SHORT COMMUNICATION

An investigation into the particle volume scattering functionvariability in a cascading reservoir system

Enner Alcantara1• Thanan Rodrigues1

• Fernanda Watanabe1• Nariane Bernardo1

Received: 30 May 2016 / Accepted: 4 June 2016 / Published online: 8 June 2016

� Springer International Publishing Switzerland 2016

Abstract This work analyzed the spectral and spatial

distribution of the particle volume scattering function, bp,in a cascading reservoir system. During fieldworks water

quality parameters and scattering data were sampled in a

predetermined stations. The bp was estimated using the

ECO-BB9 equipment that measures the volume scattering

function, b at 117� [b(117�)]. The estimated bp(117�) werecompared with the remote sensing reflectance, Rrs, and the

chlorophyll-a (Chl-a) concentration, total suspended matter

(TSM) concentration and the transparency (measured using

a Secch disk). The results showed that in a hypertrophic

environment the bp is dominated by the phytoplankton

scattering and in an oligotrophic water system, the scat-

tering by a suspended matter dominates. The bp(117�)variability from a hypertrophic to an oligotrophic aquatic

system affects the remote sensing reflectance (Rrs) spectral

shape. Due to this, the parametrization of a unique bio-

optical model to estimate the optically active components

in the water will be challenging.

Keywords Hydrologic optics � Inherent optical properties �Tropical inland waters

Introduction

Since the pioneering works of Tyler and Richardson (1958)

and Petzold (1972) a relatively few measurements of the

volume scattering function, b(h), have been made (Chami

et al. 2006; Twardowski et al. 2012). The b(h) describes theangular dependence of scattered light from an incident

unpolarized beam (units in m-1 sr-1). It is defined as the

radiant intensity, dI(h), scattered from a volume element,

dV, into a unit solid angle centered in direction h, per unitirradiance, E (Twardowski et al. 2012). Usually b(h) is

portioned into the sum of two components, the pure water

component, bw(h), and the particulate component, bp(h).The magnitude of b(h) is strongly affected and dominated

by the particle component. This component is associated

with scattering contributions by many types of particles

suspended in water.

The b(h) and the absorption coefficient, a (m-1), both

inherent optical properties (IOPs), play a fundamental role in

hydrological optics by determining the light field in aquatic

media, as well as the water leaving radiant energy to the

atmosphere crossing the water surface (Kirk 1991). The

interaction of light with aquatic particles alters the spectral

and angular characteristics of the incident light field (Zhang

et al. 2013). The b(h) is used to estimate the backscattering

coefficient, bb (m-1), which represents the integral of b(h)

within the backscattering angular range from 90� to 180�(Babin et al. 2012; Alcantara et al. 2016a). The bb and a

coefficients are directly related to the remote sensing

reflectance, Rrs, as described by Gordon et al. (1988):

Rrs kð Þ / bb kð Þbb kð Þ þ a kð Þ ð1Þ

where a(k) represents the sum of the absorption coefficients

of phytoplankton, detritus, coloured dissolved organic

matter (CDOM), and pure water, while bb(k) is representedby the sum of backscattering of particulate material and

pure water.

The aim of this letter was to analyze the variations of bpin a cascading reservoir system. To do this two fieldworks

& Enner Alcantara

enner@fct.unesp.br

1 Department of Cartography, Sao Paulo State University-

Unesp, Presidente Prudente, Sao Paulo, Brazil

123

Model. Earth Syst. Environ. (2016) 2:89

DOI 10.1007/s40808-016-0149-z

were conducted in order to obtain water quality parameters,

Rrs spectra and b(h) in a cascading reservoir system. Cas-

cading system can causes limnological modifications from

the upstream to downstream reducing the turbidity and

increasing the transparency of water, and the biotic and

abiotic factors of water accumulate until the last dam,

which receives input from all the previous water bodies

(Barbosa et al. 1999).

Materials and methods

Study area

The Barra Bonita (BB) and Nova Avanhandava (Nav)

reservoirs (Fig. 1) are placed in the middle and lower

courses of the Tiete River, Sao Paulo State, Brazil,

respectively. The BB reservoir (22�3101000S, 48�320300W)

is a storage system and began its operation in 1963

flooding an area of 310 km2, with 480 m of dam length

and 90.3 days of average residence time (Soares and

Mozeto 2006), being formed from the damming of Tiete

and Piracicaba Rivers. Nav reservoir (21�70100S,50�120600W) is a run-of-river reservoir and was created

in 1982, flooding an area of 210 km2 (at its maximum

quota), with a dam length of 2038 m and mean resi-

dence time of the water around 46 days (Barbosa et al.

1999).

Fieldwork

Two field campaigns were carried out in the end of the dry

season in both BB and Nav reservoirs. In the BB reservoir,

the field survey was accomplished between 13 and 16

October 2014 (austral spring). In the Nav reservoir, the

field campaigns occurred between 23 and 26 September

2014. A total of 18 samples were collected in BB and 19

samples in Nav.

Water sampling processing

Water samples were collected at each sampling spot and

filtered through a glass fiber filter GF/F Whatman, 47 mm

diameter and 0.7 lm pore size, to estimate the Chl-a con-

centration (lg l-1) in laboratory (Golterman 1975). To

estimate the total suspended matter, TSM (mg l-1), water

samples were also filtered through a glass fiber filter GF/F

Whatman (47 mm diameter and 0.7 lm pore size) and

stored frozen in the dark (APHA 1998).

Remote sensing reflectance (Rrs)

In situ radiometric measurements were made using three

TriOS hyperspectral radiometers (TriOS, Oldenburg, Ger-

many): two ARC-VIS sensors with a 7� field-of-view in

order to measure radiance, and one ACC-VIS sensor with a

cosine collector to measure irradiance. Both ARC and ACC

Fig. 1 Study area: a location of

Sao Paulo State in Brazil,

b Tiete River and the reservoirs

location, c samples location in

Nav and d in BB reservoirs. The

numbers 1 and 2 represents the

location of Nav and BB

reservoirs, respectively

89 Page 2 of 5 Model. Earth Syst. Environ. (2016) 2:89

123

sensors have 3.3 nm of spectral sampling, work in a

wavelength ranging from 320 to 950 nm, use an integration

time of 4 ms to 8 s, and an operation temperature ranging

from -10 to ?50 �C. Radiances (total radiance—Lt; and

diffuse radiance—Lsky, both in W m-2 sr-1) and down-

welling irradiance data [Ed (0?), in W m-2] were measured

in an azimuth angle of 90� in order to minimize the

specular reflection (Mobley 1999). To avoid shadow from

the instrument the fieldwork followed the geometry sug-

gested by Mueller (2003). The hyperspectral measurements

allowed the computing of the remote sensing reflectance

(Rrs, units in sr-1) above water, by using Eq. (2).

Rrsðh;/; k; 0þÞ ¼Ltðh;/; k; 0þÞ � 0:028� Lskyðh;/; k; 0þÞ

Edðh;/; k; 0þÞð2Þ

where h is the azimuthal angle (in degrees), / is the

zenithal angle (in degrees), k is the wavelength (in nm),

and 0? indicates that measurements were made just above

the water surface.

Particle volume scattering function [b(h)]

The b(h) was measured using a WET Labs ECO-BB9

(WET Labs, Inc. 2013). The ECO-BB9 acquires mea-

surements at nine wavelengths: 412, 440, 488, 510, 532,

595, 650, 676 and 715 nm in a single angle (117�). It isassumed that the loss of photons due to scattering is neg-

ligible and only the loss by absorption should be accounted

for and corrected according to:

bcorrection ¼ ð117�Þ ¼ bmeasuredð117�Þ expð0:0391aÞ ð3Þ

where bcorrection is the corrected total volume scattering

function (m-1 sr-1), bmeasured is the raw calculated total

volume scattering function (m-1 sr-1), and a is the corre-

sponding absorption coefficient (m-1). The spectral, a,

measurements were acquired in situ using an ac-s meter

(WET Labs AC-S device), with 10 cm optical path which

works in a spectral range from 400 to 742 nm with reso-

lution precision of approximately 4 nm.

The volume scattering function of particulates

[bp(117�)] can be obtained by subtracting the volume

scattering of water (bwater) from the bcorrection as follows:

bpð117circÞ ¼ bcorrectionð117�Þ � bwaterð117�Þ ð4Þ

Data interpolation

The bp(117�) for BB and Nav reservoir were interpolated

using the Ordinary Kriging (OK) algorithm (Isaaks and

Srivastava 1989). The semivariograms were fitted testing

several theoretical models (spherical, exponential, Gaus-

sian, linear and power) and using the weighted least square

method. The theoretical model that gave minimum stan-

dard error was chosen for further analysis. In this case, the

fitted model was based on the Gaussian model. The

adjustment on the Gaussian model suggests the existence

of smooth spatial variance pattern at the study site (Bur-

rough and Mcdonnell 1998).

Results and discussion

Water quality parameters

The Chl-a concentration in BB is average 49 times higher

than Nav reservoir, with TSM concentration 14 times

higher in BB than in Nav reservoir. Consequently, the

transparency is six times higher in Nav than in BB reser-

voir (Table 1).

Spectral Rrs and the particle volume scattering

function

Since the Nav reservoir is very clear water, with low TSM

and Chl-a concentrations, the main spectral behavior

showed high values of reflectance at shorter wavelength

and lower reflectance for longer wavelength (Fig. 2a).

This spectral behavior is expected due to the more

penetration of electromagnetic radiation in the water col-

umn for shorter wavelengths, and since the concentrations

are very low, the radiation penetrates deeper (Rijkeboer

et al. 1998). The Rrs spectra for BB reservoir highlights an

absorption feature near 680 nm is observed and is due to

the presence of cyanobacteria, and a reflectance feature

near 710 nm that is associated with Chl-a concentration

(Fig. 2b).

For both Nav and BB reservoirs the bp is higher in

shorter wavelengths, decreasing toward longer wave-

lengths. In BB reservoir two samples stations presented

Table 1 Water quality parameters descriptive statistics (SD is the

standard deviation) for measurements collected from Nav and BB

reservoirs

Chl-a (lg l-1) TSM (mg l-1) Secchi disk (m)

Nav

Min 3.41 0.50 2.45

Max 20.48 10.00 4.65

Mean 8.73 1.45 3.35

SD ±4.17 ±2.03 ±0.56

BB

Min 263.20 10.80 0.37

Max 797.80 32.80 0.78

Mean 428.70 20.80 0.56

SD ±154.50 ±4.90 ±0.09

Model. Earth Syst. Environ. (2016) 2:89 Page 3 of 5 89

123

higher bp when compared with Nav bp data. This can be

associated with the higher Chl-a and TSM concentrations

in BB than in Nav reservoir. The interpolated bp data is

presented in the next section in order to identify the spots

of higher and lower bp in both reservoirs.

Interpolated particle volume scattering function

The interpolated bp(117�) for Nav and BB reservoirs are

shown in Fig. 3a and b, respectively. These results clearly

reveal that the bp(117�) measured at 0.5 m is higher in BB

than in Nav reservoir. In addition, we can see that the Tiete

River is the main responsible for increasing the bp(117�) in

both reservoirs. However, the extension of this influence is

more pronounced in BB reservoir.

According to Alcantara et al. (2016b) in BB reservoir

when the Chl-a concentration is higher than 200 mg m-3

the Rrs spectra can be affect by the packaging effect. The

package effect is a common physiological strategy for large

phytoplankton species, such as diatoms (Bricaud et al.

1995). The package effect reduces the absorption spectra

and can increase the error of estimate Chl-a concentration

from space (Marra et al. 2007). An example of the influ-

ence of the packaging effect on the bio-optical modeling

can be accessed in the work of Watanabe et al. (2015).

Some studies suggested that the small particles are

(a)

(c) (d)

0.000

0.005

0.010

0.015

0.020

0.025

400 435 470 505 540 575 610 645 680 715

R rs(sr- 1)

λ (nm)

0.000

0.005

0.010

0.015

0.020

0.025

400 435 470 505 540 575 610 645 680 715

R rs(sr- 1)

λ (nm)

0.0000.0020.0040.0060.0080.0100.0120.0140.0160.0180.020

400 435 470 505 540 575 610 645 680 715

β p(m

-1sr-1)

λ (nm)

0.0000.0020.0040.0060.0080.0100.0120.0140.0160.0180.020

400 435 470 505 540 575 610 645 680 715

β p(m

-1sr-1)

λ (nm)

(b)

Fig. 2 bp and Rrs for Nav (a,c) and BB (b, d) reservoirs,respectively

Fig. 3 Interpolated bp(117�)for Nav and BB reservoirs

89 Page 4 of 5 Model. Earth Syst. Environ. (2016) 2:89

123

responsible for most of non-water backscattering in the

ocean (Stramski and Kiefer 1991; Ulloa et al. 1994; Zhang

et al. 2013). However, others recent studies indicated that

in open ocean larger particles may be more important in

backscattering (Westberry et al. 2010). But up to now there

is no conclusive information about these findings for inland

water. Then, more analysis needs to be realized to better

understand the bp(117�) behavior in inland water and their

relationship with the Rrs spectra and also with the optically

active components in the water.

Conclusion

This work focused on the field measurements of the

bp(117�) in a cascading reservoir system in Tiete River.

The analyses reveals a higher bp(117�) values for BB

reservoir, where the Chl-a concentration is higher enough

to be considered as a hypertrophic water body and lower in

Nav reservoir that is a oligotrophic reservoir, with very low

turbidity. The observed variability in bp(117�) will modu-

late the backscattering coefficient, that will influence the

shape of Rrs spectra. Due to this, the parametrization of a

unique bio-optical model to estimate the optically active

components in the water will be challenging.

References

Alcantara E, Watanabe F, Rodrigues T, Bernardo N, Rotta L, Carmo

A, Curtarelli M, Imai N (2016a) Field measurements of the

backscattering coefficient in a cascading reservoir system: first

results from Nova Avanhandava and Barra Bonita reservoirs

(Sao Paulo, Brazil). Remote Sens Lett 7:417–426

Alcantara E, Watanabe F, Rodrigues T, Bernardo N (2016b) An

investigation into the phytoplankton package effect on the

chlorophyll-a specific absorption coefficient in Barra Bonita

reservoir, Brazil. Remote Sens Lett 7:761–770

APHA (1998) Standard methods for the examination of water and

wastewater, 20th edn. American Public Health Association

(APHA), American Water Works Association (AWWA), Water

Environmental Federation (WEF), Washington

Babin M, Stramski D, Reynolds RA, Wright VM, Leumarie E (2012)

Determination of the volume scattering function of aqueous

particle suspension with a laboratory multi-angle light scattering

instrument. Appl Opt 51:3853–3873

Barbosa FAR, Padisak J, Espındola ELG, Borics G, Rocha O (1999)

The cascading reservoir continuum concept (CRCC) and its

application to the River Tiete-basin, Sao Paulo State, Brazil. In:

Tundisi JG, Straskraba M (eds) Theoretical reservoir ecology

and its applications. International Institute of Ecology, Academy

of Sciences and Backhuys Publishers, Sao Carlos, pp 425–437

Bricaud A, Babin M, Morel A, Claustre H (1995) Variability in the

chlorophyll-specific absorption coefficients of natural phyto-

plankton: analysis and parameterization. J Geophy Res Oceans

100:13321–13332

Burrough PA, Mcdonnell RA (1998) Principles of geographical

information systems. Oxford University Press, New York, p 333

Chami M, McKee D, Leymarie E, Khomenko G (2006) Influence of

the angular shape of the volume-scattering function and multiple

scattering on remote sensing reflectance. Appl Opt

45:9210–9220

Golterman HL (1975) Developments in water science 2. Physiolog-

ical limnology: an approach to the physiology of lake ecosys-

tems. Elsevier, Amsterdam

Gordon HR, Brown OB, Evans RH, Brown JW, Smith RC, Baker KS,

Clark DK (1988) A semianalytical radiance model of ocean

color. J Geophys Res 93:10909–10924

Isaaks EH, Srivastava MR (1989) An introduction to applied

geostatistics. Oxford University Press, New York, p 561

Kirk JTO (1991) Volume scattering function, average cosines, and the

underwater light field. Limnol Oceanogr 36:455–467

Marra J, Trees CC, O’Reilly JE (2007) Phytoplankton pigment

absorption: a strong predictor of primary productivity in the

surface ocean. Deep Sea Res 54:155–163

Mobley CD (1999) Estimation of the remote-sensing reflectance from

above-surface measurements. Appl Opt 38(36):7442–7455

Mueller JL (2003) In-water radiometric profile measurements and

data analysis protocols. In: Mueller JL, Fargion GS, McClain CR

(eds) Ocean optics protocols for satellite ocean color sensor

validation, revision 4, vol III., Radiometric measurements and

data analysis protocolsNASA Goddard Space Flight Space

Center, Greenbelt, pp 7–20 NASA/TM-2003-21621/Rev-Vol.III

Petzold TK (1972) Volume scattering functions for selected ocean

waters. Technical Report SIO 72-78 (Scripps Institute of

Oceanography, San Diego, California)

Rijkeboer M, Dekker AG, Gons HJ (1998) Subsurface irradiance

reflectance spectra of inland waters differing in morphometry

and hydrology. Aquat Ecol 31:313–323

Soares A, Mozeto AA (2006) Water Quality in the Tiete River

Reservoirs (Billings, Barra Bonita, Bariri e Promissao, SP–

Brazil) and nutrient fluxes across the sediment-water interface

(Barra Bonita). Acta Limnol Bras 18:247–266

Stramski D, Kiefer DA (1991) Light scattering by microorganisms in

the open ocean. Prog Oceanogr 28:343–383

Twardowski M, Zhang X, Vagle S, Sullivan J, Freeman S, Czerski H,

You Y, Bi L, Kattawar G (2012) The optical volume scattering

function in a surf zone inverted to derive sediment and bubble

particle subpopulations. J Geophys Res 117:C00H17

Tyler JE, Richardson WH (1958) Nephelometer for the measurements

of volume scattering in situ. J Opt Soc Am 48:354–357

Ulloa O, Sathyendranath S, Platt T (1994) Effect of the particle-size

distribution on the backscattering ratio in seawater. Appl Opt

33:7070–7077

Watanabe F, Alcantara EH, Rodrigues T, Imai NN, Barbosa C, Rotta

L (2015) Estimation of chlorophyll-a concentration and the

trophic state of the Barra Bonita hydroelectric reservoir using

OLI/landsat-8 images. Int J Environ Res Public Health

12:10391–10417

Westberry TK, Dall’Olmo G, Boss E, Behrenfeld MJ, Moutin T

(2010) Coherence of particulate beam attenuation and backscat-

tering coefficients in diverse open ocean environments. Opt

Express 18:15419–15425

WET Labs, Inc. (2013) Scattering meter (ECO-BB9) user’s guide.

Revision L. http://www.wetlabs.com/manuals

Zhang X, Huot Y, Gray DJ, Weidemann A, Rhea WJ (2013)

Biogeochemical origins of particle obtained from the inversion

of the volume scattering function and spectral absorption in

coastal waters. Biogeosciences 10:6029–6043

Model. Earth Syst. Environ. (2016) 2:89 Page 5 of 5 89

123