195

Research Needs and Applications to Reduce Erosion and Sedimentation in Tropical Steeplands (Proceedings of the Fiji Symposium, June 1990): IAHS-AISH Publ. No.192,1990.

Monetizing erosion and sedimentation costs where steeplands meet the sea

JOHN A. DIXON Environment & Policy Institute, East-West Center, Honolulu, Hawaii 96848

ABSTRACT Although the financial costs of erosion control measures are usually known or easily calculated, greater uncertainty exists over the size of potential benefits. One way to assess these benefits is to analyze the costs avoided by preventing increased erosion. In a linked economic-ecological analysis these potential costs were estimated for a tropical ecosystem containing steep uplands and an adjacent coast. Three industries are dependent on the ecosystem—logging, a coastal fishery, and a diving-based tourism industry. Estimates were made of the physical links between logging-induced soil erosion and reduced productivity of the fishery and tourism industries. Two options, continued logging and a logging ban, were analyzed. Economic analysis showed that the logging ban generated greater benefits, both financial and social. In order to enforce this ban, however, government intervention is required. Normal market forces would have led to a second best alternative with continued logging and associated erosion-caused damages.

Prescriptions of a wide array of soil conservation and erosion control measures are used throughout the world. In some places the objective is to keep soil on the land to maintain on-site productivity; in other places the objective is to prevent soil from moving into streams and causing damage to stream channels, aquatic life, irrigation canals, reservoirs, turbines, domestic water supplies or other downstream infrastructure and services. In many locations soil conservation measures are justified by the harms associated with both on-site and downstream effects.

Although the financial costs of erosion control measures are usually known or easily calculated, greater uncertainty exists over the size of potential benefits. In many cases these benefits are estimated by the costs avoided as a result of controlling erosion and sedimentation. For example, by retaining soil on upland fields, productivity declines may be avoided and the use of increasing quantitites of fertilizers or other inputs may not be necessary. These avoided costs (production declines, fertilizer used) are a measure of some of the benefits from controlling erosion.

Downstream costs attributable to soil erosion and sedimentation include:impaired water quality; sedimentation of reservoirs (and reduced power output, flood control benefits or irrigation water supplied); damage to irrigation systems, harbors, ports, river beds, mangroves", and reduced fish productivity in affected areas. In many

196

cases the downstream costs from eroded soil are larger than upstream productivity changes.

The costs of soil erosion and sedimentation, both upstream and downstream, can often be identified, quantified and placed in monetary terms. The link, however, between upland soil conservation measures and soil erosion and deposition of sediment downstream is complicated by time, various sediment-delivery factors, instream sediment storage, the impact of major storm events, and the slow movement of sediment through major systems. Some of these technical details are discussed by Mahmood in a recent World Bank Technical Paper (1987).

The economic analysis of these effects is an example of applied economic analysis within a watershed context. A number of books and reports in recent years have addressed this issue (see Easter et al., 1986; Gregersen et al., 1987; Dixon, 1989; World Bank, 1989; Magrath & Arens, 1989; and Hodgson & Dixon, 1988). The approach is fairly straightforward; its application is more complicated.

Most of the considerable uncertainty in the analytical results is due to lack of sufficient knowledge on the physical cause-and-effect links between soil conservation measures and their impacts, both upstream and downstream. In addition, the larger the watershed, the greater the uncertainty and the longer the time gap between upstream action and downstream effects.

The following case study from the island of Palawan in the Philippines illustrates one application of a linked economic-ecological analysis of steepland soil erosion and, in this case, its downstream impacts in a tropical watershed-coastal ecosystem. The tropical coastal watershed is an ecosystem of special interest for two reasons; from an ecological standpoint, multiple interlinked ecosystems meet in the coastal watershed, and from an economic perspective, coastal watershed resources are demanded by a wide variety of potentially conflicting users.

THE PALAWAN STUDY

This study evaluates alternative development plans that impinge on a coastal zone in the Philippines where two industries, tourism and fisheries, are affected by a third, the logging industry in the hinterland. In particular, the off-site ecological effects of soil erosion and sedimentation due to logging are analyzed. Projected revenues from the 3 industries are compared over a ten year period using two possible development options: 1) a complete stop to logging in the watershed area, or 2) continued logging of the watershed with resulting sediment-induced damage to the fisheries and tourism industries. Details of the study are presented in Hodgson & Dixon (1988) and Hodgson (1989)•

The detrimental side-effects of logging on watersheds have been well documented throughout the world (BIOTROP 1978; O.S. Department of State 1980; Hamilton & King, 1983; Hamilton, 1987). The major cause of erosion due to logging operations has been shown to be roads and skid trails (O'Loughlin, 1985). The erosion resulting from such networks causes high levels of suspended sediment load in streams and rivers, which in turn has been documented to lead to

197

reduced fish and invertebrate biomass and diversity (Cordone, 1956; Graynoth, 1979; Newbold et al., 1980; Erman & Mahoney, 1983). In addition, when this silt is deposited in dams and reservoirs it results in additional economic costs (Hamilton & King, 1983; Goodland & Ledec, 1986; Dixon et al., 1989).

Although the connections linking watershed erosion, silt-laden rivers and siltation of the coastal marine environment appear obvious, this problem has received relatively little scientific attention despite the high economic value of coastal marine life. Terrestrial ecologists have failed to look beyond the freshwater systems affected by erosion. In the few cases where marine scientists have studied siltation damage to tropical coastal marine species, the origin of the silt has rarely been documented. The negative effects of siltation on fisheries stocks, for example, need not be direct. Increased silt loads could act indirectly on a tuna fishery by killing planktonic larvae of coral reef fish upon which tuna feed (Brock, 1985). In this case the resulting decrease in tuna catch would not be reported until long after the sedimentation event, when the affected year-class of tuna matured and fishermen noticed a reduced catch. Unfortunately, the intricate nature of marine fisheries food webs makes it difficult to conclusively demonstrate a specific cause and effect relationship (Jones, 1982).

The study site, Bacuit Bay, El Nido, is located near the northern tip of Palawan Island in the southwest Philippines. Palawan*s 1.2 million hectares are arranged in a thin, 425 km long strip of land bounded on the east side by the Sulu Sea and on the west by the South China Sea. Palawan's timber and fisheries resources are under_increasing exploitation pressure. Almost 92? of Palawan's 12,000 km land area was forested in 1968 compared to only 70? in 1980 (Pido, 1986), and perhaps 50$ in 1987. Logging has often been followed by conversion to agriculture. Fish are another important resource of the area. The seas surrounding Palawan have been identified as one of four major spawning grounds for yellowfin and skipjack tuna in the Philippines (Wade, 1950 in Aprieto, 1981). A high percentage of the total Philippine fish catch is estimated to be taken from the waters surrounding Palawan XPido 1986).

Bacuit Bay encompasses an area of 120 km , contains 5 islands, and has an additional 9 islands located on its seaward shelf. Flourishing coral reefs surround each island and form a continuous band along the bay coastline interrupted only by river passes. Three small rivers and several continuous and seasonal streams enter the bay. The town of El Nido has about 2000 residents and is located close to the northern bay entrance. Three small villages border the bay on the north, east and south coasts.

The terrestrial ecosystem of this study includes all land surrounding and draining into Bacuit Bay, the Bacuit Bay drainage basin (Fig. 1). This basin covers 78.3 km and extends inland to the central Palawan dividing range. Slopes are steep (average 30$) and soils are highly erodible. Prior to the initiation of logging in 1985, 53? of the basin was composed of primary forest. Most of this forest land is now included within a large logging concession which encompasses most of Northern Palawan. Forest is the main land cover with less than 10? of the total area under agriculture up till

198

the time of the study. The three main industries are logging, marine fisheries and tourism.

Figure 1. Bacuit Bay, Palawan and surrounding drainage basin. An area of possible illegal logging is located to the east of El Nido. The Manlag River is the major river draining the logged portion of the basin.

K i ^ l Logging concession A Diving resort

W/j'X Drainage basin «-"" Coral reef

k . J l l l e 9 a l (?) logging

Logging

The logging operation bordering Bacuit Bay is part of a 970 km concession to Pagadanan Timber Products Incorporated (PTPI). This represents most of the remaining primary forest in northern Palawan. Logging operations in the area bordering Bacuit Bay commenced in January 1985 and were temporarily suspended by the logging company in January 1986 for one year.

199

Theoretical sustainable timber harvest in the PTPI concession is based on a 42-year selective cutting cycle with three trees planted for each tree harvested. Unfortunately, a lack of tree planting, overcutting in low volume timber stands, and excessive felling damage have already been cited as threats to the sustainability of the PTPI concession (PIADP, 1985).

The Bacuit Bay drainage basin topography is steep (30? mean slope) and the land was classified in the "severe erosion hazard" category (Bureau of Soils, 1980). Road construction within the hilly drainage basin is generally done by cutting into the hillsides leaving an exposed vertical face on the uphill side. On the opposite, downslope side, the extra fine-textured soil is pushed over the edge of the road spilling down in a wide swath called a sidecast fill slope. Both the uphill and downhill slopes are thus destabilized by this type of road construction increasing the potential for accelerated erosion, especially during the heavy rains common to this area.

Estimates of erosion in the concession area reflect these facts (Table 1). Roads, although they account for only 3 percent of the total area, are responsible for over 80 percent of total erosion. Cut-over forest has a much lower erosion rate.

Table 1. Erosion in logging concession area by land use, 1986.

Area Type Area (km ) Erosion (tons/km ) Total ('000 mt) %

Roads 1.23 14,130 17.4 84 Cut Forest 4.80 260 1.2 6 Uncut Forest 37-00 60 2.1 10

Total 43-03 20.7 100

Fisheries

Fishing is a year-around, labor-intensive occupation. Bacuit Bay and its seaward shelf are utilized by commercial fishing operations based in distant provinces as well as by local artisanal fishermen using paddle, sail or motorized outrigger boats. A wide variety of fishing methods are practiced in this area to catch a diverse array of species.

Tourism

Tourism is relatively new to Palawan; prior to 1979 there was essentially none in the Bacuit Bay/El Nido region. In 1979, a Philippine-Japanese joint venture company (Ten Knots, Inc.) chose the islands at the entrance to Bacuit Bay to set up a scuba diving resort. The resort's strategy was to rely on group tours from Japanese diving clubs for the bulk of their business while filling

200

in with local and other foreign individual and group divers during the off season. Following the widely publicized success of Ten Knots, a second diving resort was established on an adjacent island (Pangolacion) by a different company in 1984.

The El Nido resort management realized that the value of their diving location lay in the undisturbed nature of the reef and waters. Any degradation of the natural beauty of the underwater or terrestrial environment would make their investment less profitable. They drew up a list of conservation regulations to be followed by their tourist divers and worked out a cooperative agreement with the appropriate law enforcement agencies whereby the resort would provide logistic support to law officers responsible for patrolling and strictly enforcing Philippine fisheries regulations. The result has been that Bacuit Bay is one of the few areas in the Philippines where the fisheries laws are enforced.

Ecological study

In order to assess the linkages between steepland logging, soil erosion, and the impact of sediment on downstream sectors a year-long study was carried out by Hodgson (1989). In this study erosion rates were measured in the uplands and sediment delivery at both river mouth and in the bay was observed. Within the bay measurements of sediment deposition, coral cover and diversity, and fish biomass over a one year period enabled Hodgson to estimate the physical links between eroded soil in the steep uplands and the health and productivity of the bay's ecosystem. This physical link formed the basis for the economic analysis.

ECONOMICS OF MANAGEMENT ALTERNATIVES

The Bacuit Bay watershed ecosystem is an example of a complex natural system that contains a variety of resources. Multiple uses exist for the products of the system; some uses can co-exist easily and others conflict. The fisheries industry and tourism, for example, both depend on a productive marine environment. The undisturbed bay ecosystem including algal, coral reef, plankton, and fish populations produces the products desired by both the tourism industry (clear water, "good" diving) and the fishing sector (fish catch). The logging industry, on the other hand, is concerned only with timber extraction and has little or no economic interest in the marine environment. A reduction in marine resources will have no direct effect on timber production; logging activities and associated soil erosion, however, can have a major negative impact on the marine resource-dependent sectors. At present, there is good cooperation between the tourism and fisheries industries in terms of law enforcement and the operation of marine preserves and fish feeding stations for scuba divers. Assuming that this level of cooperation will continue, the fisheries industry should not have a negative impact on the tourism industry in the future.

The economic analysis evaluates the trade-offs between two options: Option 1, a ban on logging in the Bacuit Bay watershed, and

201

Option 2, no change in the present plan to log the watershed within 5 years. The analysis examines changes in the value of the production of goods (logs and fish) and services (tourism) associated with the two options. As such it is an example of the "change in productivity approach" (Dixon et al.,1988).

Although soil erosion and sedimentation are the physical link between the three economic systems, the economic analysis examines the impact of soil erosion on the two downstream sectors — fisheries and tourism. Although there may be upstream productivity impacts due to soil erosion, these were likely to be very small given the land use (forestry) and the time horizon (10 years).

The gross revenues from each industry based on market prices of the commodities (fish and logs) and advertised rates for the tourist industry are compared. In this way the analysts can identify the orders of magnitude of economic gross benefits and costs associated with soil erosion and sedimentation.

Calculation of Present Value of Gross Revenues, 1987-1996

In order to compare the benefits generated by the three sectors under the two development options, the present value of gross revenues over a ten-year period are calculated. Two discount rates are used, 10? and 15?. These rates reflect the range of rates used in analysis of government and private projects. Details of the sectoral analyses can be found in Hodgson and Dixon (1988).

The tourism industry The present value of gross revenues for Option 1, the logging ban, are based on advertised room rates, observed occupancy rate, and planned expansion. All values are in constant U.S. dollars. There are two tourist resorts operating in El Nido plus assorted small scale lodging houses and tourism enterprises.

Gross revenues and the present values of gross revenues (using a 10/5 discount rate) for the various components of the tourism industry for the period 1987 through 1991 are shown in Table 2. To calculate revenue under Option 2, tourism revenue is reduced by 10? per year from 1988 through 1991 due to predicted degradation of seawater quality and marine life upon which the diving resorts depend. Beginning in 1992, yearly tourism revenue is reduced to the 1992 value of the "Other" category alone since the dive resorts are expected to go out of business by this time. This latter prediction is considered realistic because the diving resorts in this remote location are marketing a single, costly product: pristine coral reefs and clear water, for an international clientele. The timing of the business failures will prevent the planned expansion in the tourism sector thus drastically reducing gross revenue from 1991 through 1996.

The gross revenue of the tourism industry under Option 1 is 39 million dollars more than the gross revenue under Option 2. The present value at a 10? discount over the 10 year period from Option 1 is 19 million dollars more than the present value under Option 2. (Using a 15? discount rate the total present value under Option 1 is $19,410,000 versus Option 2, $5,591,000.)

202

Table 2. Gross Revenue and Present Value of Gross Revenue from Tourism: Option 1 (logging ban) and Option 2 (continued logging), (10? discount rate).

Gross Revenue Option 1 Option 2

Present Value of Gross Revenue Option 1 Option 2

Resort 1 1987-91 1992-96

$8,897 $29,656

6,510 0

6,745 13,961

5,038 0

Total $38,553 6,510 20,706 5,038

Resort 2 1987-91 1992-96

Total

$1,834 $5,931

$7,765

1,306 0

1,306

1 ,388 2,792

4,180

1 ,010 0

1 ,010

Other tourist businesses 1987-91 $276 1992-96 $821

Total $1,097

Grand total $47,415

The fishing industry

202 160

362

8,178

The fishing

208 387

595

25,481

industry is

157 75

232

6,280

analyzed in a similar manner based on the following assumptions. Under Option 1 (logging ban), predicted fish catch and revenue is based on the 1986 catch and gross revenue. Since the larger commercial boats operate only a portion of their total fishing time in this area, these figures have been adjusted to include only fish catch from Bacuit Bay and the associated seaward shelf.

It may not be appropriate to include the value of the purse-seine revenue in the total. The primary catch of the purse-seine operations is tuna, which have generally been considered pelagic species. The extent to which tuna are biologically dependent upon coastal ecosystems such as Bacuit Bay has not been conclusively demonstrated. However, available evidence indicates a close association between tuna and inshore (reef and bay) food species, at least during part of their life cycle (Brock,1985). If this fishery is included, the 1986 total catch is 4,947 tons and gross revenue is $4,607,000.

Predicted gross revenues and the present value of the gross revenues (at 10% and 15? discount rates) under Option 1 (logging ban) and Option 2 (continued logging) for 1987-1996 are listed in Table 3 • For Option 2, the reduction in fish catch is based on a

203

multiple regression analysis of coral cover and species diversity on fish biomass (see Hodgson and Dixon,1988 for details).

Table 3- Gross revenues and present values of gross revenues (10? and 15? discount rates) from the fisheries sector under Option 1 (logging ban) and Option 2 (continued logging), 1987-1996.

Gross Revenue (x$1,000)

Present Value (x$1000) 10? 15?

discount discount rate rate

Purse-seine included Option 1 Option 2

Purse-seine excluded Option 1 Option 2

46,070 21,471

28,070 12,844

28,308 15,125

17,248 9,108

23,122 13,083

14,088 7,895

The sizeable reduction in fish catch predicted to result from continued logging under Option 2 is reflected in the gross revenue figures both with and without the tuna catch from the purse-seiners.

Logging industry Under Option 1, the logging ban would immediately reduce gross revenue from logging the Bacuit Bay drainage basin to zero. Under Option 2, the annual gross revenue is assumed to remain constant through 1991; it then drops to zero, since the area will have been completely logged.

Gross revenues and present values of gross revenues using both 10? and 15? discount rates were calculated for the logging sector as for the tourism and fisheries sectors. The results of these calculations are shown in Table 4.

Table 4. Gross revenues and present values of gross revenues (discount rate 10? and 15?) for logging industry production from Bacuit Bay drainage basin, 1987-1996. Note that production ends in 1991 as the area will be completely logged by then.

Gross Revenue (x$1,000)

Present Value 10? 15?

discount discount rate rate

Option 1 Option 2

0 12,885

0 9,769

0 8,639

204

The PTPI logging concession is planned to operate on a sustainable yield basis, with a timber rotation of 85 years and a cutting cycle of 42 years. Assuming tree growth follows the expected pattern, no new production from the Bacuit Bay drainage basin would be expected until 42 years after logging was begun in 1985. Although the economic time horizon considered here is limited to 10 years, it should be noted that major increases in the future value of wood products would have to occur over the 42 year period to allow the expected revenue from the sale of these wood products to have an economically significant present value using a 10? discount rate.

SUMMARY OF ECONOMIC ANALYSIS 1987-1996

The economic analysis has estimated the gross revenues and the present values of the gross revenues for each of the three sectors under Options 1 and 2 for the period 1987-96. Present values were calculated using both 10? and 15? discount rates. A summary of the results of this analysis is presented in Table 5.

Table 5. Ten-year sum (1987-96) of gross revenues, and present values of gross revenues (x $1000) under Option 1 (logging ban) and Option 2 (continued logging). The last column lists the difference between Options 1 and 2. Values for the fisheries sector when tuna catch is included are shown in parentheses, but they are not used to calculate the totals.

Option 1 Option 2 Option 1

Gross Revenue Tourism Fisheries (with tuna) Logging

$47,415 28,070 (46,070)

0

$ 8,178 12,844 (21,471) 12,885

$39,237 15,226 (24,599) -12,885

$ 6,280 9,108

(15,125) 9,769

$19,201 8,140

(13,183) -9,769

Total $75,485 $33,907 $41,578

Present Value (10? discount rate) Tourism $25,481 Fisheries 17,248 (with tuna) (28,308) Logging 0

Total $42,729 $25,157 $17,572

Present Value (15? discount rate) Tourism $19,511 $ 5,591 $13,920 Fisheries 14,088 7,895 6,193 (with tuna) (23,122) (13*083) (10,039) Logging 0 8,639 -8,639

Total $33,599 $22,125 $11,474

205

The results are striking; total gross revenues under Option 1 , the logging ban, are more than double gross revenues under Option 2. Since Option 1 will prevent further logging in the Bacuit Bay drainage basin, the gross revenues from logging are zero. Fisheries and tourism, however, generate large and continuing benefits. The size of the tourism benefits are expected to grow over time as demand and market increase. Fisheries benefit will remain constant. In contrast, Option 2, that allows continued logging, generates smaller and decreasing benefits. After 5 years the logs would be gone as well as a significant part of the tourism and fishery sectors. The modest logging revenues generated under Option 2 are more than offset by decreased tourism and fisheries income.

The present value of gross revenues under Option 1, calculated using a 103» discount rate, is almost double that under Option 2. Even at the higher 15% discount rate the present value ,of Option 1 is still 1.5 times the Option 2 present value.

The results of the analysis are unambiguous and indicate that a ban on logging in the Bacuit Bay drainage basin is ecologically and economically the best development alternative.

IMPLICATIONS FOR RESOURCE MANAGEMENT DECISIONS

From the perspective of planning in general, the combined ecological and economic analysis provides several valuable insights. First, that coastal zone planning and management must include the hinterland watershed because of soil/water linkages. Second, this work shows that it is possible to obtain clear and sometimes surprising answers to development questions without a large research cost in terms of time and money. Third, the results of the study suggest that valuable information and guidance can be obtained from the combination of integrated ecological research and economic evaluation. The primary feature of the integrated approach is that it considers the entire area of interest, in this case the expanded coastal zone, to be composed of a series of linked ecosystems extending from mountain top to sea floor, and regards the dynamic links between these ecosystems to be equally as important as the processes occurring solely within each ecosystem.

The integrated ecological and economics approach is applicable to many types of development planning questions. In light of Repetto's (1987) conclusion that many developing nation governments appear to be subsidizing their logging industries, the continuation of logging in other multiple resource conflict areas of the Philippines and in other tropical developing countries might benefit from a similar evaluation and analysis.

Finally, the analysis has shown that economic valuation techniques are available which can be used to value a wide variety of resources which previously could not be accounted for in a quantitative manner. Biologists, soil scientists, hydrologists, and environmental managers need to become familiar with these techniques and to apply them to the analytical process so that their voices will be heard and their recommendations will become more meaningful to decision makers.

206

REFERENCES

Aprieto, V.L. (1981) Fishery management and extended maritime jurisdiction: the Philippine tuna fishery situation. Environment and Policy Institute Research Report No. 4., East-West Center, Honolulu, HI.

BI0TR0P (1978) Proc. Symposium on the Long-Term Effects of Logging in Southeast Asia. R.S. Suparto (éd.). SEAME0 Regional Center for Tropical Biology, Bogor, Indonesia.

Brock, R.E. (1985) Preliminary study of the feeding habits of pelagic fish around Hawaiian fish aggregation devices or can fish aggregation devices enhance local fisheries productivity? Bull. Mar. Sci. 37(0:40-49.

Bureau of Soils (1980) Physiography and Soils of the Palawan Island. Ministry of Agriculture, Manila, Philippines.

Cordone, A.J. (1956) Effects of logging on fish production. Calif. Dept. of Fish and Game, Inland Fish Admin. Rept. Sacremento.

Dixon, J.A. (1989) Multilevel resource analysis and management: The Case of Watersheds in G. Schramm and J. Warford, eds. Environmental Management and Economic Development The Johns Hopkins University Press for the World Bank, Washington, D.C.

Dixon, J.A., R.A. Carpenter, L.A. Fallon, P.B. Sherman, S. Manopimoke (1988). Economic Analysis of the Environmental Impacts of Development Pro.iects. Earthscan Publications, London.

Dixon, J.A., L. Talbot and G. LeMoigne (1989) Dams and the environment: Considerations in World Bank projects. Technical Paper No. 110, The World Bank, Washington, D.C.

Easter, K.W., J.A. Dixon and M.M. Hufschmidt (1986) Watershed Resources Management: An Integrated Framework with Studies from Asia and the Pacific. Westview Press, Boulder, CO.

Erman, D.C. and D. Mahoney (1983) Recovery after logging in streams with and without bufferstrips in Northern California. Calif. Water Resources Center Contrib. 186. Univ. Calif. Davis.

Goodland, R. and G. Ledec (1986) Environmental management in sustainable economic development. Impact Assessment Bulletin 5(2):50-81.

Graynoth, E. (1979) Effects of logging on stream environments and faunas in Nelson, New Zealand. J. Mar. Freshwater Research 13:79-109.

Gregersen, H.N., K.M. Brooks, J.A. Dixon, and L.S. Hamilton (1987) Guidelines for Economic Appraisal of Watershed Management Projects FAO Conservation Guide No. 16 FAO, Rome.

Hamilton, L.S. (1987) Minimizing the adverse impacts of harvesting in humid tropical forests. In A.E. Lugo, J.R. Clark, and R.D. Child, eds. Ecological Development in the Humid Tropics: Guidelines for Planners. Winrock, Morrilton, Arkansas.

Hamilton, L. and P.N. King (1983) Tropical Forested Watersheds. Westview Press, Boulder, CO.

Hodgson, G. (1989) The effects of sedimentation on Indo-Pacific reef corals. Ph.D. Dissertation, Zoology Dept., University of Hawaii, Honolulu.

207

Hodgson, G. and J.A. Dixon (1988) Logging versus fisheries and tourism in Palawan; an environmental and economic analysis. Occasional Paper #7. East-West Center Environment and Policy-Institute. Honolulu.

Jones, R. (1982) Ecosystems, food chains and fish yields. In D. Pauly, and G.I. Murphy (eds.) Theory and Management of Tropical Fisheries. ICLARM Conference Proceedings 9. International Center for Living Aquatic Resources Management, Manila, Philippines.

Magrath, W. and P. Arens (1989) The costs of soil erosion on Java: A natural resource accounting approach. Environment Department Working Paper No. 18. The World Bank, Washington, D.C.

Newbold, J.D., D.C. Erman & K.B. Roby (1980) Effects of logging on macroinvertebrates in streams with and without buffer strips. Can. J. Fish. Aquat. Sci. 37:1076-1085.

O'Loughlin, C. (1985) The influence of forest roads on erosion and stream sedimentation—comparisons between temperate and tropical forests. Working Paper. Environment and Policy Institute, East-West Center, Honolulu, HI.

PIADP (1985) Logging Methods Review. Position Paper. Integrated Environmental Program, National Council on Integrated Area Development, Philippines.

Pido, M.P. (1986) Palawan: On natural resources conservation. Canopy International 12(2): 1-5 -

Repetto, R. (1987) Creating incentives for sustainable forest development. Ambio 16(2-3):94-99.

Sinden, J.A. and A.C. Worrell (1979) Unpriced Values: Decisions Without Market Prices. Wiley, New York.

U.S. Department of State (1980) The World's Tropical Forests: A Policy, Strategy and Program for the United States. Pub. 9117-U.S. Government Printing Office, Washington, D.C.

World Bank (1989) Asia region: Review of watershed development strategies and technologies. Asia Region TD and Environment Department, PRD, working paper. Washington, D.C.