The impact of land use change on the extent of mangroves in Aruba and their ecosystem services

(1)

The impact of land use change on the

extent of mangroves in Aruba and their

ecosystem services

Image 1: mangroves on Aruba (Van Essen instruments, n.d)

Author: Paulien Deken Student number: 12263982 Date: 29 May 2021 Place: Heerhugowaard Wordcount: 4832

Research cluster: Earth Sciences Supervisors: Kenneth F. Rijsdijk

Jim Groot

(2)

1

Abstract

Over the last 25 years the global mangrove area has decreased with 1 or 2 percent per year. The mangroves provide many ecosystem services like coastal protection, carbon sequestering and function as fish nurseries. In order to inform policy makers about the importance of the mangrove to provide management and protection policies for the mangroves it is valuable to quantify their ecosystem services. The research question answered by this study is: How did land use change in Aruba from 1986 to 2016 affected the extent of mangroves and its capacity to store carbon? This was done by using satellite images of Aruba and making a land use land cover map for the years 1986, 1996, 2006 and 2016. For the quantification of the carbon capturing capability a literature research was done. The results showed that the mangrove areas has decreased from 1986 to 2016 with 76 %. This decrease was visible around urban area and private islands that are used for tourism. With the clearing of the mangroves this also means that less carbon was captured. Therefore, Aruba contributes to the greenhouse effect instead of helping to diminishing it. Furthermore, other ecosystem services like coastal protection and their function as fish nursery will have declined with the clearing of the mangroves. This information can be used for management and monitoring of the mangroves in Aruba. Furthermore, the results could contribute to increase the knowledge about the value of the ecosystem services of the mangroves.

(3)

2

Content

Abstract ... 1 Introduction ... 3 Research question ... 4 Hypothesis ... 4 Relevance ... 4 Methodology ... 6 Satellite images ... 6

Land use classification with ArcGIS Pro ... 6

Image classification accuracy assessment ... 6

Literature review ecosystem services mangroves ... 7

Results ... 8

Image classification accuracy assessment ... 8

Mangroves over the years ... 8

Urban area over the years ... 9

Literature research on carbon sequestering capacity ... 12

Discussion ... 13

Quantification of carbon sequestering by the mangroves on Aruba ... 13

Loss of other ecosystem services ... 13

Other studies relating to mangroves ... 14

Possible limitations of the study ... 14

Conclusion ... 15 Literature list ... 16 Acknowledgements ... 18 Appendices ... 19 Appendix S1 ... 19 Appendix S2 ... 20 Appendix S3 ... 22 Appendix S4 ... 24 Appendix S6 ... 28

(4)

3

Introduction

Aruba is located in the Caribbean and has a surface area of 180 km2 with a population of 111.904 in 2021 (Oduber et al., 2015; Central bureau of statistics, 2020). Aruba is situated on 12°26’ Northern latitude and 16° 55’ Western longitude and is part of the kingdom of the Netherlands (Del Nevo, 2008). The island is small with a width of 8 to 9 km and 33 km long and its capital city is Oranjestad (Del Nevo, 2008). Aruba has a tropical steppe, semi-arid hot climate and is located in the intertropical convergence zone (Ter Horst & Becker, n.d.; Van Belle, 2016). The average precipitation is circa 471 mm during autumn which means that Aruba deals with a precipitation deficit (Oduber et al., 2015; Van Belle, 2016). The vegetation present on the island is persistent to a dry climate like cacti, scrubs and watapana (Del Nevo, 2008; Oduber et al., 2015). The temperature fluctuates between 19 and 36,5 degrees with an average of 27.9 degrees (Oduber et al., 2015. Aruba is a small island state which means it has limited natural resources and is vulnerable for climate change (Oduber et al., 2015). The island has a high biodiversity and hosts 207 bird species. This is being threatened by urbanisation which is partly due to tourism (Del Nevo, 2008). The land use history of Aruba is very diverse with drastic changes due to an oil refinery, aloe vera plantations and in recent years a flourishing tourism industry (Van Belle & Hall, 2016). The tourists come to Aruba for the tropical weather in combination with the beaches, coral reefs and astonishing nature (Oduber et al., 2015; Van Belle & Hall, 2016). One important ecosystem located in Aruba is the mangroves. Most of the mangroves in Aruba are found along the shore of the Spanish Lagoon, Mangel Halto and the small islands close to the shore. The most frequent occurring mangrove species are the Avicennia germinans and Conocarpus erectus (FAO, 2005).

Image 2 : Mangroves on Aruba in the Spanish Lagoon and on the right a Google Earth image of Aruba (Houtsma, n.d; Google Earth Pro, 2016)

Mangrove ecosystems all over the world are highly threatened (Worthington et al., 2020). The mangroves in the world are declining at an alarming rate of 1 to 2 percent per year. From the 120 countries with mangroves, 26 are extinct or heavily decreased. This is even faster than the decline of coral reefs (Duke et al., 2007). The mangroves are trees and bushes that form the transition from water to land (Houtsma, n.d.). These woody plants are important as they provide many ecosystem services. The ecosystem services differ per location due to different geomorphic settings, tidal amplitudes and climates (Worthington et al., 2020). Ecosystem services of the mangroves are among others: the provision of nurseries for fish, provision of habitat, carbon sequestering, protection of coral reefs and seagrasses, regulation of the water temperature and coastal protection (FAO, 2005; Fondo & Martens, 1998; Duke et al., 2007). The mangroves are of great importance for Aruba due to all these ecosystem services. Research by Van Belle & Hall (2016) showed that the mangroves in Aruba facilitate habitat for many organisms like birds and endangered species like the green iguana. The mangroves in Aruba are valuable for marine and terrestrial systems. Therefore research on the ecosystem services of mangroves in Aruba and their change over the years is valuable.

(5)

4 The Caribbean has lost approximately 41% m2 of mangroves in the last 25 years (Debrot et al., 2018). These islands are highly urbanised and thereby threaten the mangrove area. Especially the pressure of tourism is substantial in Aruba. In 2000 Aruba facilitated 721.000 tourists and in 2013 around 979.000 tourists (Central Bureau of Statistics, 2014). The tourist sector in Aruba accounted for 88.1% percent of the GDP in 2016. As a consequence of this industry the population is expected to grows as well. In 2010 Aruba had a population of 101.485 and in 2020 the population was 111.904 (Central Bureau of Statistics, 2014; Central Bureau of Statistics, 2020). For comparison, Texel, a small island that belongs to the Netherlands, has a surface area of 170 km2 with 13.658 inhabitants in 2020 (Centraal Bureau voor de Statistiek, 2021). The growing population means a higher demand for housing resulting in expanding urban areas. Urbanisation pressures the land scarcity that is corresponding to an island. The urban area is created at the expense of forestation and therefore affects biodiversity (Oduber et al., 2015). Shrinking of the mangroves leads to loss of ecosystem services (Fondo & Martens, 1998; Sjöling et al., 2005; Worthington et al., 2020). Johnston formulated ecosystem services as outputs, conditions, or processes of natural systems that directly or indirectly benefit humans or enhance social welfare (Johnston, 2018). There are 4 types of ecosystem services: regulating, supporting, provisioning and cultural ecosystem services (Wallace, 2017).

This research will focus on the carbon storage capacity by mangroves. This ecosystem service is of great importance due to the changing climate and the Paris agreement to decrease the emission of carbon dioxide (Alongi, 2008; Taillardat et al., 2018). Carbon storage is a regulating ecosystem service. Mangroves have the capacity to store carbon in the ground and to capture it out of the air. Mangroves have rich organic soils and high aboveground biomass and are thereby substantial carbon pools (Kauffman et al., 2011). When vegetated coastal ecosystems like the mangroves store carbon it is called blue carbon (Taillardat et al., 2018). A decrease in mangroves will result in less carbon storage (Hamilton & Friess, 2018; Duke et al., 2007). Hamilton & Friess (2018) calculated the global carbon losses from 2000 to 2012 due to deforestation of mangroves. This global loss in the researched timespan of 12 years was estimated to correspond to a value of 316,996,250 tons of CO2 that was emitted due to deforestation. This value includes the amount of carbon dioxide that was not sequestered by the mangroves. For comparison, the mangrove sequestration of carbon is estimated to be 168 ± 36 gC m−2 yr−1 while tropical forests sequester approximately 40 ± 20 gC m−2 yr−1 making mangroves a valuable carbon sink (Taillardat et al., 2018). The climate is changing due to the emission of greenhouse gasses like carbon dioxide. Therefore, the storage of carbon is increasingly important for climate mitigation (Allen et al., 2018). Thus, the quantification of the carbon storage capacity of the mangroves on Aruba is valuable.

Research question

The research question states ‘How did land use change in Aruba from 1986 to 2016 affected the extent of mangroves and its capacity to store carbon?’ This will be researched by doing a land use land cover analysis of the years 1986, 1996, 2006 and 2016 and a literature research on the ecosystem service carbon storage.

Hypothesis

It is hypothesised that the mangrove area will decrease and the urban area will increase as the global extent of mangroves had decreased with 1 to 2 percent per year ( Duke et al., 2007). Furthermore, Aruba’s population has increased in the timespan of the research which is expected to be seen in the land use cover map. The expected decrease in mangroves will correspond with a decrease in carbon storage per year.

Relevance

A research done by the FAO in 2005 affirms the knowledge gap and inaccuracy of the quantity of mangroves in Aruba due to the use of different methodologies. With the use of remote sensing the extent of mangroves and other land covers can be calculated over multiple years with the same methodology.

Van Belle & Hall (2016) state that there’s no alternative system for the mangroves and thereby it is important to research the function of the mangroves for a certain area. Especially in combination with the quantification of the impact of deforestation on ecosystem services like carbon storage. This could help to give policymakers insights into the direct impact of deforestation and the importance of conservation. This research will contribute to the conservation and protection of remaining mangrove ecosystems by policymaking in Aruba. Furthermore,

(6)

5 when understanding the dynamics of the land use change and its effect on the mangroves this can be beneficiary for global mangrove conservation.

The paper will follow the conventional structure which after the introduction will lead to the method, results section, discussion and conclusion. At the end of the paper the references, acknowledgement and appendix can be found.

(7)

6

Methodology

In order to answer the research question “How has the land use change in Aruba from 1986 to 2016 affected the extent of mangroves and its capacity to store carbon?” a land use land cover map was made of Aruba. This was done by a multi-temporal analysis of the years 1986, 1996, 2006 and 2016 to visualise the land use change over the years starting from 1986 with ArcGIS Pro. The maps were used to calculate the change of the mangroves and other land covers in hectares. After the change in the area of mangroves was estimated a literature review was done to quantify the ecosystem service carbon storage.

Satellite images

Google Earth Engine (GEE) was used to obtain the satellite images for classification with ArcGIS Pro. GEE has access to many satellite images among other the Landsat satellite data. The data for every research year was uploaded from GEE by selecting the year, the lowest cloud coverage and the area. Furthermore, for every year a composite was made to combine multiple Landsat images in order to obtain the highest quality of the images without clouds. For the years 1986 and 1996 Landsat 1-5 was used. Landsat 7 was used for 2006 and Landsat 8 for 2016. The scripts to obtain the satellite images can be found in the appendix (appendix S1). The images where exported to .tiff files in order to be accessible by ArcGIS Pro.

Land use classification with ArcGIS Pro

The .tiff files of the satellite images of Aruba made with GEE were imported to ArcGIS Pro. A new map was made for the years 1986, 1996, 2006 and 2016 where the corresponding images where imported. All the images were clipped with the same polygon that outlined Aruba. For the Landsat 1-5 and Landsat 7 images the red band was connected to spectral band 3, blue to band 2 and green to band 1. To obtain natural composition of the images from Landsat 8, red was band 4, blue band 3 and green band 2. This resulted in a natural composition for all the images. Then trainings data for the classification was created. For the classification 5 classes were selected. Namely: urban area, scrubs and woods, mangroves, desert and water. For the years 1996 and 1986 the class clouds was added due to some clouds on these images. These classes were selected by researching the land use history of Aruba and estimating the main land types with Google Earth Pro.

For every category, trainings data was created by drawing small polygons on corresponding pixels to the land cover class. For every class around 100 small polygons were created so ArcGIS Pro was trained to recognize the different classes. This number differs per class as, for example, the mangroves occupy a much smaller surface. The trainings data was made by using different sources as background information. For example, a vegetation map of Aruba (Stoffers, 1956), Google Earth Pro, tweaking the bands to create more contrast in the image and ancillary data of Aruba found in literature.

Next, with the classification wizard tool a pixel based and supervised classification was made. The classifier used for the year was chosen based of the examples of the classification maps. After the classification, the maps were reclassified when visible misclassification were spotted.

Image classification accuracy assessment

In order to check the accuracy of the classification an accuracy assessment classification was made for every year. This was done by selecting the classified map and choosing the accuracy assessment tool found under the classification tools. A new set of trainings data was made for each year and this was used as reference dataset for the accuracy assessment. When the Kappa value had a value lower than 0.85 the trainings data was redone or improved and the land use land cover classification was run again. After the classification was reliable enough proven by the confusion matrix the zonal statistics of the raster files were exported into an excel file. Finally, a layout for every year was made. The workflow for the land use maps can be found in image 2. The results of the land use maps were analysed by calculating the difference of the land cover categories over the years after the values were converted to hectares. The impact of the changes in land use on the carbon storage was assessed by the literature review.

(8)

7 Image 3: The workflow of the land use land cover analysis. This image shows the steps that were taken to make a land use land cover map and how this map was used to obtain the change in mangrove area over the years. First, satellite images were made with GEE and imported to ArcGIS Pro. Next, 2 sets of trainings data was created for the classification and the accuracy assessment. Lastly, the surface area per class was calculated with the zonal statistics tool.

Literature review ecosystem services mangroves

The next step was the analysis of literature on carbon storage by mangroves. The literature was collected and reviewed for relevance. The selected languages for the search were Dutch and English. The titles and abstracts of the paper were read to determine the usefulness of the paper for this research. Further, the reference list of the papers were scanned for other relevant papers. By doing the literature review, predictions were composed for the increase or decrease of the ecosystem services belonging to the mangroves. Comparable studies were used to calculate the carbon storage by the mangroves. There is a danger in extrapolation data from other research to this research due to differences in variables like temperature or geomorphology. However, as there is no known data for specifically the mangroves on Aruba the results found in other studies will be extrapolated to the data obtained from the land use land cover maps.

(9)

8

Results

Several observations can be made with regard to the land use maps (appendix S3, S4, S5 and S6). First of all, all the maps show that the urban area is mainly concentrated on the East and West-South coast of Aruba. Especially the land use map of 2016 shows the highly concentrated area at the coast with to the North more of a spreading between the urban area and shrubs and woods. In the North East of Aruba the Arikok natural park which contains many scrubs and woods is situated. The maps of 1996 and 1986 show some inaccuracies due to clouds. The mangroves are situated at the South part of Aruba. Most mangroves are observed to be on the small islands close to Aruba (Palm Island and Renaissance Island) and in the Spanish Lagoon. The focus of this research is on the mangroves and the urban area.

Image classification accuracy assessment

The accuracy of the land use land cover maps where calculated with the accuracy assessment tool by ArcGIS Pro. The output showed the Kappa value which is an overall accuracy value for the classified map. The land use map of 1986 has a Kappa value of 0.866756. Next, 1996 has a Kappa value of 0.969596, 2006 a value of 0.930167. Lastly, 2016 has a Kappa value of 0.986975. There were also p-values per individual class (appendix S2, S3, S4 and S5).The most accurate land use map is the map of 2016. Furthermore, all the maps had a surface area between 182 and 184 km2 what indicates that the classification of land and water was successful.

Mangroves over the years

The total area of mangroves in 1986 was found to be 697 hectares (table 1). In comparison with 2016 where the total area was 167 hectares a decrease of 76,0% was calculated (appendix S6). The strongest decline was after 1996 (figure 1). When comparing the land use maps of 1986 and 2016 most losses of mangroves are situated around the urban areas and the islands close to Aruba.

Figure 1: The area of mangroves in the years 1986, 1996, 2006 and 2016 in hectares. Table 1

Area Of Mangroves On Aruba Over The Years

Year

1986

1996

2006

2016

Area in hectares

697

674

397

167

Note: The area of mangroves on Aruba for the years 1986, 1996, 2006 and 2016 in hectares.

697 ₆₇₄ 397 167 0 100 200 300 400 500 600 700 800 1986 1996 2006 2016 in h ecta re s

(10)

9

Urban area over the years

The urban area of Aruba has grown with 22,8% percent from 1986 to 2016 (appendix S6). Noteworthy is the result that in 2006 there was less urban area than in 1996 (figure 2). This will be further elaborated on in the discussion. The land use maps show a more concentrated urban area around the South coastline with the years. Furthermore, there seems to be urban area and desert on the islands that used to be fully covered with mangroves in 1986.

Figure 2: The urban area in Aruba for the years 1986, 1996, 2006 and 2016 in hectares. Table 2

Urban Area In Aruba Over The Years

Year

1986

1996

2006

2016

Area in hectares

5829

7100

6783

7157

Note: The urban area in Aruba for the years 1986, 1996, 2006 and 2016 in hectares.

5829 7100 6783 7157 0 1000 2000 3000 4000 5000 6000 7000 8000 1986 1996 2006 2016 in h ecta re s

(11)

10 Some examples of the decrease in mangroves can be seen by the Spanish Lagoon from 1996 to 2016 (image 4). The satellite images show more urban area around the Spanish Lagoon and the mangrove area seems smaller in 2016.

Image 4: The satellite images of 1996 (left) and 2016 (right) focused on the Spanish Lagoon in Aruba. The red lines outline the mangrove area in the Spanish Lagoon.

On the Southern part of Aruba the airport can be found. Close by the airport and the parkietenbos mangroves are situated. The area of mangroves close to the Parkietenbos has decreased when looking at the satellite images (image 5). With the strongest decline between 1996 and 2006.

Image 5: Mangroves on the Southern part of Aruba in the years 1986 (upper left corner), 1996 (upper right corner), 2006 (lower left corner) and 2016( lower right corner).

(12)

11 Further, Google Earth Pro images of the small islands close to Aruba in 2016 show that there are resorts and houses build on the islands close to Aruba (image 6). These islands are called Renaissance and Palm Island.

Image 6: The upper left image is the Palm Island resort. The upper right image is a resort on Renaissance island and the image on the lower left corner are houses on Palm island. These are satellite images from Google Earth Pro of the year 2016.

(13)

12

Literature research on carbon storage capacity

The ecosystem service quantified by this research is the carbon storage capacity of the mangroves on Aruba. There have been many studies on the carbon storage capacity of mangroves. However, the burial rate of carbon differs greatly between different latitudes and temperatures. Therefore, it is important to carefully research the literature for a similar carbon rate. Tamis and Foekema (2016) researched blue carbon in the Dutch Caribbean. They state that the main carbon stock in the Caribbean was the mangroves followed by seagrasses and salt marshes. For their research they used the value 1.39 tonnes C ha-2 yr-1 to calculate the carbon burial rate in the Dutch Caribbean for mangroves. This value will be used for this research as well. This means that in 2016 the mangroves on Aruba buried 737 tonnes carbon less than in 1986 (see table 2). This means a decrease of 76%.

Table 3

Tonnes Of Carbon Stored In Mangroves On Aruba Per Year

Year 1986 1996 2006 2016

Carbon stored per year in tonnes

969 937 553 232

Note: This table shows the tonnes of carbon that can be stored by the mangroves per year. The first row shows the years and the second row shows the tonnes of carbon that could be stored. The numbers in the table are rounded to tonnes.

The carbon price per ton on May 14th 2021 is €56,12 following the European Union Emissions Trading System (EU ETS) (EMBER, 2021). This price is used as the value of a ton carbon. Therefore, the 736 tonnes of less carbon stored in 2016 could today be estimated to have a value of €41.360,44.

(14)

13

Discussion

The research question this study was designed to answer was “ How has the land use change in Aruba from 1986 to 2016 affected the extent of mangroves and its capacity to store carbon?”. The land use land cover map results showed a 22.8% increase in urban area and a 76% decrease in mangroves from 1986 to 2016. It can be speculated that the increase in urban area resulted in a decrease in mangroves. This can be supported by the fact that the population of Aruba increased from 1986 to 2016 with 85.9%. (Central bureau of statistics, 2016 & Central bureau of statistics, 1993 ). This can be seen in combination with the tourism sector being the main source of income for inhabitants. The FAO (2005) made an assessment rapport on Aruba where they stated that the deforestation of mangroves was mainly due to tourism infrastructure starting in the 1980s. From 1985 to 2000 the amount of tourists on Aruba had tripled. Starting from 1986 Aruba has focused on the more luxurious holidays and therefore the footprint per tourist increased as well (Cole & Razak, 2009). This results in more pressure on the natural area of Aruba including the mangroves. This can be interpreted by the results of this research where the urban area did show an increase. However, the results showed that in 1996 there was more urban area than in 2006 which is peculiar as the population kept increasing and therefore more houses were needed. It is expected that this result is due to an inaccuracy of the land use map of 1996. As mentioned in the results section the mangrove area had declined at the small islands close to Aruba in front of Oranjestad and the Spanish Lagoon. These islands are Renaissance island and Palm Island. The islands are private islands with resorts that can be visited. With an increasing demand for tourists it can be speculated that the mangroves on these islands will be cleared even more to accommodate tourists.

Quantification of carbon storage by the mangroves on Aruba

The second part of the research question was to quantify the capacity to store carbon by the mangroves. In line with previous studies, researching the total global area of mangroves (Goldberg et al., 2020), the mangrove ecosystem in Aruba has decreased from 1986 to 2016. The results showed a decrease that was estimated to be 76 %. With the total area of mangroves the carbon storage could be estimated. The carbon burial rate is highly dependent on many factors like the age of the mangroves and the human disturbance (Alongi, 2011). As mangroves are a substantial blue carbon stock these ecosystems are highly valuable to reduce the emissions (Hamilton & Fries, 2018). Mangroves can accumulate even more biomass with higher CO2 levels (Kathiresan, 2012). These results showed that there are actual losses concerning the population of Aruba. Not correctly managing the mangroves can have a financial loss. As the results showed, capturing less carbon in 2016 with the EU ETS carbon price of 2021 can be valued to be €41.360,44. This is not a complete value as it does not contain the social costs of carbon emission or the cost of alternative ways of reducing emission (Fuong, 2009). As a result, the storage of carbon by the mangroves can be financially beneficiary for Aruba.

This study is situated on Aruba but could be upscaled to the mangrove area on the entire Caribbean. Upscaling this research would emphasize the importance of the mangroves as an ecosystem and their ecosystem services. The consequence of upscaling this research would contribute to the knowledge on the ecosystem services of the mangroves. Especially the extent to which the global mangrove area could contribute to the lowering of the greenhouse effect could be valuable to further research. With this result the importance of the conservation and management of the mangroves on Aruba is highlighted.

Loss of other ecosystem services

This research has only quantified the carbon storage capacity of the mangroves on Aruba but there has been substantial research on the effects of deforestation on other ecosystem services as well. A study on Curaçao by Nagelkerken et al., (2002) showed that from the 17 juvenile fish species associated to mangroves, adult fish of 11 species did not occur in non-mangrove areas. Thereby showing the unique function of mangroves as fish nurseries. When the mangroves area decreases this could impact the fish species diversity in Aruba which could impact the tourism industry. Next, for coastal protection a research done by Mimura & Nunn (1998) showed that study areas where mangroves were removed quickly eroded the coast. Locations in Thailand that had not been deforested of mangroves showed 30 years later almost no loss of shoreline in contrast to the locations where the mangroves where cleared (López-Portillo et al., 2017). Fragmentation of the mangrove area also leads to less coastal protection (Lee et al., 2014). There are other options to protect the coast, namely seawalls

(15)

14 but these are very costly. In conclusion, the coast of Aruba will be more protected when there are mangroves then without the mangroves.

Other studies relating to mangroves

The results of this research speculate that the extent of the mangroves on Aruba was impacted by a change in land use. This is in line with a research done by Goldberg et al., (2020). This study showed that from 2006 to 2016 the global mangrove loss was for 62% due to human induced land use change.

The global forest resource assessment by the FAO in 2005 stated that the deforestation rate of the mangroves in Aruba in the 1990s slowed down (FAO, 2005). This does not correspond with the results of this research which showed the most deforestation happened between 1996 and 2006. The FAO did mention that the results of the mangrove area on Aruba was not clear due to the use of different methods to calculate the mangrove area which could impact their evaluation. Further, there have not been many previous studies on the mangrove area on Aruba starting from 1986. This is mostly because the mangrove area on Aruba is relatively small.

Possible limitations of the study

A limitation of this study naturally includes the resolution of the satellite images of Aruba. The resolution of all the images were 30 x 30 meters. Aruba is relatively small and therefore this resolution is quite large. The classification can be improved for further research by only classifying mangrove and non-mangrove areas with the help of the NDVI formula. However, this would mean that no conclusions can be drawn for the reasoning of the decrease or increase of the mangrove area.

The classifications for Aruba were made as accurate as possible. However, it must be pointed out that misclassification is inevitable and therefore the numbers mentioned in the results could possibly be slightly different in reality. One object that was repeatedly misclassified by the classification wizard of ArcGIS was a park and lake located on the Northwester part of Aruba. This lake seems dark on the satellite images, the same dark green colour as the mangroves. Google Earth Pro showed that this was a lake. As the classification wizard mistook this lake for all the images there is continuity in the data. Another difficult part of Aruba to classify was the shallow waters close to the island. These shallow waters were sometimes mistaken for mangroves. Many of these mistakes were removed by the reclassification tool but not all.

It is important to note that the results of carbon burial rate are dependent on a literature review where results from another study have been extrapolated to quantify the carbon storage by the mangroves on Aruba. Due to an extrapolation the results are speculative and not fully substantiated. There are multiple burial rates that differ greatly. However, the chosen burial rate was obtained from a research on the Dutch Caribbean and therefore after consideration this data was accepted to use. There was a lack of information specifically for the mangroves on Aruba to quantify other regulating and supporting services by the mangroves (Alongi, 2011).

With the results from this study policy makers could be more informed when it comes to mangrove management and protection in Aruba. The monitoring of the mangroves will be relevant for a better understanding of losses of mangrove area. This study highlights the importance of researching the ecosystem services for the mangroves in order to maintain the benefits the mangroves provide us with. Especially to show policymakers the effects of the loss of mangroves. Further research could be done to quantify other ecosystem services for the mangroves on Aruba as there has been a lack of information. This could also be done for other mangroves to benefit the global knowledge on the function of the mangroves.

(16)

15

Conclusion

To conclude, the mangrove area in Aruba has decreased with 76% from 1986 to 2016. As the urban area has increased with 22.8 %, urbanisation may have led to less mangroves. The clearing of mangroves happened mainly at the small islands close to Aruba and mangroves surrounded by urban area. On these small islands there are resorts and private beaches offered to tourists. There have been studies that researched the pressure of the tourism industry on Aruba. This research supports these findings as a loss of mangroves was seen over the years. The literature review has shown that the decrease in mangrove area may have resulted in 737 tonnes less carbon stored in 2016. This value can be quantified to have a financial value of €41.360,44. Furthermore, by clearing mangrove area there will be more carbon dioxide in the air and therefore Aruba will contribute to the greenhouse effect instead of helping to diminishing it. Other ecosystem services will likely also be negatively impacted by the decrease in mangrove area. The decrease will likely impact the mangroves function as fish nursery and coastal protection. This study emphasizes the importance of the mangroves on Aruba and highlights the need for restoration and protection in order to maintain their ecosystem services. The monitoring of mangrove areas globally can help to highlight their importance and the ecosystem service losses that are accompanied by deforestation.

(17)

16

Literature list

Allen, M., Babiker, M., Chen, Y., Taylor, M., Tschakert Australia, P., Waisman, H., Warren, R., Zhai, P., Zickfeld, K., Zhai, P., Pörtner, H., Roberts, D., Skea, J., Shukla, P., Pirani, A., Moufouma-Okia, W., Péan, C., Pidcock, R., Connors, S., … Waterfield, T. (2018). Summary for Policymakers SPM. In Aromar Revi.

Alongi, D. M. (2008). Mangrove forests: Resilience, protection from tsunamis, and responses to global climate change. In Estuarine, Coastal and Shelf Science (Vol. 76, Issue 1, pp. 1–13). Academic Press.

Alongi, D. M. (2011). Carbon payments for mangrove conservation: ecosystem constraints and uncertainties of sequestration potential. Environmental science & policy, 14(4), 462-470.

Centraal Bureau voor de Statistiek. (2021). Bevolkingsontwikkeling; regio per maand. 2021. Retrieved March 19, 2021, from https://www.cbs.nl/nl-nl/cijfers/detail/37230ned?q=texel

Central Bureau of Statistics (1993). The population of Aruba, a demographic profile. Retrieved May, 14 from: https://cbs.aw/wp/wpcontent/uploads/2013/02/05Aa_The_population_of_Aruba_a_demographic_profil e.pdf

Central Bureau of Statistics (2014) Population Projections Revised 3. Retrieved March, 23 from: www.cbs.aw Central Bureau of Statistics (2017) Quarterly Demographic Bulletin 2016. Retrieved May, 14 from:

https://cbs.aw/wp/index.php/2017/05/23/quarterly-demographic-bulletin-2016/

Central bureau of statistics. (2020.). Quarterly Demographic Bulletin 2020 – Central Bureau of Statistics. 2020. Retrieved March 19, 2021, from https://cbs.aw/wp/index.php/2020/12/17/quarterly-demographic-bulletin-2019-2/

Central Bureau of Statistics. (2021). Df.1.1. International Tourism. Retrieved March 24, 2021, from https://cbs.aw/wp/index.php/2014/11/01/df-1-1-international-tourism/

Chong, V. C. (2007). Mangroves-fisheries linkages—the Malaysian perspective. Bulletin of Marine Science, 80(3), 755-772.

Debrot, A. O., Henkens, R. J. H. G., & Verweij, P. J. F. M. (2018). Staat van de natuur van Caribisch Nederland 2017: Een eerste beoordeling van de staat (van instandhouding), bedreigingen en

managementimplicaties van habitats en soorten in Caribisch Nederland (No. C086/17). Wageningen Marine Research.

Del Nevo, A. D. R. I. A. N. (2008).Important Bird Areas in the Caribbean - ARUBA.

Duke, N. C., Meynecke, J. O., Dittmann, S., Ellison, A. M., Anger, K., Berger, U., & Dahdouh-Guebas, F. (2007). A world without mangroves?. Science, 317(5834), 41-42.

Duong, M. H. (2009). What is the Price of Carbon? Five definitions. SAPI EN. S. Surveys and Perspectives Integrating Environment and Society, (2.1).

EMBER (2021, 16 May). Daily EU ETS carbon market price (Euros). Retrieved from: https://ember-climate.org/data/carbon-price-viewer/

FAO. (2005). Forestry Department Food and Agriculture Organization of the United Nations GLOBAL FOREST RESOURCES ASSESSMENT 2005 THEMATIC STUDY ON MANGROVES ARUBA COUNTRY PROFILE. Retrieved March, 23 from http://www.fao.org/forestry/9178-0ca4d95f1be2c2e8596a4a0e51e01a2e.pdf

Fondo, E. N., & Martens, E. E. (1998). Effects of mangrove deforestation on macrofaunal densities, Gazi Bay, Kenya. In Mangroves and Salt Marshes (Vol. 2).

Goldberg, L., Lagomasino, D., Thomas, N., & Fatoyinbo, T. (2020). Global declines in human‐driven mangrove loss. Global change biology, 26(10), 5844-5855.

Hamilton, S. E., & Friess, D. A. (2018). Global carbon stocks and potential emissions due to mangrove deforestation from 2000 to 2012. Nature Climate Change, 8(3), 240–244.

(18)

17 Houtsma, S. (n.d.) The current status of mangrove forests in Spanish Lagoon (Aruba) evaluated from a

hydrological point of view.

Johnston, R. J. (2018, January 5). Ecosystem services. Encyclopedia Britannica. Retrieved from: https://www.britannica.com/science/ecosystem-services

Kathiresan, K. (2012). Importance of mangrove ecosystem. International Journal of Marine Science, 2(10). Kauffman, J. B., Heider, C., Cole, T. G., Dwire, K. A., & Donato, D. C. (2011). Ecosystem Carbon Stocks of

Micronesian Mangrove Forests.

Nagelkerken, I., Roberts, C., van der Velde, G., Dorenbosch, M., van Riel, M., Cocheret de la Morinière, E., & Nienhuis, P. (2002). How important are mangroves and seagrass beds for coral-reef fish? The nursery hypothesis tested on an island scale. Marine Ecology Progress Series, 244, 299–305.

Oduber, M., Ridderstaat, J., & Martens, P. (2015). The Connection of Vegetation with Tourism Development and Economic Growth: A Case Study for Aruba. Journal of Environmental Science and Engineering A, 4, 420–431.

Sjöling, S., Mohammed, S. M., Lyimo, T. J., & Kyaruzi, J. J. (2005). Benthic bacterial diversity and nutrient processes in mangroves: Impact of deforestation. Estuarine, Coastal and Shelf Science, 63(3), 397–406. Stoffers, A. L. (1956). The vegetation of the Netherlands Antilles (No. 135). Kemink.

Taillardat, P., Friess, D. A., & Lupascu, M. (2018). Mangrove blue carbon strategies for climate change mitigation are most effective at the national scale. Biology Letters, 14(10).

Tamis, J. E., & Foekema, E. M. (2016). Blue carbon in the Dutch Caribbean: Memo. IMARES Wageningen UR. Tran, L. X., & Fischer, A. (2017). Spatiotemporal changes and fragmentation of mangroves and its effects on fish

diversity in Ca Mau Province (Vietnam). Journal of Coastal Conservation, 21(3), 355–368. Van Belle, J.(2016). Inventarisatie Terrestrisch Milieu van het Eiland Aruba. RuG, the Netherlands.

Van Essen Instruments. (n.d). Hydrological status of the mangrove forest Het Spaans Lagoen (Aruba). Retrieved May 14, 2021 from: https://www.vanessen.com/hydrological-status-of-the-mangrove-forest-het-spaans-lagoen-aruba/

Wallace, K. J. (2007). Classification of ecosystem services: problems and solutions. Biological conservation, 139(3-4), 235-246.

Worthington, T. A., zu Ermgassen, P. S. E., Friess, D. A., Krauss, K. W., Lovelock, C. E., Thorley, J., Tingey, R., Woodroffe, C. D., Bunting, P., Cormier, N., Lagomasino, D., Lucas, R., Murray, N. J., Sutherland, W. J., & Spalding, M. (2020). A global biophysical typology of mangroves and its relevance for ecosystem structure and deforestation. Scientific Reports, 10(1).

(19)

18

Acknowledgements

After a three month period of constantly working on this research it is finally finished. I would like to thank my supervisor Kenneth F. Rijsdijk for always making me feel more inspired to continue after a meeting. Furthermore I would like to thank Sharona Jurgens for providing me with the background knowledge of Aruba and her support during the process. Thanks to Jim Groot because due to his question hours I could obtain all the data from ArcGIS and write the thesis. Lastly, I would like to thank Isis Kaiser and Suzanne Brouwer for proofreading my thesis and helping me to improve the quality of the thesis by correcting spelling errors and to improve the clarity of the arguments made.

(20)

19

Appendices

Appendix S1

Script to obtain satellite images with Google Earth Engine

This is the script used to obtain the satellite images. This script was used for all the years by changing the date when retrieving the image collection from Google Earth Engine.

Script 2006

// Create a polygon outlining Aruba named geometry

// Load a raw Landsat 5 ImageCollection for a single year for Aruba.

var collection = ee.ImageCollection("LANDSAT/LE07/C01/T1_RT").filterDate('2006-01-01', '2006-12-31').filterBounds(ee.Geometry.Point(-69.97, 12.54));

// Create a cloud-free composite with default parameters.

var composite = ee.Algorithms.Landsat.simpleComposite(collection);

// Create a cloud-free composite with custom parameters for // cloud score threshold and percentile.

var customComposite = ee.Algorithms.Landsat.simpleComposite({ collection: collection,

percentile: 90, cloudScoreRange: 10 });

// Display the composites. Map.centerObject(geometry);

Map.addLayer(composite, {bands: ['B3', 'B2', 'B1'], max: 128}, 'Aruba'); //Map.addLayer(customComposite, {bands: ['B3', 'B2', 'B1'], max: 128}, // 'Custom TOA composite');

Export.image.toDrive({ image: composite, description: 'imageToDriveExample', scale: 30, region: geometry });

(21)

20

Appendix S2

In this appendix the zonal statistics, confusion matrix and land use land cover map of Aruba in 2016 can be found.

Table 4

Zonal Statistics Of Land Use Land Cover Map Of Aruba In 2016

Land use class

AREA in m2

Desert

17469807

Scrubs and woods

92378089

Mangroves

1668731

Urban area

71571699

Water

33408918

Note: This table shows the zonal statistics of the land use land cover map of Aruba in 2016. The land use classes are noted on the left column and the left column shows the area in m2.

Table 5

Confusion Matrix Of Land Use Land Cover Map Of Aruba In 2016

ClassValue C_0

C_1

C_2

C_3

C_4

Total

U_Accuracy Kappa

Desert

21

0

2

0

23 0.913043

0 Scrubs and

woods

0

155

1

0

1

157 0.987261

0 Mangroves

0

2

9

0

11 0.818182

0 Urban area

0

99

0

99

1

0 Water

0

716

1

0 Total

21

157

10

101

717 1006

0

0 P_Accuracy

1 0.987261

0.9 0.980198 0.998605

0 0.994036

0 Kappa

0

0 0 0.986975

Note: This table shows the confusion matrix of the land use land cover map of 2016. The land use classes can be found in the 1st column. The following columns show how many points were correctly classified per land use class. The last column shows the Kappa value which shows the accuracy of the classification.

(22)

21

This is the Land Use Land Cover map of Aruba in 2016. The unit of the map is in meters and the

coordinate system is WGS 1984 UTM Zone 19N

Figure 7: Land Use Land Cover map of Aruba in 2016. This image shows the land use land cover map of Aruba in 2016. The land use types identified on this map are: desert, scrubs and woods, mangroves, urban area and water.

(23)

22

Appendix S3

Table 6

Land use class

AREA in m2

Desert

15431810

Scrubs and woods

97923237

Urban area

67832439

Water

31100125

Mangroves

3974886.7

Note: This table shows the zonal statistics of the land use land cover map of Aruba in 2006. The land use classes are noted on the left column and the left column shows the area in m2.

Table 7

ClassValue C_0

C_1

C_2

C_3

C_4

Total

U_Accuracy Kappa

Desert

31

1

2

11

0

45 0.688889

0 Scrubs and

woods

0

84

4

0

1

89 0.94382

0 Urban area

1

2

64

0

67 0.955224

0 Water

0

792

0

792

1

0 Mangroves

0

3

0

9

12

0.75

0 Total

32

90

70

803

10 1005

0

0 P_Accuracy 0.96875 0.933333 0.914286 0.986301

0.9

0 0.975124

0 Kappa

0

0 0 0.930167

(24)

23 Figure 8

This is the Land Use Land Cover map of Aruba in 2006 .The unit of the map is in meters and the

coordinate system is WGS 1984 UTM Zone 19N

Figure 8: This image shows the land use land cover map of Aruba in 2006. The land use types identified on this map are: desert, scrubs and woods, mangroves, urban area and water.

(25)

24

Appendix S4

Table 8

Zonal statistics of land use land cover map of Aruba in 1996

Land use type

AREA in m2

Desert

16033186

Scrubs and woods

87917002

Mangroves

6740866

Urban area

70999336

Water

29681967

Clouds

5124887

Table 9

ClassValue C_0

C_1

C_2

C_3

C_4

C_5

Total

U_Accuracy Kappa

Desert

31

0

5

0

36 0.861111

0 Scrubs and

woods

0

181

1

9

0

191 0.947644

0 Mangroves

0

9

0

9

1

0 Urban area

0

3

0

128

0

131 0.977099

0 Water

0

594

0

594

1

0 Clouds

0

40

1

0 Total

31

184

10

142

594

40 1001

0

0 P_Accuracy

1 0.983696

0.9 0.901408

1

0 0.982018

0 Kappa

0

0 0 0.969596

(26)

25

This is the Land Use Land Cover map of Aruba in 1996 .The unit of the map is in meters and the

coordinate system is WGS 1984 UTM Zone 19N.

Image 9: This image shows the land use land cover map of Aruba in 1996. The land use types identified on this map are: desert, scrubs and woods, mangroves, urban area, water and clouds.

(27)

26

Appendix S5

Table 10

Land use type

AREA in m2

Desert

13784179

Scrubs and woods

1.05E+08

Mangroves

6967701

Urban area

58292183

Water

31256624

Clouds

1103403

Table 11

ClassValue

C_0

C_1

C_2

C_3

C_4

C_5

Total

U_Accuracy Kappa

Desert

10

2

0

1

0

13 0.769231

0 Scrubs and

woods

0

104

0

5

0

109 0.954128

0 Mangroves

0

10

0

10

1

0 Urban area

0

5

0

28

0

33 0.848485

0 Water

0

10

0

10

1

0 Clouds

0

0 Total

10

111

10

34

10

0

175

0

0 P_Accuracy

1 0.936937

1 0.823529

1

0

0 0.925714

0 Kappa

0

0 0 0.866756

(28)

27

This is the Land Use Land Cover map of Aruba in 1986 .The unit of the map is in meters and the

coordinate system is WGS 1984 UTM Zone 19N

Image 10: This image shows the land use land cover map of Aruba in 1986. The land use types identified on this map are: desert, scrubs and woods, mangroves, urban area, water and clouds.

(29)

28

Appendix S6

In this appendix the calculations for the differences in mangrove and urban area over the years are showed. Table 12

Mangroves On Aruba Over The Years In Hectares

Year

1986

1996

2006

2016

Area in hectares

697

674

397

167

Note: The first row shows the years and the second row shows the area of mangroves in hectares.

The difference in mangrove area between 1986 and 2016: 167−697

697 ∗ 100% = −76.04% Table 13

Urban Area On Aruba Over The Years In Hectares

Year

1986

1996

2006

2016

Area in hectares

5829

7100

6783

7157

Note: The first row shows the years and the second row shows the area of urban land in hectares. The difference in urban area between 1986 and 2016: 7157−5829