332A. T. LEMOS & R. D. GHISOLFIIntroductionThe mean sea level (MSL) is defined as theaverage of the daily oscillating processes of rise andfall of tides and all the disturbing processesassociated with the meteorological effects andseasonal cycles (Pugh 2004). Understanding thefactors that influence MSL are important owing tothe impact an eventual rise or fall of sea level mayhave on human activities, especially on continentalborders. To evaluate MSL, a multidisciplinaryapproach comprising several oceanic andatmospheric processes of different spatial andtemporal scales like termohaline processes, currents,long waves, meteorology, atmospheric pressure,wind curl, evaporation, precipitation, riverdischarge, crustal movements, tides, glaciology andeustatic changes, is required (Lisitzin 1974). Anypossible change in one or more of these processeshas the ability to deform the marine surface andthus, change the sea level (Dalazoana 2005).According to the 4º Report of Assessment ofthe Intergovernmental Panel on Climate Change(IPCC 2007), the main processes affecting theoceans in a scenario of global warming are seawaterthermal expansion and melting of ice caps.According to the report, the MSL derived from tidegauge records on a global scale during the lastcentury points toward a gradual rise in sea level, andthe projections for the current century indicate anaverage elevation of approximately 1.7 mm year -1 .Despite being one of the main causes of MSLelevation, thermal expansion of oceans is toocomplicated to be measured. Different featuresrespond in different ways to thermal expansion in aneventual warming. For example, when tropical seasurface water is heated, it will expand more easilythan the deep cold waters (Pugh 2004). According toHoughton (2004), if the first 100 m of the ocean,with an average original temperature of 25ºC, showa temperature increase of 1°C, then the local depthwill increase by 3 cm. However, the first 100 m ofthe ocean include the mixing layer, a stratumsusceptible to atmospheric changes, which makes itdifficult to estimate thermal expansion. Thus, tocalculate the increase in MSL, it is necessary to useoceanic-atmospheric coupled models. Some resultsfrom these models show a gradual increase in oceanvolume as a consequence of the observed increase inatmospheric temperature since the last century (Pugh2004). The sea level rose at a rate of approximately0.3-0.7 mm year -1 in the last century (IPCC 2007),and has increased from 0.6 to 1.1 mm year -1 duringthe last decade.Accurate measurement of sea level to theorder of millimeters is a challenging task, mainlydue to the technological dependence of instrumentsand techniques used over the years. There are twomethods for measuring sea level: direct and indirect.Direct measurement is performed in situ usingmetric ruler, tide gauges, reference levels, etc. Incontrast, indirect measurement involves theestimation of sea level from altimetry usingsatellites.In Brazil, the first measurements of sea levelusing tide gauges started toward the end of the 19 thand the beginning of the 20 th century, under theresponsibility of the Navigation and HydrographicBureau (DHN). These estimates were reserved foruse on harbor applications to obtain tidalcomponents, and/or for the elaboration of nauticalcharts. Majority of the in situ measurements lastedfor no longer than a lunar month, covering onespring and neap tide.The main objective of this work is to presenta brief historic review focused on the sea level alongthe Brazilian coast. In addition, it is intended todescribe the protocols as well as the state-of-the-artand future challenges regarding MSL estimatesalong the Brazilian coast. It is important to point outthat this work is focused on the long-term absolutesea level measurements, and not on the relative sealevel within the scale of decades, which is used inengineering surveys with local application.Materials and MethodsEquipment and protocols for measuring MSLSea level mensuration can be carried outdirectly and indirectly. The method is considered tobe direct when the equipment is installed in situ, likethe tide stakes and tide gauges. As these methods arerelatively cheap, easy to handle and do not requiresophisticated technology, they were initially used toassess the MSL. Indirect method of sea levelmeasurement makes use of altimetry satelliteestimates and represents a new technology (in usesince the 1990s) as well as a more accuratetechnique than the direct estimates. Nevertheless,both the methods have their advantages anddisadvantages, and the readers are referred to thereports by UNESCO (1985, 1994, 2002, 2006) forfurther information.In Brazil, direct estimates are generally usedto measure sea level. Nowadays, the two mostcommon and recommended tide gauges are the floattide gauge and the radar tide gauge. The float tidegauge is basically a weight floating inside a tubeimmersed partially in water. The tube prevents thefloat from moving under the action of winds andwaves for short periods. The float keeps itself linkedPan-American Journal of Aquatic Sciences (2010), 5(2): 331-340
Long-term mean sea level measurements along the Brazilian coast333to a cable and one pulley. When the cable isdisplaced on the pulley, an encoder transforms thismovement into a measurement of sea leveloscillation. Despite the equipment being simple toinstall and widely used, it is susceptible to manyerrors, such as sedimentation in the installation site,biological encrustations and crustal movements. Theradar tide gauge is a new technology recentlyapplied in Brazil. The main advantage of this systemis its installation (it is located outside the water) andthus, is not susceptible to temperature and densityoscillations. It measures the distance between theair–sea interface and the equipment through anacoustic signal. However, a major disadvantage ofthis system is the energy demand in case of use inlong-term research.Protocols for measuring MSL using tide gaugesSince tide gauges are being currently used inBrazil to measure MSL, we will further discuss theprotocols associated with this technique.The choice of system to measure sea leveldepends mostly on the purpose for which the data isto be used. Aspects such as costs, accuracy, locationand duration of measurements must be taken intoaccount to make the best decision possible. Forexample, harbor operations demand an accuracy ofabout 0.1 m (Pugh 2004). Hence, there is no need ofthe equipment to be sophisticated and a low cost tidegauge may be used accordingly.Scientific studies focusing on MSL, on theother hand, require a more accurate estimate ofabout 0.01 m. In this case, the installation must havean adequate number of reference levels (RRNNs) tomonitor possible changes in the ruler(s) position, tocarry out a topographic–geodesic monitoring of thereference levels and tide gauges, besides a digitaldata record (Fig. 1). These items will be discussedlater.Nevertheless, once the choice is made, it isessential to follow its basic recommendationsaccordingly to obtain a valid and usefulmeasurement.The use of metric ruler is the simplest wayto measure sea level. Despite being quite susceptibleto positioning mistakes, it is still used as acalibration reference to verify and/or correcteventual vertical displacements of already installedtide gauges.Tide gauges are relatively simple to install,widely known and do not require sophisticatedtechnology to operate. However, there are manyerrors and some disadvantages associated with them.For example, they are for local use (i.e., theinformation is restricted to the point being sampled),are subject to operational errors (e.g., biologicalincrustation), crustal movements, meteorologicalfactors, geographic positioning of the levels’references, etc.There are many types of tide gauges. Theycan be classified, for example, according to thephysics employed to obtain the information. Thereaders can refer to UNESCO (2002) for a completelist of tide gauges and their operating systems. Toensure accuracy of results, all of them commonlyfollow the same basic installation requirements. Anymeasurements of height should have a referencelevel relative to that plane. A reference plane is welldefined locally on a stable surface, free of anyinfluence of vertical and horizontal movements,Figure 1. Scheme of the system of measurement using a tide gauge. Adapted from UNESCO (2002).Pan-American Journal of Aquatic Sciences (2010), 5(2): 331-340