, while also being toohigh during the dry season. HOP and CPC at this station both pro-duce the best discharge estimates. The NCEP-2 also produces over-estimates of the discharge at the Cucu station. Similarly to HOP,GPCP provides the closest mean monthly discharge time series atthat station. Because of their upstream location, the runoff com-puted for the Taraqua and Cucu sub-basins affects the meanmonthly discharges at Curicuriari and Serrinha stations. Most data-sets, except for HOP and GPCP, significantly underestimate ob-served discharges in the wet season at these four locations.

4. Conclusions

This study focuses on two main objectives: the evaluation of theability of six datasets to reproduce rainfall in the Rio Negro basin

2

4

6 Taraqua

0

2

4

6ET(mm/day)

Caracarai

2

3

4

5

1/98 1/99 1/00 1/01 1/02

Manaus

HOP

CPC

GPCP

TMPA

ERA-40

NCEP-2

Fig. 10. Top: spatial distribution of mean annual evapotranspiration for the 19982002 period derived from the MGB-IPH model forced with the six datasets considered inthis intercomparison (HOP, CPC, GPCP, TMPA, ERA-40 and NCEP-2). ETmin, ETmax and ETmed represent, respectively, the minimum, maximum and average annual

evapotranspiration over the Negro River basin. Bottom: monthly evapotranspiration time series (in mm/day) averaged for three sub-basins (Caracara, Taraqua and Manaus).



14/19

during the 19982002 period, as given by the Hybam ObservatoryPrecipitation dataset (HOP); and the simulation of the water bal-ance by forcing the MGB-IPH hydrological model. The HOP datasethas been chosen as a reference as it is gauge-based and includes alarge number of stations. The other five rainfall datasets are gauge-based (CPC), satellite-based (GPCP and TMPA) and reanalysis-based (ERA40 and NCEP-2).

The rainfall datasets have been compared at the annual,

monthly and daily temporal scales and also at the basin-wideand catchment scales. Mean annual rainfall over the basin variedsubstantially from a dataset to another, and, in particular ERA-40and satellite products had the largest underestimations. TheEastWest gradient of annual rainfall was not correctly reproducedby some datasets. For instance, the CPC datasets underestimatedrainfall in the western most part of the basin because rainfall inter-polation was performed using data from relatively dry Colombianstations that are located in the Andes, outside of the Amazon basin(Silva et al., 2007). This has been confirmed by the comparison ofmonthly rainfall in six sub-basins of the Rio Negro: the best agree-ment between datasets has been found in the north-eastern Cara-cara basin, while more significant differences have been observedin the western Taraqua basin where comprehensive observed data

are not found in all of the datasets. This explains why CPC achievesbetter scores in this region when compared to HOP. Moreover, theanalysis of daily events through various indicators [frequency biasindex (FBI), probability of detection (POD), false alarm ratio (FAR)and equitable threat score (ETS)] has confirmed the notable supe-riority of the CPC dataset which is explained by its rain gauge basis.Among the non-gauge-based datasets, TMPA has provided the bestresults for most rainfall thresholds. Finally, the reanalysis datasetsexhibited substantial deficiencies in representing the temporal andspatial distribution of rain events in the Rio Negro basin resultingin overestimated rainfall (as shown in Figs. 5 and 6) and inconsis-tent rain events (as shown in Figs. 7 and 8) over the entire thebasin.

In conclusion, limited data availability and the use of unrepre-

sentative rain-gauge stations located outside the Amazon basinin the interpolation process are the most limiting factors for

constructing precipitation datasets over poorly gauged and unga-uged regions. This problem has been reduced considerably byusing the HOP dataset as it includes stations from the north-wes-tern and western parts of the Negro River basin that are missingin other datasets. For that reason, the reference parameter set usedto run the MGB-IPH model has been chosen as the one obtained bycalibrating the model with the HOP dataset. This procedure cer-tainly biased the water balance towards the HOP dataset, but it

is argued that this rainfall dataset is likely the most complete data-set comprised of rain gauges which has ever been used over thisregion.

The analysis of the water balance simulated by the MGB-IPHmodel forced by the different data sets showed that the meanevapotranspiration (ET) is in good agreement among the data-sets. The similarity among the mean and the spatially distributedET derived from the different datasets reveals an energy limitedevapotranspiration in most parts of the basin. The exception isthe Caracara basin which is the driest region of the study area.In term of spatial distribution, soil moisture content and runoffare more sensible to rainfall datasets. While soil moisture con-tent tends to be relatively constant among the datasets, thoughit varies in space, water discharge varies considerably in space

and among datasets. The HOP and CPC gauge-based datasetsprovided the most consistent runoff at the six gauge stationsconsidered in the study. They best reproduced the mean values,the annual cycle and interannual variability. This has beenshown for the Caracara basin where all datasets generallybehave reasonably well. CPC is not as good as HOP in thewestern part of the Rio Negro basin, where most of the datasetsprovided bad results. The GPCP satellite-based dataset has pro-vided good results despite its relatively coarse resolution. How-ever, as expected considering the rainfall biases found in thefirst part of the analysis, TMPA and ERA-40 underestimate waterdischarges.

Defining the best precipitation dataset to be used in general sit-uations is a difficult or even impossible task. Assessments must be

performed to identify the best datasets for specific regions. Thisintercomparison identified limitations of datasets that incorpo-

ERA-40

NCEP-2

TMPA

GPCP

HOP

0

100

200

300

400

CPC

0

100

200

300

400

J F M A M J J A S O N DJ F M A M J J A S O N DJ F M A M J J A S O N D

P,Q,E

T,W(m

m)

Precipitation (P) Runoff (R) Evapotranspiration (ET) Soil Moisture (W)

Fig. 11. Monthly mean precipitation (P), evapotranspiration (ET), runoff (R) and soil moisture content (W) averaged over the Negro River basin for the 19982002 period asderived from the hydrological modeling forced with the different precipitation databases.



15/19

rated information from gauged data, satellite observations or large

scale atmospheric model outputs for the hydrological modeling ofthe Negro River basin. Improved understanding of rainfall variabil-

ity during long time periods will also potentially help atmospheric

models to improve estimates of precipitation, thus improving pre-diction of future rainfall variability.

0

5,000

10,000

15,000Caracarai

0

2,500

5,000

7,500Taraqua

Cucui

0

5,000

10,000

15,000

0

10,000

20,000

30,000Curicuriari

Serrinha

0

13,000

26,000

39,000

1-1-98 1-1-99 1-1-00 1-1-01 1-1-02

Obs HOP TMPA ERA-40

Q (m3/s)

Fig. 12a. Comparison of observed and simulated discharge values obtained with the MGB-IPH model using a daily time step and forced by the HOP, TMPA and ERA-40precipitation datasets at five gauge stations (Caracara, Taraqua, Cucu, Curicuriari and Serrinha).



16/19

Other satellite-based datasets that could be included in further

analyses are the Climate Prediction Center Morphing techniqueprecipitation dataset, CMORPH (Joyce et al., 2004) and the Precip-

itation Estimation from Remotely Sensed Information using Artifi-

cial Neural Networks, PERSIANN (Sorooshian et al., 2000). LikeTMPA and GPCP, these datasets are also based on passive micro-

0

5,000

10,000

15,000Caracarai

0

2,500

5,000

7,500Taraqua

Cucui

0

5,000

10,000

15,000

0

10,000

20,000

30,000Curicuriari

Serrinha

0

13,000

26,000

39,000

1-1-98 1-1-99 1-1-00 1-1-01 1-1-02

Obs CPC GPCP NCEP-2

Q (m3/s)

Fig. 12b. Comparison of observed and simulated discharges obtained with the MGB-IPH model using a daily time step and forced by the CPC, GPCPand NCEP-2 precipitationdatasets at five gauge stations (Caracara, Taraqua, Cucu, Curicuriari and Serrinha).



17/19

wave and infrared data and constitute the latest generation of highspatiotemporal resolution precipitation data derived from satelliteobservations.

Future spatial missions such as the Global Precipitation Mea-surement (GPM) and the Franco-Indian Megha Tropiques mission

are global efforts with the objective of obtaining more accurateprecipitation estimates at high spatio-temporal resolutions. Theavailability of such new data will require a better understandingof the potential and limitations of the available satellite precipita-tion products for hydrological studies. International initiativessuch as those proposed by the International Precipitation WorkingGroup (IPWG), the Program to Evaluate High Resolution Precipita-

tion Products (PEHRPP) and the Distributed Model Intercompari-son Project (DMIP) promote the assessment and understanding ofprecipitation fields and their impacts on the water cycle at themeso, regional and global scales via hydrological modeling. Theseglobal collaborative efforts should support more comprehensivestudies of present and future water resources availability and pro-vide guidelines for future development and water management.

Acknowledgments

The first author would like to thank CNPq and CAPES (Projeto516/05) for financial support. The study benefitted from data madeavailable by Agncia Nacional de guas (ANA, Brazil), Ministerio deMedio Ambiente y de los Recursos Naturales (MARN, Venezuela),Instituto de Hidrologa, Meteorologa y Estudios Ambientales(IDEAM, Colombia) and Observatoire de Recherche en Environne-ment Hybam (INSU). Grateful acknowledgements are also due toG. Cochonneau (IRD), B. Collischonn (ANA), J.-L. Guyot (IRD) andA. Laraque (IRD) for their help in data acquisition and processing,to M. Th. van Genuchten (UFRJ), A. Boone (CNRM/Mto France)and three anonymous reviewers for their suggestions and revisionof the manuscript, and to W. Collischonn for providing the MGB-IPH model. The authors also would like to acknowledge GSFC/DAAC, NASA for providing TMPA data and V.B.S. Silva (NOAA) forproviding the CPC dataset for South America. The GPCP-1dd data-set was obtained from the NOAA National Climatic Data Center(http://www.ncdc.noaa.gov/), the NCEP/NCAR Reanalysis datasetfrom the NOAA Climate Diagnostics Center (http://www.cdc.noaa.-

gov/), and the ECMWF ERA-40 dataset from the ECMWF Data Ser-ver (http://www.ecmwf.int/).

Table 5

Values of the NashSutcliffe (NS) efficiency coefficient for predicted streamflows.

Station HOP CPC GPCP TMPA ERA-40 NCEP-2


Table 6

Values of the relative error (RE) for predicted streamflow volumes.



Table 7

Correlation coefficients (r) between observed and predicted streamflow using

different rainfall products.



Cucu

0

3500

7000

10500

14000Caracara

0

3000

6000

9000

12000Taraqua

0

1500

3000

4500

6000

Curicuriari

0

7500

15000

22500

30000Serrinha

0

9000

18000

27000

36000

J F M A M J J A S O N DJ F M A M J J A S O N D J F M A M J J A S O N D

J F M A M J J A S O N D J F M A M J J A S O N D

HOP CPC GPCP TMPA ERA-40 NCEP-2 OBS

Q(m3/s)

Q(m3/s)

Fig. 13. Mean monthly discharges at five gauge stations (Caracara, Taraqua, Cucu, Curicuriari and Serrinha).



18/19

References

Adam, J.C., Lettenmaier, D.P., 2003. Adjustment of global gridded precipitation forsystematic bias. J. Geophys. Res. 108, 114.

Adler, R.F., Huffman, G.J., Chang, A., Ferraro, R., Xie, P.-P., Janowiak, J., Rudolf, B.,Schneider, U., Curtis, S., Bolvin, D., Gruber, A., Susskind, J., Arkin, P., Nelkin, E.,2003. The Version-2 Global Precipitation Climatology Project (GPCP) MonthlyPrecipitation analysis (1979-present). J. Hydrometeorol. 4, 11471167.

Adler, R.F., Kidd, C., Petty, G., Morissey, M., Goodman, H.M., 2001. Intercomparison

of global precipitation products: the third precipitation intercomparison project(PIP-3). Bull. Am. Meteorol. Soc. 82 (7), 13771396.Allasia, D.G., Silva, B., Collischonn, W., Tucci, C.E.M., 2006. Large basin simulation

experience in South America. IAHS Publication n. 303, pp. 360370.Arvor, D., Dubreuil, V., Ronchail, J., Meirelles, M., 2008. Apport des donnes TRMM

3B42 ltude des prcipitations au Mato Grosso. Climatologie 5, 4970.Betts, A.K., Ball, J.H., Bosilovich, M., Viterbo, P., Zhang, Y.-C., Rossow, W.B., 2003a.

Intercomparison of water and energy budgets for five Mississippi subbasinsbetween ECMWF reanalysis(ERA-40) and NASAData Assimilation Office fvGCMfor 19901999. J. Geophys. Res. 108, 8618. doi:10.1029/2002JD003127.

Betts, A.K., Ball, J.H., Viterbo, P., 2003. Evaluation of the ERA-40 surface waterbudget and surface temperature for the Mackenzie River basin. J. Hydrometeor.4, 11941211.

Betts, A.K., Ball, J.H., Viterbo, P., Dai, A., Marengo, J., 2005. Hydrometeorology of theAmazon in ERA-40. J. Hydrometeor. 6, 764774.

Cochonneau, G.,Sondag, F.,Guyot, J.L., Boaventura, G.,Filizola, N.,Fraizy, P., Laraque,A., Magat, P., Martinez, J.M., Noriega, L., Oliveira, E., Ordonez, J.J., Pombosa, R.,Seyler, F., Sidgwick, J., Vauchel, P., 2006. Lobservatoire de recherche enenvironnement ORE HYBAM sur les grands fleuves amazoniens. In: Demuth,

S., Gustard, A., Planos, E., Scatena, F., Servat, E. (Eds.), 5th FRIEND WorldConference Climate Variability and Change, Hydrological Impacts. La Havane,Cuba, 11/2006, IAHS, 308 4450.

Collischonn, W., Allasia, D., Silva, B.C., Tucci, C.E.M., 2007. The MGB-IPH Model forLarge-Scale Rainfall-Runoff Modelling. Hydrol. Sci. J. 52 (5), 878895.

Collischonn, B., Collischonn, W., Tucci, C.E.M., 2008. Daily hydrological modeling inthe Amazon basin using TRMM rainfall estimates. J. Hydrol.. doi:10.1016/j.jhydrol.2008.07.032.

Costa, M.H., Foley, J.A., 1998. A comparison of precipitation datasets for the Amazonbasin. Geophys. Res. Lett. 25 (2), 155158.

Dinku, T., Anagnostou, E.N., 2005. Regional Differences in Overland RainfallEstimation from PR-Calibrated TMI Algorithm. J. Appl. Meteorol. 44, 189205.

Espinoza Villar, J.C., Ronchail, J., Guyot, J.-L., Cochonneau, G., Filizola, N., Lavado, W.,Noriega, L., de Oliveira, E., Pombosa, R., Romero, H., Vauchel, P., 2009. Spatio-temporal rainfall variability in the Amazon basincountries (Brazil, Peru, Bolivia,Colombia, and Ecuador). Int. J. Climatol. 29 (11), 15741594. doi:10.1002/joc.1791.

Fekete, B.M., Vrsmarty, C.J., Roads, J.O., Willmott, C.J., 2004. Uncertainties inprecipitation and their impacts on runoff estimates. J. Clim. 17, 294304.

Fernandes, K., Fu, R., Betts, A.K., 2008. How well does the ERA40 surface waterbudget compare to observations in the Amazon River basin? J. Geophys. Res.113, D11117. doi:10.1029/2007JD009220.

Figueroa, S.N., Nobre, C.A., 1990. Precipitation distribution over central and westerntropical South America. Climanalise 6, 3640.

Gebremichael, M., Krajewski, W.F., 2004. Assessment of the statisticalcharacterization of small-scale rainfall variability from radar: analysis ofTRMM ground validation datasets. J. Appl. Meteor. 43, 11801199.

Gebremichael, M., Krajewski, W.F., Morrissey, M., Langerud, D., Huffman, G.J., Adler,R., 2003. Error uncertainty analysis of GPCP monthly rainfall products: a data-based simulation study. J. Appl. Meteor. 42, 18371848.

Getirana, A.C.V., Bonnet, M.-P., Martinez, J.-M., 2009a. Evaluating parameter effectsin a DEM burning process based on land cover data. Hydrol. Process. 23 (16),23162325. doi:10.1002/hyp.7303.

Getirana, A.C.V., Bonnet, M.-P., Rotunno Filho, O.C., Guyot, J.-L., Seyler, F., Mansur,W.J., 2010. Hydrological modelling and water balance of the Negro River basin:evaluation based on in situ and spatial altimetry data 24, 32193236.doi:10.1002/hyp.7747.

Getirana, A.C.V., Bonnet, M.-P., Rotunno Filho, O.C., Mansur, W.J., 2009b. Improvinghydrological information acquisition from DEM processing in floodplains.Hydrol. Process. 23 (3), 502514. doi:10.1002/hyp.7167.

Getirana, A.C.V., Bonnet, M.-P., Roux, E., Calmant, S., Rotunno Filho, O.C., Mansur,W.J., 2009c. Hydrological monitoring of large poorly gauged basins: a newapproach based on spatial altimetry and distributed rainfall-runoff model. J.Hydrol. 1, 45. doi:10.1016/j.jhydrol.2009.09.049.

Getirana, A.C.V., Frappart, F., Ramillien, G., Bonnet, M.P., Rotunno Filho, O.C.,Mansur, W.J., 2009d. Negro River basin water balance: comparison betweenrainfall-runoff model outputs and GRACE-derived products. Joint InternationalConvention of 8th IAHS Scientific Assembly and 37th IAH Congress Water: Avital resource under stress How Science can help, September 6-12, 2009,Hyderabad, India.

Gonalves, L.G.G., Shuttleworth, W.J., Nijssen, B., Burke, E.J., Marengo, J.A., Chou, S.C.,Houser, P., Toll, D.L., 2006. Evaluation of model-derived and remotely sensedprecipitation products for continental South America. J. Geophys. Res. 111,D16113.1D16113.13. doi:10.1029/2005JD006276.

Guyot, J.-L., Molinier, M., Guimares, V., Cudo, K.J., de Oliveira, E., 1993. BalanoHidrico da bacia do rio Negro. In: X Simpsio Brasileiro de Recursos Hdricos,

Gramado, RS.

Hiez, G. 1977. Lhomognit des donnes pluviomtriques. Cahier ORSTOM, srieHydrologie, 14, 129172.

Hong, Y., Adler, R., Hossain, F., Curtis, S., 2007. A first approach to global runoffsimulation using satellite rainfall estimation. Water Resour. Res. 43, W08502.doi:10.1029/2006WR005739.

Huffman, G.J., Adler, R.F., Morrissey, M., Bolvin, D.T., Curtis, S., Joyce, R., McGavock,B., Susskind, J., 2001. Global precipitation at one-degree daily resolution frommulti-satellite observations. J. Hydrometeorol. 2, 3650.

Huffman, G., Adler, R., Bolvin, D., Gu, G., Nelkin, E., Bowman, K., Hong, Y., Stocker, E.,Wolff, D., 2007. The TRMM multisatellite precipitation analysis (TCMA): quasi-

global, multiyear, combined-sensor precipitation estimates at fine scales. J.Hydrometeorol. 8, 3855.

Joyce, R.J., Janowiak, J.E., Arkin, P.A., Xie, P., 2004. CMORPH: A method that producesglobal precipitation estimates from passive microwave and infrared data athigh spatial and temporal resolution. J. Hydrometeorol. 5, 487503.

Juarez, R.I.N., Li, W., Fu, R., Fernandes, K., Cardoso, A.O., 2009. Comparison ofprecipitation datasets over the tropical South American and African continents.J. Hydrometeorol. 10, 289299. doi:10.1175/2008JHM1023.1.

Kalnay, E., Kanamitsu, M., Kister, R., Collins, W., Deaven, D., Gandin, L., Iredell, M.,Saha, S., White, G., Woollen, J., Zhu, Y., Leetmaa, A., Reynolds, B., Chelliah, M.,Ebisuzaki, W., Higgins, W., Janowiak, J., Mo, K.C., Ropelewski, C., Wang, J., Jenne,R., Joseph, J., 1996. The NCEP/NCAR 40-years reanalyses project. Bull. Am.Meteorol. Soc. 77, 437471.

Kanamitsu, M., Ebisuzaki, W., Woolen, J., Yang, S.-K., Hnilo, J.J., Fiorino, M., Potter,G.L., 2002. NCEP-DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc. 83,16311643.

Karam, H.N., Bras, R.L., 2008. Estimates of net atmospheric moisture fluxconvergence over the Amazon basin: a comparison of reanalysis products. J.Hydrometeorol. 9, 10351047. doi:10.1175/2008JHM887.1.

Kouwen, N., Soulis, E.D., Pietroniro, A., Donald, J., Harrington, R.A., 1993. Groupedresponse units for distributed hydrologic modelling. J. Water Resourc. Plan.Manage. 119 (3), 289305.

Krysanova, V., Mller-Wohlfeil, D-I., Becker, A., 1998. Development and test of aspatially distributed hydrological/water quality model for mesoscalewatersheds. Ecol. Model. 106, 261289.

Liang, X., Lettenmaier, D.P., Wood, E.F., Burges, S.J., 1994. A simple hydrologicallybased model of land surface water and energy fluxes for general circulationmodels. J. Geophys. Res. 99 (7), 1441514428.

Malhi, Y., Pegoraro, E., Nobre, A.D., Pereira, M.G.P., Grace, J., Culf, A.D., Clement, R.,2002. Energy andwater dynamics of a central Amazonian rainforest. J. Geophys.Res., 107, D20): 8061.

Marengo, J.A., 2005. Characteristics and spatio-temporal variability of the Amazonriver basin water budget. Climate Dynamics 24, 1122. doi:10.1007/s00382-004-0461-6.

Martinez, J.-M., Le Toan, T., 2007. Mapping of flood dynamics and spatialdistribution of vegetation in the Amazon floodplain using multitemporal SARdata. Remote Sens. Environ. 108 (3), 209223. doi:10.1016/j.rse.2006.11012.

New, M., Hume, M., Jones, P., 2000. Representing twentieth century spacetimeclimate variability: II. Development of 19011996 monthly grids of terrestrialsurface. J. Climatol. 13, 22172238.

Nijssen, B., Lettenmaier, D.P., 2004. Effect of precipitation sampling error onsimulated hydrological fluxes and states: Anticipating the Global PrecipitationMeasurement satellites. J. Geophys. Res. 109, D02103.

Nijssen, B., ODonnell, G., Lettenmaier, D.P., Lohmann, D., Wood, E.F., 2001.Predicting the discharge of global rivers. J. Clim. 14, 33073323.

Rao, V.B., Santo, C.E., Franchito, S.H., 2002. A diagnosis of rainfall over South Americaduring the 1997/98 El Nio event. Part I: validation of NCEPNCAR reanalysisrainfall data. J. Clim. 15, 502511.

Ribeiro Neto, A., Vieira da Silva, R., Collischonn, W., Tucci, C.E.M., 2005. HydrologicalModelling in Amazonia - Use of the MGB-IPH Model and Alternative Data Base.Proceedings of VII IAHS Scientific Assembly, Foz do Iguau, Brasil.

Roads, J.O., 2003. The NCEPNCAR, NCEPDOE, and TRMM tropical atmospherehydrologic cycles. J. Hydrometeor. 4, 826840.

Rudolf, B., Hauschild, H., Rueth, W., Schneider, U., 1994. Terrestrial precipitationanalysis: operational method and required density of point measurements. In:Desbois, M., Desalmand, F. (Eds.), Global Precipitation and Climate Change,

NATO ASI Series, Springer Verlag, 1(26), 173186.Rudolf, B., Schneider, U., 2005. Calculation of gridded precipitation data for the

global land-surface using in-situ gauge observations. Proc. 2nd Workshop of theInt. Precipitation Working Group IPWG, Monterey October 2004, EUMETSATEUM P44, 231-247.

Salati, E., Marquez, J., Molion, L.C., 1978. Origem e distribuio das chuvas naAmaznia. Interciencia 3, 200205.

Sapiano, M.R.P., Stephenson, D.B., Grubb, H.J., Arkin, P.A., 2006. Diagnosis ofvariability and trends in a global precipitation dataset using a physicallymotivated statistical model. J. Clim. 19 (17), 41544166. doi:10.1175/JCLI3849.1.

Shuttleworth, W.J., 1993. Evaporation. In: Maidment, D., (Ed.), Handbook ofHydrology. McGraw-Hill, New York, USA, pp. 4.14.53.

Shuttleworth, W.J., 1988. Evaporation from Amazonian forest. Proc. Roy. Soc. 233,321346.

Silva, V.B.S., Kousky, V.E., Shi, W., Higgins, R.W., 2007. An improved griddedhistorical daily precipitation analysis for Brazil. J. Hydrometeorol. 8, 847861.

Sorooshian, S., Hsu, K., Gao, X., Gupta, H.V., Imam, B., Braithwaite, D., 2000.

Evaluation of PERSIANN system satellite- based estimates of tropical rainfall.Bull. Am. Meteor. Soc. 81, 20352046.

http://dx.doi.org/10.1029/2002JD003127http://dx.doi.org/10.1016/j.jhydrol.2008.07.032http://dx.doi.org/10.1016/j.jhydrol.2008.07.032http://dx.doi.org/10.1002/joc.1791http://dx.doi.org/10.1002/joc.1791http://dx.doi.org/10.1029/2007JD009220http://dx.doi.org/10.1002/hyp.7303http://dx.doi.org/10.1002/hyp.7747http://dx.doi.org/10.1002/hyp.7167http://dx.doi.org/10.1016/j.jhydrol.2009.09.049http://dx.doi.org/10.1029/2005JD006276http://dx.doi.org/10.1029/2006WR005739http://dx.doi.org/10.1175/2008JHM1023.1http://dx.doi.org/10.1175/2008JHM887.1http://dx.doi.org/10.1007/s00382-004-0461-6http://dx.doi.org/10.1007/s00382-004-0461-6http://dx.doi.org/10.1016/j.rse.2006.11012http://dx.doi.org/10.1175/JCLI3849.1http://dx.doi.org/10.1175/JCLI3849.1http://dx.doi.org/10.1175/JCLI3849.1http://dx.doi.org/10.

Documents

Getirana Et Al, 2011