6. Exploring The Impacts Of Ocean - Climate Variability On Marine Resources Along The East African Coast: A Prototype Multi-Disciplinary Project
M. Jury, T. McClanahan, J. Maina and L. Rydberg - West Indian Ocean Coastal Analysis Project Team
SUMMARY
The western rim of the Indian Ocean receives a flow of water from the south equatorial current. Although rich in biodiversity, fish catches are low and population pressures are growing. A multi-disciplinary, multi-national project is proposed to analyse the coastal ocean circulation, its forcing and impacts. Project research will relate fluctuations in the large-scale marine environment to living resources along the coast. This will be accomplished by examining links between past physical oceanographic data, climatic conditions, marine resource availability and fish catches in the east African region, from Lamu to Mozambique islands 2 – 15 S. Outcomes are expected to provide platforms for improved resource management in the region.
INTRODUCTION
The exploitation of marine resources and pollution inputs threaten the fragile coastal environment and contribute to global change, in particular to loss of habitat. The importance of oceans to the world economy is reflected in fisheries, petrochemicals, shipping and tourism activities. The western Indian Ocean is well exploited for recreational activities and impacts on coastal waters are increasing. It is vital that marine and coastal ecosystem resources be used intelligently through informed management. This is especially applicable to fishing and breeding grounds in coral ecosystems, as well as the coastal habitats, wetlands and estuaries that contribute to productivity of the coastal ocean (fig 1).
Close to a third of the world's population lives in countries next to the Indian Ocean, yet this region produces 10% of global fish catch (Okemwa & Sted 1995). The western Indian Ocean washes the shores of numerous countries, the great majority of which are in a developing state with annual GDPs of ~ $ 3 billion. The coastal zone is of critical importance to eastern African people. It is the location of significant human habitation, the focus of subsistence agricultural and fisheries activities and the target of economic development. Increasingly, this combination of factors results in coastal habitats being degraded, natural resources being over-exploited, and growing conflict in resource use, especially around the coastal cities.
In recognition of regional obligations to support various Framework Conventions and global efforts to implement operational ocean monitoring (GOOS 1998), a multi-disciplinary, multi-national research project is proposed to investigate the impacts of ocean climate on east African marine resources. The project will make use of ocean general circulation model (oGCM) products, satellite and insitu data to study the water properties and circulations in the west Indian Ocean in relation to coastal fish catch. Climate change and its attendant thermal warming and sea level rise are likely to impact sensitive low-lying coastal areas and coral reef habitats. Thus, there is a need to investigate the capacity for ecosystems to adapt to a changing environment and to improve forecasts of ocean - climate events that could impact marine resources.
*Univ. Zululand, South Africa, + CRCP, Kenya, ^ Univ. Gothenburg, Sweden
Fig. 1 – Satellite image of 4-yr mean ocean chlorophyll (left) from Seawifs at 9 km resolution. Green = higher productivity. At right is the mean SST from IR satellite mapped at 50 km resolution.
MOTIVATION FOR COASTAL OCEAN ANALYSIS
The coastal zone occupies a relatively small percentage of the marine environment and contains complex habitats. Within it, anthropogenic and terrestrial fluxes impact upon biological production, carbon burial and mineralisation (LOICZ 1994). In the west Indian Ocean the shelf is narrow and there are significant riverine inputs buffered by wetlands prior to export to the open ocean. In east Africa there are new urban and tourist developments that create a demand for resources and create localised impacts in respect of toxins, salinity, sediments, etc. Global warming will contribute to increased storminess and shoreline recession. There is a need to investigate the physical and chemical oceanographic processes that place limits on biological productivity.
The impacts of coastal ecosystem functioning upon marine resources and socio-economic benefits are significant in relation to sources (contaminant dispersion) and receptors (vulnerability). The cost / benefit and feasibility of management actions need to be considered in relation to ecosystem habitats (LOICZEA, 1997). Change is inevitable and a multi-disciplinary coastal analysis project can develop the necessary scientific and technical skills for understanding environmental influences on marine resources. Here we provide an example of a prototype project that will initially focus on Kenya, Tanzania, the Comoros and northern Mozambique; countries who receive a flow of water from the south equatorial current of the Indian Ocean.
PROJECT AIMS
Marine resource utilisation cuts across a diverse range of people and significantly impacts socio-economic development in the region. However, it also creates challenges in the application of technology, design of sustainable scientific research, gathering of data and especially in the development of management strategies. This in turn demands a high level of scientific and technical knowledge and information that should be at the disposal of planners, managers, users, decision-makers and sponsors.
There is a lack of technical capacity and institutional coordination in the east African region to identify and assess marine resources within the shelf zone (Okemwa & van der Elst, 1996). This can result in poor planning and loss of potential benefit to the countries and their people. In turn this compromises regional development. Hence our project aims to:
The project will contribute to a larger IO-GOOS initiative to analyse the Indian Ocean and its resources to the mutual benefit of surrounding nations.
PROJECT BACKGROUND
Studies on the shelf oceanography of the Western Indian Ocean are underdeveloped. Transport estimates on the bifurcation of the south equatorial current rely on data from few expeditions (Schott 1983; Swallow et al. 1983, 1991). Recent advances in our understanding of the ocean have taken place through modelling and satellite data, yet the east African shelf waters have not received much attention. The main scientific question to be answered is:
- how does the regional circulation (ocean currents / atmospheric monsoon) impact upon the coastal environment and biological productivity?
Fish catch is low along the coast and confined to a few regions. The tuna fisheries in the mid-ocean thermocline ridge exceeds 1 million T pa. Similar catches may be possible in shelf waters with additional investment underpinned by scientific research. Significant expertise already exists in fisheries biology and coastal management, however studies on physical environmental influences on marine resources are limited. The proposed project addresses this deficiency and compliments regional work on coastal observations (WIOMAP) and ocean modelling.
Warm El Nino events in the western Indian Ocean generate high surface water temperatures (and sea levels) and low salinities (due to high rainfall), the impacts of which vary from place to place and generally increase offshore. Coastal currents vary in association with monsoonal winds that tend to increase from the north during El Nino events. Rossby and Kelvin waves entering and leaving the coast at different latitudes help drive long-shore 'pulses', whose impacts on marine resources are unknown. Using oGCM and satellite data available via the internet, the structure of various oceanic fields can be determined (figs 5 & 6), but there is a need for further analysis to validate the dynamical models, resolve coastal features and determine resource responses.
The project will foster an understanding of how coastal currents, SST's (fig 2) and the shifting monsoon circulation and rainfall (fig 3) relate to marine productivity and fish stocks. To build regional expertise researchers from participating countries will be drawn into the project. The work will focus on ocean - climate impacts on coastal (mainly fisheries) resources, and look at potential downstream improvements required in the network of coastal observations.
Fig 2 - Time series of SST near Zanzibar illustrating a warming trend
since 1950, based on ship data extracted from COADS.
Fig 3 - Time series of east African rainfall, an index of freshwater inputs
to the shelf. Extracted from CRU data via the IRI website.
Fig 4 - Comparison of sea levels for Seychelles and Zanzibar, from the GLOSS website, with seasonal mean removed. El Nino events create higher sea levels offshore, but the coastal signal is weak.
Fig 5 - Salinity (left) and zonal currents at 50 m depth based on data assimilated by the oGCM for December 2001. Westward (10S) and eastward (eq.) currents create a northward conveyor belt of low salinity, consistent with low productivity near the coast in figure 1.
SCIENTIFIC ISSUES
Monsoon winds along the east African coast exert a seasonal influence on the ecology, yet the uptake of remote influences needs to be distinguished if fluctuations in coastal resources are to be attributed to predictable processes. Seasonal variations are dependent on water masses and vertical motions. Vertical displacements of the thermocline are known to occur in response to large-scale waves impinging from mid-basin, themselves dependent on the Indian Ocean dipole and its association with the global El Niño. The coast causes significant distortions, for example cape effects that generate increased winds and currents, and river inflows that alter nutrients, turbidity and salinity.
COASTAL FISHERIES DATA
The catch of fish and invertebrates in east Africa has been compiled by FAO since the 1950s and stored in various databases. There has been very little analysis of the factors that influence the catches, both human effort and oceanographic processes. The data for total catch indicate a stable situation in Kenya and Mozambique with a few shortfall years, however significant fluctuations occur in Tanzania, with a general upward trend. These data are for total catch and there are more than 30 species of fish catch provided by FAO and national data bases that reveal more complex dynamics. The upward trend is probably due to increasing fishing effort, but estimates of catch per unit effort (CPUE) are problematical. The environmental processes influencing catches of various species in each of the countries has not been adequately studied. Some preliminary results are given below.
SCIENTIFIC DATA AND WORKPLAN
The work on marine resources will focus on the monsoon-driven ocean circulation along the coast between the islands of Lamu (2 S) and Mozambique (15 S), extending offshore to the Comoros (45 E). The investigation will seek to establish how the variability of the ocean climate impacts coastal fisheries. The mapping of areas of environmental influence on living resources will be carried out using NCEP climate fields and oGCM assimilated ocean data at global and basin scale (eg. 20 N - 30 S, 30 - 110 E).
A particular thrust will be studies of interactions between the large-scale ocean dynamics and the coastal environment, for example how small variations in sea surface height contribute to large variations in currents, both seasonally and from year-to-year. oGCM data are available (via the web) at monthly intervals for the period 1960 – present derived from an ocean data assimilation system operated at the Univ. Maryland, USA. All available ocean observations are inserted into the first-guess field of the ocean model and discrepancies are iteratively minimized. Ocean temperature observations are assimilated from ship measurements and, since 1981, with satellite data. XBT profiles are dense along key routes: from Madagascar to the NE and from Mombassa to the east. An important addition is the assimilation of satellite altimetry data since 1990 that, together with drifter data, underpins the accuracy of current estimates. The oGCM is forced by wind stress from NCEP reanalysis fields, incorporating satellite scatterometer data in recent years. The horizontal resolution of the model is around 1o (100 km) in the tropics, so although the vertical resolution is around 20 m, details of the coastal environment are not well captured. Hence satellite and insitu data will be employed to describe the coastal uptake of large-scale signals such as the Indian Ocean dipole.
A key technology input to the project will be satellite data on SST, ocean colour, etc. These are routinely available as composite images every week at < 10 km resolution (via the web). These will be evaluated for spatial structure and gradients in the coastal zone that affect marine resources. Fluctuations of sea surface temperature, river runoff and other ocean parameters and their relationship with historical fish catch will be analysed with a view to developing predictive models. A further thrust of the work will be studies on the impacts of climatic events on the coastal zone, by consideration of sea levels, marine winds, river flows, coastal erosion, etc. This component will make use of station data from the meteorological services, with whom collaboration is already at an advanced stage.
The project will add value to data from fisheries institutions in the region, through desktop studies conducted by the project scientists and visiting researchers from participating countries (eg. east Africa and Sweden). These studies will be performed at the Univ. Zululand and CRCP, Mombasa, using available data on physical, chemical and biological attributes. These will be analysed and related to the variability of fisheries and associated coastal resources. Once the fishery indices are formulated, we will investigate statistical relationships with environmental fields. In parallel we will consider the local oceanographic processes that contribute to the fluctuation of marine resources.
EXPECTED OUTCOMES
Through the work of the project scientists and visiting researchers, we will investigate links between coastal fisheries and nearshore environmental conditions. Desktop studies using available historical data will compare the ocean-climate with fisheries distributions and fluctuations in the period 1960-2000. Statistical associations will be studied and predictive models for fish catch will be developed. Annual progress reports will be produced. Academic journal articles will be co-authored between the project scientists and visiting researchers. Results will be made available to fisheries managers and coastal planners, to ensure wide application. The final report will contain recommendations for downstream enhancements to the coastal monitoring network to improve the local data-base.
PRELIMINARY RESULTS
The influence of coastal oceanography on East African coastal resources has been poorly explored despite the existence of disparate time-series data that has been collected and stored in national and international databases for more than two decades. There is some older proxy data from coral cores, such as oxygen isotope and barium data. Poor collection and reporting may still be a problem with some of the data until today, notably Kenya and Mozamibique. We compare differences in the sources of data, national statistics, FAO and the Wildlife Conservation Society program and find some discrepancies, effects of the sampling and compiler as well as similar trends of the data from different sources. The Kenyan data shows very little change, except an occasional drop in total catch and some change in fish categorization is associated with changes in the compiler and the way the data are organized. The Mozambique data shows a steady increase that is probably associated with increased effort. In order to determine fish abundance rather than fishing effort, we can detrend the catch with the human population numbers, which seems to be well associated with increasing effort and catch, and this is supported by studies on effort from Kenya. Another trend in the fish catch is a notable decrease in the mean trophic level for Tanzania, suggesting overfishing may be changing ecological structure of the fishes and associated catches. The detrended and normalized data can then be related to oceanographic parameters that include surface height, temperature, salinity, current speeds, and depth of the thermocline. The time series of satellite ocean colour is too short to pick up interannual trends but does reflect the monsoon seasons. Tanzania catch data has the greatest interannual variability (fig 7) with an 8-10 year cycle that is related to the number of tropical cyclone days near Madagascar (r = -.37 for n = 38) and composite maps indicate that SST and rainfall are low (figs 8 and 9), hence salinity is increased, in the year preceding a higher catch. A full analysis of the data, particularly the various trophic levels of fish is underway and may further illuminate the relationship between ocean-climate and marine resources in the West Indian Ocean.
Fig 7- Detrended Tanzania fish catch
Figures 8 and 9 Composite analysis of SST and precipitable water in the year preceding high catches.
REFERENCES