GEOTHERMAL FEATURES OF MOZAMBIQUE COUNTRY UPDATE

G. G. M.Q.W. JONES+ and A. RODRIGUEZ++

Key words: geothermal exploration. geothermal manifestations, heat flow,Mozambique

ABSTRACT

Geological aspects of Mozambique havebeen described since the beginning of thiscentury. Tectonic trends related to the RiftValleys have always attracted attention, nlyby researchers involved in studies related tomineral exploitation, but geothermalmanifestations have usually been considered asno more than natural curiosities. Recentgeochemical and geophysical evaluations confum the possible existence of significantamounts of geothermal energy. This isimportant in a developing country Mozambique, where the economics andreliability of equipment play important roles when comparing energy sources.

INTRODUCTION

Mozambique has a diversity of energyresources, but relies on imported petroleum andpetroleum products for 75% of its commercialconsumption. Hydroelectric potential exceeds

MW, coal is being mined in remarkablevolumes, gas has been found in several areas, and there is renewed interest in oil exploration.Energy is an important element inMozambique's development strategy because it is a source of foreign exchange and a catalystfor industrial progress. However, logistic problems, shortage of manpower, andfinancial limitations are obstacles to normaloperations and to policy and planning decisions.

The main objective for the medium and long term is to become self-sufficient in energy,production. To attain this objective,Mozambique is attempting to establish

conservation programs and to developdomestic energy resources for the national andinternational market (see alsoBank, 1987).

THE OBJECTIVES OF MOZAMBICANENERGY POLICY

Table I provides estimates for thecontribution of different energy forms inMozambique (Mozambican Government Paper,1981;U.N.,

TABLE I1979

Hydroelectricpower 44Fuelwood 30

12coal 10AgricultureWaste 4

While oil represents only 12% of energyneeds, it has a considerable portion of theNational Hard Currency Budget.

To emerge from the existingunderdeveloped situation, the current policy isto develop large amounts of cheap andrenewable energy in order to meet requirements of large industrial and agriculture schemes.

A permanent attention to technologicalevolution of new and renewable sources ofenergy (including geothermal, wind, waves andsolar energy) is in the objectives of Mozambicanenergy policy (see also S.A.D.C.C., 1982).

Geothermal energy can therefore be

et

considered as a potential energy source withinthe frame of current economic trends.

PREVIOUS STUDIES ON GEOTHERMALENERGY IN MOZAMBIQUE

East Africa is intersected by a system ofrift faults running north-south and characterizedby relatively high seismicity (Kebede andKulhanek, 1992)and volcanic manifestations.

These tectonic disturbancies strongly affect crustal permeability and allow thedevelopment of intense hydrothermal fluidcirculations.

Hydrothermal fields are well known inEast Africa and some Countries, have beeninvestigated for siting of geothermal power plants as in Ethiopia, Kenya, the Republic ofDjibouti, and Zambia (see also ELC, 1982).

A preliminary evaluation of localgeothermal potential in unexplored areas wascompleted by McNitt (1978; whoconsidered the presence of recent volcanism,hot springs, and other geologicalindicators. On the basis of the distributionof hotsprings the Mozambican geothermal potential was evaluated at 25 electrical Thisestimate is probably conservative since it wasbased on a preliminary investigation of Koenig(1981) which identified only 26 hot springs inMozambique, while our map shows that there are at least 38 emergences. (Fig. A)

Other preliminary considerations ongeothermal potential of Mozambique were madeby BRGM Aquater and

Nacional de Geologia (1981). On thebasis of volcanological and geological features,the BRGM (1980) proposal described the following areas as promising: the Zambesi andChire Valleys, the Cabo Delgado area, theCarinde-Zambesia area, the Manica area, andthe Zambesi-Limpopoarea. The Aquater (1980) proposal focussed attention on volcanologicalfeatures of Sofala, and ManicaProvinces and on the presence of hot springs. The Nacional de Geologia report deltwith proposals of BRGM, Aquater and theUnited Nations which are compared with local

knowledge mainly described in theMetalogenica da Republica Popular de Mozambique. On the basis of its evaluation the

Nacional de Geologia assigned to the following areas for funher study: the

Urema Valleys, the Zambesi Valley, theNiassa Lake area, and the areas of Ilha deMozambique, Pebane,Espungabera and Lugenda-Rovuma.

A complete check on available and pastdocumentation on hot springs, heat flow andgeological features has been carried out, and,this has resulted in a better knowledge ofgeothermal features of Mozambique.

GEOLOGICAL AND GEOTHERMAL FEATURES

Mozambique occupies ofwhich one-tenth lies within the East African riftsystem or within grabens and fracture zonesmarginal to the rift.

In terms of surface geology (Afonso,1976; Afonso, 1978; the

can divided into:

1 - Crystalline and metamorphic terrains, mostly of Precambrian age (but includingMesozoic and Late Paleozoic bodies)forming the northern and western half ofMozambique;

2 - Late Mesozoic and Cenozoic sedimentarycover, forming a wedge thickening to theeast and south that was in part depositedon the crystalline basement.

The rift system cuts through the crystalline terrain, and either terminates againstthe Cenozoic cover or is buried beneath it.Earthquake seismicity suggests southwardcontinuation of the rift zone beneath theCenozoic cover (Kebede and Kulhanek, 1992)Tertiary and Quaternary development of the riftsystem is indicated in northern Mozambique bysteep fault scarps, across which Quaternarysediment is juxtaposed with largelycrysstalline rock. This is especially true further north in neighbouring Malawi.

252

Late Mesozoic to Early Tertiary alkaline chemical analyses are also given. The locationinstrusive and extrusive bodies are principally of sample sites and the temperature of the waterlocated along marginal fracture systems. Basalt are shown in Fig. A and in Table and carbonatite probably of Late The squared diagram (Fig. B) ofTertiary age, are found along the western Langelier Ludwig (1942) suggests two distinctboundary of the rift system. Late TertiaryQuaternary volcanics also appear to be locatedalong faults marginal to the western riftboundary (along with the Late Mesozoic a) A group of chloride-sulphate alkalineigneous bodies) and along the waters, essentially located in the Zambesiaborder between crystalline basement and Late district and towards the south in the SofalaTertiary-Quaternary cover northeastern and Manica districts. The watersMozambique. A third possible locus of Late belong to this group. They have sodiumTertiary-Quaternary volcanism is a graben as the major cation and chloride ad theextending from the Tanzanian border in to major anion. The highest emergence north-eastern Mozambique (BRGM, 1980). temperature is 80 'C; silica content doesn't

exceed

water groups:

At least 38 thermal springs are nowrecognized in Mozambique (Fig. A). The mostinteresting geothermal area is within the rift just A group of acid carbonate-alkaline watersnorth of Metangula on Lake Niassa where located in the southern part of the countryvigorously boiling water was reported at the around Maputo. They are typicallake's edge in the years prior to the rise in water waters with moderate total dissolved ionlevel (de Freitas, 1959). Today, there concentrations and cold emergenceevidence of hot spring source, and thermal temperatures. Sodium is the mostdisturbances due to the vapor source affect the abundant cation. This group also includeswaters of the lake within a radius of several the Lake Niassa-Malawi thermalhundred meters. No geochemical data are manifestations, which are characterized byavailable for this source, so the chemical high emergence temperatures (up to 78composition of neighbouring Malawi springs and very low concentrations ofwere utilized for preliminary calcium and magnesium. The silicaevaluations. contents of these latter springs are higher

than those of the group.

b)

Several lower temperature springs below60' C) issue from Mesozoic crystalline terrainalong and west of major faults in the The relationship between alkalinity and

areas, near the border with the total content of dissolved solids Zimbabwe. Numerous other thermal springs shown in Fig. C, the existence of twohave been reported on the Zimbabwe side of the main water types, with the highter salinityborder, including a boiling spring and values being observed in the chloride-sulphateVerhagen, 1976). waters. Salinity values even higher than those

shown in Fig. B are indicated in Table amongHYDROGEOCHEMISTRY the incomplete analyses. Even in these waters

the relationship because chloride and sulphateions and salinity point to the same pattern.Published major element chemistry of a

set of water samples from Mozambique (deFreitas, 1959, Fernandes, 1975, Serrano Pinto,

together with the composition of fourthermal springs located along the western shoreof Lake Malawi (Muller, are reported inTable The composition of water samplesfrom Zambia (Legg, 1974)and some incomplete

253

et

Fig. A - Locality map showing sample sites and water temperature. Main mineralized areas arealso indicated.

MAP OF THE H O T SPRO F M O Z A M B I Q U E

NGS

LEGEND

HOT SPRINGS

NORMRL SPRINGS

HOT SPRINGS

VOLLEY FRULTS

OF DEEP

T I N . RORE

IRON.

COPPER. NICKEL

BAUXITE.MONTMORILLONITE. PERLITE

TOHNS

254

et

Table Chemical for the springs discussed in this study. Ion are in Silica anddissolved are expressed in References: Mozambique waters (samples 2242) from

(1959) and Senano Pinto data Malawi springs (samples MALI-MALA) from Muller and(1973);data for Zambia (samples Kapisya-Gwisho) from (1974).

Table List of Mozambique's waters discussed in this pages and closest place-names. Four samples bornMalawi (Mal and sixteen samples from Zambia Gwisho) are also listed.

Villa

23

Garcia5

Panic

N'

256

s 2

1213

16

a

SI 181920212223

2526

28

CABODELCADO

2932

303134373838

4 042

39

TETE

33

36

A plot of versusin considered by Giggenbach

(1988) as a good method ofbetween isochemical dissolution of igneousrocks and waters having

equilibriumwith crustal asa function of temperature. In such a diagram

the Mozambique thermal waters plotclose to the equilibrium line with apparentequilibrium temperatures in the range

Cold bicarbonate waters plot in theupper part of the diagram, far from the equilibrium line, indicating excess magnesiumand sodium. High magnesium concentrationsare generally considered indication of lowtemperature. As these waters are likely to haveformed by direct reaction between CO, -

charged waters and the aquifer rocks, theirmagnesium enrichment reflects highequilibrium CO, contents. the other handfree CO, is reported to widely occur in thesewaters. Low temperature reactions may havealso altered potassium concentrations by uptakeof K from secondary clays. The thermal springsof Malawi occupy positions in Fig. D thatsuggest close approach to full equilibrium atmoderately high temperature

The trilinear diagram of Fig. E(Giggenbach, 1988) indicates the degree towhich water approaches equilibriumThe equilibrium condition is establishedthe intersections of two set of isotherms describing the temperature dependence of

256

et

Table water samples from Mozambique, and

T'C

2223 10.1324 2.0225 2.2126 8.962728 1.4329 0.8130 0.1631 63.00 8.8932 46.00 78.70 33 33.0034 600035 600036 47.00373839 73.00 45.7040 75.00 69.70

80.00 18.0242 78.00 7.31

78.00 19.90

M A L 2 78.00 4.61

M A L 3 65.00 12.26MAL4 52.00 11.74

4.3351.00 34.7849.00 3.13

42.61

70.00 17.3967.00 21.7474.W 21.74

11.7465.00 3.2250.00 13.0460.00

4.3552.0094.W 13.9172.00 28.26

K

0.030.11 1.130.05 0450.01 0.48

3.19

0.09 0.080.04 0.390.090.01 600.22 35.40

084.851.5820.30

0.38 4.740.81 8.802.09 42.84 1.73 63.200.49 3.680.21 0.30

0.59 0.85

0.03 006

0.080.12

0.13 0.255.59

0.13 1.851.660.77 18.460.46 2.300.77 2.001.02 2.300.23 17.470.26 0.830.38 1.100.36 1.800.100.26 0.330.87 2.74

4.74

0.40 281.37 67.2040 4960.34 38 62.93 95.00.92 79 50.13 25.20.22 41.00550.08 87048 290036 7 2 0

120 66881.2

0.20 63.681 087.2

008 95 60.09 97.2

025

0.01

0.06 9080.0

0.08 30.024.0

0.08 250.82 30.00.08 40.00.41 40.00.08 40.00.080.08 20.00.08 40.00.33 20.0

40.00.08 40.00.08 40.0

40.00.16 40.0

77

90I34

729356

78

128

I30

I36

79707279929292926392639292929292

81

92

I2277936212682

I23

I26

122

83737783949494946994699494949494

4587

39

97

6224

47

87988498

97

48394048616161613161316161616161

27

40847423437792869

7865756375798485

89

8174

3021233042424242

42144142414242

a-Cbr

4

3526-22

-3631

21

29

27

27313636

57

32

-16-24-22-16

-30-3

-30

-35-2

-14-24

6.39-22

9-27

7

33

6

-39-33

15376213

219194329I747

1%

83

7967

I6464

184164

164

149

77

21

671074169

42

89

60

91

I48

47

113132

73

3263

146

93

14880143

143147

180

142

257

Figure B

800 -

600

400-

X

CL +

A

A

so

200 .

30

0 10 20

A "

AA A

A

Malawi springs

30 40

H C 0 3

Figure C and dissolved

A Cold waters IThermal

Malawi springs

Zambia springs

Total Dissolved Solids

258

Figure D -Plot of versus The shaded area representsdissolutionof basalt granite and averagecrustal rock The solidline indicatestheexpected composition of waters in thennodynamic equilibrium with an average crustal rock as afunction of temperature.

..-

AAA

0.8 - .A cold waters

thermal waters

0.6

,0.4 - .Zambia springs

0.2

0.0 70.0 0.2 0.6 0.8 1

K

FigureE - Trilinear plot of Na, K and Mg content drawn according to Giggenbach (1988). Black dots represent Zambian samples, represent Malawian samples,black squares representMozambican samples. white dot is the seawater.

0 20 80

259

et

Figure F Plot of silica vs. tcmperaturc. Silica expected forchemical equilibrium with to and chalcedony (TCH) are also drawn. Black

arc to Zambia samples.

200

100

0

Temperature

0 50 100 1 5 0 200

260

Martinelli et

quartz conductive and quartz adiabaticgeothermometers are no higher than (Tab.

(Lkn) and ratios according

Lkn = = 1.75with T the absolute

Lkm = = 14.0 -temperature in K.

to:

spring and Lake Malawithermal springs plot on the equilibrium l ie,indicating attainment of water-rockat 80°C. Among the samples falling in the fieldof "partially mature waters" we distinguish:a) two springs, Maganja da Costa andNamacurra, in the middle of the triangle, whichhave emergence temperatures of and

Niaondive spring which is close tothe lower boundary of the field suggestingattainment of partial equilibrium at highertemperature.

The remaining thermal springs and thegroup of the bicarbonate-alkaline waters fall inthe field of "immature waters", close to the Mgcomer. However, they lie above thedissolution line, suggesting removal of alkalinemetals.

As suggested by Giggenbach the Na-K-Mg geothermometer may be applied withconfidence for samples on or close to the

line. Therefore, the most reliabledeep temperature indicated by thegeothermometer seems to be around (Tab.

Dissolved silica in Mozambican watersranges from 30 to 90 ppm, with an average value of 70 ppm. Fig. F depiets silicaconcentrations in thermal waters versus temperature. Silica concentrations expected forchemical equilibrium with respect to quartz andchalcedony are also drawn.

The data points are closer to thechalcedony curve, suggesting that this mineralphase could be controlling dissolved silicaAccording to Arnorsson's data for Icelandic wells, geothermal waters beloware in equilibrium with chalcedony, whilesaturation with respect to quartz ingeothermal waters above 180°C. Even

estimated reservoir temperatures based on

Temperatures obtained from silicacontent are considerably lower than those fromthe alkali geothermometers. However, they aresimilar to estimated temperatures of the fullyequilibrated water samples in Fig. E. Similarconclusions can be drawn for water samplesfrom Zambia It is unlikely that mixingphenomena are responsible for a generalizedreduction of concentration inMozambique, Malawi and Zambia Low silicaconcentrations are more probably attributable toregional petrochemical of relatively silica-poor rocks.

Despite these problems, generalagreement obtained using independent chemicalgeothermorneters seems to indicate theexistence of aquifers at temperature of about100°C and these are worthy of further accuratestudy as sources of low-medium enthalpygeothermal energy. The fact that a number ofsamples reach the full equilibrium line suggeststhat there are consistent circulation path flows and related fluid availability.

PRELIMINARY ASSESSMENT OFBOTTOM HOLE TEMPERATURE DATAFROM PETROLEUM WELLS

To date there has been little opportunity for heat flow research in Mozambique, butbottom hole temperature measurementsin petroleum wells provide information whichhas been used with reasonable successelsewhere have Chapman et al., 1984;Deming and Chapman, 1988; Lam et al., 1982;Majorowicz and Jessop, 1981; Majorowicz etal., 1986; Speece et al., Here we analyzeBHT data from wells in Cretaceous-Tertiarysedimentary basins of coastal Mozambique toestimate geothermal gradients and heat flow.Problems associated with the use of BHT data for this purpose include the accuracy of theBHT data, estimation of equilibriumtemperatures from these measurements, andassigning thermal conductivities. Unfortunately,the available data (Koenig 1981, Salman at al.1985) contain insufficient information for a

261

et

Figure G. Petroleumwell localitiesin Mozambique (open circles,numbers refer to individualinTable IV. heat flowdata in southeast Africa (closed circles,heat flow in datafrom et al., 1987; Chapman and Pollack, 1977;Ebinger et 1987; Jones, 1988,

Nyblade et 1990; Pollack et al., 1990; Sebagenzi et .. 1993).M, Maputo; B,

2

3

I I I I

2332 . .

42. I'

,.47

30

3967

I

3 8

I I I I

262

et

thorough analysis, and the results must beregarded as preliminary.

BHT data from 35 wells theMozambique basin (Fig.G)were extracted fromreports by Empresa Nacional de

de Mqambique andElecmcidade de Mqambique (Koenig,1981; Salman et al., 1985). The reports containlittle information on the history of temperaturemeasurement, except that mostappear to have been made after a stabilizationperiod of 6-10 hours. It was therefore notpossible to estimate equilibrium temperaturesfrom temperature-time plots, and nor was thegradient correction of the American Associationof Petroleum Geologists attempted because ofthe lack of local control data (see Speece et al.,1985for discussion). Instead, a thermal gradientwas calculated from observed BHTs and acorrection was applied to this value (see below).

The entire data set from theMozambique is plotted in Figure H. Aleast squares fit to the data yields a slope of20.8 In an attempt to reduce the scatterwe have discarded points that appear to beinconsistent. These include: (1) a number of relatively shallow points which yield spuriousthermal compared with deeper BHTsand which may be affected by ground water movement in upper levels of the basin, (2) somethe deepest BHT data in a which also yield

results that may be associated to factthat some of the deeper wells penetrateStormberg lava below the basin's sediments,and (3) all results from two wells, the firsthaving erratic and high temperatures,which were suspected as (Koenig,1981) and the second having that wereoffset to higher temperature by about 25compared with the rest of the data set. Detailsof the screening process be discussedelsewhere. The "cleaned" data set is plotted inFigure K which shows considerably less scatter,Figure H although the least squares thermalgradient, 21.1 is not significantly different.

We also calculated temperaturegradients for eleven wells with two or moreBHT measured over long depth intervals. The

results are summarized in Table IV. Wells thatbe in extensions of the rift zone (marked

"a") have a slightly higher meangradient, compared with theremaining results, but there istoo much uncertainty regarding the data and thetectonic setting of the wells to makeconclusions. Cross reference with Figure Gdoes not reveal any other systematicpattern.

In an attempt to estomate afor drilling transients, we note that gradientsobtained from uncorrected BHTs are typically

lower than obtained fromcorrected BHT data (Lam et al., 1982).However, unless BHT data have goodtemperature-time such corrections stilltend to underestimate equilibrium temperature(Beck and Shen, 1989). A more realistic estimate may be obtained from Majorowicz andJessop (1981). who show that reliable BHTsfrom wells preserved for temperatureobservations in the westem Canadian basin yielda thermal that is approximately 35%higher than the result for BHTs measured within10 hours. As most of the data discussed herewere recorded after a stabilization period of10 hours, we regard 35% as a reasonablecorrection for Mozambique. The thermalgradient for the data in Figure K is therefore increased to 28.5

BHT data are also available from onewell drilled by ESSO Exploration Inc. in theRovuma basin in north Mozambique (site 12,Figure G).The data define a linear trend, andyield a least squares gradient of 39.1(Table IV). A 35% correction yields a thermal

of 52.8

The Cretaceous and Tertiary sedimentsin Mozambique are similar to those in basins inthe coastal plains of Tanzania and Kenya,where thermal conductivitieshave been reportedfor eleven wells 1975; Nyblade et

These conductivity values werecorrected for porosity and temperature effects.The reported mean values fall ,in the range from1.5-2.5 W and average at 1.95W

. To make a fust order estimate of the heatflow in Mozambique, we have applied aconductivity value of 2.0 W to the above

263

1

el

Figure H Plot of observed BHT data from the Mozambique basin. The slopeof the regression line is 20.8

160

2000 3000 50001000 ' .Depth,

Figure K -Plot of sclccted, but not corrected, from Mozambiquebasin (see text). of the regression line is 21.1

264

Table IV. Mean Thermal for Wells with RelativelyWell

No Name Gradient. N

4232. 18.8 42 3117 21.0 23 Temene 3205 15.54 17.4 25 Pande 17.9 76 3111 19.7 37 3838 22.2 128 4115 25.1 49 3230 18.3 2

3328 21.419.2 3

12 3292 39.1 Sa, wells possibly in of gradicn! i s wilh otherwells, but data are IO higher by 25°C c.Rovuma basin.

Table V.Heat Row in Tectonic Southeast Africa.

Unit Heat Flow, N

Tanzanian cratonZimbabwe

Kaapvaal craton(northern sector)

belt

(undifferentiated)Mozambique belt

belts

KenyaTanzaniaMozambique basin Rovuma basin

Zambezi riftrift

Malawi rift.Kivu rift'TanganyikaKenya rift'

-

8 Nyblade et al..32-65 10 Ballard et 1987;

et

25-76

49-57

44-76

35-68

54-9751-87

73-77

23-172

17-15156-101

16-187

15 Ballard et 1987. Jones.

3 Ballard et 1987;Nyblade et

I5 and

9 Nyblade etSebagenzi 1993

84

57

7775

22

S12

79 2

29-67

Nyblade etNyblnde etThis studyThis study

Nyblade et al..et

et 1987et 1973et 1971

Morgan.1973 .

al.,

for sediment blanketing effects; b,

265

et

resulting heat flow forthe Mozambique basin, 57 is clearlycompatible with results in the MozambiqueChannel, South Africa, and Zimbabwe (FigureG, Table V). Data from the Rovuma basin implya much higher heat flow, Thisresult is derived from only one well and must obviously be treated with caution, but it may besignificant that such high values are also foundin the coastal plains of Tanzania (up to 87

and Kenya (up to 97 (Nyblade etal., 1990).

HEAT FLOW AND GEOTHERMAL RESERVOIRS.

The range of heat flow and mean valuesfor different tectonic units surrounding Mozambique are listed in Table V. Heat flow inthe Mozambique Channel is typical of that inocean basins. On the continent, the heat flow inArchaean terrains is more uniform and lowerthan that in younger terrains, an observationthat is consistent (Jones, 1987; Nybladeet al., Nyblade and Pollack, 1993). Heat flow in the Tanzanian craton and theProterozoic Mozambique belt are lower thanusually observed in provinces of equivalent age, and the implications of this are discussed byNyblade et al. (1990). Mean values for theMesozoic coastal basins and rifts are slightlyenhanced compared with Proterozoic belts;these means are influenced by occasional highvalues which may have implications forgeothermal resources. The greatest variability ofheat flow and the highest values are found in themodem rifts, although the means are notsignificantly enhanced above those for Mesozoicregions. Some very high values in the Niassalake area (Ebinger et al., 1987) suggest that themodem rifts may be the most profitable localities for further investigations relating togeothermal energy resources. High local heat flow anomalies are commonlylinked with hydrothermal circulation asevidenced from geochemical Thegeological setting and geochemicalcharacteristics suggest similarities betweenMozambican, Malawian, ZambianZimbabwean hot fluid reservoirs (Muller andForstner, 1973: Legg, 1974; Mazor andVerhagen, 1976). Furthermore the partial or

total lack of cap rocks in Mozambicangeothermal reservoirs may strongly influencethe areal extension of heat flow anomalies andthe research strategy for future geothermaldevelopmet (see also Dickson and Fanelli,1990).

CONCLUSION

Geochemical and geophysicalconsiderations have allowed a better recognition of high heat flow areas in Mozambique, and abetter identification of deep fluid circulationcharacteristics.The most promising areas for geothermalenergy development are the andCentral Provinces of the Country. The localavailability of geothermally interesting fluids

the possibility of small-scale power generation, and warrants more detailed studiesand eventual exploratory drilling. Small scalegeothermal power plants are particularlysuitable for Mozambique because of theirrelatively low to natural hazardsand relatively low costs when compared withother energy sources. Current trends ofgeothermal energy generation in East AfricanRift Countries these conclusions.

ANNEX

Mozambique’s Update Report is summarizedin tables

Thanks are due to Santos Garcia,Nacional Geologia e

and J.M. EDM,de Mqambique for kind

data supply and suggests during theItaly

MozambiqueCooperationThanks are also to E. ,DAL. Milan for

kind remarks and to A.di Scienze della Terra,

di for mapdrawing.

!!

266

REFERENCES

R.S., (1976) para oconhecimento da deMqambique, Maputo, 37 pp.

R.S., (1978) - A geologia deMaputo, 191 pp.

AQUATER, (1980) - Study for high and lowenthalpy in Province and adjacent areas of tete, Sofala and Monica Provinces Preliminary Geothemal

S. in Campo, 16

S., (1975) - Application of the silicageothermometer in low temperaturehydrothermal areas in Iceland. Am. J.

S., H. N. N. J. SKINNER,(1987) heat flow in Botswanaand Namibia, Geophys. Res., 92, 6300.

BECK, A. E., P. Y. (1989) - a morerigorous approach to geothermic problems, Tectonophysics,

BRGM, (1980) - dexplorationgeothermique an Mozambique, technicalreport, Orleans Cedex, 12pp.

CHAPMAN, D. S., H. N. (1977) -Heat flow and heat production in Zambia:evidence for lithospheric thinning mcenual Africa, Tectonophysics, 41, 100.

D. S., T. H. M. S. BAUFR,M. D. (1984) - Heat flow in the Uinta determined bottom-holetemperature (BHT) data, Geophysics, 49,

DE A.J., (1959) - A geologia e odesenvolvimento econbmico e social de

Junta de Extemo,Provincia de 396 pp.

E. T., R. P. VON H. WONG,(1971) Lake waterchemistry, sediments, geologic structure,Naturwissenschaften,

E. T., R. P. H. K. WONG,W. G. DENSER, H. W.(1973) - Lake Kivu: structure, chemistry,and biology of an East African rift lake,Geol.

D., D. S. CHAPMAN, (1988) - Heatflow in the Utah-Wyomingthrust belt fromanalysis of bottom-hole temperature datameasured in oil and gas wells, J. Geophys.

272,763-784.

453-466.

Res.,R.H. and (Editors), (1990)

Small geothennical resources, a guide to

Centre on small energyresources,274 Rome - Italy.

depreliminar de energia geotennica RPM,technical report, 6pp.

C. J., B. D.(1987) Tectonic model of the Malawirift, Africa, sedimentary basins within the.Dead Sea and other rift zones,Tectonophysics, 141,215-235.

(1982) Geothermal studies of the EastAfrican Rift Technicalreport,8 pp.

EVANS, T. R., (1975) heat flowstudies in eastem Africa and the NorthSea, thesis, 486 pp., Univ. ofLondon, England.

A.M., (1975) - Hidrologicade Mqambique, Maputo 15pp.

W.F., (1988) - Geothermal solute Derivation of Na-K-Mg-Ca

geoindicators. Geochim. Acta,

F. KULHANEK O., (1992) -seimicity of the East African rift systemand its implications, Tectonophysics, 209,51-54.J.B., (1981) Evaluation of the potential

for the discovery and devolpment ofgeothermal resources in the People'sRepublic of Mozambique. Departement ofTechnical Cooperation for Development,United Nations, New York, technicalreport, 16pp.M. W., (1987) - Heat flow and heatproduction in the Namaqua mobile belt,South Africa, Geophys. Res.,6289.

JONES, M. Q. W., (1988) - Heat flow in the basin and environs and its

for the South African shield geotherm and lithosphere thickness, J.Geophys. Res., 93,3234-3260.

JONES, M. W., - Heat flow in SouthAfrica, Handbook Geol. Surv. of S.

Government Pretoria, 174

JONES, M. Q. W., - Heat flow anomaly mLesotho: implications for the southemboundary of the Kaapvaal

development and utilization.

DE (1981)

52,2749-2765.

261

et

Geophys. Res. Lett.,19,2031-2034.S., (1985) - problems de

geologia, deMqambique.Maputo, 19-83.

H. L., F. W. JONES, C. (1982)Geothermal gradients in the area

of Alberta, Can. J. Earth.

W.F., LUDWIG H.F., (1942) -Graphical methods f a indicating themineral character of natural waters. Am.Water Works Jour.

A.M., (1974) - A reconnaissance survey ofthe hot and mineralized springs ofZambia, Lusaka, technical report, 60 pp.

Regional heat flow patterns in the western

Tectonophysics,J. A., F. W. JONES, A. M.

(1986) - Geothermics of theWilliston basin in Canada in relation to

and hydrocarbon occurrences, Geophysics, 51,767-779.

springs of Rhodesia: their noble gases,isotopic and chemical composition Journal

Mc J.R., (1978) - The United Nationsto geothermal resource

assessment, 1-242.Mc J.R., (1982) - The Geothermal Potential

of East Africa,Geothermal 9

MORGAN, P., (1973) - heat flow studiesin and Kenya, thesis, Univ.of London, England.

MOZAMBICAN PAPER, (1981) -Energy Presentation, technical report, 16

G.AND U., (1973) - Recentiron ore formation in Lake Malawi, Africa.Mineral. Deposita

A. A., H. N. POLLACK, D. L.JONES, F.

(1990) - heat flow in east andsouthern Africa, J. Geophys. Res., 95,

NYBLADE, A. A. ANDH. N. POLLACK, (1993) - Aglobal analysis of heat flowPrecambrian terrains: implications for thethermal structure of Archean and

J. A., A. M. JESSOP, (1981) -

Canadian sedimentary

E., AND VERHAGEN B.T., (1976) - Hot

Proterozoic lithosphere, J. Geophys. Res.,

POLLACK, H. N., S. HURTER, J. R. JOHNSON,(1990) The new global heat flow datacompilation, Trans. AGU,(1982) - Energy

Policy Papers Mozambique,41.G., A., N.,

and A., (1985)- Heat flow in Mozambique basin.sedimentary basins of Mozambique andand gas prospectivity, EHN, Maputo,241.

M. N., G. P.(1993) - First heat flow densitydeterminations from southeastern (central Africa), J. Afr. 16,

SERRANO M., (1983) - para oconhecimento das da

da Popular deMqambique, Garcia de Geol.,

M. A., T. D. J. L. H.N. POLLACK, (1985) - Analysis oftemperatures in sedimentary basins: theMichigan basin, Geophysics, 50,1334.

BANK, (1987) - MozambiqueIssues and Options in the Energy Sector,Report No. 6128-MOZ, 215 pp.NATIONS, (1990-91) - African Statistical

yearbook, Vol. 3, 320 pp..

413-243.

119-126.

et

TABLE PRESENT AND PLANNED PRODUCTION OF ELECTRICITY

-iy Rod.

485in 1995

Undain

Rod

bur nor underin

1995

iy iy

280 435 2078 50

Iby

Nuclear+-I

169

et

TABLE 4. SUMMARY TABLE OF GEOTHERMAL DIRECT HEAT USES

250,000,000Therapeutic !use I

Total 500,000,000

270

TABLE 6. INFORMATION ABOUT GEOTHERMAL

v vapor

NRP

I2 93

5 5

5 3

7 35

7 26

531

5 3 2

5 3 3

536

537

S41

54

37 25

I6

3 5 5 9

36

38

3 7 31

38 97

33 35

35 25

11

’982

!43

1786

1147

i26

160

87

63

72

78

67

61-145

81-166

87-141

96-146

97-154

271

et

TABLE 9. ALLOCATION OF PROFESSIONAL PERSONNEL TO GEOTHERMALACTIVITIES to personnel with a

(1)Public

(3)

(4)Aid

(6)

212

et

1975 - 1984

TABLE 10. TOTAL INVESTMENTS IN GEOTHERMAL IN

0.02

0.02

100

100