Land Cover Changes in the Caroni Swamp, Trinidad from 1942-2007

Prepared by

Rahanna A. Juman Wetlands Ecologist

andDeanesh Ramsewak

Remote Sensing Officer

July 2011

© 2011 Copyright The Institute of Marine Affairs

All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording of any information storage or retrieval system without the permission in writing to the Director of the IMA. The Institute encourages fair use of this document and proper citation is mandatory.  Furthermore, the Institute of Marine Affairs gives no representations or a warranty either expressed or implied, as to the appropriateness of the data for a particular purpose, and as such, is not liable for inferences, recommendations or conclusions drawn by third parties. 

All queries must be addressed to:

DirectorInstitute of Marine AffairsHilltop LaneChaguaramasORdirector@ima.gov.tt

i

ABSTRACT

Caroni Swamp, the largest mangrove dominated wetland in Trinidad and Tobago, has

been impacted by human activities since the early 1900‟s. This study examines the

change in land cover and land use from 1942-2007. Remote sensing technology and

geographic information systems (GIS) together with extensive field surveys in 2004 and

2005 were used to demarcate the boundary of the wetland as well as the various land use

and land cover classes. Land cover maps for were produced for 1942, 1957, 1986, 1994,

2003 and 2007 from aerial photos and satellite imagery. Caroni Swamp occupied

approximately 21.7 km of the Gulf of Paria coastline, extending from the Beetham and

Churchill Roosevelt Highways in the north to the Chandernagore River in the South. In

2005, the wetland was estimated at 9,648.4 ha inclusive of the WASA Sewage Treatment

Plant, Beetham Landfill and the El Socorro Industrial Estate.

Major hydrological alteration in this wetland in the 1920‟s resulted in large freshwater

marsh and rice fields. From 1942 to 1957, marshland and agriculture cover increased,

but after this period there was a decline in both due to salt water intrusion and freshwater

diversion. Although mangrove forest was cleared for built development, it coverage has

consistently increased in the Swamp from 1957, with the exception of 2003 when there

was a slight decrease by less than 100 ha. Natural wetland communities (mangrove,

marsh and open water) increased from 1942-2003, but declined in 2007, as built

development more than doubled. The report provides spatial coverage, and quantified

land cover from 1942-2007. It also identifies reasons for the changes in land cover and

uses.

ii

ACKNOWLEDGEMENT

The authors wish to acknowledge the following individuals who have contributed to this

study: Suresh Sookbir, Neal O‟Connor, Lamani Patino, Jonathan Gomez, Al Small,

Addison Titus, Marc Bejai, Sarah Hosein, Jillian St. Bernard, Kahlil Hassanali, Adam

Jehu and Ramesh Gayah.

3

TABLE OF CONTENT

Page No.

1.0 INTRODUCTION ……………………………………………………... 1

2.0 GOAL AND OBJECTIVES …………………………………………... 7

3.0 METHODOLOGY …………………………………………………….. 8

4.0 RESULTS …………………………………………………………….......10

5.0 DISCUSSION ………………………………………………………….. 24

6.0 REFERENCES ………………………………………………………… 32

4

LIST OF FIGURES

Page No.

Figure 1: Location of Caroni Swamp, Trinidad ………………………… 2

Figure 2: Map of the major rivers in the Caroni River Basin ………….. 3

Figure 3: Land cover in the Caroni River Basin in 2003-2004 ………… 5

Figure 4: Land cover map of Caroni Swamp as determinedusing 2003 aerial photography ………………………………. 11

Figure 5: Land cover map in the Caroni Swamp in 1942 and 1957 ……. 13

Figure 6: Land cover map in the Caroni Swamp in 1957 and 1986 ……. 15

Figure 7: Land cover map in the Caroni Swamp in 1986 and 1994 ……. 17

Figure 8: Land cover map in the Caroni Swamp 1994-2003 …………… 18

Figure 9: Land Cover map in the Caroni Swamp 2003-2007 ………….. 21

Figure 10: Chart showing mangrove, marsh and agriculture coveragefrom 1042-2007 ………………………………………………. 23

Figure 11: Map showing channels within the Caroni SwampRamsar site …………………………………………………….. 28

Figure 12: Map of Caroni Ramsar Site showing privately owned lands ….. 31

5

LIST OF TABLES

Page No.

Table 1: Land cover and plant communities in the CaroniSwamp in 2003 …………………………………………………… 10

Table 2: Land cover and plant communities in the Caroni Swamp1942-2007 ………………………………………………………… 14

Table 3: Land cover of natural wetland types in the Caroni Swamp1942-2007 ………………………………………………………… 23

LIST OF PLATES

Plate 1: The straightened Caroni River and mangrove diebackobserved on the southern side ……………………………………. 19

Plate 2: Narrow cut that connects the Caroni river to the Blue River …….. 20

Plate 3: Red mangroves extending into marshland dominatedby sedges (Eleocharis sp.) in the Caroni Swamp …………………. 22

Plate 4: Infilling of marshland along the Uriah ButlerHighway for built development ………………………………….. 22

1

1.0 INTRODUCTION

Human transformation of land to yield goods and services represents the most substantial

changes to ecosystems worldwide (Millennium Ecosystems Assessment, 2005; Worm et

al., 2006). Material demands of production and human consumption alter land use and

land cover, biodiversity, and hydro-systems locally to regionally; and waste discharge

affects local to global biogeochemical cycles and climate (Grimm et al. 2008). While it

is more difficult to quantify alterations to marine ecosystems, changes are substantial as

about 60% of the world population lives within 100 km of the ocean (Vitousek et al.,

1997). Coastal wetlands that mediate interactions between land and sea have been altered

over large areas; approximately 50% of mangrove ecosystem globally have been

transformed or destroyed by human activities (World Resource Institute, 1996).

Trinidad and Tobago, like other Small Islands Developing States (SIDS), has a small land

mass, high population density and concentrated economic activities along the coast which

makes it more vulnerable to environmental disasters. SIDS river basins are generally of

comparatively small size because of the predominance of island areas, and thus, impacts

are characterized by a short response time from basin development activities to coastal

responses as compared to the larger basins of South America (Kjerfve, 2002). In almost

all Caribbean watersheds, population growth has overwhelmed any attempt to implement

rational coastal management. In Trinidad and Tobago for instance, the population has

almost doubled from 600,000 in 1950 to 1.2 million in 2000 (CSO, 2000) with

approximately 70% living on the west coast of Trinidad.

The Caroni River Basin, the hydrometric area that encompasses the Caroni Swamp is

situated in the northwestern section of Trinidad and covers about 883.4 km2, equivalent

to 22% of the land surface area of the island (Figure 1) (Juman et al., 2002). The Caroni

and associated Rivers discharge into the Caroni Swamp, the largest mangrove forest in

Trinidad and Tobago (Juman et al., 2010). The swamp is important economically for

oyster and fish harvesting, for hunting and for ecotourism. Major commercial fisheries

are based on demersal stock in the adjacent eastern Gulf of Paria.

2

Figure 1: Location of Caroni Swamp, Trinidad

The Caroni River, which is the largest river in Trinidad, drains the Northern and Central

Ranges to the west and has a catchment area of about 600 km2 (Juman et al., 2002). The

major part of the Caroni River water supply comes from perennial tributaries of the

Northern and Central Ranges. Twelve rivers flow into the Caroni Swamp on its northern

side from the Northern Range, and six rivers flow in from the southern side from the

Caroni Plain and the northern side of the Central Range (Juman et al., 2002) (Figure 2).

3

Figure 2: Map of the major rivers in the Caroni River Basin

The gently sloping foothills of the northern and central ranges and the non-flood prone

areas of the Caroni floodplain are used extensively for built development, both industrial

and residential. The Caroni River Basin represents the most populated part of the

4

country, housing 33% of the national population (CSO, 1995). The area has a population

density of 439 persons km-2 which is higher than the national average. The river basin

contains some of the most fertile land in the country (Brown et al., 1966). Increased

competition for lands in the flatter areas of the basin has led to built development on the

fertile lowlands, encroachment into the wetland area and additional development on

steeper slopes. Figure 3 is a land cover map for the Caroni Basin based on data collected

during 2003-2004.

Caroni Swamp is situated southwest of the capital, Port of Spain. It is the second largest

wetland in Trinidad after Nariva Swamp and consists of mangrove and herbaceous marsh

interrupted by numerous channels and lagoons (Bacon, 1993). This swamp was altered in

the 1920‟s under the Cipriani Reclamation Scheme which was initiated to facilitate rice

cultivation. The scheme involved hydrological alterations to the swamp; the North South

Drain was excavated, 10 east–west channels were cut to join the NS drain, two tide

exclusion sluices were constructed, 7 east-west canals on the south side of the Blue River

were cut to increase water circulation and 2 embankments were built (Bacon, 1970).

The Reclamation Scheme was officially abandoned in 1954 by which time the canals and

embankments were already in a state of disrepair (Bacon, 1970). Further flood protection

works which began in the 1950‟s and the construction of the Caroni Arena Dam has

resulted in an overall reduction of freshwater inflow into the Caroni Swamp (Phelps,

1997). This has reduced freshwater storage in the wetland and caused salinity levels to

increase. In addition to saltwater penetration, the Caroni Swamp receives water polluted

with sewage, wastewater from industry and agriculture run-off (Phelps, 1997).

5

Figure 3: Land cover in the Caroni River Basin in 2003-2004

In the 1970‟s, the eastern borders of the Caroni Swamp between the North-South

Embankment and the Uriah Butler Highway were predominantly freshwater herbaceous

wetland (Bacon, 1970) This area know by local residents as the „Reeds‟ was largely an

artifact of embankment work carried out under the Cipriani Drainage Scheme (Bacon,

1970). Bacon (1970) hypothesized that there was likely to have been a large lagoon,

6

bordered by freshwater marsh along the eastern border of the wetland. Distributary

patterns and water movements were altered by the drainage work, leading to

impoundment of land run-off to the east and influencing the location and extent of

colonization by freshwater species. As the tide exclusion banks deteriorated in the late

1960‟s, the wetland began to revert to its former structure (Bacon, 1970). Ramcharan et

al., (1982) produced a vegetation map for Caroni Swamp, but the legend is difficult to

decipher.

Between 1922 and 1985, more than 500 ha of mangrove forest were lost for the

construction of roads, WASA sewage ponds, landfill, and as a results of dredging to

widen the river (Gerald, 1985). In 2001, the IMA detected from satellite imagery large

areas of mangrove dieback in the Caroni Swamp. The total area of mangrove die-off

detected was estimated at 170 ha, and this is within the Caroni Swamp Forest Reserve,

which was proclaimed in 1936. More recently, private landowners have cleared wetland

and there is encroachment on the eastern boundary from unregulated housing (squatting).

The management of this wetland has become very challenging since most of the impacts

on the wetland emanate from outside its boundaries and there are issues of land tenure.

The Caroni Swamp has been impacted by human activities within the catchment areas

and within its borders. The plant communities have changed over time. This study

examines the changes in land cover and land use from 1942-2007 based upon aerial

photography and satellite imagery.

7

2.0 PROJECT GOAL AND OBJECTIVES

The goal of this study was to investigate land use and cover changes in Caroni Swamp

from 1942 to 2007. The specific objectives were:

1. To map the historical and “up-to-date” areal extent and physical features

of the Caroni Swamp using Remote Sensing and GIS tools.

2. To characterize the plant communities within the Caroni Swamp and map

their spatial extent and distribution.

3. To determine spatial changes in the plant community and identify the

human impacts on the ecology of the swamp.

8

3.0 METHODOLOGY

A literature search was conducted to obtain information on the Caroni River Basin and the

associated Caroni Swamp. Documentation was sought primarily from the libraries of the

following agencies:

Institute of Marine Affairs

University of the West Indies

Wildlife Section, Forestry Division, Ministry of Public Utilities and the

Environment

Environmental Management Authority

Land and Water Development Division, Ministry of Agriculture, Lands and

Marine Resources

The boundaries of Caroni Swamp were defined using physical and biological

characteristics: topography, hydrology, and vegetation type. Extensive ground-truthing

surveys were conducted between November 2004 and April 2005 using 2003 aerial

photography to delineate the boundary of the wetland. Additional surveys were

conducted in October 2008 and February 2009 using IKONOS satellite imagery (2007).

Areal cover was determined using maps and aerial and satellite imagery relevant to the

study area. Data obtained included the following:

1:25 000 Topographic Map Sheets (Nos. 23, 24, 33 & 34)

1:50 000 Aerial Photographs (1942)

1:12 500 Aerial Photographs (1957)

1:10 000 Aerial Photographs (1986)

1:12 500 Aerial Photographs (1994)

1: 25 000 Aerial Photographs (2003)

Landsat Imagery (Aug 2001)

2007 IKONOS Imagery

9

Aerial photographs were geo-referenced and registered in the Naparima Datum – UTM

Projection – to be compatible with the Topographic map sheets. For each epoch, aerial

photographs were mosaicked to produce a seamless data layer. The satellite images (as

well as other data layers) were also registered in the Naparima Datum to ensure spatial

correspondence.

The data layers were imported into a GIS for mapping, change detection and analysis.

Features mapped and assessed for changes over time include the following:

Extent and size of the wetland system (based on a set of criteria)

Mangrove and other vegetation types

Agricultural areas

Water channels

Land use (agriculture, urban, recreation, grazing, waste etc.)

The major plant communities (natural and agricultural) within the system were identified,

and classified following the classification for Caribbean wetlands provided by Bacon

(1993). Data collected was compared to past studies (Bacon, 1970; Bissessar, 1998). The

status of the mangrove swamp was investigated, and impacts identified.

10

4.0 RESULTS

Caroni Swamp Boundary

The boundary of Caroni Swamp as delineated in 2005 using 2003 aerial photography

(1: 25,000) is shown in Figure 4. The wetland boundary is contiguous, hydrologically

connected and below the 25 ft (7.5 m) contour. Caroni Swamp was estimated at

9,648.4 ha inclusive of the WASA Sewage Treatment Plant, Beetham Landfill and the El

Socorro Industrial Estate. It occupied approximately 21.7 km of the Gulf of Paria

coastline, extending from the Beetham and Churchill Roosevelt Highways in the north to

the Chandernagore River in the South (Figure 4).

The land-cover and plant communities within the Swamp in 2003 are provided in

Table 1. The mangrove forest was the dominant plant community occupying about 45 %

of the wetland area (Figure 4). It occupied the western side of the wetland where

conditions were brackish to marine. Marsh lands and rice fields were the other main

plant communities. Rice fields accounted for 1,279 ha of the agriculture lands

Table 1: Land cover and plant communities in the Caroni Swamp in 2003

Land Cover Class Area (ha)

Mangrove 4,343.1

Marsh 1,836.6

Pond and open water 536.3

Mangrove die-back 110.7

Backfilled area 38.0

Agriculture 2,289.2

Built Development 346.4

Total 9,648.4

11

Figure 4: Land cover map of Caroni Swamp as determined using 2003 aerial photography

12

Caroni Swamp has experienced land cover / land use changes dating back to the early

1900‟s. In 1942, a larger portion of the swamp was covered with marsh vegetation and

agriculture (primarily rice cultivation) and the mangrove forest extended north to the

Railway which is presently the Priority Bus Route (Figure 5; Table 2). The Beetham

Gardens community and Highway, the WASA Sewage Treatment Plant, the Beetham

Landfill did not exist. Construction of the Churchill–Roosevelt Highway began in

December 1941 and was completed in March 1942, however it did not extend far west

into the Caroni Swamp.

By 1957, there was a large increase in agriculture (> 600 ha), and slight increases in built

development and marsh vegetation (Table 2). There was a small decrease in mangrove

coverage as more (East-West) drains were cut and pond and open water areas increased.

While the agriculture land expanded on the eastern edge of the swamp taking up marsh

area, agriculture land on the northern edge reverted to marsh, and mangrove area on the

southeastern edge was colonized by marshes (Figure 5). By 1957, the Churchill–

Roosevelt Highway extended north to the Eastern Main Road in Barataria. Construction

of the Princess Margaret Highway began and was completed by 1958 (Figure 5). It was

extended and later renamed the Uriah Butler Highway in 1988.

There were major changes in the Caroni Swamp between 1957 and 1986. Within those

29 years, more than 500 ha of mangroves were cleared for the Construction of the

Beetham Landfill, WASA Sewage Treatment Plant, Beetham Highway and from

widening the Caroni River (Figure 6). However, mangrove coverage increased in the

swamp by 523 ha as mangrove trees expanded into marshland. Therefore, more than

1000 ha of mangroves spread into marshland while approximately 500 ha were cleared

on the northern edge.

Marshland coverage decreased considerably by more than 500 ha. Besides marsh being

over grown by mangroves, marsh on the northern end of the wetland was also cleared for

the El Soccoro Industrial Estate. At the same time, some agriculture land was left to lie

fallow and reverted to marsh. Agriculture decreased considerable by 413 ha; some lands

13

Figure 5: Land cover map in the Caroni Swamp in 1942 and 1957

14

Table 2: Land cover and plant communities in the Caroni Swamp 1942-2007

Land Cover Class AREA (hectares)

Mangrove 3575.9 3560.5 4083.5 4444.3 4343.1 4681.1

Marsh 1535.8 1693.0 1155.1 1162.4 1836.6 1012.6

Pond /open water 156.8 337.0 783.0 612.1 536.3 676.3

Agriculture 3086.0 3724.7 3311.3 2603.6 2289.2 2692.5

Built Development 6.9 55.8 215.2 363.4 346.4 842.1

Highway 8.0 18.5 44.3 89.5 75.4 82.4

Railway 17.7 NA NA NA NA NA

Mangrove die-back 18.7 NA NA NA 110.7 12.1

Back fill NA NA NA NA 38.0 NA

Solid Waste Land fill NA NA 47.5 68.5 72.6 73.7

NA – Not applicable

1942 1957 1986 1994 2003 2007

15

Figure 6: Land cover map in the Caroni Swamp in 1957 and 1986

reverted to marshland and others were converted into built development. Built

development increased by 159 ha.

Between 1986 and 1994, mangrove forest continued to expand and increased in coverage

by 361 ha (Figure 7). Mangrove forest continued to spread eastward into marshland, as

well as in the northwestern side around the sewage ponds and landfill. Ponds and open

water areas were overgrown by mangroves and they decreased by 170 ha. Dredged spoilt

that was dropped at the mouth of the Caroni River, was also colonized by mangroves.

Although marshlands were lost to mangrove, marsh coverage in the swamp increased

slightly by 6 ha, as some agriculture lands reverted to marsh, and the WASA sewage

ponds were covered with marsh vegetation. Agriculture decreased considerably in the

swamp during this period by more than 700 ha. Built development increased by 148.2 ha

as new developments were built along the fringes of the Uriah Butler Highway on land

that was previously used for agriculture.

By 2003, the agricultural land on the northeastern side of the wetland, and between

Caroni and Blue Rivers, west of the Uriah Butler Highway had reverted to marshland

(Figure 8). Between 1994 and 2003, marshland increased by 674.2 ha while agricultural

land decreased by 314 ha (Figure 8). Open water areas within the mangrove swamp

decreased by about 75 ha and these areas were occupied by marsh vegetation (Figure 8).

Marsh vegetation also occupied areas that were once built up as built development

decreased by 14 .1 ha.

During this same period (1994-2003), the Caroni River was straightened and widened,

and mangroves along the banks were cleared. An island was created and an area of dead

mangrove approximately 110.7 ha was observed south of the Caroni River (Plate 1).

Overall, mangrove coverage decreased by 101.2 ha, even though mangrove dieback

accounted for 110.7 ha. This indicates that there were about 9.5 ha of new growth.

Mangrove continued to extend into marshes on the eastern side of the wetland, west of

the Uriah Butler Highway.

16

17

Figure 7: Land cover map in the Caroni Swamp in 1986 and 1994

18

Figure 8: Land cover map in the Caroni Swamp 1994-2003

19

Plate 1: The straightened Caroni River and mangrove dieback observed on the southern side (Photo by R. Karim)

Between 2003 and 2007, mangrove coverage increased by 338 ha, built development

more than doubled (increased from 346.4 to 842.1 ha), and agricultural land increased by

403.3 ha (Figure 9). The area of mangrove dieback observed in 2003 had re-grown,

however some new dieback was observed further east on the southern side of the Caroni

River. Although not shown in Figure 9, a narrow cut was made connecting the Caroni

River to the Blue River (Plate 2); this is close to the area where the die-off is seen (Figure

9).

Plate 2: Narrow cut that connects the Caroni River to the Blue River

During the same period marshland decreased by 824 ha and open water area decreased by

140 ha. Mangrove forest extended further east overgrowing marshlands (Plate 3), and

filled in some of the open water/ pond. The marsh areas observed on the western side of

the swamp among the mangroves in 2003 have either been overgrown with mangroves or

have reverted to open water areas. Marshland has also been converted into built

development and agriculture lands. Built development continues to encroach upon

marshland along the Uriah Butler Highway (Plate 4).

20

Figure 9: Land Cover map in the Caroni Swamp 2003-2007

21

22

Plate 3: Red mangroves extending into marshland dominated by sedges (Eleocharis sp) in the Caroni Swamp

Plate 4: Infilling of marshland along the Uriah Butler Highway for built development

23

Mangrove forest is the major vegetation community in the Caroni Swamp. This ecotype

has increased considerable from 1957 to 2007 (Figure 10). In the 65 years span of this

study (1942-2007), mangrove coverage in the swamp has increased by 1105 ha while

marshland has decreased by 523 ha, agriculture has decreased by 393.5 ha, and built

development within the swamp has increased by 835 ha. The solid waste landfill has

increased in size since its construction from 47.5 ha in 1986 to 73.7 ha in 2007 (Table 2).

Figure 10: Chart showing mangrove, marsh and agriculture coverage from 1942-2007

Generally, the natural wetland communities (mangrove, marshes and open water/ pond)

increased between 1942 and 2003, as agriculture lands reverted to marsh and mangrove

colonized mudflats and dredge spoilt on the seaward side. However between 2003 and

2007, natural wetland coverage declined by about 346 ha as built developed and

agriculture lands increased (Table 3).

Table 3: Land cover of natural wetland types in the Caroni Swamp 1942-2007

Wetland class

1942 1957

Year

1986 1994 2003 2007

Mangrove 3575.9 3560.5 4083.5 4444.3 4343.1 4681.1

Marsh 1535.8 1693.0 1155.1 1162.4 1836.6 1012.6

Open-water/ pond 156.8 337.0 783 612.1 536.3 676.3

Total 5,268.5 5,590.5 6,021.6 6,218.8 6,716.0 6,370.0

24

5.0 DISCUSSION

Wetlands were once considered wastelands which would better serve nations when filled

for built development or drained for agriculture (Field, 1995). This perception is

changing as research conducted during the past four decades have highlighted the values

of wetlands and the services they provide. The Millennium Ecosystem Assessment

(2005) classified wetland services as provisioning (food, water and fuel), regulating

(erosion, flood control, climate regulation), cultural and supporting (tourism and

recreation), and has acknowledged their tremendous economic benefits (Blumenfield et

al., 2009). Yet, wetland coverage worldwide continues to decline to accommodate

growing populations.

Knowledge of the wetland‟s current and past extent, condition and uses is essential for

managers, policy and decision makers. This can help them to make effective

management interventions, so as to maintain and if necessary, restore ecosystem services.

This study examined the land-use / land cover changes that occurred in the Caroni

Swamp from 1942-2007 using remote sensing and GIS technology. The 2003 aerial

photographs were extensively ground-truthed and the different land use/land cover types

classified. These classifications were used as a guide to visual interpretation in order to

map the land cover in earlier imagery.

The first documented alteration to Caroni Swamp was under the Cipriani Reclamation

Scheme during 1921-1922; although aerial photography prior to, and during this period

were not available to confirm changes. The original drainage pattern in the Swamp was

modified to reclaim the eastern section of the Swamp for agriculture, specifically rice

cultivation, and to improve drainage from the swamp to the coast (Bacon, 1970; Phelps,

1997). This involved the:

i. excavation of the North-South Drain, construction of an embankment and

installation of two sluice gates to isolate the reclaimed area from tidal movement;

25

ii. division of the reclaimed areas into polders with embankments founded on

mangrove mat and divided into 9 East-West drainage channels that discharges

into the N-S Drain;

iii. construction of an East-West channel west of the N-S embankment to intercept

the natural watercourses running from northeast to southwest in the Swamp and

link them to the Blue River;

iv. construction of 7 east-west canals on the south side of the Blue River to increase

water circulation;

v. construction of a Cipriani Canal linking the Caroni River to the Blue River near

the coast;

vi. construction of flood relief channel south of the Caroni River, and parallel to it;

vii. installation of a pumping station at the northern end of the N-S Drain for drainage

of reclaimed area.

After this hydrological work, the eastern section of the swamp was isolated by the

embankment and salinity was gradually lowered so that it was possible to begin

agriculture activities.

During 1940-1955, the government had developed two areas for rice cultivation; the

Caroni Irrigation Scheme; east and west of the Uriah Butler Highway and the Caroni

Savannah Scheme which occupied the same area reclaimed under the Cipriani Scheme.

Agriculture had increased by 639 hectares from 1942-1957 (Table 2). In the interest of

agriculture, the Uriah Butler Highway (formerly Princess Margaret Highway) was

constructed to connect the Churchill Roosevelt Highway to the Southern Main Road at

Chaguanas. The highway partially blocked the free flow of flood waters of the Caroni

River arriving from the east; thereafter drainage in the swamp could take place only

through culverts and under the bridges provided integrally with the new highway (Phelps,

1997).

26

In 1942, a larger portion of the Caroni Swamp was covered with marsh as well as

agriculture (Figure 5). Bacon (1970) provided a detailed description of this freshwater

marsh community referred to as the „Reeds‟ by locals since it comprised extensive tracts

of sedges and grasses. Beard (1946) and Bacon (1970) postulated that the construction of

the tidal exclusion embankments along the western side of the N-S drain may have led to

the development of the marshland, therefore this section of the swamp may have

developed artificially. In the 1942 image, the mangrove forest was found mainly west of

the N-S embankment, except in the south along the banks of the Madame Espagnole

River. The mangrove also extended north to the Railway (Figure 5).

The Cipriani Reclamation scheme was officially abandoned in 1954, by which time the

N-S embankments and the enclosing polder embankments were already in a state of

disrepair, allowing saline water to enter the reclaimed area once again (Bacon, 1970).

The new swamp drainage channel constructed under the Scheme provided a conduit for

saline water to enter (Phelps, 1997). In the 1957 image, mangroves had started to grow

into the marsh area (Figure 5) and rice cultivation had expanded.

In the 1950‟s, government also implemented the first flood protection scheme for the

Caroni River Basin as built development within the Basin increased (Phelps, 1997). The

Caroni River was widened and deepened, and flood embankments constructed. The flow

capacity of the Caroni River downstream of the Uriah Butler Highway increased

substantially by widening and deepening the River and dredging the mouth. The breach

in the embankment at the N-S drain was repaired in the 1966, and this eliminated

freshwater flow into the Swamp from the Caroni River.

The freshwater supply to the southern section of the Swamp by the Guayamare and

Cunupia Rivers was cut off when the joint flow of these rivers were directed solely to the

Madame Espagnole by sealing the eastern end of the lower section of the Guayamare

(Phelps, 1997) (Figure 11). The quantity of freshwater flowing in the Swamp decreased

consistently through this drainage work as well as the irrigation scheme for agriculture.

27

Freshwater was diverted from the Caroni River to the Guayamare River to provide an

irrigation supply to the Caroni rice projects.

Freshwater diversion and saltwater encroachment have changed the „Reed‟ plant

community (Bacon et al., 1997; Bissessar, 1998). This has now been replaced by

mangroves, open saline water bodies, and by saline marshes, dominated by sedges,

Eleocharis spp. and the fern Acrostichum aureum. The spread of mangroves into the

eastern marshland has been rapid over the past four decades (Figures 6-9). These

marshes may eventually be replaced by mangrove forest.

Prior to the 1960‟s, the major changes to Caroni Swamp were hydrological in nature;

channels were cut, embankments were built, water was diverted away from the wetland,

and the Caroni River was dredged and widened. Conduits were constructed to drain the

wetland, and these same conduits carried saline water upstream when the embankment

fell into disrepair in the 1950‟s (Figure 11). Highways (Churchill Roosevelt and Uriah

Butler) were constructed and this also impeded the flow of the rivers into the wetland.

There were minor changes in the natural wetland communities (Table 3), while

agricultural land increased and built development started.

Beyond, the 1950‟s there were major land cover/ land use changes in the Caroni Swamp.

The mangrove forest on the northern side of the Caroni River was reclaimed for the

construction of the Beetham landfill, WASA sewage treatment plant and ponds, and the

Beetham Gardens. Built development expanded along the fringes of the wetland and

while agriculture land reverted to marshes, marsh lands were overgrown by mangroves

(Figure 9). In 2007, there was a small increase in agriculture land (≈ 400 ha), which had

consistently declined form 1957-2003.

28

Figure 11: Map showing channels within the Caroni Swamp Ramsar site

29

The Caroni River Basin has also experienced extensive land-use/land cover changes

within the past few decades and this has serious consequences for the Caroni Swamp, the

receiving environment for the land-based runoff. In addition to saltwater penetration,

Caroni Swamp receives sediment laden water polluted with sewage, wastewater from

industry and agriculture run-off (Donawa, 1976; Deonarine 1980, Siung-Chang, 1987;

Phelps, 1997; IMA 1999; IMA (unpublished). This has affected the quality of the habitat,

and the shellfish harvested in the swamp. For instance there was a country-wide ban on

mangrove oyster (Crassostrea rhizophorae) in 1992 because of the threat of cholera.

Fish kills and mangrove dieback seem to be a regular occurrence in Caroni Swamp

(Bacon, 1970; Siung-Chang, 1987; IMA unpublished). Major mangrove die-off events

were reported by Bacon (1970) and observed in 2001 Landsat Imagery. In 2001, 170 ha

of mangroves were recorded and although the cause of the die-off was undetermined, it

coincided with the widening, deepening and straightening of the Caroni River, the raising

of the southern bank (IMA, unpublished) and hypersaline conditions. Interstitial salinity

in the dead area was recorded at 82‰.

Frequent hydrological work in the Caroni Swamp to mitigate flooding in the associated

catchment have negatively impacted on the health of the mangroves, since channels are

widened and dredged, mangrove trees are removed, and the dredge spoilt placed on the

bank inhibits the natural flushing of the system. Mangrove productivity is a function of

water turnover in the forest (Pool et al., 1975); reduced tidal flushing results in higher

soil salinity (Cintrón et al., 1978; Santos et al., 1997), decrease in nutrients and/ or the

accumulation of toxic substances such as hydrogen sulphide (Nickerson and Thibodeau,

1985). Mangroves have an optimum salinity range for maximum growth; at extreme

level mangrove species suffer damage and even mortality (FAO, 1994).

Management of Caroni Swamp

In 1936, approximately 2,833 ha of Caroni Swamp were proclaimed a Forest Reserve. In

1953, a further 136 ha were designated a Wildlife Sanctuary for breeding Scarlet Ibises,

and this was later extended to 200 ha (Bacon & Ffrench, 1972). Fishing, oyster

30

collecting and hunting had been permitted in all areas except the Wildlife Sanctuary, but

in 1987 the major part of Caroni Swamp, south of the Blue River, was declared a

Prohibited Area (Forests Act, Chap 66:01) and these activities became prohibited. Only

licensed tour guides were allowed to take tours into the Prohibited Area.

In 2005, the Caroni Swamp was declared a Ramsar Site, a wetland of international

importance, as it is an essential habitat for numerous commercial species, as well as rare

and endangered species (Figure 11). Within the Ramsar site boundary, there are privately

owned land (Figure 12) and development of these lands especially along the Uriah

Highway would further encroach on the wetland communities. The Forestry Division is

the government agency with responsibility for managing the Caroni Swamp and while

they provide game wardens to patrol the swamp to discourage poaching, there is no

comprehensive management plan for the conservation and restoration of this wetland.

31

Figure 12: Map of Caroni Ramsar Site showing privately owned lands

32

6.0 REFERENCES

Bacon, P. R., 1970. The Ecology of Caroni Swamp, Trinidad. Special Publication, Central Statistical Office, Trinidad and Tobago, 68 pp.

Bacon, P. R. 1993 Mangroves in the Lesser Antilles, Jamaica and Trinidad and Tobago. In L. D. Lacerda (Project Co-ordinator)- Conservation and Sustainable Utilization of Mangrove Forests in Latin America and Africa Regions. International Tropical Timber Organization/ International Society of Mangrove Ecosystems, Technical Reports Vol. 2: Part 1- Latin America 272pp.

Bacon, P. R., Bissessar, V. and Cuffy, G., 1997. Study of the Feasibility of Marsh Restoration in the Caroni Swamp, Trinidad- Wetland Ecology. Report to the Inter American Development Bank and the Wildlife Section, Forestry Division, Government of Trinidad and Tobago, on the UWI- Zoology/ IDB Project; 84pp.

Bacon, P.R. and Ffrench, R. P., 1972. The Wildlife Sanctuaries on Trinidad and Tobago. Wildlife Conservation Committee, Ministry of Agriculture, Lands and Fisheries, Trinidad and Tobago; 80pp.

Beard, J. S., 1946. The Natural Vegetation of Trinidad. Clarendon Press, Oxford, UK; 152 pp.

Blumenfield, S., Lu, C., Chistophersen, T. and Coastes, D. 2009. Water, Wetlands and Forests. A Review of Ecological, Economic and Policy Linkages. Secretariat of the Convention on Biological Diversity and Secretariat of the Ramsar Convention on Wetlands, Montreal and Gland. CBD Technical Series No 47.

Brown, C. B., Hansell, J. R. F., Hill, I.D., Stark, J. and Smith, J. W. 1966. Land Capability Survey of Trinidad and Tobago # 1; 129pp.

33

Central Statistical Office (CSO) of Trinidad and Tobago, 1995. Population and Housing Census 1990.

Cintrón, G., Lugo, A. E., Pool, D. J. and Morris, G. 1978. Mangroves of arid environments in Puerto Rico and adjacent islands. Biotropica 10: 110-121.

Deonarine, G., 1980. Studies on the Biomagnification of Some Chlorinated Hydrocarbons in a Neotropical Mangrove Swamp. M. Phil Thesis, Zoology Department, University of the West Indies, Trinidad; 312pp.

Donawa, A. L., 1976 Draft Report of pollution Ecologist. . Caroni River Basin Study.

Government of Trinidad and Tobago, WASA. Planning Associates/ Burgess and Niple; 71pp.

FAO, 1994 Mangrove forest management guidelines FAO Forestry Paper 117, Rome.

Field, C. D. 1995. Journey amongst mangroves. International Society for Mangrove Ecosystem, Okinawa, Japan, 140pp.

Gerald, L. 1985 The Changing Caroni Swamp. Naturalist 6, 14-17.

Grimm, N. B, Faeth, S. H., Golubiewski, N. E., Redman, C. L., Wu, J., Bai, X & Briggs,J. M. 2008 Global Changes and the Ecology of Cities. Science Vol. 319; 756- 760.

Institute of Marine Affairs, 1999. Trinidad and Tobago Water Sector Institutional Strengthening Design and Implementation of the Surface Water Quality Monitoring Programme for the Caroni River Basin Draft Final Report - Surface Water Quality in the Caroni River Basin. Document prepared for the Water and Sewage Authority, Trinidad and Tobago; 215pp.

34

Juman, R. A. & Ramsewak, D. 2010. The state of mangrove forests in Trinidad and Tobago. Research Report, Institute of Marine Affairs, Hilltop Lane, Chaguaramas; 124 pp

Juman, R., Bacon, P. and Gerald, L. 2002. Caroni Basin Case Study, Caribbean Basins, LOICZ (Land-Ocean Interaction in the Coastal Zone) Global Change Assessment and Synthesis of River Catchment / Island Coastal Sea Interaction and Human Dimensions, with a desktop study of Oceania Basins.

Kjerfve, B., Wiebe, W.J., Kremer, H. H., Salomons, W. and Marshall Crossland, J. I. (Caribbean); Marcom, N., Harvey, N. and Marshall Crossland, J. I (Oceania); 2002. Caribbean Basins: LOICZ Global Change Assessment and Synthesis of River Catchment/ Island-Coastal Sea Interaction and Human Dimensions, with a desktop study of the Oceania Basins, LOICZ Reports and Studies No. 27, ii + 174 pp, LOICZ Texel, The Netherlands

Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-being: Wetlands and Water Synthesis. World Resource Institute, Washington D.C.

Nickerson, N. H. and Thibodeau, F. R. 1985. Association between pore water sulphide concentrations and the distribution of mangroves. Biogeochemistry 1: 183- 192.

Phelps, H. O., 1997. Caroni Swamp Rehabilitation Study- Hydrology and Engineering.Report to the Inter American Development Bank; 61pp.

Pool, D., Lugo, A. E. and Snedaker S. C. 1975. Litter productions in mangrove forest of southern Florida and Puerto Rico. In Proceedings of the International Symposium on Biology and Management of Mangroves Edited by G. E. Walsh, S. C. Snedaker and H. J Teas. Honolulu, Hawaii: East West Center.

Ramcharan, E.K., De Souza, G. and Ffrench, R. 1982 Inventory of the living resources of coastal wetlands of Trinidad. Draft technical report, Institute of Marine Affairs, Trinidad and Tobago, July 1982. 91 pp.

35

Santos, M. C. F. V., Zieman, J. C. and Cohen, R. R. H. 1997. Interpreting the upper mid- littoral zonation patterns of mangroves in Maranhao (Brazil) in response to microtopography and hydrology. In Mangrove ecosystems studies in Latin America and Africa. Edited by B. Kjerfve, L. D. de Lacerda and E. H. S. Diop. Paris: UNESCO and Tokyo: ISME

Siung-Chang, A. M., Norman, P. E. and Dalipsingh, R., 1987. Caroni River Study - Organic Pollution. Institute of Marine Affairs, Chaguaramas, Trinidad; 18pp.

Vitousek, P. M., Mooney, H. A., Lubchenco, J. & Melillo, J. M., 1997. Human Domination of Earth‟s Ecosystem. Science Vol. 227 No. 5325 pp 494-499

World Resources Institute. 1996. World Resources 1996-1997 Oxford University Press, New York

Worm, B., Barbier, E.B., Beaumont, N., Duffy, J. E., Folke, C., Halpern, B.S., Jackson, J.B.C., Lotze, H.K., Micheli, F., Palumbi, S.T et al. 2006. Science 314: 787- 790