Ecosystem Services of marina in the

Dissertation by Hanan Almahasheer

In Partial Fulfillment of the Requirements

For the Degree of

Doctor of Philosophy

King Abdullah University of Science and Technology, , Kingdom of

November, 2016 2

EXAMINATION COMMITTEE PAGE

The dissertation by Hanan Almahasheer is approved by the examination committee.

Committee Chairperson: Xabier Irigoien Committee Members: Carlos Duarte, Danielle Daffonchio, Catherine Lovelock

3

© November, 2016 Hanan Almahasheer All Rights Reserved 4

ABSTRACT

Ecosystem Services of Avicennia marina in the Red Sea

Hanan Almahasheer

The Red Sea is an arid environment, without riverine inputs, oligotrophic waters and extreme temperature and salinity. Avicennia marina is the dominant vegetation in the shores of the Red Sea. However, little is known about their distribution, dynamics, and services. Therefore, the aim of this Ph.D. was to obtain the basic information needed to evaluate their role in the coastal ecosystems and quantify their services. With that objective we 1) estimated the past and present distribution of in the Red Sea, 2) investigated the growth, leave production and floration 3) examined the growth limiting factors 4) measured the nutrients and heavy metal dynamics in the leaves and 5) estimated carbon sequestration. We found an increase of about 12% in the last 41 years, which contrasts with global trends of decrease. The extreme conditions in the Red Sea contributed to limit their growth resulting in stunted trees. Hence, we surveyed Central Red Sea mangroves to estimate their node production with an average of 9.59 node y-1 then converted that number into time to have a plastochrone interval of 38 days. As mangroves are taller in the southern Red Sea where both temperature and nutrients are higher than the Central Red Sea, we assessed nutrient status Avicennia marina propagules and naturally growing leaves to find the leaves low in nutrient concentrations (N < 1.5 %, P < 0.09 %, Fe < 0.06) and that nutrients are reabsorbed before shedding the leaves (69%, 72% and 35% for N, P, and Fe respectively). As a result, we conducted a fertilization experiment (N, P, Fe and combinations) to find that iron additions alone led to significant growth responses. Moreover, we estimated their leaf production and used our previous estimates of both the total cover in the Red Sea along with plastochrone interval to assess their total nutrients flux per year to be 2414 t N, 139 t P and 98 t Fe. We found them to sequester 34 g m-2 y-1, which imply 4590 tons of carbon sequestered per year for the total mangroves covered by the Red Sea.

5

ARABIC ABSTRACT

ﻣﻠﺨﺺ ﺧﺪﻣﺎت اﻟﻨﻈﺎم اﻟﺒﻴﺌﻲ ﻟﻨﺒﺎت ُاﻟﻘﺮم ﻓﻲ اﻟﺒﺤﺮ اﻷﺣﻤﺮ ﺣﻨﺎن اﻟﻤﻬﺎﺷﻴﺮ

ﻳﻌﺪ اﻟﺒﺤﺮ اﻷﺣﻤﺮ ﻣﻦ اﻟﺒﻴﺌﺎت اﻟﻘﺎﺣﻠﺔ اﻟﺘﻲ ﺗﻔﺘﻘﺮ ﻟﻠﻤﻮارد اﻟﻨﻬﺮﻳﺔ، وﺗﺘﻤﻴﺰ ﺑﻔﻘﺮ اﻟﻤﻐﺬﻳﺎت اﻟﻤﺎﺋﻴﺔ ودرﺟﺎت اﻟﺤﺮارة اﻟﻘﺼﻮى واﻟﻤﻠﻮﺣﺔ. إن ُاﻟﻘﺮم ﻣﻦ اﻟﻨﺒﺎﺗﺎت اﻟﻤﻨﺘﺸﺮة ﻋﻠﻰ ﺳﻮاﺣﻞ اﻟﺒﺤﺮ اﻷﺣﻤﺮ ﻏﻴﺮ أﻧﻨﺎ ﻻ ﻧﻌﺮف اﻟﻜﺜﻴﺮ ﻣﻦ اﻟﻤﻌﻠﻮﻣﺎت ﻋﻦ ﻣﺼﺎدر ﺗﻮزﻳﻌﻬﺎ وﺣﺮﻛﻴﺘﻬﺎ وﺧﺪﻣﺎﺗﻬﺎ. ﻟﺬا ﻓﺈن اﻟﻬﺪف ﻣﻦ رﺳﺎﻟﺔ اﻟﺪﻛﺘﻮراﻩ ﻫﻮ اﻟﺤﺼﻮل ﻋﻠﻰ اﻟﻤﻌﻠﻮﻣﺎت اﻷﺳﺎﺳﻴﺔ اﻟﻼزﻣﺔ ﻟﺘﻘﻴﻴﻢ دورﻫﺎ ﻓﻲ اﻟﻨﻈﻢ اﻟﺒﻴﺌﻴﺔ اﻟﺴﺎﺣﻠﻴﺔ، وﺗﻘﺪﻳﺮ ﺧﺪﻣﺎﺗﻬﺎ. ﻟﺬا ﻗﻤﻨﺎ ﻣﻦ ﺧﻼل ذﻟﻚ اﻟﻬﺪف ﻋﻠﻰ (1) ﺗﻘﺪﻳﺮ ﻣﺎﺿﻲ وﺣﺎﺿﺮ ﻧﺒﺎﺗﺎت اﻷﻳﻜﺎت اﻟﺴﺎﺣﻠﻴﺔ ﻓﻲ اﻟﺒﺤﺮ اﻷﺣﻤﺮ و (2) دراﺳﺔ اﻟﻨﻤﻮ واﻧﺘﺎج اﻷوراق واﻹزﻫﺎر (3) وﻓﺤﺺ ﻋﻮاﻣﻞ ﺣﺼﺮ اﻟﻨﻤﻮ (4) وﻗﻴﺎس اﻟﻤﻐﺬﻳﺎت، وﺧﺼﺎﺋﺺ اﻟﻤﻌﺎدن اﻟﺜﻘﻴﻠﺔ ﻓﻲ اﻷوراق و(5) ﺗﻘﺪﻳﺮ ﻋﺰل اﻟﻜﺮﺑﻮن. وﺟﺪﻧﺎ زﻳﺎدة ﺑﻤﺎ ﻳﺮﺑﻮ ﻋﻠﻰ 12 ﻓﻲ اﻟﻤﺎﺋﺔ ﺧﻼل اﻟﻮاﺣﺪ واﻷرﺑﻌﻴﻦ ﻋﺎﻣﺎ اﻷﺧﻴﺮة ﻣﻤﺎ ﻻ ﻳﻨﺴﺠﻢ ﻣﻊ اﻻﺗﺠﺎﻫﺎت اﻟﻌﺎﻟﻤﻴﺔ ﻓﻲ اﻟﻨﻘﺼﺎن. أﺳﻬﻤﺖ اﻟﻈﺮوف اﻟﻘﺎﺳﻴﺔ ﻓﻲ اﻟﺒﺤﺮ اﻷﺣﻤﺮ ﻓﻲ اﻟﺤﺪ ﻣﻦ ﻧﻤﻮﻫﺎ، ﻣﻤﺎ أدى اﻟﻰ ﻇﻬﻮر أﺷﺠﺎر ﻏﻴﺮ ﻣﻜﺘﻤﻠﺔ اﻟﻨﻤﻮ. ﻟﺬﻟﻚ أﺟﺮﻳﻨﺎ ﻣﺴﺤﺎ ﻟﻨﺒﺎﺗﺎت اﻷﻳﻜﺎت اﻟﺴﺎﺣﻠﻴﺔ ﻓﻲ ﻣﻨﻄﻘﺔ وﺳﻂ اﻟﺒﺤﺮ اﻷﺣﻤﺮ ﻟﺘﻘــــــــﺪﻳﺮ اﻧﺘﺎج اﻟﻌﻘﺪ ﺑﻤﻌﺪل 9.59 ﻋﻘﺪة ﻟﻜﻞ ﺳﻨﺔ ﺛﻢ ﺗﻢ ﺗﺤﻮﻳﻞ ذﻟﻚ اﻟﺮﻗﻢ اﻟﻰ زﻣﻦ ﻟﻠﺤﺼـــــﻮل ﻋﻠﻰ ﻓﺎﺻﻞ زﻣﻨﻲ ﻟﻔﺘﺮة اﻟﻨﻤﻮ ﺑﻮاﻗــــــــــﻊ 38 ﻳﻮﻣﺎ. وﺑﻤـــــــﺎ أن ﻧﺒﺎﺗــــــــــــــﺎت اﻷﻳﻜﺎت اﻟﺴﺎﺣﻠﻴﺔ أﻃـــﻮل ﻓﻲ اﻟﻤﻨﻄﻘﺔ اﻟﺠﻨﻮﺑﻴـــﺔ ﻣﻦ اﻟﺒﺤﺮ اﻷﺣﻤﺮ ودرﺟـــــــﺔ اﻟﺤــــــــــــﺮارة واﻟﻤﻐﺬﻳــــــــــﺎت أﻋﻠـــــﻰ ﻣﻦ ﻣﻨﻄﻘــــــــﺔ وﺳﻂ اﻟﺒﺤﺮ اﻷﺣﻤﺮ. ﻗﻤﻨﺎ ﺑﺈﺟــﺮاء ﺗﻘﻴﻴــــــﻢ ﻟﻮﺿــﻊ اﻟﻤﻐﺬﻳـــــــــﺎت ﻟــﺒﺬور ﻧﺒــــﺎت اﻟﻘـــــﺮم وأوراﻗـــﻪ اﻟﻨــــــﺎﻣﻴــــــــــــــﺔ ﺑﺼـﻮرة ﻃﺒﻴﻌﻴــــﺔ ﻟـﻴﺘﺒﻴﻦ ﻟﻨــــــــﺎ أن اﻷوراق ﻣﻨﺨﻔﻀــــﺔ ﻣــــــــﻦ ﺣﻴﺚ ﻣﺤﺘـــــــــــﻮى اﻟﻨﺒـــــــــﺎت ﻣـــــﻦ اﻟﻌﻨﺎﺻــــــــــــﺮ اﻟﻐــﺬاﺋﻴـــــــــــــــــﺔ (N < 1.5 %, P < 0.09 %, Fe < 0.06) وأن اﻟﻤﻐـــــــــــــــﺬﻳﺎت ﻳﻌــــــــــﺎد اﻣﺘﺼﺎﺻﻬﺎ ﻗﺒــــــــﻞ ﺗﺴﺎﻗـــﻂ اﻷوراق (%69 و %72 و %35 ﺑﺎﻟﻨﺴﺒﺔ ﻟﻠﻨﻴﺘﺮوﺟﻴﻦ و اﻟﻔﻮﺳﻔﻮر و اﻟﺤﺪﻳﺪ ﻋﻠﻰ اﻟﺘﻮاﻟﻲ ). وﺑﻨﺎء ﻋﻠﻰ ذﻟﻚ، أﺟﺮﻳﻨﺎ ﺗﺠﺮﺑﺔ اﺧﺼﺎب ﻟﻠﻨﻴﺘﺮوﺟﻴﻦ واﻟﻔﻮﺳﻔﻮر واﻟﺤﺪﻳﺪ واﻟﺘﻮﻟﻴﻔﺎت، واﺳﺘﺨﻠﺼﻨﺎ ﻣﻨﻬﺎ أن اﻻﺿﺎﻓﺎت اﻟﺤﺪﻳﺪﻳﺔ ﻟﻮﺣﺪﻫﺎ ﻗﺪ أدت إﻟﻰ اﺳﺘﺠﺎﺑﺎت ﻧﻤﻮ ﻣﻌﻨﻮﻳﺔ. ﻛﻤﺎ ﻗﻤﻨﺎ ﺑﺘﻘﺪﻳﺮ إﻧﺘﺎج أوراﻗﻬﺎ واﺳﺘﺨﺪﻣﻨﺎ ﺗﻘﺪﻳﺮاﺗﻨﺎ اﻟﺴﺎﺑﻘﺔ اﻟﺨﺎﺻﺔ ﺑﻤﺠﻤﻮع ﻏﻄﺎء ﻧﺒﺎﺗﺎت اﻷﻳﻜﺎت اﻟﺴﺎﺣﻠﻴﺔ ﻓﻲ اﻟﺒﺤﺮ اﻷﺣﻤﺮ، واﻟﻔﺎﺻﻞ اﻟﺰﻣﻨﻲ ﻟﻔﺘﺮة اﻟﻨﻤﻮ ﻟﺘﻘﻴﻴﻢ ﻣﺠﻤﻮع اﻟﺘﺪﻓﻘﺎت اﻟﻐﺬاﺋﻴﺔ ﺳﻨﻮﻳﺎً ﺑﻮاﻗﻊ 139 ﻃﻦ ﻣﻦ اﻟﻔﻮﺳﻔﻮر و 2414 ﻃﻦ ﻣﻦ اﻟﻨﺘﺮوﺟﻴﻦ و 98 ﻃﻦ ﻣﻦ اﻟﺤﺪﻳﺪ، وأﺧﻴﺮاً ﺛﺒﺖ ﻟﻨﺎ ﻋﺰل ﺑﻮاﻗﻊ 34 ﺟﻢ ﻟﻜﻞ ﻣﺘﺮ ﻣﺮﺑﻊ ﻟﻜﻞ ﺳﻨﺔ، ﻣﻤﺎ ﻳﻌﻨﻲ وﺟﻮد 4590 ﻃﻦ ﻣﻦ اﻟﻜﺮﺑﻮن اﻟﻤﻌﺰول ﺳﻨﻮﻳﺎ ﻣﻦ إﺟﻤﺎﻟﻲ ﻧﺒﺎﺗﺎت اﻷﻳﻜﺎت اﻟﺴﺎﺣﻠﻴﺔ اﻟﺘﻲ ﻏﻄﺎﻫﺎ اﻟﺒﺤﺮ اﻷﺣﻤﺮ.

6

ACKNOWLEDGMENTS

First and foremost, Praise be to Allah

Although only my name appears on the cover of this dissertation, many great people have supported me to finish it, my thanks and appreciation to all of them for their help during this unforgettable experience.

I owe my deepest gratitude to my supervisor Prof. Xabier Irigoien, I have been fortunate to have an advisor who was so generous with his time and immense knowledge, not only that he assisted me in each step to complete this dissertation but also guided me when my steps faltered. I can’t thank him enough for encouraging me throughout this experience.

I am also truly grateful to my co-supervisor Prof. Carlos Duarte. His guidance, insightful comments, and constructive criticisms have been a valuable input to my research. I am forever thankful to him for holding me to a higher research standard.

I thank Prof. Catherine Lovelock and Prof. Danielle Daffonchio for accepting to be in my examination committee.

I am extremely grateful to King Abdullah University of Science and Technology for the scholarship and financial support.

My sincere thanks to both Eng. Abdulaziz Aljowair from King Abdulaziz City for Science and

Technology and Dr. Oscar Serrano from the University of Western for sharing their expertise with me. 7

I thank KAUST central workshops, Coastal and Marine Resources Core Labs, in particular,

Dr. Nabeel Alikunhi, Dr. Zenon Batang and to all boat captains for the significant help I received from them during my field work. To the coast guards for letting me use their boat when ours did not work. The Analytical core lab for providing me the training to analyze my samples. KAUST greenhouse Dr. Muppala Reddy for always letting me borrow anything

I needed. And to the Presidency of Meteorology and Environment for providing weather data. To Mr. Abdullah Alwatied from the Saudi Wildlife Authority for sharing his knowledge. I also thank Dr. Joao Curdia for helping me in the Nursery work. And I warmly appreciate the help from Mrs. Christine Nelson, Ana Sofia Viegas, and Lily Chen regarding any administrative matters.

I thankfully acknowledge KAUST school grade 5 and 11 students and their teachers Mr.

Euan Riddell and Mrs. Caitlin McQuaid for helping me mangrove seedlings in the beach. I consider that my greatest achievements at KAUST and a memory I will cherish forever.

I am forever thankful to my friends Widyan Alamoudi and Alanoud Albugami for their

friendship, emotional support, and the extra hands whenever I needed in the lab.

Finally, my deep and sincere gratitude to my beloved parents, brothers, and sisters for the

encouragement and support throughout the whole period. This journey would not have

been possible without their love, patience, and faith. I dedicate this milestone to them.

8

Table of Contents

EXAMINATION COMMITTEE PAGE ...... 2 ABSTRACT ...... 4 ABSTRACT ...... 5 ACKNOWLEDGMENTS ...... 6 Table of Contents ...... 8 List of Figures ...... 11 List of Tables ...... 14 List of Supplementary Tables ...... 17 List of Supplementary Figures ...... 18 Introduction ...... 19 Mangroves ...... 19 1. Definition and history ...... 19 2. Species and distribution ...... 20 3. Morphological and physiological characteristics ...... 21 4. Importance and contribution to ecosystem services ...... 22 5. Losses and threats ...... 24 The Red Sea ...... 25 1. Location and formation ...... 25 2. Climate and Environmental characteristics ...... 26 3. Red Sea Mangroves ...... 27 References ...... 28 The Impact of this Research and the Objectives ...... 36 Chapter One ...... 38 Decadal Stability of Red Sea Mangroves ...... 38 Abstract ...... 39 Introduction ...... 40 Methods ...... 42 1. Satellite imagery ...... 42 2. Data processing ...... 42 3. Accuracy assessment ...... 44 4. Error estimates ...... 45 Results ...... 45 1. Ground-referencing data...... 45 2. Estimating mangrove cover ...... 46 3. Losses and gains over time ...... 51 Discussion ...... 55 Acknowledgements ...... 57 References ...... 58 Supplementary Materials ...... 62 9

Chapter Two ...... 72 Phenology and Growth dynamics of Avicennia marina in the Central Red Sea ...... 72 Abstract ...... 73 Introduction ...... 74 Methods ...... 75 1. Internodal measurements in Central Red Sea mangroves ...... 75 2. Estimating plastochrone interval ...... 81 3. Monthly phenological observations ...... 81 4. Statistical analysis ...... 81 Results ...... 82 Discussion ...... 88 Acknowledgements ...... 92 Author Contributions statement ...... 92 Competing financial interests ...... 92 References ...... 93 Supplementary Materials ...... 96 Chapter Three ...... 98 Nutrient Limitation of Central Red Sea Mangroves ...... 98 Abstract ...... 99 Introduction ...... 100 Methods ...... 103 1. Study area ...... 103 2. Nutrients status in Central Red Sea mangrove stands ...... 106 3. Seedling Fertilization Experiments...... 107 4. Measurements and chemical analysis...... 113 5. Statistical analysis...... 114 Results ...... 114 Discussion ...... 125 1. Nutrient concentration and stoichiometric ratios ...... 125 2. Nutrient requirements and critical concentrations ...... 126 3. Nutrient inputs to the Red Sea ...... 128 Conclusion...... 130 Acknowledgements ...... 131 Conflict of Interest Statement...... 131 Author Contributions statement ...... 131 References ...... 132 Supplementary Materials ...... 138 Chapter Four ...... 141 Nutrient reabsorption and flux of Avicennia marina in an ultra-oligotrophic environment ...... 141 Abstract ...... 142 Introduction ...... 143 Methods ...... 144 1. Sampling across mangrove stands ...... 144 2. Estimating density and flux ...... 146 3. Chemical analysis ...... 146 4. Statistical analysis ...... 147 Results ...... 148 10

1. Nutrient content, concentration, and reabsorption ...... 148 2. Leaf production and nutrient flux ...... 156 Discussion ...... 158 References ...... 160 Supplementary Materials ...... 164 Chapter Five ...... 166 Heavy metal dynamics in Red Sea Mangrove leaves...... 166 Abstract ...... 167 Introduction ...... 168 Methods ...... 170 1. Study location and leaves sampling ...... 170 2. Chemical analysis and quality control ...... 172 3. Statistical analysis ...... 172 Results ...... 173 Discussion ...... 180 Conclusions ...... 184 Acknowledgements ...... 185 Author Contributions statement ...... 185 References ...... 185 Supplementary Materials ...... 189 Chapter Six ...... 214 Carbon sink capacity of Red Sea mangroves ...... 214 Abstract ...... 215 Introduction ...... 216 Methods ...... 218 1. Study location ...... 218 2. Collecting and processing the samples ...... 220 3. Analyzing the samples ...... 220 4. Dating ...... 222 5. Statistical analysis ...... 224 Results ...... 224 Discussion ...... 234 References ...... 236 General Discussion ...... 239 References ...... 246

11

List of Figures

Chapter One Figure 1a: Mosaic images of mangrove distribution (green areas) in the Red Sea ...... 46 Figure 1b: Mangrove cover along the Red Sea shorelines...... 47 Figure 2: Estimate of mangrove cover (Km2) along the Asian and African shores of the Red Sea and the total cover assessed for the three different periods...... 48 Figure 3: Pareto Plot describing the size distribution for individual mangrove patches (Km2) in the Red Sea in 1972 (red), 2000 (green) and 2013 (blue). The solid colored lines show the fitted Pareto regression for each of the time periods (1972, log (%) N>x = -4.03 – 0.69 log Size, R2 =0.96; 2000, log (%) N>x = -4.07 – 0.70 log Size, R2 =0.96; 2013, log (%) N>x = -3.95 – 0.68 log Size, R2 =0.96 ), the dotted lines indicate the 10% and 1 % size percentiles and the insert shows the median, error bars are the ( ±C.L) patch size for each period...... 50 Figure 4: Overlapped images showing mangrove gains (green), losses (red), and unchanged areas (yellow) over the time intervals (A) between 1972 and 2000, and (B ) between 2000 and 2013...... 52 Figure 5: Examples of Red Sea mangrove vegetation over the study period. (A) dynamics of the largest mangrove stand in the Red Sea (Tabuk, N W. Saudi Arabia). (B) mangrove expansion assoicated to the rehaplitation project in (Centeral Red Sea coast of Arabia). (C) decline of mangrove in a costal lagon at Alith...... 53 Figure 6: Map for the drivers of loss and gains in mangrove communities. The symbols from left to right are: (camel: the losses, decay and overgrazing), (factory: represent the pollution), (house for the costal development) and (tree: for the rehablitation/afforistation projects)...... 54 Chapter Two Figure 1: Study sites in Central Red Sea. The sites are located in the kingdom of Saudi Arabia. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative Divisions layer provided by Esri Data and Maps, and DeLorme Publishing Company. Redistribution rights are granted http://www.esri.com/~/media/Files/Pdfs/legal/pdfs/redist_rights_103.pdf?la=en...... 76 Figure 2: An illustration of counting nodes and measuring internodal length. Photo by H.A and the artist work by I.Gromhico...... 78 Figure 3: Examples of standaraized internodal length for a selected in each study site in the Central Red Sea...... 80 Figure 4. Number of nodes produced by the main axis of Avicennia marina trees (nodes/year) in five stands sampled in the Central Red Sea...... 83 Figure 5: Number of sub-branches produced annually by Avicennia marina trees in two stands sampled in the Central Red Sea (Thuwal)...... 84 Figure 6: Elongation rate (cm y-1) of the main axis of Avicennia marina trees in five populations sampled in the Central Red Sea...... 85 Figure 7: Phenology patterns and mean percent reproductive branches in a Avicennia marina populations in the Central Red Sea (Thuwal)...... 86 Figure 8. Climate diagram for temperature, humidity and rainfall in the Central Red Sea. 87 12

Chapter Three Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company...... 105 Figure 2: Experimental setup for (A) the experiment testing for N, P and Fe limitation, alone and in combination conducted in 2014, and (B) the experiment testing for the role of iron vs. the chelant used in the first experiment (EDTA) conducted in 2015. The colors representing the treatments are used for the same treatment in all figures...... 112 Figure 3: Mean (± SE) for height, number of nodes and leaves of mangrove seedlings over time under control and different nutrient addition treatments across experiments testing for (A) N, P and Fe addition and (B) components of response to iron addition (Fe vs. EDTA)...... 117 Figure 4: Mean (± SE) for root development of mangrove seedlings over time under control and different nutrient addition treatments across experiments testing for (A) N, P and Fe addition and (B) components of response to iron addition (Fe vs. EDTA)...... 119 Figure 5: Mean (± SE) for leaf Chl a concentration over time of fully developed mangrove apical leaf under control and different nutrient addition treatments across experiments testing for (A) N, P and Fe addition and (B) components of response to iron addition (Fe vs. EDTA). Bars showing different letters within a sampling event identify significantly different treatments (P < 0.05), as indicated by post Tukey’s HSD multiple comparison test...... 120 Figure 6: The relationship between carbon and nutrient concentration (mmol nutrient g-1 DW) in fully developed apical leaves of Avicennia marina of seedlings. The solid line represents the reciprocal regression equation fitted across data derived from all treatments combined (mmol C = 34.071879 - 7.4611856*Recip mmol N), (mmol C = 36.763021 - 0.2475941*Recip mmol P) and (mmol C = 26.25834 + 0.0034299*Recip mmol Fe)...... 122 Figure 7: Mean (± SE) nutrients stoichiometric ratios over time of fully developed apical leaves receiving different nutrient additions (n=3 ). The dashed line represents the value for propagules...... 124 Chapter Four Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company...... 146 Figure 2: Accumulation rate of nutrients concentration (Difference in mg g DW-1 to the value of the first leaf) with the age of Avicennia marina leaves in the Central Red Sea. One tree each site. The fitted lines present nitrogen in blue, phosphorous in red, and green for iron...... 150 Figure 3: Accumulation rate of nutrients content (difference in mg leaf-1 to the value of the first leaf) with the age of Avicennia marina leaves in the Central Red Sea. One tree each site. The fitted lines present nitrogen in blue, phosphorous in red, and green for iron...... 153 Figure 4: Accumulation rate of N: P ratio vs. time in Avicennia marina leaves in the Central Red Sea. One tree per site. Based on elemental content. The line is a smoothed line...... 155 Figure 5: Regression analysis for the Avicenna marina girth and total number of meristems produced in (a) Khor Alkharar and (b) Thuwal Island...... 157 13

Chapter Five Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company...... 171 Figure 2: The increase or decrease of metal elements with the age of Avicennia marina leaves in the Central Red Sea. The slopes of the fitted linear regressions provide an estimate of metal accumulation rate (units mg metal gDW-1 day-1). The red line is the fitted linear regressions. The elements are ordered based on their goodness of the fit...... 177 Figure 3: The increase of metal elements with the age of Avicennia marina leaves in the Central Red Sea. The slopes of the fitted linear regressions provide an estimate of metal accumulation rate (units mg metal leaf-1 day-1). The red line is the fitted linear regressions. The elements are ordered alphabetically...... 178 Chapter Six Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company...... 219 Figure 2: Sediment grain size fractions, expressed as a percentage of the total sediment dry from four different locations in the Central Red Sea. Sediments were classified as; ...... 226 Figure 3: Concentration profiles of total and excess 210Pb in mangrove sediments in central Red Sea...... 228 Figure 4: Vertical profile of the % Corg and sediment and carbon density g cm-3 in mangrove sediments in central Red Sea...... 233 ... 234 General Discussion Figure 1:5: FlowVertical chart profile summarizing of δ13C and results δ15N obtained in mangrove from sedimentsCentral Red in Sea central mangroves. Red Sea...... 246

14

List of Tables

Chapter One Table 1: The error matrix for the presence and absence of mangroves in predicted images (Landsat data) over actual (Google maps)...... 44 Table 2: Estimate of mangrove cover growth rate and the % of increase per year, along the Asian and African shores of the Red Sea and the total cover assessed for the three different periods...... 51 Chapter Two Table 1: Mean ± SE (n) of the internodal length (cm y-1) of the main axis of Avicennia marina trees over five consecutive phonological years in five populations sampled in the Central Red Sea. The annual growth was calculated as the sum of internodal lengths produced during each annual cycle. The letters correspond to Tukey posthoc HSD multiple comparison testing for significant differences between years for each location, the same letters means no significant differences within one location (P > 0.05)...... 85 Table 2: Reported number of nodes produced annually along the main axis by different mangrove species around the world...... 89 Chapter Three Table 1: Mean (± SE) for nutrient concentrations (mmol g DW-1) in Avicennia marina leaves from four different locations in the Central Red Sea. R2 and F values correspond to an ANOVA that tested for significant differences between locations. * P between 0.01 and 0.05, ** P < 0.01. Nutrients linked with the same letter did not differ significantly among themselves (Tukey HSD multipile comparision post-hoc test, P > 0.05)...... 107 Table 2: Incoming solar radiation to natural growing mangrove and our experiment ...... 108 Table 3: Mean (± SE) growth rate (cm seedling-1 day-1), leaf and node production rate (number of leaves and nodes seedling-1 day-1), and root development (g DW seedling-1day- 1) of Red Sea mangrove seedlings under different nutrient addition treatments. The slopes of the fitted linear regressions are between seedling height and time. Treatments linked with the same letter did not differ significantly among themselves (Tukey HSD multipile comparision post-hoc test, P > 0.05)...... 118 Table 4: Atomic stoichiometric ratios across marine primary producers compared to Avicennia marina...... 126 Chapter Four Table 1: Mean (± SE) for nutrients concentration (mg g DW-1) in Avicennia marina leaves from four different locations in the Central Red Sea, and the results from ANOVA and Tukey HSD post hoc tests for differences in nutrient concentrations among locations. * = 0.05 > P > 0.01. ** = P < 0.01. Locations sharing the same superscript letters do not differ among themselves in nutrient concentration for a particular element...... 148 Table 2: Mean (± SE) for nutrients content (mg leaf-1) in Avicennia marina leaves from four different locations in the Central Red Sea, and the results from ANOVA and Tukey HSD posthoc tests for differences among in nutrient content among locations. * = 0.05 > P > 0.01. ** = P < 0.01. Locations sharing the same superscript letters do not differ among themselves in nutrient content for a particular element...... 149 15

Table 3: Intercept (± SE) and Slope (± SE) for nutrients concentration (mg g DW-1 to the initial value of the first leaf) in Avicennia marina leaves in the Central Red Sea. The slopes are per tree (one tree each site)...... 151 Table 4: Intercept (± SE) and Slope (± SE) for content (mg leaf-1 to the initial value of the first leaf) in Avicennia marina leaves in the Central Red Sea. The slopes are per tree (one tree each site)...... 154 Table 5: Mean (± SE) for nutrients resorption in Avicennia marina leaves from four different locations in the Central Red Sea and the results from ANOVA to tests for differences between locations. * = 0.05 > P > 0.01. ** = P < 0.01...... 155 Table 6: Mean of nutrients fluxes (mg element m -2 year -1) in Avicennia marina leaves from four different locations in the Central Red Sea ...... 158 Table 7: Comparison of nutrients resorption in Avicennia marina leaves from different locations worldwide...... 159 Chapter Five Table 1: Mean ± St. Error for the element concentrations (mg g DW-1). Results of ANOVA by location and sites for each elements showing F and significant differences at **P < 0.01; *P between 0.01 and 0.05. Letters indicate significant differences of Tukey test HSD post hock between locations for each element separately...... 174 Table 2: Mean ± St. Error of the element for content (mg leaf-1), Results of ANOVA by location and sites for each elements showing F and significant differences at **P < 0.01; *P between 0.01 and 0.05. Letters indicate significant differences of Tukey test HSD post hock between locations for each element separately...... 175 Table 3: Correlation for elements concentration mg g DW-1 and probability (** p<0.01, * P between 0.01 and 0.05, and no star P>0.05)...... 179 Table 4: Comparing different mangrove species and locations around the globe, the -1 and mg kg-1), species abbreviation: (A.i, Acanthus ilicifolius ;A.c, Aegiceras corniculatum ; A.a, Avicennia alba; A.o, Avicenniaselected results officinalis are about; A.m ,the Avicennia leaves only marina and the;B.c units, Bruguiera are in (μgcylindrical g ;B.g, Bruguiera gymnorhiza ;C.d, Ceriops decandra ;C.t, Ceriops tagal ;E.a, Excoecaria agallocha ;H.f, Heritiera fomes ;H.t, Hibiscus tiliaceus ;K.c, Kandelia candel ;L.r, Lumitzera racemose ;R.a, Rhizophora apiculate ;R.m, Rhizophora mucronata ;R.s, Rhizophora stylosa ;R, Rhizophora mangle ;S.c, Sonneratia caseolaris )...... 182 Chapter Six Table 1: Mean (± SE) for % sediment grain sizes and texture from four different locations in the Central Red Sea. R2 and F value correspond to an ANOVA testing for significant differences between locations. * P between 0.01 and 0.05, ** P < 0.01. locations with different letters indicate significant differences (Tukey HSD multiple comparison post-hoc test, P < 0.05)...... 225 Table 2: Mean (± SE) of organic carbon in 10 cm stock sediment from four different locations in the Central Red Sea, along with estimation of 210Pb sediment accretion rate for the last 100–150 years and carbon sequestration rate. R2 and F value correspond to an ANOVA testing for significant differences between locations. * P between 0.01 and 0.05, ** P < 0.01. locations with different letters indicate significant differences (Tukey HSD multiple comparison post-hoc test, P < 0.05)...... 229 Table 3: Mean (± SE) of organic carbon in 1m stock sediment from four different locations in the Central Red Sea, along with estimation of 14C sediment accretion rate for the last 16

1000 years. Both estimations resulted a carbon sequestration rate. R2 and F value correspond to an ANOVA testing for significant differences between locations. * P between 0.01 and 0.05, ** P < 0.01. locations with different letters indicate significant differences (Tucky HSD multiple comparison post-hoc test, P < 0.05)...... 230 Table 4: Mean (± SE) of sediment characteristics from four different locations in the Central Red Sea. R2 and F value correspond to an ANOVA testing for significant differences between locations. * P between 0.01 and 0.05, ** P < 0.01. locations with different letters indicate significant differences (Tukey HSD multiple comparison post-hoc test, P < 0.05)...... 231 Table 5: Mean (± SE) of marine plants and sediment in the Central Red Sea...... 232 Table 6: The sources of organic carbon in sediment obtained using marine plant and ...... 232 General Discussion Tablesediment 1: Percent δ15N and of mangrove δ13C values. and coral reef published papers in the RED Sea vs. global. . 240

17

List of Supplementary Tables Chapter One Table S 1: Satellite images data used in the study for 2013 ...... 62 Table S 2: Satellite images data used in the study for 2000 ...... 63 Table S 3: Satellite images data used in the study for 1972 ...... 64 Table S 4: Map showing 500 random points applied over the costliness of the Red Sea to test the presence and absence of mangrove communities...... 66 Table S 5: Satellite images data used to estimate error in the assessment of mangrove area ...... 67 Table S 6: Error assessment (as Coefficient of Variation, %, of the estimated mangrove area in each image, CV) for three to five locations assessed for each of the study periods. N = number of replicated image per location...... 68 Table S 7: The mean of size distribution along with the upper and lower confident limits . 69 Table S 8: Estimate of mangrove cover, showing the Mean patch size (the ratio between the total cover and the number of patches...... 69 Table S 9: Drivers of loss and gain ...... 70 Chapter Two Table S 1: Annual node production y-1 for each single interannual cycle ...... 96 Table S 2: Sub branching production for one single interannual cycle ...... 96 Table S 3: Internodal length y-1 for each single interannual cycle ...... 96 Chapter Three Table S 1: Data for the weather conditions during the experiment. Data were provided by the Presidency of Meteorology and Environment (PME)...... 138 Table S 2: Mean (± SE) nutrient concentration (mmol g DW-1) in fully developed apical leaves (n=3) of Avicennia marina seedlings grown under different experimental nutrient addition treatments. R2 and F value correspond to an ANOVA testing for significant differences between treatments over time. * P between 0.01 and 0.05, ** P < 0.01. Treatments linked with the same letter did not differ significantly among themselves (Tuckey HSD multiple comparison post-hoc test, P > 0.05)...... 139 Table S 3: Mean (± SE) for nutrients concentrations (mmol g DW-1) in the experimental fully developed apical leaves of Avicennia marina seedlings comparing Iron with EDTA (n=3). R2 and F value correspond to an ANOVA testing for significant differences between treatments. * P between 0.01 and 0.05, ** P < 0.01. Treatments linked with the same letter did not differ significantly among themselves (Tuckey HSD multiple comparison post-hoc test, P > 0.05)...... 140 Chapter Five Table S 1: Raw data for heavy metals content in Avicennia marina leaves ...... 189 Table S 2: Raw data for heavy metals concentration in Avicennia marina leaves ...... 193 Table S 3: The detailed slopes calculated per tree for heavy metals content ...... 196 Table S 4: The detailed slopes calculated per tree for heavy metals concentration ...... 204 Table S 5: The detailed slopes calculated per tree for heavy metals content ...... 212

18

List of Supplementary Figures

Chapter Four Figure S 1: Slope (± SE) for the regression analysis of leaves nutrients concentration (mg g DW-1 to initial value of the first leaf)...... 164 Figure S 2: Slope (± SE) for the regression analysis of leaves nutrients content (mg leaf-1 to initial value of the first leaf)...... 164

19

Introduction

Mangroves

1. Definition and history

Mangrove is a tree or shrub, generally growing half a meter above mean sea level (Duke

1993), found in coastal areas, lagoons, estuaries and deltas to form the main vegetation in

tidal and saline wetlands. They are present in a range of wet to dry subtropical and tropical

climates (Ball 1996, Chen and Twilley 1999). And grow well in loose mud or silt rich in

humus, humid weather with flow of freshwater bringing nutrients (Kathiresan and

Bingham 2001).

The origin of the word mangrove is unclear, probably coming from Portuguese mangue and English grove or manggi-manggi a Malaysian word meaning above the soil (MacNae

1968). MacNae also defined the term ‘mangrove’ as both the plant and its associated community, whereas the term ‘mangal’ is the forest community. Tomlinson (Tomlinson

1986) gave a clear, widely accepted classification for true mangroves or associates:

Growing in mangrove environment and not extending to the terrestrial communities, morphological and physiological adaptation to the habitat, taxonomically different from terrestrial relatives.

The first historical description of mangroves was in the Red Sea and the Arabian Gulf, in the fourth century BC by both Theophrastus, a student of Plato, and Aristotle in his book

Periphyton historia (Enquiry into Plants), and by the admiral Nearchus, the commander of the fleet of Alexander the Great where he described the mangrove of Tylos

“”(Saenger 2002). 20

2. Species and distribution

Depending on the author's mangroves comprise approximately 70 to 84 species belonging from 27 to 39 genera in 20 to 26 families (Saenger 2002, Alongi 2009).

Mangrove species are divided into two global hemispheres; the Atlantic East Pacific (AEP) with fewer species compared to the Indo-West Pacific (IWP) (Duke 1993). Most extensively are in 42%, 20%, North and Central America 15%, Oceania 12% and finally

South America 11% (Giri, Ochieng et al. 2011). In terms of countries; they are mostly concentrated in Indonesia, Australia, Brazil and Nigeria with a 43% of the world mangrove

forests (Alongi 2002).

Mangroves can be found between 30°N and 30°S, but the largest areas are between 5° N

and 5° S latitude (Giri, Ochieng et al. 2011). Because they decrease with increasing latitude

(Alongi 2002) at higher latitudes above 32° N and 40° S they are replaced by salt marshes

herbs (Stuart, Choat et al. 2007). Factors like low temperature (Stuart, Choat et al. 2007),

water logging and soil salinity (Koch and Snedaker 1997) and nutrient limitation (Koch and

Snedaker 1997, Lovelock, Feller et al. 2004) limit their growth and distribution. Also, their

biomass expansion is determined by rainfall, tides, waves and river flow (Alongi 2009).

Avicennia L. is found in tropical, subtropical and temperate regions, whereas other

groups such as Rhizophora L. are limited to tropical areas, making Avicennia the widest

distributed mangrove on the world (Duke 1991). Quisthoudt (Quisthoudt, Schmitz et al.

2012) inferred that the latitudinal range limits of the genus Avicennia and Rhizophora is not defined by temperature only as both reach their thermal limits towards higher latitudes since mangrove limits have warmer months in the northern than in the southern hemisphere. 21

3. Morphological and physiological characteristics

Mangroves possess morphological, physiological and dynamic characteristics making them unique: Tidal dispersal of propagules, fast turnover of foliage and very effective nutrient retaining mechanism (Alongi 2002). They are facultative halophytes (Krauss and

Ball 2013), with the capability of tolerating salt stress (Gab-Alla, Khafahi et al. 2010).

Particularly, these trees can grow in water ranging widely in salinity from marine to fresh water. No other trees can do it, and therefore when the environment becomes unsuitable for mangroves, the vegetation cannot be replaced.

Mangrove trees are tolerant to comparatively high salinity (Gab-Alla, Khafahi et al.

2010) because they have protective traits such as storing salt in the stem and roots bark, increasing leaf succulence, a waxy epidermis which decreases the transpiration rates, in addition to transferring the salts into senescent leaves (Parida and Jha 2010). Mangroves fundamentally eliminate salt through three mechanisms: salt excluders with an ultrafiltration mechanism of the root cortical cell membrane, salt secretors via glands and salt accumulators in leaf cells (Aziz and Khan 2001). Avicennia is used as an example of the last mechanism, but it has other mechanisms leading to high tolerance (Parida and Jha

2010). It also can tolerate high levels of interstitial salinity as shown by Dodd et al. (Dodd,

Blasco et al. 1999) in dispersed populations of Avicennia marina in the Arabian Gulf where mangroves survived salinities reaching (45‰–70‰).

Avicennia leaves have fine hairs secreted from the lower surface while their upper surface is shiny with some salt glands (Yasseen and Abu-Al-Basal 2008). The mechanism of reducing the accumulation of salt is by loading ions in the leaf hypoderm, then a dynamic elimination of salt through glands (Griffiths, Rotjan et al. 2008). A study by (Tan, Lin et al. 22

2013) demonstrated that salt crystals were observed in the leaves after treating the shoot with 500mM NaCl and the X-ray microanalysis confirmed that these crystals were mostly sodium and chloride. These salt glands help them maintain mineral balance and water status under extreme salinity (Esteban, Fernández-Marín et al. 2013). However, the glands are only located in the young leaves and as the leaf becomes older these glands decline

(Drennan and Pammenter 1982).

The ultrafiltration mechanism acts through a negative hydrostatic pressure by transpiration in plants to overcome the soil osmotic pressure (Parida and Jha 2010), While the accumulation mechanism acts by increasing ions in leaf tissue then reserve it to the vacuoles depending on Na+/H+ anti-porters system along with V-type H+- ATPase and H+-

PPase. The sodium sequestration in the vacuoles is essential for osmotic adjustment

because it reduces sodium concentration in the cytosol (Mimura, Kura-Hotta et al. 2003).

The response of antioxidant enzymes is another strategy in A. marina to overcome stressful

environmental conditions. In a salt stress experiment of leaves, it was detected that

superoxide dismutase (SOD) activity increases which decrease the harmful production of

reactive oxygen species (ROS) to protect mangroves (Jithesh, Prashanth et al. 2006). All

these properties of A. marina make them a good model for salt stress research because they

preserve active leaves under harsh conditions (Cheeseman, Herendeen et al. 1997).

4. Importance and contribution to ecosystem services

Ecosystem services are profits gained from an ecosystem that contributes to human

wellbeing directly or indirectly (Hsieh, Lin et al. 2015). The Millennium Ecosystem

Assessment (MEA) reported a synthesis about the ecosystems and human well-being and

categorized the ecosystem services into four major groups 1) provisioning, e.g., food and 23 water, 2) regulating, e.g., climate change and water purification, 3) supporting, e.g., primary production and soil formation, 4) cultural, e.g., recreation and education (Assessment

2005).

Mangroves forests rank amongst the most productive ecosystems in the world (Coulter,

Duarte et al. 2001). As the highest carbon rich forest in the tropics accounting for 49–98% of carbon storage in estuarine and oceanic mangrove sites (Donato, Kauffman et al. 2011).

Also, preventing their loss has the potential of decreasing the global emissions for a low

cost of about $4 to $10 ton 1 CO2 which is within the current range of prices on the − European carbon trading system (Siikamäki, Sanchirico et al. 2012). And although they are

low in number mangrove forests play a key role in coastal ecosystems providing at least

US$ 1.6 billion y-1 in support of the coastal livelihoods (Polidoro, Carpenter et al. 2010),

mainly because of their enormous root systems along with detritus and nutrient abundance

provide a shelter, feeding and nursery area for vertebrates and invertebrates (Rönnbäck

1999, Nagelkerken, Blaber et al. 2008). Their root stabilizes the sediment and helps to

protect the coast from erosion and prevent the offshore seagrass beds and coral reef from

being washed out (Ewel, Twilley et al. 1998) Mangroves and salt marshes help protecting

seagrass meadows from the nitrogen supplies increases due to urbanization (Valiela and

Cole 2002).

The annual estimations of fisheries captured in mangroves ranged from US$750 to

16750 per hectare (Rönnbäck 1999). In the USA the value of the commercial annual fish

harvest derived from mangroves has been estimated at US$6,200 per Km2 while in

Indonesia it was estimated at US$60,000 per Km2 (Miththapala 2008). In India, rich

mangrove areas supported a catch of 11 kg shellfish ha 1 day 1 and 4.5 kg finfish ha 1 day 1, − − − − 24 compared with the mangrove-poor areas (Kathiresan and Rajendran 2002). Mangroves

also act as a nursery for coral reef fish where adults live as far as 30 km from the

mangroves (McMahon, Berumen et al. 2012).

Mangroves are resistant to some of the climate change components such as the rise in atmospheric Co2 and temperature along with the sea temperature (Macintosh,

Mahindapala et al. 2012) and they offer protection of coastal areas against tides (Mazda,

Magi et al. 1997), currents, storms and sea level rise (Khan and Kumar 2009). Studies

suggest that wave height will be reduced by 50 to 99% across 500 m width of mangrove

forest (McIvor, Möller et al. 2012).

They can temporarily reduce the amount of pollution from soil and water by

accumulating heavy metals in roots, shoot, and leaves, although the dynamics of heavy

metals in leaves before shedding remain unknown. They also play a key role as a buffer for

the marine ecosystems system by trapping sediments, nutrients, and contaminants of

terrestrial origin (Peters, Gassman et al. 1997). They provide shade, increase humidity and

reduce soil water evaporation (Macintosh, Mahindapala et al. 2012) as well as water

purification services through their roots (Kim, Seo et al. 2016) and medicinal uses

(Bandaranayake 1998).

5. Losses and threats

Mangroves have been historically overexploited. The lack of knowledge about the role

of mangroves led to the idea of converting the mangrove forests to productive lands for

agriculture and aquaculture (Mastaller 1997). Globally FAO estimates that 3.2 million ha of

mangrove forests have been lost from 1980 to 2010 (FAO 2007, FAO 2010). This data are

confirmed by some independent and local assessments that estimate losses from 35 to 86 25

% in the last two decades (Valiela, Bowen et al. 2001, Duke, Meynecke et al. 2007, Giri,

Ochieng et al. 2011). With an annual loss rate of about 0.7% to 2% (Murray, Pendleton et al.

2011), this loss is fast in developing countries where more than 90% of the world’s mangroves are found (Duke, Meynecke et al. 2007). Seventy-five percent of the world mangroves are in 15 countries where only 6.9% of the mangrove areas are protected under the existing protected areas network (IUCN I-IV) (Giri, Ochieng et al. 2011). And > 40% of the evaluated mangrove-endemic vertebrates are in danger of extinction (Luther and

Greenberg 2009).

The reduction of these communities are mainly due to local human impacts like coastal construction, contamination, loss of water quality and shrimp farms development (Tawfiq and Olsen 1993, Macintosh and Ashton 2002, PERSGA/GEF 2004) and in some areas for.e.g the Red Sea camels grazing on leaves (Parvaresh, Abedi et al. 2011). But also to global phenomena such as earthquakes, tsunamis, coastal erosion (Kumar 2009) and climate

change components such as sea-level rise (Gilman, Ellison et al. 2008).

The Red Sea

1. Location and formation

The Red Sea is approximately 2000 Km long from lat.12.5° N to 30° N (Bailey 2010).

Surrounded by seven countries; Saudi Arabia, , , Israel, Jordan, Djibouti, and

Eritrea, the north part is divided between the Gulf of Suez west and the Gulf of Aqaba east, while in the south part there is a narrow and relatively shallow connection to the Indian

Ocean through Bab Al-Mandab (Bailey 2010). The Red Sea is bounded by mountains

throughout much of its length, which are separated from the coast by a narrow coastal

plain. Mountains are tallest in southern Saudi Arabia, reaching a maximum of 3700 m 26

(Bruckner 2011). The coast is formed of shallow banks and coral reefs, however only after few kilometers; the depth extends to hundreds of meters reaching the deepest > 2000m with a steep-walled axial thought only 5-30 km wide (Braithwaite 1987).

2. Climate and Environmental characteristics

The climate of the region is arid or semi-arid (Bailey 2010), with the desert surrounding

both coasts in the Red Sea, and no permanent rivers in the basin. Temperatures at shores

have a minimum of 8ºC in the north with a diurnal change of 20ºC, to 14ºC in the Central

region whereas the south is fully tropical (PERSGA 2004). The Red Sea is a spatially and

temporally heterogeneous with salinity ranging from 35 to 41 ppt, temperature from 21 to

34 °C and chl a from 0.5 to4.0 mg m-3 (Sofianos and Johns 2003, Raitsos, Pradhan et al.

2013). However, the salinity may rise to hypersaline 80–180 ppt and even hypersaline 80–

300 ppt in coastal lagoons and intertidal pools The high rates of the evaporation makes the

Red Sea one of the highest saline seas of the world, nevertheless, the connection of Bab al

Mandab Straits to the Indian Ocean minimize the salinity to about 35‰ in the South, whereas the north salinity increases to approximately 41‰ (Bailey 2010). The Red Sea is

an oligotrophic sea (Kimor, Gordon et al. 1992) and has arguably one of the longest coral

reef systems of the world (Berumen, Hoey et al. 2013). This reef may be a source of

chlorophyll a in the water by nutrients or chlorophyll-rich detritus and sediment (Acker,

Leptoukh et al. 2008).

After the Arabian Gulf, the Red Sea presents the harshest environmental conditions in

the tropics, due to the high temperature and salinity. Also, it is considered that the Red Sea

and the Gulf of Aden marine environments are generally in healthy condition (Gladstone, 27

Tawfiq et al. 1999). The tide ranges from 0.9m in the south to 0.6m in the north whereas in

the Central area is between 20 to 30 cm (Sultan, Ahmad et al. 1996).

3. Red Sea Mangroves

In 1775, Avicennia marina was first described in the Red Sea by Pher Forsskal, the

student of Linnaeus in his book Aegyptiaco-arabica, who named it Sceura marina, later changed to Avicennia marina because his mentor had previously described Avicennia officinalis in India in 1753. Avicennia was named after the famous scientist Avicennia or Ibn

Sina, the author of a medical book used in Europe for several centuries (Saenger 2002).

Avicennia marina is common around the Indian Ocean, with the southern limits of the distribution in and New Zealand. The northern limit of the distribution is probably in the Red Sea, where A. marina is present along both East and West coasts. Along the Saudi coast mangroves may be divided into two sections taking ALlith as the border:

“the north area with spare distributions Sharm Zubeir, the shoreline between Al-Wajh and

Umm-Lajh, Al-Wajh Bank, near Yanbu, between and Mastura, area south of

Jeddah, and Qishran Bay north of Al-Lith) and the southern area with dense distributions

(Khor Amiq, Shuqaiq and Jizan mangroves and some offshore islands”(PERSGA 2004).

Mangroves are found in the form of fragmented stands in tidal zones of Saudi Arabia on the Red Sea and Arabian Gulf coastlines. They consist mainly of Avicennia marina (Forssk.)

Vierh, although in the Red Sea there are also a few stands of Rhizophora mucronata Lam.

(El-Juhany 2009). They occupy isolated areas which contribute to their preservation

(Mandura 1997). The scientific literature about Red Sea mangroves is modern and more limited than in other regions (Zahran, Younes et al. 1983, Mohamed 1984, Price, Medley et al. 1987, Saifullah 1996, Price, Jobbins et al. 1998, Rouphael, Turak et al. 1998, Gladstone, 28

Tawfiq et al. 1999, PERSGA/GEF 2004, Ahmed and Khedr 2007, El-Juhany 2009, Ali A. Gab-

Alla; Ishrak 2010, Price, Ghazi et al. 2014, Sabeel, Ingels et al. 2014). In the Red Sea, their

distribution has bee studied through satellite imagery (El-Juhany 2009, Kumar, Khan et al.

2010, Kumar, Khan et al. 2010, Kumar, Khan et al. 2011). However, to our knowledge, there

are not detailed studies in their distribution, neither about the changes in the last decades.

Further, mangroves growth in the Red Sea has been described as limited, compared to

other tropical mangroves, due to extreme salinity with low rainfall and low nutrient supply

(Mandura 1997). A study conducted by Saifullah (Saifullah 1996) revealed that the

conditions in the southern part of the Red Sea are fairly more favorable because of higher

nutrient concentrations. But as for the distribution, to our knowledge growth rates and

population dynamics of mangroves in the Red Sea have not been quantified.

References

Acker, J., G. Leptoukh, S. Shen, T. Zhu and S. Kempler (2008). "Remotely-sensed chlorophyll a observations of the northern Red Sea indicate seasonal variability and influence of coastal reefs." Journal of marine systems 69(3): 191-204.

Ahmed, E.-K. A. and A. Khedr (2007). "Zonation Pattern of Avicennia marina and Rhizophora mucronata along the Red Sea Coast, Egypt." World Applied Sciences Journal 2(4): 283-288.

Ali A. Gab-Alla; Ishrak, K. K. W., M. Morsy 3 and Fouda, Moustafa M (2010). "Ecology of Avicennia marina mangals along Gulf of Aqaba, South Sinai, Red Sea " Egypt J. Aquat. Biol. & Fish 14.

Alongi, D. (2009). The energetics of mangrove forests, Springer Science & Business Media.

Alongi, D. M. (2002). "Present state and future of the world's mangrove forests." Environmental conservation 29(03): 331-349.

Assessment, M. E. (2005). Ecosystems and human well-being: Synthesis, Island press Washington, DC:. 29

Aziz, I. and M. A. Khan (2001). "Experimental assessment of salinity tolerance of Ceriops tagal seedlings and saplings from the Indus delta, Pakistan." Aquatic Botany 70(3): 259- 268.

Bailey, G. (2010). The Red Sea, coastal landscapes, and hominin dispersals. The evolution of human populations in Arabia, Springer: 15-37.

Ball, M. C. (1996). Comparative ecophysiology of mangrove forest and tropical lowland moist rainforest. Tropical forest plant ecophysiology, Springer: 461-496.

Bandaranayake, W. (1998). "Traditional and medicinal uses of mangroves." Mangroves and salt marshes 2(3): 133-148.

Berumen, M. L., A. Hoey, W. Bass, J. Bouwmeester, D. Catania, J. E. Cochran, M. T. Khalil, S. Miyake, M. R. Mughal and J. Spaet (2013). "The status of coral reef ecology research in the Red Sea." Coral Reefs 32(3): 737-748.

Braithwaite, C. J. (1987). "Geology and palaeogeography of the Red Sea region." Red Sea: 22-24.

Bruckner, A. (2011). Khaled bin Sultan Living Oceans Foundation habitat mapping and characterization of coral reefs of the Saudi Arabian Red Sea: 2006–2009.Final Report Part I Panoramic Press Phoenix.

Cheeseman, J. M., L. B. Herendeen, A. T. Cheeseman and B. F. Clough (1997). "Photosynthesis and photoprotection in mangroves under field conditions." Plant, Cell & Environment 20(5): 579-588.

Chen, R. and R. R. Twilley (1999). "Patterns of mangrove forest structure and soil nutrient dynamics along the Shark River Estuary, Florida." Estuaries 22(4): 955-970.

Coulter, S. C., C. M. Duarte, M. S. Tuan, N. H. Tri, H. T. Ha, L. H. Giang and P. N. Hong (2001). "Retrospective estimates of net leaf production in Kandelia candel mangrove forests." Marine Ecology Progress Series 221: 117-124.

Dodd, R. S., F. Blasco, Z. A. Rafii and E. Torquebiau (1999). "Mangroves of the : ecotypic diversity in cuticular waxes at the bioclimatic extreme." Aquatic Botany 63(3-4): 291-304.

Donato, D. C., J. B. Kauffman, D. Murdiyarso, S. Kurnianto, M. Stidham and M. Kanninen (2011). "Mangroves among the most carbon-rich forests in the tropics." Nature Geoscience 4(5): 293-297.

Drennan, P. and N. W. Pammenter (1982). "Physology of salt excretion in the mangrove Avicennia marina (Forsk).Vierh." New Phytologist 91(4): 597-606. 30

Duke, N. (1991). "A systematic revision of the mangrove genus Avicennia (Avicenniaceae) in Australasia*." Australian Systematic Botany 4(2): 299-324.

Duke, N. C. (1993). "Mangrove floristics and biogeography." Tropical mangrove ecosystems: 63-100.

Duke, N. C., J.-O. Meynecke, S. Dittmann, A. M. Ellison, K. Anger, U. Berger, S. Cannicci, K. Diele, K. C. Ewel and C. D. Field (2007). "A world without mangroves?" Science 317(5834): 41-42.

El-Juhany, L. (2009). "Present status and degradation trends of mangrove forests on the southern Red Sea coast of Saudi Arabia." American-Eurasian Journal of Agricultural and Environmental Science 6(3): 328-340.

Esteban, R., B. Fernández-Marín, A. Hernandez, E. Jiménez, A. León, S. García-Mauriño, C. Silva, J. Dolmus, C. Dolmus and M. Molina (2013). "Salt crystal deposition as a reversible mechanism to enhance photoprotection in black mangrove." Trees 27(1): 229-237.

Ewel, K., R. Twilley and J. Ong (1998). "Different kinds of mangrove forests provide different goods and services." Global Ecology & Biogeography Letters 7(1): 83-94.

FAO (2007). The world’s Mangroves 1980-2005, FAO, Food and Agriculture Organization of the Rome.

FAO (2010). Global forest resources assessment 2010, Food and Agriculture Organization of the United Nations Roma.

Gab-Alla, A., I. Khafahi, W. Morsy and M. Fouda (2010). "Ecology of Avicennia marina mangals along Gulf of Aqaba, South Sinai, Red Sea " Egypt J. Aquat. Biol. & Fish 14(2): 79- 93.

Gilman, E. L., J. Ellison, N. C. Duke and C. Field (2008). "Threats to mangroves from climate change and adaptation options: a review." Aquatic botany 89(2): 237-250.

Giri, C., E. Ochieng, L. L. Tieszen, Z. Zhu, A. Singh, T. Loveland, J. Masek and N. Duke (2011). "Status and distribution of mangrove forests of the world using earth observation satellite data." Global Ecology and Biogeography 20(1): 154-159.

Gladstone, W., N. Tawfiq, D. Nasr, I. Andersen, C. Cheung, H. Drammeh, F. Krupp and S. Lintner (1999). "Sustainable use of renewable resources and conservation in the Red Sea and Gulf of Aden: issues, needs and strategic actions." Ocean & coastal management 42(8): 671-697.

Griffiths, M. E., R. D. Rotjan and G. S. Ellmore (2008). "Differential salt deposition and excretion on leaves of Avicennia germinans mangroves." Caribb J Sci 44: 267-271. 31

Hsieh, H.-L., H.-J. Lin, S.-S. Shih and C.-P. Chen (2015). "Ecosystem functions connecting contributions from ecosystem services to human wellbeing in a mangrove system in Northern Taiwan." International journal of environmental research and public health 12(6): 6542-6560.

Jithesh, M., S. Prashanth, K. Sivaprakash and A. K. Parida (2006). "Antioxidative response mechanisms in halophytes: their role in stress defence." Journal of Genetics 85(3): 237- 254.

Kathiresan, K. and B. L. Bingham (2001). "Biology of mangroves and mangrove ecosystems." Advances in marine biology 40: 81-251.

Kathiresan, K. and N. Rajendran (2002). "Fishery resources and economic gain in three mangrove areas on the south‐east coast of India." Fisheries Management and Ecology 9(5): 277-283.

Khan, M. A. and A. Kumar (2009). "Impact of “urban development” on mangrove forests along the west coast of the Arabian Gulf." e-journal Earth Sci India 2: 159-173.

Kim, K., E. Seo, S.-K. Chang, T. J. Park and S. J. Lee (2016). "Novel water filtration of saline water in the outermost layer of mangrove roots." Scientific reports 6.

Kimor, B., N. Gordon and A. Neori (1992). "Symbiotic associations among the microplankton in oligotrophic marine environments, with special reference to the Gulf of Aqaba, Red Sea." Journal of Plankton research 14(9): 1217-1231.

Koch, M. S. and S. C. Snedaker (1997). "Factors influencing Rhizophora mangle L. seedling development in Everglades carbonate soils." Aquatic Botany 59(1): 87-98.

Krauss, K. W. and M. C. Ball (2013). "On the halophytic nature of mangroves." Trees 27(1): 7-11.

Kumar, A. (2009). "Reclaimed islands and new offshore townships in the Arabian Gulf: potential natural hazards." Current science 96(4): 480.

Kumar, A., M. Khan and A. Muqtadir (2011). "Distribution of mangroves along the Red Sea coast of the : part 3: Coast of Yemen." Earth Sci India 4(II): 29-38.

Kumar, A., M. A. Khan and A. Muqtadir (2010). "Distribution of mangroves along the Red Sea coast of the Arabian Peninsula: Part–1: the northern coast of western Saudi Arabia." Earth Sc. Ind 3: 28-42.

Kumar, A., M. A. Khan and A. Muqtadir (2010). "Distribution of mangroves along the Red Sea coast of the Arabian Peninsula: Part–2: The Southern Coast of Western Saudi Arabia." Earth Sc. Ind 3: 154-162. 32

Lovelock, C., I. C. Feller, K. McKee, B. Engelbrecht and M. Ball (2004). "The effect of nutrient enrichment on growth, photosynthesis and hydraulic conductance of dwarf mangroves in Panama." Functional Ecology 18(1): 25-33.

Luther, D. A. and R. Greenberg (2009). "Mangroves: a global perspective on the evolution and conservation of their terrestrial vertebrates." BioScience 59(7): 602-612.

Macintosh, D., R. Mahindapala and M. Markopoulos (2012). "Sharing Lessons on Mangrove Restoration." Bangkok, Thailand: Mangroves for the Future and Gland, Switzerland: IUCN.

Macintosh, D. J. and E. C. Ashton (2002). "A Review of Mangrove Biodiversity Conservation and Management." Centre for Tropical Ecosystems Research, Denmark: University of Aarhus.

MacNae, W. (1968). "A general account of the fauna and flora of mangrove swamps and forests in the Indo-West-Pacific region." Advances in marine biology 6: 73-270+ 271 diag.

Mandura, A. (1997). "A mangrove stand under sewage pollution stress: Red Sea." Mangroves and Salt marshes 1(4): 255-262.

Mastaller, M. (1997). Mangroves: the forgotten forest between land and sea, Tropical Press.

Mazda, Y., M. Magi, M. Kogo and P. N. Hong (1997). "Mangroves as a coastal protection from waves in the Tong King delta, Vietnam." Mangroves and Salt marshes 1(2): 127-135.

McIvor, A., I. Möller, T. Spencer and M. Spalding (2012). "Reduction of wind and swell waves by mangroves." Natural Coastal Protection Series: Report 1: 27.

McMahon, K. W., M. L. Berumen and S. R. Thorrold (2012). "Linking habitat mosaics and connectivity in a coral reef seascape." Proceedings of the National Academy of Sciences 109(38): 15372-15376.

Mimura, T., M. Kura-Hotta, T. Tsujimura, M. Ohnishi, M. Miura, Y. Okazaki, M. Mimura, M. Maeshima and S. Washitani-Nemoto (2003). "Rapid increase of vacuolar volume in response to salt stress." Planta 216(3): 397-402.

Miththapala, S. (2008). Mangroves, Ecosystems and Livelihoods Group Asia, IUCN.

Mohamed, B. F. (1984). "Ecological observations on mangroves of the Red Sea shores of the ." Hydrobiologia 110(1): 109-111.

Murray, B. C., L. Pendleton, W. A. Jenkins and S. Sifleet (2011). "Green payments for blue carbon: Economic incentives for protecting threatened coastal habitats." Nicholas Institute for Environmental Policy Solutions, Report NI 11: 04.

Nagelkerken, I., S. Blaber, S. Bouillon, P. Green, M. Haywood, L. Kirton, J.-O. Meynecke, J. Pawlik, H. Penrose and A. Sasekumar (2008). "The habitat function of mangroves for terrestrial and marine fauna: a review." Aquatic Botany 89(2): 155-185. 33

Parida, A. K. and B. Jha (2010). "Salt tolerance mechanisms in mangroves: a review." Trees 24(2): 199-217.

Parvaresh, H., Z. Abedi, P. Farshchi, M. Karami, N. Khorasani and A. Karbassi (2011). "Bioavailability and concentration of heavy metals in the sediments and leaves of grey mangrove, Avicennia marina (Forsk.) Vierh, in Sirik Azini creek, Iran." Biological trace element research 143(2): 1121-1130.

PERSGA (2004). "The Regional Organization for the Conservation of the Environment of the Red Sea and Gulf of Aden (PERSGA) Status of Mangroves in the Red." PERSGA Technical Series, (11).

PERSGA (2004). "Standard survey methods for key habitats and key species in the Red Sea and Gulf of Aden." PERSGA Technical Series, Jeddah(10): 302

PERSGA/GEF (2004). "Status of Mangroves in the Red Sea and Gulf of Aden." PERSGA Technical Series, Jeddah(11).

Peters, E. C., N. J. Gassman, J. C. Firman, R. H. Richmond and E. A. Power (1997). "Ecotoxicology of tropical marine ecosystems." Environmental Toxicology and Chemistry 16(1): 12-40.

Polidoro, B. A., K. E. Carpenter, L. Collins, N. C. Duke, A. M. Ellison, J. C. Ellison, E. J. Farnsworth, E. S. Fernando, K. Kathiresan, N. E. Koedam, S. R. Livingstone, T. Miyagi, G. E. Moore, V. Ngoc Nam, J. E. Ong, J. H. Primavera, S. G. Salmo, J. C. Sanciangco, S. Sukardjo, Y. Wang and J. W. Yong (2010). "The loss of species: mangrove extinction risk and geographic areas of global concern." PLoS One 5(4): e10095.

Price, A., S. Ghazi, P. Tkaczynski, A. J. Venkatachalam, A. Santillan, T. Pancho, R. Metcalfe and J. Saunders (2014). "Shifting environmental baselines in the Red Sea." Marine pollution bulletin 78(1): 96-101.

Price, A., P. Medley, R. McDowall, A. Dawson-Shepherd, P. Hogarth and R. Ormond (1987). "Aspects of mangal ecology along the Red Sea coast of Saudi Arabia." Journal of natural history 21(2): 449-464.

Price, A. R., G. Jobbins, A. R. D. Shepherd and R. F. Ormond (1998). "An integrated environmental assessment of the Red Sea coast of Saudi Arabia." Environmental Conservation 25(01): 65-76.

Quisthoudt, K., N. Schmitz, C. F. Randin, F. Dahdouh-Guebas, E. M. Robert and N. Koedam (2012). "Temperature variation among mangrove latitudinal range limits worldwide." Trees 26(6): 1919-1931.

Raitsos, D. E., Y. Pradhan, R. J. Brewin, G. Stenchikov and I. Hoteit (2013). "Remote sensing the phytoplankton seasonal succession of the Red Sea." PLoS One 8(6): e64909. 34

Rönnbäck, P. (1999). "The ecological basis for economic value of seafood production supported by mangrove ecosystems." Ecological Economics 29(2): 235-252.

Rouphael, T., E. Turak and J. Brodie (1998). "Seagrasses and Mangroves of Yemen’s Red Sea." Protection of Marine Ecosystems of the Red Sea Coast of Yemen. UN Publication: 41- 49.

Sabeel, R. A. O., J. Ingels, E. Pape and A. Vanreusel (2014). "Macrofauna along the Sudanese Red Sea coast: potential effect of mangrove clearance on community and trophic structure." Marine Ecology.

Saenger, P. (2002). Mangrove ecology, silviculture and conservation, Springer.

Saifullah, S. M. (1996). " Mangrove Ecosystem of Saudi Arabian Red Sea Coast -An Overview " Journal of King Abdul Aziz University. 7(1): 263 -270

Siikamäki, J., J. N. Sanchirico and S. L. Jardine (2012). "Global economic potential for reducing carbon dioxide emissions from mangrove loss." Proceedings of the National Academy of Sciences 109(36): 14369-14374.

Sofianos, S. S. and W. E. Johns (2003). "An oceanic general circulation model (OGCM) investigation of the Red Sea circulation: 2. Three‐dimensional circulation in the Red Sea." Journal of Geophysical Research: Oceans 108(C3).

Stuart, S., B. Choat, K. Martin, N. Holbrook and M. Ball (2007). "The role of freezing in setting the latitudinal limits of mangrove forests." New Phytologist 173(3): 576-583.

Sultan, S., F. Ahmad and N. Elghribi (1996). "Sea level variability in the central Red Sea." Oceanologica acta 19(5).

Tan, W. K., Q. Lin, T. M. Lim, P. Kumar and C. S. Loh (2013). "Dynamic secretion changes in the salt glands of the mangrove tree species Avicennia officinalis in response to a changing saline environment." Plant, cell & environment.

Tawfiq, N. and D. A. Olsen (1993). "Saudi Arabia's response to the 1991 Gulf oil spill." Marine Pollution Bulletin 27: 333-345.

Tomlinson, P. (1986). The botany of mangroves. Cambridge tropical biology series, Cambridge University Press, Cambridge.

Valiela, I., J. L. Bowen and J. K. York (2001). "Mangrove forests: One of the world's threatened major tropical environments." Bioscience 51(10): 807-815.

Valiela, I. and M. L. Cole (2002). "Comparative evidence that salt marshes and mangroves may protect seagrass meadows from land-derived nitrogen loads." Ecosystems 5(1): 92- 102. 35

Yasseen, B. and M. Abu-Al-Basal (2008). "Ecophysiology of Limonium axillare and Avicennia marina from the coastline of Arabian Gulf-." Journal of Coastal Conservation 12(1): 35-42.

Zahran, M., H. Younes and H. (1983). "On the ecology of mangal vegetation of the Saudi Arabian Red Sea coast." J. Univ. Kuwait (Sci.) 10(1): 87-99. 36

The Impact of this Research and the Objectives

Globally, mangrove forests have severely declined mainly due to human impacts

through industrial pollution, organic inputs, filling, littering, and reclamation. In addition to

the impacts of natural disasters and climate change. These impacts, which are evident

today, will remain for a long period unless our understanding of mangrove ecology

advances sufficiently as to allow the formulation of effective rehabilitation projects. The

harsh conditions of the Red Sea offer a unique opportunity to understand the mangroves

ecology and to study their resilience under extreme conditions.

Out of the four major services provided by mangroves (coastal protection, nursery,

phytoremediation, and carbon sequestration) we have identified two as a focus for this

Ph.D. project: phytoremediation and carbon sequestration. The Red Sea is a narrow (200

km on average) semi-enclosed sea with a limited tidal range (about 20 cm). Mangroves do

not cover the whole coast and their role in protecting the coastline is limited. However, sea level rise is an issue to be considered in the future.

On the other hand industrial and urban development in the Saudi Coast of the Red

Sea (but also in the Gulf) is increasing the level of contaminants in coastal sediments. There is, therefore, an increasing interest on the potential capacities of mangroves for phytoremediation. Further, the Kingdom of Saudi Arabia is one of the main producers of oil in the world with also one of the highest per capita consumption. CO2 capture by vegetation

in Saudi Arabia is extremely limited due to the arid conditions and mangrove forests are

the only coastal forests in the Kingdom. However, mangroves are one of the ecosystems

with the highest capacity to absorb and bury CO2 in the sediment, and it is, therefore, 37 important for the country to evaluate the CO2 absorption capacity of this delicate ecosystem that is at risk.

This research focusses on the study of the species Avicennia marina in the Red

Sea.The objective is to fill the gaps in the knowledge of its growth dynamics and ecology to

evaluate better some of the ecological services they provide (phytoremediation and carbon sequestration) and to serve as a basis to improve rehabilitation projects.

38

Chapter One

Decadal Stability of Red Sea Mangroves

Hanan Almahasheer1,2,, Abdulaziz Aljowair 3, Carlos M. Duarte1 and Xabier Irigoien1

1 King Abdullah University of Science and Technology (KAUST), Red Sea Research Center, Thuwal 23955-6900, Kingdom of Saudi Arabia

2 Biology Department, University of (UOD), Dammam 31441-1982, Kingdom of Saudi Arabia

3 Space Research Institute, King Abdul Aziz City for Science and Technology (KACST), 11442-6086, Kingdom of Saudi Arabia

This is the final, accepted copy of the manuscript that was published in Estuarine, Coastal and Shelf Science, 2016. And the data set of this manuscript that was published in PANGAEA-Data Publisher for Earth & Environmental Science, 2015.

Citation: Almahasheer, H., A. Aljowair, C. M. Duarte and X. Irigoien (2016). "Decadal stability of Red Sea mangroves." Estuarine, Coastal and Shelf Science 169: 164-172.

Almahasheer, H., A. Aljowair, C. M. Duarte and X. Irigoien (2015). Mangrove cover in the Red Sea (1972-2013) [data set], PANGAEA-Data Publisher for Earth & Environmental Science.

39

Abstract

Across the Earth, mangroves play an important role in coastal protection, both as nurseries and carbon sinks. However, due to various human and environmental impacts, the coverage of mangroves is declining on a global scale. The Red Sea is in the northern-most

area of the distribution range of mangroves. Little is known about the surface covered by

mangroves at this northern limit or about the changes experienced by Red Sea mangroves.

We sought to study changes in the coverage of Red Sea mangroves by using multi-temporal

Landsat data (1972, 2000 and 2013). Interestingly, our results show that there has been no

decline in mangrove stands in the Red Sea but rather a slight increase. The area covered by

mangroves is about 69 Km2 along the African shore and 51 Km2 along the Arabian

Peninsula shore. From 1972 to 2013, the area covered by mangroves increased by about

0.29% y-1. We conclude that the trend exhibited by Red Sea mangroves departs from the

general global decline of mangroves. Along the Red Sea, mangroves expanded by 12% over

the 41 years from 1972 to 2013. Losses to Red Sea mangroves, mostly due to coastal

development, have been compensated by afforestation projects.

Keywords: Remote sensing, GIS, Mapping, Satellite, Landsat, NDVI and Distribution.

40

Introduction

Mangroves form highly productive ecosystems, providing habitat for marine and

terrestrial species (Nagelkerken, Blaber et al. 2008), protecting coastal areas from storms

and sea level rises (Koch, Barbier et al. 2009), and acting as intense carbon sinks (Donato,

Kauffman et al. 2011). Unfortunately, mangroves are found within the most threatened

ecosystems on Earth; about 1/3 of their global area has been lost since World War II

(Alongi 2002). This decline continues at an annual rate of about 2.1% (Valiela, Bowen et al.

2001). A recent assessment of regional mangrove trends, however, showed that reported

rates were highly variable between regions (Friess and Webb 2014). These reports identify

aquaculture and urban developments as the main drivers of mangrove decline across these

regions (FAO 2007).

The Red Sea is adjacent to the northern limit of the Indo-Pacific mangrove, located

in the Sinai Peninsula at 28°N. Along the Red Sea, mangroves experience some of the most

difficult conditions in their distribution range, including no permanent freshwater inputs,

salinities over 40 ppt, sea surface temperatures over 31 oC in summer and recent abrupt warming of the sea (Raitsos, Hoteit et al. 2011). On the other hand, urban development and aquaculture in the Red Sea are comparatively limited and, because the desert nature of the coast, direct anthropogenic impacts have been relatively contained. The Red Sea therefore provides a good model to study the resilience of mangrove ecosystems to harsh environmental conditions with as yet limited direct anthropogenic impacts. Even so, information on the area covered by Red Sea mangroves and on changes in this area over time is scarce and often reported only in grey literature at national levels. The world atlas 41 of mangroves (Spalding, Spalding et al. 2010) reports that Red Sea mangroves are scattered along the coast of Djibouti, cover only 500 ha in Egypt, while in Sudan they are found around creeks and near-shore islets. ’s mangroves are patchy and distributed along

approximately 380 km of shoreline (De Grissac and Negussie 2007), while mangroves are

abundant along Yemen’s Red Sea coast although mostly absent along the Gulf of Aden

coastline in Yemen (Spalding, Spalding et al. 2010). Mangroves are found as fragmented

stands in the intertidal zones of the Red Sea coast of Saudi Arabia (Kumar, Khan et al.

2010). An additional study (El-Juhany 2009) reported that mangrove stands covered an

area of 36.15 Km2 between the southern border of Saudi Arabia in Jazan to the Jeddah’s southern corniche. One-third (35%) of this area was in Jazan (12.77 Km2), while the

remaining two-thirds (65%) was in Makkah and Asir (23.38 Km2) Provinces.

Although FAO (FAO, 2007) described the status of mangroves in the Red Sea from

1980-2005, a comprehensive assessment of the area covered by mangroves and of the

stability of this area along the Red Sea is still lacking. Assessments of the areas occupied by

mangroves and of any changes to these areas are fundamental to the estimation of the

ecological services provided by the mangroves in the region. In addition, such assessments

inform conservation plans and provide basic information to improve our understanding of

the resilience of mangrove ecosystems to harsh environmental conditions and warming.

Here, we report changes in the status and distribution of Red Sea mangroves over 41 years,

from 1972 to 2013, using Landsat images to determine the area and stand structure of

mangroves in the Red Sea and their dynamics over the past four decades.

42

Methods

1. Satellite imagery

We used Erdas Imagine V9.3 and ArcGIS 10.2 to assess mangrove vegetation along the

Arabian Peninsula and African coastlines of the Red Sea based on the analysis of Landsat

images for three periods: 1972, when the first satellite in the series (Landsat 1) was

launched (Chander, Markham et al. 2009) to 2013 (Landsat 8), and including 2000

(Landsat 7). Briefly, the Landsat 1 mission carried a Multispectral Scanner (MSS) sensor

with a resolution of 60 m, whereas the sensors on board the Landsat 6 and 7 missions were

Enhanced Thematic Mapper (ETM) and Enhanced Thematic Mapper plus (ETM+), respectively, both with resolutions of 30 m (Chander, Markham et al. 2009). The Landsat 8 mission also carried an Operational Land Imager (OLI) and a Thermal Infrared Sensor

(TIRS), with resolution of 30 m (Lulla, Duane Nellis et al. 2013). Details on the images used in the study are provided in the supplementary materials (Tables S1, S2, and S3). The data set is available in PANGAEA (Almahasheer, Aljowair et al. 2015)(http://www.pangaea.de).1

2. Data processing

Vectors were drawn to delineate the coastline because mangroves in the Red Sea only

occur along the coastline because there are no permanent rivers and estuaries.

Furthermore, the desert reaches to the coast around the Red Sea and large vegetation along the coastline is limited to mangroves. This means that classification problems are limited to presence/absence of vegetation. We applied an atmospheric correction to the data in which the pixels were converted to top of atmosphere (TOA) spectral radiances using the radiance rescaling factors provided in the metadata file: L = MLQcal + AL

λ 1 http://doi.pangaea.de/10.1594/PANGAEA.855896 (doi:10.1594/PANGAEA.855896) 43

(USGS_Landsat_Missions 2013), except for the 1972 images, which were corrected by the local data provider. The Normalized Difference Vegetation Index (NVDI) was used to estimate the vegetation in the coastal fringe, through unsupervised classification. Briefly,

NVDI uses near-infrared and red light reflected by the vegetation and captured by the

sensor of the satellite to measure absorbance of red light by chlorophyll and the reflection

of near-infrared by the mesophyll leaf structure (Pettorelli, Vik et al. 2005). NDVI values range from -1 to +1, where any value below zero does not correspond to green vegetation

(Hunink, Veenstra et al. 2010). Hence, the images were classified using the NDVI >0 as

mangrove and NDVI 0 as non-mangrove. The robustness of this classification was verified

through the ground ≤referencing (see below). This index was applied only to the areas where mangroves were expected to occur (i.e., vegetation along the coastline and coastal vegetation on islands). Inland and open-sea areas were excluded because mangroves do not occur in such areas (Giri, Ochieng et al. 2011). We generated vegetation thematic images and shape files assuming that that any green vegetation farther than 1 kilometer from the coast was not mangrove. Images were mosaicked and the surface of the mangroves was estimated using ArcGIS from the shape files and retrieving the area for each mangrove stand located along the Red Sea coast. Mapcarta and British Admiralty

Maps (numbers 158, 171, 10, 116 and 1157) were used as sources for the location names.

Moreover, we used four high-resolution images (GeoEye Satellite) of the Central Red Sea

and Google maps to verify that no other type of vegetation besides mangroves can be found

along the Red Sea coast.

44

3. Accuracy assessment

The assessment of mangrove vegetation was crosschecked with ground-referencing data in various ways: using reported locations where mangroves occurred in 1972 and

2000 (Price, Medley et al. 1987), visiting a number of locations along a 90 Km strip of coastline between Thuwal and Khor Alkharar between December 2014 and March 2015, and using Google Earth products to verify the classification of mangrove stands in remote locations.

To estimate the accuracy of our estimates, we applied a 100-meter buffer around the coast, then randomly selected 500 points to be crosschecked with the vegetation shape files for 2013. Out of the 500 points, 158 were positive for vegetation in the shape files and 342 were negative. Then we visually checked each of the 500 points on Google Earth to determine that 16 out of the 158 were not mangroves (i.e. false positives) and 26 out of the

342 classified as non-mangrove did have mangroves (false negatives). These results were used to calculate the accuracy of the classification (Congalton and Green 2008, Fatoyinbo and Simard 2013), resulting in an accuracy of 91.6% (Table 1, Fig. S4).

Table 1: The error matrix for the presence and absence of mangroves in predicted images

(Landsat data) over actual (Google maps).

Predicted Predicted Non- Total Mangrove Mangrove (True) (False) Actual Mangrove (Positive) 142 (TP) 16 (FP) 158 Actual Non-Mangrove (Negative) 26 (FN) 316 (TN) 342 Total 168 332 500 45

The overall accuracy was 91.6% 4. Error estimates

We also assessed the error associated with our estimates of stand size by analyzing two to three replicate images along with the original image for different locations in each of the study periods (Supplementary material, Table S5). The resulting estimate of uncertainty around the areas covered by mangrove stands was then propagated, by drawing, for each stand, replicated area estimates randomly from a normal distribution with the mean equal to the estimated area and the standard deviation calculated from the estimated coefficient

of variation of area estimates for each year, was calculated from the average and standard

deviation and reported as a percentage. (Supplementary material, Tables S1, S2, S3, S5 and

S6). This error propagation allowed us to calculate a standard error for the estimates of the total area covered by Red Sea mangroves in each year and was used to assess the significance of changes to the total area over time.

We examined the frequency distribution of mangrove stands along the Red Sea using a

Pareto distribution (Vidondo, Prairie et al. 1997), after removing all stands smaller than

0.00026 Km2 in size to account for the difference in pixel size between the 1972 and subsequent images used in this study.

Results

1. Ground-referencing data

Ground referencing data confirmed the reliability of the classification of the presence or

absence of mangroves (with an overall accuracy of 91.6%, Table 1, Fig. S4). The coefficient

of variation in the assessment of the area covered by individual mangrove stands ranged

from 23 to 35.9 % (Table S6), which is substantial due to the small size of individual 46 patches (median patch area of Red Sea mangroves 0.00721, 0.00878 and 0.00948 Km2 for

1972, 2000 and 2013, respectively, Table S7). Our error estimates were relatively high due

to the small size of individual mangrove stands along the Red Sea, which made them

conducive to edge effects. In addition, uncertainty derived from the interference of high

tides in detecting the mangroves increased the error estimates.

2. Estimating mangrove cover

The mapped Red Sea mangroves included stands located all along the African and

Arabian Peninsula coasts of the Red Sea, extending all the way to the strait of Bab-el-

Mandeb where the Red Sea opens to the Indian Ocean (Fig. 1a) (28.20893°N to 27.67161°N in 1972, 28.207348°N to 27.671217°N in 2000 and 28.207302°N to 27.671293°N in 2013).

The images showed that the abundance of mangroves increase from north to south along the Red Sea (Figs. 1a, b and Table S8).

Figure 1a: Mosaic images of mangrove distribution (green areas) in the Red Sea

47

Figure 1b: Mangrove cover along the Red Sea shorelines.

The estimated total area of Red Sea mangroves increased significantly (Tukey HSD posthoc test, P < 0.05) from 120 ± 0.54 Km2 in 1972 to 132 ± 0.94 Km2 in 2000, and no significant change, within the power of our analysis, was detected between 2000 and 2013

(Tukey HSD posthoc test, P > 0.05) when the area was estimated at 135 ± 0.86 Km2 (Fig. 2).

The increase between 1972 and 2000 was mainly due to an increase in mangrove stands along the African shore, where the area of mangroves increased from 69 Km2 to 77 Km2, compared with a change from 51 Km2 to 55 Km2 along the Arabian Peninsula (Figs.1b and

2). 48

Figure 2: Estimate of mangrove cover (Km2) along the Asian and African shores of the Red

Sea and the total cover assessed for the three different periods.

We identified about 5,000 mangrove patches along the Red Sea (2234 in 1972, 5765 in

2000 and 5157 in 2013). These patches had linear dimensions greater than 60 m in 1972

and greater than 30 m in 2000 and 2013, the pixel size of the Landsat images used. The

much greater number of patches in 2000 and 2013 compared with in 1972 indicates both

the appearance of new patches and the detection of smaller patches not identified in the

1972 images. Indeed, the number of patches detectable with the Landsat 1 resolution

would be 2233 in 1972, then 2381 and 2234, 2000 and 2013, respectively. The increased

resolution in 2000 and 2013 had, however, only a small effect on the area estimates, as

shown by the mangrove areas that were estimated when patches too small to be detected

in 1972 images were excluded (128.7 Km2 in 2000 and 131.8 Km2 in 2013). They

contributed only 2.7 % and 2.0 % of the total mangrove area detected in 2000 and 2013, 49 respectively (total area 132.4 Km2 in 2000 and 135.1 Km2 in 2013). Hence, the increase in

mangrove area between 1972 and 2000 (Fig. 3) remains significant even after accounting

for the increased resolution in 2000 and 2013. Future efforts at mapping mangroves in the

Red Sea should use high-resolution satellites, such as GeoEye with 0.5 m resolution, to

resolve smaller patches than those we could resolve here is recommended to be used if the

temporal changes are not needed for the study.

The size distribution of Red Sea mangrove patches was highly skewed and conformed

to a Pareto distribution with a cut-off represented by steeper decline than expected for a

Pareto distribution for patches larger than 0.4 Km2, which comprised less than 3 % of all

patches (Fig. 3). Most of the mangrove stands were composed of small mangrove patches,

with only 10% of the patches larger than 0.096 Km2 and only 1% of them larger than 0.87

Km2 (Fig. 3). When patches smaller than 0.00026 Km2, too small to be detected in the 1972 images, were excluded to avoid biasing the comparison, the median size as well as the 90% and 99% percentiles of the patches tended to increase significantly (Student’s t-test ,

P<0.05) over time (Fig. 3, Supplementary material, Table S7).

50

2

1 0.8 0.7 0.6 0.5

0.4

0.3

0.2

0.1

0.08 0.07 0.06 0.05

0.04

0.03

0.02

0.01 0.008 0.007 0.006 0.005

0.004

0.003 10%>0.0969 1%>0.87875

0.002

0.001 0.001 0.002 0.003 0.005 0.01 0.02 0.03 0.04 0.06 0.1 0.2 0.3 0.4 0.5 0.7 1 2 3 4 5 2

Figure 3: Pareto Plot describing the size distribution for individual mangrove patches

(Km2) in the Red Sea in 1972 (red), 2000 (green) and 2013 (blue). The solid colored lines

show the fitted Pareto regression for each of the time periods (1972, log (%) N>x = -4.03 –

0.69 log Size, R2 =0.96; 2000, log (%) N>x = -4.07 – 0.70 log Size, R2 =0.96; 2013, log (%)

N>x = -3.95 – 0.68 log Size, R2 =0.96 ), the dotted lines indicate the 10% and 1 % size

percentiles and the insert shows the median, error bars are the ( ±C.L) patch size for each

period.

51

3. Losses and gains over time

Examination of the balance between losses and gains showed that there was a prevalence of increases, particularly along the African shore (Table 2, Fig 4). The largest mangrove stand along the Red Sea, located along the Alwajh Bank (Tabuk, Saudi Arabia), remained in an approximately steady state over the 41 years of study (1972 = 3.8 Km2,

2000 = 4.5 Km2, 2013 = 4.3 Km2, Fig. 5A). A successful rehabilitation project in Yanbu,

Saudi Arabia, resulted in a 50-fold increase in the area covered by mangroves (1972 =

0.011 Km2, 2000 = 0.543 Km2, 2013 = 0.562 Km2) in this region (Fig. 5B). In contrast, a

decline was detected in a costal lagoon in Alith, Saudi Arabia, where one of the largest

shrimp farms in the world was established in 1986. Mangrove stands in the lagoon

occupied 2.33 km2 in 1972 and declined to 1.91 Km2 in 2000 and 0.18 Km2 in 2013 (Fig.

5C). Thorough documentation of potential drivers of losses and gains is presented in

Supplementary material, Table 9, with an illustration of these drivers in Fig. 6.

Table 2: Estimate of mangrove cover growth rate and the % of increase per year, along the

Asian and African shores of the Red Sea and the total cover assessed for the three different periods.

African Asian Total 1972-2000 ( 28 years) Km2 8.51 3.97 12.48 Km2 y-1 0.30 0.14 0.45 % y-1 0.42 0.27 0.35 2000-2013 (13 years) Km2 1.50 1.12 2.63 Km2 y-1 0.12 0.09 0.20 % y-1 0.15 0.15 0.15 1972-2013 (41 years) Km2 10.01 5.09 15.11 Km2 y-1 0.24 0.12 0.37 % y-1 0.33 0.23 0.29 52

Figure 4: Overlapped images showing mangrove gains (green), losses (red), and unchanged areas (yellow) over the time intervals (A) between 1972 and 2000, and (B ) between 2000 and 2013. 53

Figure 5: Examples of Red Sea mangrove vegetation over the study period. (A) dynamics of the largest mangrove stand in the Red Sea (Tabuk, N W. Saudi Arabia). (B) mangrove expansion assoicated to the rehaplitation project in Yanbu (Centeral Red Sea coast of

Saudia Arabia). (C) decline of mangrove in a costal lagon at Alith.

54

Figure 6: Map for the drivers of loss and gains in mangrove communities. The symbols from left to right are: (camel: the losses, decay and overgrazing), (factory: represent the pollution), (house for the costal development) and (tree: for the rehablitation/afforistation projects).

55

Discussion

Red Sea mangrove forests are dominated by small patches, with a median size of only

0.007 to 0.009 Km2. The fact that the median patch size was similar between years despite

a major increase in resolution of the images between 1972 and 2013 indicates that the

estimate is robust and independent of the change in resolution of the images. Although the area covered is relatively modest, mangroves are the most important vegetated habitats in

the region, which is characterized by inland deserts along both the African and Arabian

Peninsula shores.

The largest mangrove stand in the Red Sea occurs along the Alwajh Bank (Tabuk, Saudi

Arabia) where extensive mangrove and seagrass beds can be found (Bruckner 2011). Along the African shore, the largest stands are found along the coast of Eritrea. We found mangrove stands to be patchy in the northern Red Sea and to increase, in area and patch

size, to the south, as reported in the past (Mandura 1997). A range of factors may explain

this pattern, including higher minimum air temperatures, lower salinity, higher rainfall and

freshwater supply and increased nutrient concentrations towards the south (Saifullah

1996). Also, a shift in sediment composition from stony corals in the north to fine sediment in the south would favor mangrove growth (Price, Medley et al. 1987).

Mangrove forests are experiencing a decline around the world due to logging, land

reclamation and conversion of mangrove forests into aquaculture farms (Polidoro,

Carpenter et al. 2010), causes connected with the growth rate of the human population

with 60% of the population living in coastal regions (Green, Mumby et al. 1996). Although

Red Sea nations are also experiencing population growth 1.8 to 5.1 % per annum (PERSGA 56

2002), our analysis did not provide evidence of major losses of mangrove stands over the

past four decades. On the contrary, the areas covered by Red Sea mangrove stands

increased significantly, by 15 Km2 (11 Km2 when excluding the patches too small to have

been detected in 1972), or 12 %, during the study period. This expansion is supported by a

tendency for the size of individual stands to expand between 1972 and 2000. Two-thirds

of the area gained by mangroves is located along the African shore, particularly along

Eritrea, where large mangrove stands are located. The recent expansion of mangroves

along the Red Sea coast of Saudi Arabia is in contrast to the major loss of mangroves in

Tarut Bay in the Arabian Gulf Coast of Saudi Arabia, where over 55 % of the area of

mangroves was lost from 1972 to 2011 (Almahasheer, Al-Taisan et al. 2013).

The expansion of mangroves between 1972 and 2000 is the net result of the balance between losses and gains. Whereas a significant expansion of mangrove area was observed between 1972 and 2000, losses and gains between 2000 and 2013 compensated each other within the uncertainty of our analysis, so no significant change in total mangrove area in the Red Sea was detected in the later period. Afforestation projects contributed to gains in several areas, including those in Yanbu, Saudi Arabia, and the Manzanar mangrove restoration project conducted in the mid 1990’s in Hirgigo, Eritrea (De Grissac and

Negussie 2007). Losses were mostly due to coastal development and pollution, particularly along the Saudi coast, whereas along the coasts of Yemen and Africa, losses were due to overgrazing by camels and logging (Table S9). Whereas camel grazing would degrade mangroves, but generally not lead to losses, events of intense degradation have been reported, such as the heavy impact of camel grazing on 500 ha of mangrove around Port of

Sudan by camel grazing (Bojang 2009). Moreover, camel grazing, which is likely to be a 57 pressure of mangroves intrinsic to the Red Sea, may prevent gains by removing seedling recruits. Compensatory afforestation projects associated with some projects, such as the construction of the King Abdullah University of Science and Technology in Thuwal, Saudi

Arabia, proved effective in compensating for the losses during the development phase. One of the mangrove stands experiencing losses was that located in Alith, Saudi Arabia, where a large aquaculture farm was built in the late 1980’s. However, the decline cannot be directly related to the construction of the aquaculture facilities, as suggested by (Gladstone, Tawfiq et al. 1999), because the mangrove ponds were excavated inland and the decline of mangroves at Alith occurred between 2000 and 2013, 10 years after the construction and

the start of operations of the aquaculture farm. Similarly, 58% of African mangroves were

destroyed before shrimp aquaculture developed in the region (Sadek, Rafael et al. 2002).

In summary, the results presented here show that Red Sea mangroves depart from the

general declining trend of mangroves elsewhere, exhibiting an expansion, by 12% in area,

over 41 years from 1972 to 2013. This expansion is the net result of the balance between

losses and gain, where afforestation projects have played an important role in maintaining,

and expanding mangroves along the Red Sea. However, coastal development in the region,

including major infrastructure projects for ports and tourism, may represent a threat in the

future unless mangrove conservation receives prominent attention as a required milestone

at the planning stage.

Acknowledgements

We thank the Space Research Institute in King Abdul-Aziz city for Science and Technology

(KACST) for providing the raw data. We also thank Virginia Unkefer for reviewing the 58 manuscript and M. Campbell for help with accuracy assessment. This research was supported by the baseline funding provided by King Abdullah University of Science and

Technology to Xabier Irigoien.

References

Al-But’hie, I. M. and M. A. E. Saleh (2002). "Urban and industrial development planning as an approach for Saudi Arabia: the case study of and Yanbu." Habitat International 26(1): 1-20.

Aleem, A. (1990). "Impact of human activity on marine habitats along the Red Sea coast of Saudi Arabia." Hydrobiologia 208(1-2): 7-15.

Almahasheer, H., W. Al-Taisan and M. K. Mohamed (2013). "Mangrove Deterioration in Tarut Bay on the Eastern Province of the Kingdom of Saudi Arabia." Pakhtunkhwa J. Life Sci 01(02): 49-59.

Almahasheer, H., A. Aljowair, C. M. Duarte and X. Irigoien (2015). "Mangrove cover in the Red Sea (1972-2013)."

Alongi, D. M. (2002). "Present state and future of the world's mangrove forests." Environmental conservation 29(03): 331-349.

Badr, N. B., A. A. El-Fiky, A. R. Mostafa and B. A. Al-Mur (2009). "Metal pollution records in core sediments of some Red Sea coastal areas, Kingdom of Saudi Arabia." Environmental monitoring and assessment 155(1-4): 509-526.

Bojang, E. (2009). "The relevance of mangrove forests to African fisheries, wildlife and water resources." Food And Agriculture Organisation of the United Nations. Accra, Ghana.

Bruckner, A. (2011). "Khaled bin Sultan Living Oceans Foundation Habitat Mapping and Characterization of Coral Reefs of the SaudiArabian Red Sea: 2006-2009." Panoramic Press, Arizona: 140 pp.

Chander, G., B. L. Markham and D. L. Helder (2009). "Summary of current radiometric calibration coefficients for Landsat MSS, TM, ETM+, and EO-1 ALI sensors." Remote sensing of environment 113(5): 893-903.

Congalton, R. G. and K. Green (2008). Assessing the accuracy of remotely sensed data: principles and practices, CRC press.

De Grissac, A. and K. Negussie (2007). "State of the Coast Eritrea, 2006-2007." Coastal Marine and Island Biodiversity Conservation Project, UNDP-Eritrea: 22. Eritrea‟ s 59

Donato, D. C., J. B. Kauffman, D. Murdiyarso, S. Kurnianto, M. Stidham and M. Kanninen (2011). "Mangroves among the most carbon-rich forests in the tropics." Nature Geoscience 4(5): 293-297.

El-Juhany, L. (2009). "Present status and degradation trends of mangrove forests on the southern Red Sea coast of Saudi Arabia." American-Eurasian Journal of Agricultural and Environmental Science 6(3): 328-340.

El Raey, M. (2010). "Impact of sea level rise on the Arab region." Arab Climate Initiative/UNDP, http://www.arabclimateinitiative.org/knowledge-center.html# background-papers (last accessed 30 December 2011).

FAO (2007). The world’s Mangroves 1980-2005, FAO, Food and Agriculture Organization of the Rome.

Fatoyinbo, T. E. and M. Simard (2013). "Height and biomass of mangroves in Africa from ICESat/GLAS and SRTM." International Journal of Remote Sensing 34(2): 668-681.

Friess, D. A. and E. L. Webb (2014). "Variability in mangrove change estimates and implications for the assessment of ecosystem service provision." Global ecology and biogeography 23(7): 715-725.

Giri, C., E. Ochieng, L. Tieszen, Z. Zhu, A. Singh, T. Loveland, J. Masek and N. Duke (2011). "Status and distribution of mangrove forests of the world using earth observation satellite data." Global Ecology and Biogeography 20(1): 154-159.

Gladstone, W., N. Tawfiq, D. Nasr, I. Andersen, C. Cheung, H. Drammeh, F. Krupp and S. Lintner (1999). "Sustainable use of renewable resources and conservation in the Red Sea and Gulf of Aden: issues, needs and strategic actions." Ocean & coastal management 42(8): 671-697.

Green, E., P. Mumby, A. Edwards and C. Clark (1996). "A review of remote sensing for the assessment and management of tropical coastal resources." Coastal management 24(1): 1- 40.

Hunink, J., T. Veenstra, W. van der Hoek and P. Droogers (2010). Q fever transmission to humans and local environmental conditions, FutureWater.

Kaust (2010). "Mangrove Restoration Program." http://facilities.kaust.edu.sa/Services/Pages/Mangrove-Restoration-Program.aspx.

Koch, E. W., E. B. Barbier, B. R. Silliman, D. J. Reed, G. M. Perillo, S. D. Hacker, E. F. Granek, J. H. Primavera, N. Muthiga and S. Polasky (2009). "Non-linearity in ecosystem services: temporal and spatial variability in coastal protection." Frontiers in Ecology and the Environment 7(1): 29-37. 60

Kumar, A., M. A. Khan and A. Muqtadir (2010). "Distribution of mangroves along the Red Sea coast of the Arabian Peninsula: Part–1: the northern coast of western Saudi Arabia." Earth Sc. Ind 3: 28-42.

Lulla, K., M. Duane Nellis and B. Rundquist (2013). "The Landsat 8 is ready for geospatial science and technology researchers and practitioners." Geocarto International 28(3): 191- 191.

Mandura, A. (1997). "A mangrove stand under sewage pollution stress: Red Sea." Mangroves and Salt marshes 1(4): 255-262.

Mandura, A. and A. Khafaji (1993). Human impact on the mangrove of Khor Farasan Island, southern Red Sea coast of Saudi Arabia. Towards the rational use of high salinity tolerant plants, Springer: 353-361.

Nagelkerken, I., S. Blaber, S. Bouillon, P. Green, M. Haywood, L. Kirton, J.-O. Meynecke, J. Pawlik, H. Penrose and A. Sasekumar (2008). "The habitat function of mangroves for terrestrial and marine fauna: a review." Aquatic Botany 89(2): 155-185.

PERSGA (2002). "The Red Sea and Gulf of Aden Regional Network of Marine Protected Areas. Regional Master Plan." PERSGA Technical Series, Jeddah(1).

Pettorelli, N., J. O. Vik, A. Mysterud, J.-M. Gaillard, C. J. Tucker and N. C. Stenseth (2005). "Using the satellite-derived NDVI to assess ecological responses to environmental change." Trends in Ecology & Evolution 20(9): 503-510.

Polidoro, B. A., K. E. Carpenter, L. Collins, N. C. Duke, A. M. Ellison, J. C. Ellison, E. J. Farnsworth, E. S. Fernando, K. Kathiresan, N. E. Koedam, S. R. Livingstone, T. Miyagi, G. E. Moore, V. Ngoc Nam, J. E. Ong, J. H. Primavera, S. G. Salmo, J. C. Sanciangco, S. Sukardjo, Y. Wang and J. W. Yong (2010). "The loss of species: mangrove extinction risk and geographic areas of global concern." PLoS One 5(4): e10095.

Price, A., P. Medley, R. McDowall, A. Dawson-Shepherd, P. Hogarth and R. Ormond (1987). "Aspects of mangal ecology along the Red Sea coast of Saudi Arabia." Journal of natural history 21(2): 449-464.

Raitsos, D. E., I. Hoteit, P. K. Prihartato, T. Chronis, G. Triantafyllou and Y. Abualnaja (2011). "Abrupt warming of the Red Sea." Geophysical Research Letters 38(14).

Rouphael, T., E. Turak and J. Brodie (1998). "Seagrasses and Mangroves of Yemen’s Red Sea." Protection of Marine Ecosystems of the Red Sea Coast of Yemen. UN Publication: 41- 49.

Royal-Commission-Yanbu (2013). "Sustainability report. Yanbu Industrial City." http://www.rcjy.gov.sa/en- US/Yanbu/MediaCenter/DocumentCenter/Documents/SUSTAINABILITY%20REPORT.pdf 61

Sadek, S., R. Rafael, M. Shakouri, G. Rafomanana, F. L. Ribeiro and J. Clay (2002). "Shrimp aquaculture in Africa and the Middle East: the current reality and trends for the future." Report prepared under the World Bank, NACA, WWF and FAO Consortium Program on Shrimp Farming and the Environment. Work in Progress for Public Discussion. Published by the Consortium.

Saifullah, S. M. (1996). " Mangrove Ecosystem of Saudi Arabian Red Sea Coast -An Overview " Journal of King Abdul Aziz University. 7(1): 263 -270

Spalding, M., M. Spalding, M. Kainuma and L. Collins (2010). World atlas of mangroves, Earthscan.

Spurgeon, J. (2002). "Rehabilitation, conservation and sustainable utilization of mangroves in Egypt." Socio-economic assessment and economic valuation of Egypt’s mangroves. Final Report. Ministry of Agriculture & Land Rrclamation. Ministry of State for Environment Food and Agriculture Organization of the United Nations.

USGS_Landsat_Missions (2013). "Using the USGS Landsat 8 Product." Available online: http://landsat.usgs.gov/Landsat8_Using_Product.php

Valiela, I., J. L. Bowen and J. K. York (2001). "Mangrove forests: One of the world's threatened major tropical environments." Bioscience 51(10): 807-815.

Vidondo, B., Y. T. Prairie, J. M. Blanco and C. M. Duarte (1997). "Some aspects of the analysis of size spectra in aquatic ecology." Limnology and Oceanography 42(1): 184-192.

62

Supplementary Materials

Table S 1: Satellite images data used in the study for 2013

LANDSAT_SCENE_ID Zone PATH ROW SENSOR_ID ACQUISITION_DATE LC81720452013267LGN00 36 172 45 OLI_TIRS 9/24/2013 LC81730412013242LGN00 36 173 41 OLI_TIRS 8/30/2013 LC81730422013242LGN00 36 173 42 OLI_TIRS 8/30/2013 LC81730432013226LGN00 36 173 43 OLI_TIRS 8/14/2013 LC81730442013226LGN00 36 173 44 OLI_TIRS 8/14/2013 LC81740402013249LGN00 36 174 40 OLI_TIRS 9/6/2013 LC81740412013249LGN00 36 174 41 OLI_TIRS 9/6/2013 LC81740422013265LGN00 36 174 42 OLI_TIRS 9/22/2013 LC81750392013176LGN00 36 175 39 OLI_TIRS 6/25/2013 LC81750402013256LGN00 36 175 40 OLI_TIRS 9/13/2013 LC81750412013256LGN00 36 175 41 OLI_TIRS 9/13/2013 LC81760392013231LGN00 36 176 39 OLI_TIRS 8/19/2013 LC81760402013231LGN00 36 176 40 OLI_TIRS 8/19/2013 LC81670512013296LGN00 37 167 51 OLI_TIRS 10/23/2013 LC81680462013223LGN00 37 168 46 OLI_TIRS 8/11/2013 LC81680472013191LGN00 37 168 47 OLI_TIRS 7/10/2013 LC81680482013143LGN01 37 168 48 OLI_TIRS 5/23/2013 LC81680492013239LGN00 37 168 49 OLI_TIRS 8/27/2013 LC81680502013271LGN00 37 168 50 OLI_TIRS 9/28/2013 LC81690452013342LGN00 37 169 45 OLI_TIRS 12/8/2013 LC81690462013246LGN00 37 169 46 OLI_TIRS 9/3/2013 LC81690472013246LGN00 37 169 47 OLI_TIRS 9/3/2013 LC81690482013246LGN00 37 169 48 OLI_TIRS 9/3/2013 LC81690492013246LGN00 37 169 49 OLI_TIRS 9/3/2013 LC81700432013221LGN00 37 170 43 OLI_TIRS 8/9/2013 LC81700442013253LGN00 37 170 44 OLI_TIRS 9/10/2013 LC81700452013173LGN00 37 170 45 OLI_TIRS 6/22/2013 LC81700462013173LGN00 37 170 46 OLI_TIRS 6/22/2013 LC81700472013269LGN00 37 170 47 OLI_TIRS 9/26/2013 LC81700482013173LGN00 37 170 48 OLI_TIRS 6/22/2013 LC81710412013228LGN00 37 171 41 OLI_TIRS 8/16/2013 LC81710432013212LGN00 37 171 44 OLI_TIRS 7/31/2013 LC81710442013212LGN00 37 171 44 OLI_TIRS 7/31/2013 LC81710452013276LGN00 37 171 45 OLI_TIRS 10/3/2013 LC81710462013324LGN00 37 171 46 OLI_TIRS 11/20/2013 LC81710472013260LGN00 37 171 47 OLI_TIRS 9/17/2013 63

LC81720412013235LGN00 37 172 41 OLI_TIRS 8/23/2013 LC81720422013171LGN00 37 172 42 OLI_TIRS 6/20/2013 LC81720432013251LGN00 37 172 43 OLI_TIRS 9/8/2013 LC81720442013139LGN01 37 172 44 OLI_TIRS 5/19/2013 LC81730402013226LGN00 37 173 40 OLI_TIRS 8/14/2013 LC81660502013145LGN00 38 166 50 OLI_TIRS 5/25/2013 LC81660512013145LGN00 38 166 51 OLI_TIRS 5/25/2013 LC81670482013104LGN01 38 167 48 OLI_TIRS 4/14/2013 LC81670492013136LGN01 38 167 49 OLI_TIRS 5/16/2013 LC81670502013296LGN00 38 167 50 OLI_TIRS 10/23/2013

Table S 2: Satellite images data used in the study for 2000

LANDSAT_SCENE_ID Zone PATH ROW SENSOR_ID ACQUISITION_DATE LE71720452000144SGS00 36 172 45 ETM 5/23/2000 LE71730432000215SGS00 36 173 43 ETM 8/2/2000 LE71730442000215SGS00 36 173 44 ETM 8/2/2000 LE71740402000142SGS00 36 174 40 ETM 5/21/2000 LE71740411999219EDC00 36 174 41 ETM 8/7/19991 LE71740422000174SGS01 36 174 42 ETM 6/22/2000 LE71750392000229SGS00 36 175 39 ETM 8/16/2000 LE71750402000229SGS00 36 175 40 ETM 8/16/2000 LE71750412000229SGS00 36 175 41 ETM 8/16/2000 LE71760392000140EDC00 36 176 39 ETM 5/19/2000 LE71760402000284EDC00 36 176 40 ETM 10/10/2000 p173r041_7x20001005.ETM 36 173 41 ETM+ 10/5/2000 p173r042_7x20001005.ETM 36 173 42 ETM+ 10/5/2000 LE71670512000253SGS00 37 167 51 ETM 9/9/2000 LE71680462000148SGS00 37 168 46 ETM 5/27/2000 LE71680472000260SGS01 37 168 47 ETM 9/16/2000 LE71680482000036SGS00 37 168 48 ETM 2/5/2000 LE71680492000036SGS00 37 168 49 ETM 2/5/2000 LE71680502000132SGS00 37 168 50 ETM 5/11/2000 LE71690452000347SGS00 37 169 45 ETM 12/12/2000 LE71690462000155SGS00 37 169 46 ETM 6/3/2000 LE71690482000027EDC00 37 169 48 ETM 1/27/2000 LE71690492000027EDC00 37 169 49 ETM 1/27/2000 LE71690502000027EDC00 37 169 50 ETM 1/27/2000 LE71690512000027EDC00 37 169 51 ETM 1/27/2000 LE71700432000258SGS00 37 170 43 ETM 9/14/2000 LE71700442000194EDC00 37 170 44 ETM 7/12/2000 64

LE71700452000306SGS00 37 170 45 ETM 11/1/2000 LE71700462000306SGS00 37 170 46 ETM 11/1/2000 LE71700472001084SGS00 37 170 47 ETM 3/25/20011 LE71700482001084SGS00 37 170 48 ETM 3/25/20011 LE71710432000297SGS00 37 171 43 ETM 10/23/2000 LE71710452000169SGS01 37 171 45 ETM 6/17/2000 LE71710462000169SGS01 37 171 46 ETM 6/17/2000 LE71710472000153SGS00 37 171 47 ETM 6/1/2000 LE71720442000176SGS00 37 172 44 ETM 6/24/2000 p169r047_7x19990905.ETM 37 169 47 ETM+ 9/5/19991 p171r044_7x20001023.ETM 37 171 44 ETM+ 10/23/2000 p172r042_7x20000710.ETM 37 172 42 ETM+ 7/10/2000 p172r043_7x20000710.ETM 37 172 43 ETM+ 7/10/2000 LE71660512001232SGS00 38 166 51 ETM 8/20/20011 LE71670482000253SGS00 38 167 48 ETM 9/9/2000 LE71670492000253SGS00 38 167 49 ETM 9/9/2000 LE71670502000253SGS00 38 167 50 ETM 9/9/2000 p166r050_7x20001223.ETM 38 166 50 ETM+ 12/23/2000 1 Closest images for the year were used because the images of the coresponding year were not availiable

Table S 3: Satellite images data used in the study for 1972

LANDSAT_SCENE_ID Zone PATH ROW SENSOR_ID ACQUISITION_DATE LM11850451972311AAA03 36 185 45 MSS 11/6/1972 LM11860411972276AAA04 36 186 41 MSS 10/2/1972 LM11860421973288AAA05 36 186 42 MSS 10/15/19731 LM11860431973252AAA04 36 186 43 MSS 9/9/19731 LM11860441972258AAA04 36 186 44 MSS 9/14/1972 LM11860451972276AAA04 36 186 45 MSS 10/2/1972 LM11870401972259AAA05 36 187 40 MSS 9/15/1972 LM11870411972259AAA05 36 187 41 MSS 9/15/1972 LM11870421972313AAA05 36 187 42 MSS 11/8/1972 LM11880401972278AAA05 36 188 40 MSS 10/4/1972 LM11880411972296AAA05 36 188 41 MSS 10/22/1972 LM11890391973075AAA04 36 189 39 MSS 3/16/19731 LM11890401973075AAA04 36 189 40 MSS 3/16/19731 LM11800481972270AAA05 37 180 48 MSS 9/26/1972 LM11800491972270GDS03 37 180 49 MSS 9/26/1972 LM11800501972306AAA05 37 180 50 MSS 11/1/1972 65

LM11810461972343AAA04 37 181 46 MSS 12/8/1972 LM11810471972343AAA04 37 181 47 MSS 12/8/1972 LM11810481972271AAA02 37 181 48 MSS 9/27/1972 LM11810491972253AAA02 37 181 49 MSS 9/9/1972 LM11820461972326AAA04 37 182 46 MSS 11/21/1972 LM11820471972272AAA05 37 182 47 MSS 9/28/1972 LM11820481972272AAA05 37 182 48 MSS 9/28/1972 LM11820491972308AAA04 37 182 49 MSS 11/3/1972 LM11830441972327AAA05 37 183 44 MSS 11/22/1972 LM11830451972273AAA05 37 183 45 MSS 9/29/1972 LM11830461972273AAA05 37 183 46 MSS 9/29/1972 LM11830471972273AAA05 37 183 47 MSS 9/29/1972 LM11830481972309FAK03 37 183 48 MSS 11/4/1972 LM11830491972309FAK03 37 183 49 MSS 11/4/1972 LM11840431973052AAA05 37 184 43 MSS 2/21/19731 LM11840441972328AAA05 37 184 44 MSS 11/23/1972 LM11840451972274XXX01 37 184 45 MSS 9/30/1972 LM11840461972274XXX01 37 184 46 MSS 9/30/1972 LM11840471972346FAK03 37 184 47 MSS 12/11/1972 LM11850421972257AAA04 37 185 42 MSS 9/13/1972 LM11850431972257AAA04 37 185 43 MSS 9/13/1972 LM11850441972329GDS03 37 185 44 MSS 11/24/1972 LM11780501972322FAK03 38 178 50 MSS 11/17/1972 LM11780511972322AAA05 38 178 51 MSS 11/17/1972 LM11780521972322AAA05 38 178 52 MSS 11/17/1972 LM11790481972341AAA04 38 179 48 MSS 12/6/1972 LM11790491972287AAA04 38 179 49 MSS 10/13/1972 LM11790501972287AAA04 38 179 50 MSS 10/13/1972 LM11790511972287AAA04 38 179 51 MSS 10/13/1972 LM11800471972360AAA04 38 180 47 MSS 12/25/1972 1 Closest images for the year were used because the images of the coresponding year were not availiable

66

Table S 4: Map showing 500 random points applied over the costliness of the Red Sea to test the presence and absence of mangrove communities.

67

Table S 5: Satellite images data used to estimate error in the assessment of mangrove area

LANDSAT_SCENE_ID Satellite Path/Row status Sum ACQUISITION DATE LM11780511972358AAA02 Landsat1 178/81 (166/51) replicate 21.59 12/23/1972 LM11780511972250AAA05 Landsat1 178/81 (166/51) replicate 27.64 9/6/1972 LM11780511972214AAA02 Landsat1 178/81 (166/51) replicate 21.59 8/1/1972 LM11780511972322AAA05 Landsat1 178/81 (166/51) original 23.89 11/17/1972 LM11820461973032GMD03 Landsat1 182/46 (169/46) replicate 3.57 2/1/19731 LM11820461972326AAA04 Landsat1 182/46 (169/46) replicate 2.80 11/21/1972 LM11820461972254AAA04 Landsat1 182/46 (169/46) replicate 2.30 9/10/1972 LM11820461973014AAA04 Landsat1 182/46 (169/46) original 3.90 1/14/19731 LM11840431972364AAA04 Landsat1 184/43 (171/43) replicate 0.19 12/29/1972 LM11840431972346FAK07 Landsat1 184/43 (171/43) replicate 0.23 12/11/1972 LM11840431973070GDS05 Landsat1 184/43 (171/43) replicate 0.13 3/11/1972 LM11840431973052AAA05 Landsat1 184/43 (171/43) original 0.11 2/21/19731 LE71660512000358SGS00 landsat7 166/51 replicate 29.11 12/23/2000 LE71660512000198EDC00 landsat7 166/51 replicate 13.65 7/16/2000 LE71660512000086SGS00 landsat7 166/51 replicate 23.42 3/26/2000 LE71660512001232SGS00 landsat7 166/51 original 16.18 8/20/20011 LE71680502000244SGS00 landsat7 168/50 replicate 17.30 8/31/2000 LE71680502001294SGS00 landsat7 168/50 replicate 26.23 10/21/20011 LE71680502000132SGS00 landsat7 168/50 original 23.24 12/5/2000 LE71690462000027EDC00 landsat7 169/46 replicate 9.59 1/27/2000 LE71690462000347SGS00 landsat7 169/46 replicate 6.80 12/12/2000 LE71690462000155SGS00 landsat7 169/46 original 7.08 1/3/2000 LE71710432000361SGS00 landsat7 171/43 replicate 1.81 12/26/2000 LE71710432000169SGS01 landsat7 171/43 replicate 0.92 1/17/2000 LE71710432000025EDC00 landsat7 171/43 replicate 0.42 1/25/2000 LE71710432000297SGS00 landsat7 171/43 original 1.72 10/23/2000 LE71720422000320EDC00 landsat7 172/42 replicate 5.82 12/15/2000 LE71720422000176SGS00 landsat7 172/42 replicate 10.84 6/24/2000 p172r042_7x20000710.ETM landsat7 172/42 original 13.47 7/10/2000 LC81660512013353LGN00 landsat8 166/51 replicate 27.70 12/19/2013 LC81660512013113LGN01 landsat8 166/51 replicate 21.69 4/23/2013 LC81660512014068LGN00 landsat8 166/51 replicate 22.12 3/9/20141 LC81660512013145LGN00 landsat8 166/51 original 19.91 5/25/2013 LC81690462013102LGN01 landsat8 169/46 replicate 5.00 4/12/2013 LC81690462014057LGN00 landsat8 169/46 replicate 6.75 2/26/20141 LC81690462013246LGN00 landsat8 169/46 original 2.63 9/3/2013 LC81710432013356LGN00 landsat8 171/43 replicate 1.13 12/22/2013 68

LC81710432013164LGN00 landsat8 171/43 replicate 1.28 1/13/2013 LC81710432013212LGN00 landsat8 171/43 original 1.62 7/31/2013 LC81720422014046LGN00 landsat8 172/42 replicate 4.91 2/15/20141 LC81720422013100LGN01 landsat8 172/42 replicate 1.97 4/10/2013 LC81720422013171LGN00 landsat8 172/42 original 9.12 6/20/2013 1 Closest images for the year were used because the images of the coresponding year were not availiable

Table S 6: Error assessment (as Coefficient of Variation, %, of the estimated mangrove area in each image, CV) for three to five locations assessed for each of the study periods. N = number of replicated image per location.

Year Landsat Path/Row N CV Mean CV 1972 178/811 (166/51) 4 12.1 23.0 1972 182/462 (169/46) 4 23.1 1972 184/433 (171/43) 4 33.7 2000 166/51 4 34.1 33.5 2000 168/50 3 20.4 2000 169/46 3 19.6 2000 171/43 4 54.6 2000 172/42 3 38.7 2013 166/51 4 14.7 35.9 2013 169/46 3 43.1 2013 171/43 3 18.7 2013 172/42 3 67.3

1 This the Landsat 1 Path and Row correspond to Landsat 7 and 8 (166/51) 2 This the Landsat 1 Path and Row correspond to Landsat 7 and 8 (169/46) 3 This the Landsat 1 Path and Row correspond to Landsat 7 and 8 (171/43)

69

Table S 7: The mean of size distribution along with the upper and lower confident limits

1972 Quantile Estimate Lower 95% Upper 95% Actual Coverage 50% 0.00721 0.0072 0.00722 95.09 90% 0.09369 0.08235 0.1152 95.14 99% 0.92723 0.70432 1.3383 95.67 2000 50% 0.00878 0.00834 0.00945 95.09 90% 0.09198 0.08021 0.10845 95.23 99% 0.84678 0.71123 1.26596 95.09 2013 50% 0.00948 0.00911 0.01025 95.09 90% 0.10379 0.09196 0.12308 95.12 99% 0.90896 0.75443 1.16381 95.66

Table S 8: Estimate of mangrove cover, showing the Mean patch size (the ratio between the

total cover and the number of patches.

A B A+B Mean Patch Size in % Number of Patches 107 944 1051 2.27 Total cover in Km2 1.53 22.34 23.88 C D C+D

Number of Patches 232 824 1056 1.56 Total cover in Km2 3.20 13.33 16.53 E F E+F

Number of Patches 2041 1010 3051 3.10 Total cover in Km2 73.88 20.80 94.69

70

Table S 9: Drivers of loss and gain

Driver of Shore Reference Details Change Coastal Arabian (Gladstone, Shrimp farming contributed to the decline of Development Peninsula Tawfiq et al. Saudi Arabia costal mangroves. 1999, El Raey 2010) (Aleem North coast of Jeddah and southern bank of 1990) Obhor creek, Saudi Arabia; major disruption of marine habitat by dumping sediment, rocks and boulders into the sea for construction of houses, swimming beaches and artificial islands (Mandura 9.5% of the mangrove in Khor Farasan island and Khafaji in Saudi Arabia was destroyed due to cutting 1993) and reclamation (Badr, El- By 1987; 8% of costal habitat affected by Fiky et al. coastal development effecting the inter-tidal 2009) and sub-tidal habitat Africa (Spurgeon Egyptian mangrove declined over the past 2002) years due to costal development Pollution Arabian (Gladstone, Power and plant built in Saudi Peninsula Tawfiq et al. Arabia, Yemen and Jordan, Releasing heated 1999) and hypersaline water into the sea. With a potential environmental impact on coastal environments. (Aleem Organic pollution from sewage effluents and 1990) Oil pollution from habours in Obhor Creek ( Jeddah) Thermal pollution from the desalination plant in Jeddah which causes seawater salinity and temperature to rise above normal effecting the marine habitat (Badr, El- Increased industrialization led to increased Fiky et al. metal pollution in some areas of the Saudi 2009) Arabian coastline. Africa (Gladstone, urea discharges from the Suez plant, chemical Tawfiq et al. spills from ships, and high population growth 1999) rates in Egypt, uncontrolled spillages of port of Sudan, and a direct sewage pumping near Djibouti city, these regional issues are participating in declining the seagrass , mangroves and fisheries. 71

Losses, decay Arabian (Aleem Observations of Mangrove community in and Peninsula 1990) Shuaiba, Saudi Arabia reported to be denser overgrazing in the late seventies than present day. (Rouphael, Yemen’s Rhizophora decline due to logging Turak et al. for use as building material 1998) (Rouphael, 50% of mangrove forest north Al-Hudaydah, Turak et al. Yemen, either heavily grazed or dead, but it is 1998) not clear if the grazing was the cause of death. (Rouphael, Mangroves in a small island in Khawr Turak et al. Kathibwere, Yemen were damaged by logging 1998) in 1998. (Rouphael, Camels, sand drift, and over flooding Arabian Turak et al. wadies are major threats led to mangrove 1998) decline in Yemen. Africa (Bojang About 500 ha mangrove stand around port of 2009) Sudan reported to be in badly degraded due to camel grazing, logging, salt ponds blocking freshwater inflow to mangroves, and highway constructions.

Mangrove Arabian Ministry of In 2012; 300,000 seedling planted in Sharm rehabilitation Peninsula Agriculture Yanbu, Saudi Arabia (restoration) (information) In 2012; 100,000 seedlings planted in Amlj ( and/ or shabaan), Saudi Arabia afforestation In 2003, 50,000 seedlings planted Jazan, Saudi Arabia (Kaust 2010) Restoration program at KAUST was initiated in 2010; with 100.000 seedlings planted

(Al-But’hie Protection program developed in late 70’s and Saleh and early 80’s for marine environment (e.g. 2002, Royal- mangrove and coral reefs), in the Yanbu, Commission- Saudi Arabia Yanbu 2013) Africa (De Grissac Afforestation in Eritrea in the mid 90’s in in and Negussie Hirgigo (10 km south Massawa). 2007)

72

Chapter Two

Phenology and Growth dynamics of Avicennia marina in the Central Red Sea

Hanan Almahasheer1,2,, Carlos M. Duarte1 and Xabier Irigoien1

1 King Abdullah University of Science and Technology (KAUST), Red Sea Research Center, Thuwal 23955-6900, Kingdom of Saudi Arabia

2 Biology Department, University of Dammam (UOD), Dammam 31441-1982, Kingdom of Saudi Arabia

This manuscript was accepted in Scientific Reports

73

Abstract

The formation of nodes, stem elongation and the phenology of stunted Avicennia marina was examined in the Central Red Sea, where Avicennia marina is at the limit of its

distribution range and submitted to extremely arid conditions with salinity above 38 psu and water temperature as high as 35◦C. The annual node production was rather uniform

among locations averaging 9.59 node y-1, which resulted in a plastocron interval, the

interval in between production of two consecutive nodes along a stem, of 38 days.

However, the internodal length varied significantly between locations, resulting in growth

differences possibly reflecting the environmental conditions of locations. The reproductive

cycle lasted for approximately 12 months, and was characterized by peak flowering and

propagule development in November and January. These phenological observations

provide a starting point for research and restoration programs on the ecology of

mangroves in the Central Red Sea, while the plastochrone index reported here would allow

calculations of the growth and production of the species from simple morphological

measurements.

Keywords: nodes, buds, flowers, propagules, temperature, humidity, and mangroves.

74

Introduction

Mangroves are woody trees and shrubs that occupy the intertidal zone in the tropics and subtropics (Tomlinson 1986, Feller, Lovelock et al. 2010, Hogarth 2015), where they contribute to primary production (Duke 2011) and provide a range of ecosystem services

(Barbier, Hacker et al. 2011). Forming lush vegetation in the wet tropics, mangroves are often stunted in regions with limited freshwater runoff, including karstic (Lara-Domínguez,

Day Jr et al. 2005) and arid (Cintrón, Lugo et al. 1978) regions.

Avicennia marina has the widest latitudinal as well as longitudinal distribution among mangroves(Tomlinson 1986). Thus Avicennia marina occupies coastal areas across a broad range of environmental conditions, therefore suggesting that the species must have considerable growth plasticity(Morrisey, Swales et al. 2010). Avicennia marina is the dominant mangrove species in the Red Sea(El-Juhany 2009), where its growth is limited due to lack of rivers and freshwater input, low nutrient supply, high salinity, and hot summer air temperature (Mandura 1997). Indeed, Avicennia marina are stunted in the

Central Red Sea (about 2 to 3 m in height), compared to 16 m tall trees reported in

Australia(Mackey 1993).

The Red Sea represents, therefore, an end member in terms of the combination of growth conditions to support Avicennia marina growth. However, the phenology and growth dynamics of Avicennia marina in the Red Sea has only been assessed in a single study, reporting the reproductive cycle of the Southern Red Sea Coast of Saudi

Arabia(Mandura, Saifullah et al. 1987). Indeed, there is a paucity of information on the phenology of the species, with a limited number of reports worldwide (Wium-Andersen 75 and Christensen 1978, Burns and Ogden 1985, Duke 1990, Clarke and Myerscough 1991,

Clarke 1994, Hegazy 1998, Ochieng and Erftemeijer 2002, Homer 2009, Wang’ondu, Kairo et al. 2010, Robert, Jambia et al. 2014). This is a surprising gap, as this species plays a major role as a source of food and nursery grounds, as an intense carbon sink and in offering coastal protection in tropical coastlines around the world (Polidoro, Carpenter et al. 2010).

The harsh conditions in the Red Sea, particularly in summer, may affect the phenology and growth of Avicennia marina, which may deviate from those reported elsewhere.

Here we report the phenology and growth dynamics of monospecific stands of stunted

Avicennia marina in the Central Red Sea. In particular, we estimate their annual growth, including the node and branch production as well as the phenology of their reproductive stages: budding, flowering, and fruiting. The phenology of this species is an essential biological trait required to underpin ecological research on this important species and to plan restoration projects, as well as to provide a baseline to assess responses to climate change (Poloczanska, Brown et al. 2013). Whereas the internodal production of the congeneric species Avicennia alba has been reported for SE Asia at 17.6 nodes per year

(Duarte, Thampanya et al. 1999), that of Avicennia marina has not been reported anywhere

along its extended range.

Methods

1. Internodal measurements in Central Red Sea mangroves

Measurements of internodal length sequences, which allow inferences on the growth

dynamics of mangroves (Gill and Tomlinson 1971), were conducted in five contrasting

Central Red Sea mangrove stands, supporting different levels of human impacts (Fig. 1).

Mangroves in the Thuwal area were cleared by the construction of the King Abdullah 76

University for Science and Technology (KAUST) in 2007 and there was a new plantation program in 2010 that allows us to validate the age estimates of the mangroves. Both

Thuwal-Island and Khor Alkharar are areas undisturbed by human activities. On the contrary, Petro Rabigh suffers industrial impacts from petrochemical activities and the

Economic city area is impacted by the construction of the city (Fig. 1).

Figure 1: Study sites in Central Red Sea. The sites are located in the kingdom of Saudi

Arabia. The map was produced with ArcMap Version 10.2. Background map credits: the

World Administrative Divisions layer provided by Esri Data and Maps, and DeLorme 77

Publishing Company. Redistribution rights are granted http://www.esri.com/~/media/Files/Pdfs/legal/pdfs/redist_rights_103.pdf?la=en.

We reconstructed mangrove growth dynamics based on the cycles in internodal length

(Duarte, Thampanya et al. 1999). To study internodal growth, we selected 48 trees in Khor

Alkharar, 75 trees in Economic city, 50 trees in Petro Rabigh, 52 trees in Thuwal-Island and

30 trees in Thuwal-Kaust (total, n=255 tree) to assess the internode production and length

and derive annual growth. The main criteria to select the trees was the possibility to clearly

follow the main axis. Where possible we counted and measured the nodes along the main

axis from the apical meristem to the node that could be safely identified nearest to the

hypocotyl (Fig. 2). Overall, the average height of the selected trees was 184.7 cm with a size

distribution ranging from 105 cm to 370 cm. The internodal series measured for individual

trees assessed averaged 33 internodes, with a maximum of 51 internodes in Thuwal-Kaust

and a minimum of 17 internodes in Khor Alkharar.

78

Figure 2: An illustration of counting nodes and measuring internodal length. Photo by H.A and the artist work by I.Gromhico.

79

Mangrove plants have been shown to produce a fixed number of nodes annually, and to display an annual cycle of internodal length that can be used to discern the number of nodes they produce annually from records of internodal length along the main axis of mangrove branches and saplings using the reconstruction technique formulated by Duarte et al. (1999)(Duarte, Thampanya et al. 1999). Because multiple effects have been found to

add noise to the sequences of internodal lengths in mangrove trees resulting in both short

term (e.g. storms) or long term (e.g. interannual variations)(Duarte, Thampanya et al.

1999), we applied a long-pass and a short-pass filter to remove such noise for the

internodal measurements, thereby highlighting variability at annual scales. The long-term

filter involved subtracting the sequences of raw internodal length measurements from

values corresponding to a running average of n= 14 internodes centered at the specific

node being filtered to remove long term variability. Then the resulting values were filtered

for short-term noise by subtracting a running average of n=3 internodes centered at the

specific node being filtered. For fifty (out of 255) trees the n=3 or n=4, depending on the

amount of short-term noise, the node production was calculated for each location after

counting the number of nodes between consecutive maxima along the filtered sequence

and averaging the values for each tree (Fig. 3)(Duarte, Thampanya et al. 1999), and then

averaging those for each population examined. The annual growth was calculated as the result of the sum of internodal lengths during each annual cycle. In general, the sequences of internodal lengths collected along stems allowed the identification of four to five phenological years, thereby also allowing to assess growth variability over the past 4 to 5 years.

80

One annual

Figure 3: Examples of standaraized internodal length for a selected plants in each study site in the Central Red Sea. 81

2. Estimating plastochrone interval

To directly examine the plastochrone interval (the average time in between the production of a node (Gill and Tomlinson 1971), thirty A. marina propagules collected from Thuwal-Island were sprouted in February 2014 and grown in a nursery with brackish water for 91 days. We monitored the number of nodes produced over time by the seedlings grown in a nursery (Almahasheer 2016). We then calculated the plastochrone interval (as

the average ratio between the time elapsed and number of nodes produced) to validate the

estimate derived from the reconstructive analysis of growth cycles.

3. Monthly phenological observations

We monitored monthly between January 2014 and May 2015 a total of 96 axillary

shoots distributed over 32 trees in two different stands 5 kilometers apart (Thuwal-Island

and Thuwal-Kaust). These shoots were tagged and we counted the number of sub-branches with buds, flowers and propagules in each shoot, to link the Avicennia marina phenology to weather we used data from the Presidency of Meteorology and Environment (PME) in the

Kingdom of Saudi Arabia. We also calculated the annual branching production from the time series of the number of branches along each of the tagged stems using the same filters that we used earlier for the node production in the main axis of the tree, because according to (Duke and Pinzon 1992) counting the nodes between apical meristem to the hypocotyl is often equal from the hypocotyl to the axillary shoot. Seventeen (out of 96) branches were excluded from the analysis because of failure to detect a clear annual growth cycle.

4. Statistical analysis

Statistical analyses, including descriptive statistics, general linear models to test the differences between trees and locations, as well as Student's t-test and Tukey HSD posthoc 82 test to assess pairwise differences were carried out using JMP, a computer program for statistics developed by the SAS Institute.

Results

All of the examined trees showed very clear interannual cycles of internodal length

along their main axis when the data were filtered to remove long-term and short-term

variability (Fig. 3). The node production ranged from 9.3 to 9.8 node y-1 and did not differ

significantly among populations (F =0.93 and P=0.44, Fig. 4), resulting in a range of

plastochrone intervals of 39.20 to 37.90 days. There were no significant differences in the

number of internodes produced per year between locations (Tukey HSD post hoc test, P <

0.05, Fig. S1), which averaged 9.59±0.08, N=255 across stands, corresponding to an

average plastochrone interval of 39 days. This result was validated by the results derived

from the seedlings grown in the nursery, for which the plastochrone interval was

calculated to be 38 days. Further, validation was provided by the annual production of total

sub-branches in the axillary shoot, which produced on average 9.44 ± 0.13 nodes sub- branch y-1 (N=71), corresponding to a plastochrone interval of 39.20 ± 0.56 days, without

significant differences between the two locations (F =0.0576, P<0.0001 and Student’s t-test,

P < 0.05, Fig. 5, Fig. S2). 83

Figure 4. Number of nodes produced by the main axis of Avicennia marina trees

(nodes/year) in five stands sampled in the Central Red Sea.

84

Figure 5: Number of sub-branches produced annually by Avicennia marina trees in two stands sampled in the Central Red Sea (Thuwal).

In contrast to the annual node production the internodal length varied significantly between locations (F =15.85 and P<0.0001, Fig. 6), with plants at Thuwal-Kaust developing significantly longer internodes, followed by both Thuwal-Island and Khor Alkharar, then

Petro Rabigh, and finally Economic city (Tukey HSD post hoc test, P < 0.05, Fig. 6, Fig. S3).

As a result, annual mangrove stem elongation varied across sites (F=9.10 and P<0.0001), with a higher annual growth in Thuwal-Kaust, followed by Thuwal-Island, Khor Alkharar and Petro Rabigh, and then the lowest growth in Economic city (Tukey HSD post hoc test, P

< 0.05, Table. 1). However, no significant differences in node production or internodal length were observed over the years where these properties were regenerated (F=0.66 and

P=0.5745).

85

Figure 6: Elongation rate (cm y-1) of the main axis of Avicennia marina trees in five

populations sampled in the Central Red Sea.

Table 1: Mean ± SE (n) of the internodal length (cm y-1) of the main axis of Avicennia

marina trees over five consecutive phonological years in five populations sampled in the

Central Red Sea. The annual growth was calculated as the sum of internodal lengths

produced during each annual cycle. The letters correspond to Tukey posthoc HSD multiple

comparison testing for significant differences between years for each location, the same

letters means no significant differences within one location (P > 0.05).

Location 1st 2nd 3rd 4th 5th Average (2014-2015) (2013-2014) (2012-2013) (2011-2012) (2010-2011) Total, cm Khor Alkhararab 41.99±2.75 46.24±3.68 28.50±4.87 34.50±13.00 42.69 (48)a (26)a (3)a (2)a (79) Economic cityc 26.70±1.71 28.59±2.17 27.11±2.71 29.51±7.59 27.54 (75)a (57)a (19)a (6)a (157) Petro Rabighb 38.29±2.37 43.03±2.98 36.22±5.10 29.4 (1)a 39.71 (50)a (35)a (10)a (96) Thuwal-Islandab 43.85±2.92 45.09±2.88 50.29±5.60 52.63±20.88 45.34 (52)a (43)a (15)a (2)a (112) *Thuwal-kaust a 49.35±4.38 48.44±3.39 54.56±6.02 44.75±19.75 49.84 (30)a (26)a (13)a (2)a (71) *the measurements in Thuwal-Kaust were conducted one year earlier than those at other locations

Bud production was initiated in June and the reproductive cycle concluded with the

release of propagules in April (Fig. 7), thereby elapsing over a total 10 months. Propagules

development lagged flowering by about 3 months, which is the approximate time elapsed

from pollination to propagules maturation (Fig. 7). The buds peaked in October with an

average (±SE) of 50.8 ± 5.1% of the sub-branches producing them, followed by a flowering

peak in November, with 25.1±4.4% productive sub-branches flowering. Propagules 86 production peaked in January, with 39.4±5.0 % productive sub-branch (Fig. 7). Bud and flower production are initiated at the time of the summer solstice (20-23 June), when temperature is highest, but the peak of buds and flower production occur in the Autumn, when atmospheric humidity is highest (Figs. 8).

Figure 7: Phenology patterns and mean percent reproductive branches in a Avicennia marina populations in the Central Red Sea (Thuwal). 87

Figure 8. Climate diagram for temperature, humidity and rainfall in the Central Red Sea.

88

Discussion

The stem elongation of Avicennia marina seedlings provides a record of internodal length that allows the reconstruction of its growth patterns and calculation of the plastochrone interval, a key property to convert biological into chronological time (Gill and

Tomlinson 1971), consistent with findings for other mangrove species (Duarte, Thampanya et al. 1999). The annual node production estimated from the analysis was 9.59 nodes y-1,

corresponding to an average plastochrone interval of 38 days. This estimate is consistent

with that derived from the observation of annual sub-branch production of 39 days, and

that directly observed for seedlings growing in a nursery of 37 days. Hence, the 38 days

plastochrone interval provides a robust estimate to convert number of internodes into time

elapsed, which can be used to derive mangrove seedling stem elongation and production.

This is important as the plastochrone interval for Avicennia marina had not been resolved

to date, despite being a key mangrove species with a broad distribution (Smith 2013). The

production of 9.59 nodes y-1 for Avicennia marina in the Central Red Sea, is within the

range of reports for other species, spanning from 3.8 nodes y-1 for Rhizophora seedlings in

Panama (Duke and Pinzon 1992) to 30.3 nodes y-1 for Sonneratia caseolaris in Thailand

(Thampanya, Vermaat et al. 2002) (Table 2). Moreover, Avicennia marina node production

is below that for the cognatic Asian species Avicennia alba (Table 2).

89

Table 2: Reported number of nodes produced annually along the main axis by different mangrove species around the world.

Species Nodes y-1 Location Reference

Avicennia alba 13.2 Thailand (Thampanya, (transplanted) Vermaat et al. 2002) Avicennia alba 17.6 SE Asia (Duarte, (natural) Thampanya et al. 1999) Avicennia marina 9.59 Central Red Sea This study (natural) Avicennia marina 9.44 Central Red Sea This study (natural) branch y-1 Avicennia marina 9.63 Central Red Sea This study (planted) Rhizophora sp 5.4* Caribbean coast of Panama (Duke and (planted) Pinzon 1992) nodes shoot-1 y-1 Rhizophora sp 5.5 Philippines (Padilla, (natural) Fortes et al. 2004) Rhizophora apiculata 7.3 SE Asia (Duarte, (natural) Geertz- Hansen et al. 1998) Rhizophora apiculate 8.03 SE Asia (Duarte, (natural) Thampanya et al. 1999) Rhizophora mucronata 6.5 Thailand (Thampanya, (transplanted) Vermaat et al. 2002) Sonneratia caseolaris 28.8 SE Asia (Duarte, (natural) Thampanya et al. 1999) Sonneratia caseolaris 30.3 Thailand (Thampanya, (transplanted) Vermaat et al. 2002) *reclculated from 3.8 ± 0.3 to 7.0± 0.6 90

Contrary to node production, which was conserved across A. marina stands in the

Central Red Sea, the internodal length varied between locations, resulting in annual changes in stem growth and production. Whereas the plastochrone interval is believed to be under genetic control (Gill and Tomlinson 1971), as supported by the absence of significant differences among the stands studied here, the plasticity in the length of internodes across stands suggests that this trait is under environmental control, possibly including freshwater inputs and nutrient limitation, which contributes to the dwarfing of

Avicennia(Alongi 2011), such as that the species displays in the Red Sea. Indeed, recent experiments found iron to be limiting the growth of A. marina seedlings in Central Red

Sea(Almahasheer 2016). Duke (Duke 1990) reported that the timing and success of

Avicennia reproductive cycle increase by a factor of two or three for each additional 10 °C

up to a maximum temperature of about 34 °C. Therefore the low growth rates in the Red

Sea are unlikely to be attributable to high temperature alone. Either temperature or

photoperiod determines the onset of the budding and flowering season, as increasing

temperature with peak solar radiation triggers flower development and high humidity is

conducive to propagules development. Confirming the role of photoperiod, temperature

and humidity in A. marina phenology requires, however, experimental validation, as these are strongly correlated in the field. In addition, seawater salinity increases from the south at about 36.5 psu near the strait of Bab el Mandeb, to the north at about 41-40 psu, at the southern tip of the Sinai Peninsula(Alraddadi 2013). Yet, the salinity in a coastal lagoon near Jeddah varied between 40.5 psu in April and 41.8 psu in August(Al-Barakat 2010) .

Therefore, the low growth rate might be attributed to the high salinity as well. 91

Even though tree production may vary interannually, there seems to be a balanced breeding system keeping a productive population with strong recruitment and dispersal potential (Clarke and Myerscough 1991). The variability in the reproductive branches of A. marina in Central Red Sea (i.e. when comparing Jan 2014 and Jan 2015) is typical for plants, where individual branches do not flower in consecutive years but branches in the same tree are offset in the biannual cycle so the tree flowers and produces propagules every year

(Homer 2009). Fruits follows a similar pattern which suggests a limitation in the number of shoot axes able producing them (Clarke 1994).

The duration of the reproductive cycle of A. marina in this study at 22°N (approximately

12 months) is similar to that of A. marina in the southeast coast of Australia (approximately

12 months) at 35°S (Clarke and Myerscough 1991), and longer than the reproductive cycle reported in Kenya at 4°S, where bud initiation to fruit falling lasted 6 to 8 months(Wang’ondu, Kairo et al. 2010). Because generally flowering/fruiting cycles are connected to latitude (Duke, Bunt et al. 1984) with the floral initiation being later with increasing latitude (Clarke 1994) it has been suggested that the reproduction of Avicennia marina is less successful at higher latitudes because of the shorter summer decreasing the growth period (Duke 1990).

In conclusion, A. marina mangroves in the Central Red Sea produce about 9.59 nodes y-1

per axis and have a reproductive period of 12 months. The node production did not vary

between locations whereas internodal length varied significantly between locations,

resulting in growth differences probably reflecting the environmental conditions of each

location (i.e. the limitation in nutrient supply and fresh water inputs). The results

presented on the basic phenology of Avicennia marina in the Central Red Sea provide an 92 important baseline for research on the ecology of the species and is, therefore, an important resource for many applications, including planning of restoration projects, which require the availability of propagules and, therefore, should be informed by knowledge of the reproductive phenology of the species. In addition, the plastochrone index of Avicennia marina in the Central Red Sea reported here would allow calculations of the growth and production of the species from simple morphological assessments, as demonstrated by

Duarte et al. (1999)(Duarte, Thampanya et al. 1999), therefore allowing rapid assessments

of growth conditions and plant performance for monitoring purposes.

Acknowledgements

The research reported in this paper was supported by King Abdullah University of Science

and Technology (KAUST) in the Kingdom of Saudi Arabia. We thank the Costal and Marine

Resources core lab in KAUST Nabeel Alikunhi and Zenon Batang, as well as the Red Sea

Research Centre Joao Curdia and Katherine Rowe for helping in the field. And we thank

Ivan Gromhico, KAUST, for the artist work on Fig. 2. And the Presidency of Meteorology and

Environment (PME) in the Kingdom of Saudi Arabia for providing the weather data.

Author Contributions statement

H.A., C.M.D and X.I designed the study, H.A. carried out the field measurements, H.A. and

C.M.D did the statistical analysis and H.A., C.M.D and X.I. wrote the manuscript.

Competing financial interests

The author(s) declare no competing financial interests. 93

References

Al-Barakat, A. (2010). "Some Hydrographic Features of Rabigh Lagoon along the Eastern Coast of the Red Sea." Marine Scienes 21(1).

Almahasheer, H. (2016). "Ecosystem Services of Avicennia marina in the Red Sea." P.hD. Dissertation. King Abdullah University of Science and Technology.

Alongi, D. (2011). "Early growth responses of mangroves to different rates of nitrogen and phosphorus supply." Journal of Experimental Marine Biology and Ecology 397(2): 85-93.

Alraddadi, T. (2013). Temporal changes in the Red Sea circulation and associated water masses, University of Southampton, Ocean and Earth Science, PhD Thesis, 198 pp.

Barbier, E. B., S. D. Hacker, C. Kennedy, E. W. Koch, A. C. Stier and B. R. Silliman (2011). "The value of estuarine and coastal ecosystem services." Ecological Monographs 81(2): 169-193.

Burns, B. and J. Ogden (1985). "The demography of the temperate mangrove [Avicennia marina (Forsk.) Vierh.] at its southern limit in New Zealand." Australian journal of ecology 10(2): 125-133.

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

Clarke, P. and P. Myerscough (1991). "Floral biology and reproductive phenology of Avicennia marina in south-eastern Australia." Australian Journal of Botany 39(3): 283-293.

Clarke, P. J. (1994). "Base-Line Studies of Temperate Mangrove Growth and Reproduction; Demographic and Litterfall Measures of Leafing and Flowering." Australian Journal of Botany 42(1): 37-48.

Duarte, C. M., O. Geertz-Hansen, U. Thampanya, J. Terrados, M. D. Fortes, L. Kamp-Nielsen, J. Borum and S. Boromthanarath (1998). "Relationship between sediment conditions and mangrove Rhizophora apiculata seedling growth and nutrient status." Marine Ecology Progress Series (MEPS) 175: 277-283.

Duarte, C. M., U. Thampanya, J. Terrados, O. Geertz‐Hansen and M. D. Fortes (1999). "The determination of the age and growth of SE Asian mangrove seedlings from internodal counts." Mangroves and Salt Marshes 3(4): 251-257.

Duke, N. (1990). "Phenological trends with latitude in the mangrove tree Avicennia marina." The Journal of Ecology: 113-133.

Duke, N. (2011). Mangrove Islands. Encyclopedia of Modern Coral Reefs. D. Hopley, Springer Netherlands: 653-655. 94

Duke, N., J. Bunt and W. Williams (1984). "Observations on the floral and vegetative phenologies of north-eastern ." Australian Journal of Botany 32(1): 87-99.

Duke, N. C. and Z. S. M. Pinzon (1992). "Aging Rhizophora seedlings from leaf scar nodes: a technique for studying recruitment and growth in mangrove forests." Biotropica: 173-186.

El-Juhany, L. (2009). "Present status and degradation trends of mangrove forests on the southern Red Sea coast of Saudi Arabia." American-Eurasian Journal of Agricultural and Environmental Science 6(3): 328-340.

Feller, I. C., C. Lovelock, U. Berger, K. McKee, S. Joye and M. Ball (2010). "Biocomplexity in mangrove ecosystems." Annual Review of Marine Science 2: 395-417.

Gill, A. M. and P. B. Tomlinson (1971). "Studies on the growth of red mangrove (Rhizophora mangle L.) 3. Phenology of the shoot." Biotropica: 109-124.

Hegazy, A. K. (1998). "Perspectives on survival, phenology, litter fall and decomposition, and caloric content of Avicennia marina in the Arabian Gulf region." Journal of arid environments 40(4): 417-429.

Hogarth, P. J. (2015). The biology of mangroves and seagrasses, Oxford University Press.

Homer, L. E. (2009). "Population structure and distance of gene flow in Avicennia marina (Forsk.) Vierh.(Avicenniaceae) on a local/regional scale in the Northern Rivers of New South Wales, Australia."

Lara-Domínguez, A. L., J. W. Day Jr, G. V. Zapata, R. R. Twilley, H. A. Guillén and A. Yanez- Arancibia (2005). "Structure of a unique inland mangrove forest assemblage in fossil lagoons on the Caribbean Coast of Mexico." Wetlands Ecology and Management 13(2): 111- 122.

Mackey, A. (1993). "Biomass of the mangrove Avicennia marina (Forsk.) Vierh. near Brisbane, south-eastern Queensland." Marine and Freshwater Research 44(5): 721-725.

Mandura, A. (1997). "A mangrove stand under sewage pollution stress: Red Sea." Mangroves and Salt marshes 1(4): 255-262.

Mandura, A., S. Saifullah and A. Khafaji (1987). "Mangrove Ecosystem of Southern Red Sea Coast of Saudi Arabia." Proc.Saudi Biol.Soc 10: 165-193.

Morrisey, D. J., A. Swales, S. Dittmann, M. A. Morrison, C. E. Lovelock, C. M. Beard and R. Gibson (2010). The ecology and management of temperate mangroves, CRC Press, Boca Raton(USA).

Ochieng, C. A. and P. L. Erftemeijer (2002). "Phenology, litterfall and nutrient resorption in Avicennia marina (Forssk.) Vierh in Gazi Bay, Kenya." Trees 16(2-3): 167-171. 95

Padilla, C., M. Fortes, C. Duarte, J. Terrados and L. Kamp-Nielsen (2004). "Recruitment, mortality and growth of mangrove (Rhizophora sp.) seedlings in Ulugan Bay, Palawan, Philippines." Trees 18(5): 589-595.

Polidoro, B. A., K. E. Carpenter, L. Collins, N. C. Duke, A. M. Ellison, J. C. Ellison, E. J. Farnsworth, E. S. Fernando, K. Kathiresan, N. E. Koedam, S. R. Livingstone, T. Miyagi, G. E. Moore, V. Ngoc Nam, J. E. Ong, J. H. Primavera, S. G. Salmo, J. C. Sanciangco, S. Sukardjo, Y. Wang and J. W. Yong (2010). "The loss of species: mangrove extinction risk and geographic areas of global concern." PLoS One 5(4): e10095.

Poloczanska, E. S., C. J. Brown, W. J. Sydeman, W. Kiessling, D. S. Schoeman, P. J. Moore, K. Brander, J. F. Bruno, L. B. Buckley and M. T. Burrows (2013). "Global imprint of climate change on marine life." Nature Climate Change 3(10): 919-925.

Robert, E. M., A. H. Jambia, N. Schmitz, D. J. De Ryck, J. De Mey, J. G. Kairo, F. Dahdouh- Guebas, H. Beeckman and N. Koedam (2014). "How to catch the patch? A dendrometer study of the radial increment through successive cambia in the mangrove Avicennia." Annals of botany 113(4): 741-752.

Smith, T. J. (2013). Forest Structure

Tropical Mangrove Ecosystems, American Geophysical Union: 101-136.

Thampanya, U., J. E. Vermaat and C. M. Duarte (2002). "Colonization success of common Thai mangrove species as a function of shelter from water movement." Marine Ecology Progress Series 237: 111-120.

Tomlinson, P. (1986). The botany of mangroves. Cambridge tropical biology series, Cambridge University Press, Cambridge.

Wang’ondu, V., J. Kairo, J. Kinyamario, F. Mwaura, J. Bosire, F. Dahdouh-Guebas and N. Koedam (2010). "Phenology of Avicennia marina (Forsk.) Vierh. in a disjunctly-zoned mangrove stand in Kenya." Western Indian Ocean Journal of Marine Science 9(2): 135-144.

Wium-Andersen, S. and B. Christensen (1978). "Seasonal growth of mangrove trees in southern Thailand. II. Phenology of Bruguiera cylindrica, Ceriops tagal, Lumnitzera littorea and Avicennia marina." Aquatic Botany 5: 383-390.

96

Supplementary Materials

Table S 1: Annual node production y-1 for each single interannual cycle

Khor Economic Petro Thuwal- Thuwal- Alkarar city Rabigh island kaust

18 18 15 16 15 17 17 Distribution 14 16 15 16 14 13 15 14 15 13 14 14 12 13 12 13 13 11 12 11 12 12

10 11 11 11 10

10 10 9 10 9 9 9 8 9 8 8 8 8 7 7 7 7

6 6 7 6 6 5 5 5 6 5 4 4 Mean 9.40 9.46 9.10 9.53 9.77 Median 9 9 9 10 10 Std Err Mean 0.15 0.11 0.14 0.12 0.16 Upper 95% Mean 9.70 9.69 9.38 9.79 10.10 Lower 95% Mean 9.09 9.24 8.81 9.28 9.44 N 166 321 182 226 139

Table S 2: Sub branching production for one single interannual cycle

Thuwal-island Thuwal-kaust Distribution 14 13 13 12 12 Tab 11 11

10 10 le S 9

9 8

7 8 3:

6

Mean 9.45 9.45 Inte Median 9 9.5 Std Err Mean 0.16 0.18 rno Upper 95% Mean 9.78 9.83 Lower 95% Mean 9.11 9.07 dal N 40 46 len gth y-1 for each single interannual cycle 97

Khor Economiccity Petro Thuwal- Thuwal- Alkarar Rabigh island kaust

120 110 100 90 90 Distribution 100 90 80 80 100 90

80 70 70 80 80 70 60 60 70

60 50 50 60 60

50 40 40 50

40 40 30 30 40

30 20 30 20 20 20 20 10 10

10 0 10 0 Mean 40.36 26.92 38.95 44.39 49.56 Median 37.75 23.9 37.1 42 45.8 Std Err Mean 1.68 1.01 1.44 1.51 2.10 Upper 95% Mean 43.70 28.91 41.79 47.37 53.72 Lower 95% Mean 37.02 24.92 36.10 41.40 45.41 N 130 266 153 191 123

98

Chapter Three Nutrient Limitation of Central Red Sea Mangroves

Hanan Almahasheer1,2,, Carlos M. Duarte1 and Xabier Irigoien1

1 King Abdullah University of Science and Technology (KAUST), Red Sea Research Center, Thuwal 23955-6900, Kingdom of Saudi Arabia

2 Biology Department, University of Dammam (UOD), Dammam 31441-1982, Kingdom of Saudi Arabia

This manuscript was accepted in Frontiers, Marine Science.

99

Abstract

As coastal plants that can survive in salt water, mangroves play an essential role in large marine ecosystems (LMEs). The Red Sea, where the growth of mangroves is stunted, is one of the least studied LMEs in the world. Mangroves along the Central Red Sea have characteristic heights of ~2 m, suggesting nutrient limitation. We assessed the nutrient status of mangrove stands in the Central Red Sea and conducted a fertilization experiment

(N, P and Fe and various combinations thereof) on four-week-old seedlings of Avicennia marina to identify limiting nutrients and stoichiometric effects. We measured height, number of leaves, number of nodes and root development at different time periods as well as the leaf content of C, N, P, Fe and Chl a in the experimental seedlings. Height, number of nodes and number of leaves differed significantly among treatments. Iron treatment resulted in significantly taller plants compared with other nutrients, demonstrating that iron is the primary limiting nutrient in the tested mangrove population and confirming

Liebig’s law of the minimum: iron addition alone yielded results comparable to those using complete fertilizer. This result is consistent with the biogenic nature of the sediments in the

Red Sea, which are dominated by carbonates, and the lack of riverine sources of iron.

Keywords: Avicennia marina, Ratios, Chl a content, Carbon, Nitrogen, Phosphorous, and

Iron.

100

Introduction

Mangroves play a key role in coastal ecosystems as a source of food, nursery grounds, and a carbon sink as well as for coastal protection (Polidoro, Carpenter et al.

2010). Mangroves often grow in river mouths and deltaic areas, where they receive abundant nutrient supply from riverine discharge, but they also grow in areas devoid of riverine inputs, such as carbonate shores in the Caribbean and islands in South East Asia and the Red Sea, where they are often nutrient limited and, as a result, have a dwarf stature

(Feller 1995, Duarte, Geertz-Hansen et al. 1998, Lovelock, Feller et al. 2004). In particular, nitrogen, phosphorus, or iron is often reported to limit mangrove growth, while other essential nutrients are abundant in seawater, making them available to mangroves (Alongi

2011). Indeed, low availability of nitrogen, phosphorus, and iron has been suggested as the cause for the absence of mangroves along shorelines that are otherwise suitable for mangrove growth (Sato, Negassi et al. 2011). Mangrove fertilization experiments have demonstrated site-specific nutrient limitation of nitrogen and phosphorous (Feller 1995,

Koch and Snedaker 1997, Feller, McKee et al. 2003, Feller, Whigham et al. 2003, Lovelock,

Feller et al. 2004, Lovelock, Feller et al. 2006, Naidoo 2009, Alongi 2011), whereas an experimental study conducted on mangroves in north Queensland, Australia, demonstrated iron deficiency (Alongi 2010). Previous work has not considered these three nutrients together, which is the focus of this work.

Mangroves cover ~135 km2 of the Red Sea coastline (Almahasheer, Aljowair et al.

2016), providing natural coastal vegetation along an otherwise arid topography that is devoid of rivers and significant rainfall. A recent assessment (Almahasheer, Aljowair et al.

2016) established that the area of the Red Sea covered by mangroves has increased by 12% 101

since 1972. High concentrations of nutrients are mainly found in the southern (Churchill,

Bower et al. 2014) and northern (Triantafyllou, Yao et al. 2014) reaches of the Red Sea and

generally taper in the central area (Eshel and Naik 1997). The Central Red Sea is the most

oligotrophic part of the basin, which is clearly evident in the Chl a concentrations there,

which are the lowest in the Red Sea (Raitsos, Pradhan et al. 2013). The monsoons and

prevailing regional winds play a key role in driving the mechanisms that supply nutrients

and consequently regulate the seasonal and interannual variability of Chl a concentrations

in the Red Sea (Raitsos, Yi et al. 2015). The lack of riverine input and negligible

precipitation mean that the main nutrient source is the monsoonal-driven horizontal intrusion of nutrient-rich waters from the Indian Ocean (Murray and Johns 1997, Johns and

Sofianos 2012, Churchill, Bower et al. 2014, Raitsos, Yi et al. 2015). The intruding water flows northward along the African and Arabian coasts of the Red Sea, hardly reaching the

Central Red Sea (Churchill, Bower et al. 2014). Additional nutrients are supplied by sub- surface mixing below the nutricline in the Northern Red Sea (Triantafyllou, Yao et al. 2014,

Yao, Hoteit et al. 2014) and dust deposition (Brindley, Osipov et al. 2015).

The low nutrient concentrations in the Red Sea resulting from the absence of riverine inputs and limited hydrographic exchange with the Indian ocean suggest that any mangroves growing there will be nutrient limited (Thompson, Field et al. 2013). Indeed, the dwarf status of the trees, with an approximate height of 2 to 3 m in the Central Red Sea area (Almahasheer unpubl. results) is characteristic of nutrient-limited mangroves (Krauss,

Lovelock et al. 2008). Low nutrient inputs from the land in this area may also result in nutrient limitation. Alternatively, it has been suggested that the reduced size of mangroves could result from other stresses, such as the extremely high temperature and salinity (Yao, 102

Hoteit et al. 2014) characteristic of the Red Sea region (Douabul and Haddad 1970).

However, mangroves are comparatively taller in the southern area of the Red Sea where salinity is higher than the ocean average, sea surface temperature is higher than in the

Central Red Sea and nutrient concentrations are much higher than in the central area,

(Mandura, Saifullah et al. 1987, Mandura 1997), indicating that nutrient availability is the primary factor limiting growth.

Although mangroves support a very tight nutrient economy to avoid nutrient loss

(Feller 1995), their reproduction is a major vector for nutrient loss as the seedlings are dispersed away from the mother tree. Effective reproduction relies on developing propagules that contain enough nutrients to support the survival and dispersal of seedlings

(McKee 1995). The nutrient status of the seedling is particularly critical and affects the tree’s resistance to light, high salinity and floods (Koch 1997). Acute nutrient limitation, likely experienced by mangrove trees in the Central Red Sea, may reduce the mother tree’s capacity to supply sufficient nutrients to the propagule, thus hindering seedling growth and performance. The assessment of nutrient limitation during seedling growth in mangroves is therefore particularly important.

Here, we assess the nutrient status monospecific stands of Avicennia marina trees in the ultra-oligotrophic Central Red Sea and experimentally examine the response of seedlings to the addition of nutrients. We ascertain if nutrient deficiency in the mother trees affects the capacity of propagules to fully support early growth in seedlings.

Specifically, we evaluate the nutrient status of mangrove trees by assessing nutrient concentrations in leaves across a wide range of stands in the Central Red Sea area, from pristine stands to stands receiving nutrient inputs from industrial activity. We also 103 experimentally test the responses of Avicennia marina seedlings to the addition of nutrients

(N, P and Fe and various combinations) to assess which nutrients limit their growth and development.

Methods

1. Study area

The Red Sea has about 135 Km2 of mangroves distributed up to 28.207302oN, the northern biogeographical boundary of mangroves. Unlike the coverage of mangroves in other regions, the area covered by mangroves in the Red Sea remains relatively stable

(Almahasheer, Aljowair et al. 2016). Nutrient inputs to the Red Sea are dominated by inputs from the Indian Ocean (Murray and Johns 1997, Johns and Sofianos 2012, Churchill,

Bower et al. 2014, Raitsos, Yi et al. 2015), leading to a gradient of oligotrophication toward the north (Ismael 2015), concurrent with an increase in salinity due to high evaporation losses (Talley 2002). On the other hand, the sea surface temperature declines from south to north (Rushdi 2015). The Central Red Sea is an arid environment characterized by high temperatures (Supplementary Materials Table S1 presents the weather conditions during the experiment), sparse rainfall (Edwards 1987); the mean annual (sporadic) rainfall in

Jeddah is 55 mm), high salinity (Bruckner 2011) and low nutrient inputs and concentrations (Mandura 1997). The tidal range on the Central Red Sea is only 20 to 30 cm

(Sultan, Ahmad et al. 1996), resulting in mangrove habitats that develop as a narrow

(typically one to three trees wide) fringe along the shore, adjacent to sand flats that may occasionally be flooded, with desert further inland. The mangrove fringe supports small halophytes. Recent urban and industrial development of the Saudi shore in the Central Red 104

Sea area has resulted in locally elevated nutrient inputs, providing an opportunity to assess

how such inputs may affect the nutrient status of mangroves.

The sampled sites in the study area (Fig. 1) were selected to capture the diversity of

environments in the region. In particular, we sampled mangrove leaves from four locations

in the Central Red Sea, including Thuwal Island, Economic City lagoon, Petro Rabigh and

Khor Alkharar (Table 1, Fig. 1). Away from anthropogenic sources of nutrients, the

mangrove stands at Khor Alkharar, Economic City lagoon and Thuwal Island are pristine:

King Abdullah Economic City, about 40 km south of Rabigh, is a new and fast-growing urban development; Khor Alkharar lagoon lies on the coastal plain northwest of Rabigh and is connected to the adjacent Red Sea by a narrow inlet on its northwestern side (Gheith and

Abou-ouf 1996, Al-Farawati 2011); on Thuwal Island, Avicennia marina grows on shallow

soils of weathered coral (Balk, Keuskamp et al. 2015). In contrast, the city of Rabigh hosts a

large petrochemical complex and the mangrove stand are likely influenced by the industrial

environment. 105

Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative

Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company. 106

2. Nutrients status in Central Red Sea mangrove stands

We assessed the nutrient status of mangrove stands in the Red Sea by examining the concentrations of nitrogen, phosphorous and iron, the nutrients most often reported to limit mangrove growth (Feller 1995, Koch and Snedaker 1997, Feller, McKee et al. 2003,

Feller, Whigham et al. 2003, Lovelock, Feller et al. 2004, Lovelock, Feller et al. 2006, Naidoo

2009, Alongi 2011). We analyzed nutrient concentrations in mature leaves as well as the stoichiometric ratios of the nutrients to carbon. Whereas examination of leaf nutrient concentration alone does not suffice to establish nutrient budgets (Lü, Freschet et al. 2012), leaf nutrient concentration robustly indicates the nutrient status of plants (Duarte 1990,

Duarte 1992), including mangroves (Duarte, Geertz-Hansen et al. 1998, Feller, Whigham et al. 2003). Moreover, leaf nutrient concentration is an important ecosystem property because it affects the decomposition rates of leaf litter (Enriquez, Duarte et al. 1993).

Deeming that quadrant or transect sampling would not achieve the representativeness of the leaf nutrient concentration required by this study, we used a nested sampling strategy to characterize leaf nutrient concentration at each location. First, we selected two to three disjoint stands at each location. From each of these stands, we randomly selected one tree and sampled 6 to 10 leaves from the main stem of the tree. In total, this yielded 15 to 30 leaves from each location (Table 1). We collected the mangrove leaves in March 2015, following propagule release, which causes significant nutrient loss and may render the plants vulnerable to nutrient limitation.

107

Table 1: Mean (± SE) for nutrient concentrations (mmol g DW-1) in Avicennia marina leaves from four different locations in the Central Red Sea. R2 and F values correspond to an

ANOVA that tested for significant differences between locations. * P between 0.01 and

0.05, ** P < 0.01. Nutrients linked with the same letter did not differ significantly among

themselves (Tukey HSD multipile comparision post-hoc test, P > 0.05).

Location N C N P Fe C: N: P : Fe Rows (Atomic) Khor Alkarar 15 36.99±0.35a 0.76±0.06ab 0.005±0.001b 2017 C:39 0.021±0.002a N:1 P:0.3 Fe Economic city 30 35.43±0.17b 0.76±0.05ab 0.008±0.001a 2004 C:36 0.023±0.002a N:1 P:0.5 Fe Petro Rabigh 27 36.20±0.16a 0.96±0.08a 0.010±0.001a 1837 C:40 0.030±0.004a N:1 P:0.5 Fe Thuwal 19 36.68±0.29a 0.67±0.04b 0.008±0.001ab 1816 C:29 0.025±0.002a N:1 P:0.4 Fe R2 0.24 0.12 0.05 0.12 F ratio location 9.21** 4.06** 1.52ns 4.29**

3. Seedling Fertilization Experiments

We conducted two experiments to examine the response of seedlings to the addition of nutrients. The experiments were designed to test whether propagules collected from nutrient-deficient stands on Thuwal Island contained enough nutrients to support seedling growth or if their growth and development could be increased by nutrient input. In addition, we identified the limiting nutrient by using various combinations of treatments of the three nutrients. Two varied treatment experiments (see below) were conducted, the first in 2014 and the second in 2015. The experiments were conducted in a nursery 108 covered with a green mesh to reduce incoming solar radiation by around 90 % (Quantum

Meter MQ-100, µmol PAR m-2 s-1) . The average radiation in the nursery (mean ± SE = 8.3 ±

2.5 % of the incoming radiation) was slightly greater than that received under natural growing conditions in Thuwal (5.5 ± 1.7 %, Table 2). The plants were exposed to solar radiation and temperature regimes comparable to those in the Thuwal Island mangrove stand where the propagules were produced. We carefully controlled nutrient conditions in both the sediment and water. In particular, we washed the sediments and diluted the seawater to reduce external nutrient inputs in the control treatment, thereby measuring seedling growth as a response to added nutrients in the different treatments. The control treatment thus captures the growth and development capacity of Avicennia marina seedlings as determined by the nutrients allocated by the mother plant to the propagules.

Accordingly, the response of the seedling to various nutrient treatments indicates which nutrients are deficient in the propagules relative to those required to support nutrient- sufficient growth of the resulting seedlings because propagule production is likely to be a major vector for nutrient loss in the mangrove stands studied.

Table 2: Incoming solar radiation to natural growing mangrove and our experiment

Nursery Natural Time Inside Outside % Time shaded sunny % 7:50 AM 23 249 9.2 11:30 AM 70.0 1700.0 4.1 7:50 AM 21 246 8.5 11:30 AM 100.0 1670.0 6.0 7:50 AM 20 251 8.0 11:30 AM 95.0 1690.0 5.6 12:00PM 150 1733 8.7 11:30 AM 160.0 1840.0 8.7 12:00PM 147 1657 8.9 11:30 AM 110.0 1750.0 6.3 12:00PM 178 1788 10.0 11:30 AM 90.0 1800.0 5.0 5:30 PM 15 227 6.6 11:30 AM 50.0 1910.0 2.6 5:30 PM 16 220 7.3 11:30 AM 96.4 1765.7 5.5 5:30 PM 16 219 7.3 Average 65.1 732.2 8.3 Average 96.4 1765.7 5.5

109

The nursery was fitted with raceways periodically flooded with brackish water using pumps to simulate the tidal cycle (see below). The experiments were initiated in February

2014 and March 2015, using Avicennia marina seedlings germinated from propagules collected from healthy mature trees on Thuwal Island. The collected propagules were fully formed, with a hard pericarp and an average dry weight of 2.98 ± 0.12 g DW (n=33). Five propagules were dried for nutrient analysis. The raceways were washed with 1% acid followed by nutrient-free water to avoid contamination before the propagules were germinated. The propagules were soaked in brackish water for three days until their covers shed naturally. All the propagules used in the experiment germinated on the same day. The sediment used to accommodate the seedlings in the raceway consisted of sand collected from Thuwal that was washed with nutrient-free water to reduce the concentrations of available nutrients. The control treatment was thereby rendered nutrient depleted. The brackish water was a mixture of equal volumes of seawater collected near Thuwal and nutrient-depleted fresh water collected from King Abdullah University of Science and

Technology (KAUST), which was found to provide suitable growth conditions (Connor

1969, Clough 1984). Previously, between September and November 2013, we conducted a preliminary test of the growth response of seedlings to various salinity regimes (0, 25, 50,

75, 100% seawater) and confirmed that a 50:50 mixure of Red Sea seawater and fresh water provided the best growth conditions. Indeed, this finding is consistent with the occurrence of Red Sea mangroves near groundwater inputs in the Red Sea; many mangrove stands are located near Wadis (dry river beds with associated groundwater discharge). The raceways were flooded with brackish water by pumps for 6 hours daily, following (Clarke and Johns 2002). The excess water then slowly flowed back into the tanks located under 110 the raceways (Fig. 2). This flooding simulated the intertidal cycle while also avoiding excess algal growth in the raceways. Furthermore, to avoid water loss due to evaporation and evapotranspiration, we replaced the water in the tanks every month with freshly prepared brackish water as described above. Tanks, valves, and raceways were washed monthly during the experiment.

Propagules were planted at half height in the sediment (Bovell 2011) in individual pots.

The experimental treatments were initiated 4 weeks after planting. By this time, the seedlings were equal in height and had developed 4 leaves each (Clough 1984, Ball, Chow et al. 1987). We used 30 replicated seedlings per treatment in the first experiment and 21 replicated seedlings per treatment in the second experiment, with the treatments applied independently to each replicated seedling.

The first experiment aimed at assessing the roles of nitrogen, phosphorus and iron, alone and in combination, in promoting seedling growth beyond the growth capacity corresponding to the nutrients received from the mother plant, whereas the second experiment aimed at separating the possible effects of iron and the chelant used in combination in the first experiment, which was found to significantly enhance seedling growth (see Results). A chelant is an organic ligand incorporating metal ions in a structure that facilitates metal uptake by plants (Salt, Smith et al. 1998, Hart 2000). Specifically,

Ethylenediaminetetraacetic acid (EDTA) has been found to facilitate Fe uptake. We used the nutrient concentrations of (Ball, Chow et al. 1987) for nitrogen and phosphorus and

(Steiner and van Winden 1970) for Fe-EDTA. The first experiment, initiated in March 2014, included the following final nutrient concentrations in six treatments applied to the brackish water used to flood the raceways: (1) nitrogen (as 2 mmol NH4NO3 L-1); (2) 111

phosphorus (0. 2 mmol NaH2PO4 L-1); (3) iron (0. 2 mmol Fe-EDTA L-1); (4) nitrogen,

phosphorous, and iron, each supplied at the preceding concentrations; and (5) a complete

fertilizer, including 4 mmol Ca (NO3)2 L-1, 0. 2 mmol L-1 NaH2PO4, 1 mmol L-1 MgSO4, 0. 2

mmol Fe-EDTA L-1, 2 mmol NH4NO3 L-1, 0.025 mmol H3BO3 L-1, 0.002 mmol MnSO4 L-1, 0.002

mmol ZnSO4 L-1, 0.0005 mmol CuSO4L-1, 0.0005 mmol H2MoO4 L-1, and 0.025 mmol MgCl2 L-

1; and (6) a control without nutrient addition. In the second experiment, the following four

treatments were tested: (1) iron (0. 2 mmol FeCl2, L-1); (2) EDTA (0. 2 mmol EDTA L-1); (3)

Fe + EDTA (0.2 mmol Fe-EDTA L-1); and (4) the control without iron or EDTA addition.

Nutrient concentrations in the brackish water were adjusted monthly. The nutrient

concentrations used in the treatments were higher than in the water in a lagoon near

Thuwal, where the (mean ± SE) concentrations of NO2 were 0.062±0.006 µmol L-1 and

NO2+NO3 were 0.181±0.017 µmol L-1 (Banguera-Hinestroza, Eikrem et al. 2016). They were

also higher than values found in coastal water from Duba (north) and Jazan (south) along

the Saudi coast of the Red Sea (NO3 = 2.351± 0.386 µmol L-1 and PO4 = 0.239± 0.039µmol L-

1; (Pearman, Kürten et al. 2016).

112

A

Control Fe Fe+P+N Mixed Nutrients N P

B

Control Fe EDTA FeEDTA

Figure 2: Experimental setup for (A) the experiment testing for N, P and Fe limitation, alone and in combination conducted in 2014, and (B) the experiment testing for the role of iron vs. the chelant used in the first experiment (EDTA) conducted in 2015. The colors representing the treatments are used for the same treatment in all figures. 113

4. Measurements and chemical analysis

The following parameters were measured once per month in the first experiment and

every month and a half in the second experiment: (a) C, N, P, Fe and Chl a concentrations in

the first pair of fully developed apical leaves of three seedlings per treatment; and (b)

growth rate, as determined by the elongation rate along the main meristem, number of

nodes, number of leaves and root development for each seedling in the experiment. Chl a

was extracted from 1-cm diameter discs cut from fresh leaves using a QIAGEN TissueLyser

II and extracted overnight in 10 ml 80% acetone. The Chl a concentration in the extract was

measured spectrophotometrically based on absorbance at 663 nm following (Wellburn

1994). The other apical leaf was washed with nutrient-

oven and ground for carbon and nutrient analyses. Carbonfree water, and driednitrogen at 60˚ concentrations C in a drying

were determined on 2 mg samples of powderized leaf material wrapped in a pre-

combusted aluminum capsule using a FLASH 2000 CHNS Analyzer (Zimmermann, Keefe et

al. 1997). Iron and phosphorous concentrations were determined after digesting 0.50 mg of

the same powderized leaf with 5 ml concentrated HCL and a few drops of H2O2 in a Digi

PREP digestion system, following three temperature steps: 30˚C for 30 min, 50˚C for 30-Q min.water and and 75˚C analyz fored 45 by min. Inductively The samples Coupled were Plasma allowed-Optical to cool, Emission diluted Spectrometry to 25 ml with (Varian Mili

Inc. model 720-ES).

Iron and phosphorous analyses were run in duplicate. Two different standards

(Inorganic Ventures and PerkinElmer’s Pure Plus) were used to assess the accuracy of the

measurements, with recovery rates of 100.1% for iron and 99.1% for phosphorous (n=10)

in the first experiment (2014) and 98.8% for iron (n=10) in the second (2015). The 114 recovery was, on average, 98.7% for phosphorous and 104.9% for iron (n=20) in the analyses of leaves collected in the survey (n=30). We also used a standard reference material from The National Institute of Standards and Technology (NIST) to verify the accuracy of the phosphorous and iron concentrations, resulting in 94.6% accuracy in apple leaves and 81.9% accuracy in peach leaves for the second experiment and 73.3% accuracy in apple leaves and 79.2% accuracy in peach leaves in the analyses run for the leaves collected in the survey. To analyze carbon and nitrogen concentrations, we used a standard reference material (BBOT) and a calibration standard (Sulfanilamide) to obtain an average recovery of 99% for nitrogen and 102% for carbon (n=14) for the first experiment and an average of 101.1% for nitrogen and 100.7% for carbon for the leaves collected in the survey (n=30).

5. Statistical analysis.

Seedling growth rates were calculated from the slopes of the fitted linear regressions on the relationship between the development metrics and time. Statistical analysis, including descriptive statistics, linear regression analyses, general linear models to test for the effects of the different experimental treatments or differences among stands, and Tukey's HSD

(honest significant difference) posthoc test to assess pairwise differences were carried out using JMP, a computer program for statistical analyses developed by SAS Institute.

Results

Nutrient concentrations in Avicennia marina leaves were low (N < 1.5 %, P < 0.09 %,

Fe < 0.06) and differed significantly between locations (ANOVA, P < 0.05, Table 1). The plants at Petro Rabigh exhibited increased concentrations of N, P and Fe in the leaves compared with the other populations sampled. Those in Khor Alkharar Lagoon exhibited 115

the lowest leaf nutrient concentrations (except for nitrogen in the leaves collected from

leaves collected from Thuwal Island; Table 1). The C: N: P: Fe ratios were extremely high,

indicative of acute nutrient depletion relative to carbon (Table 1).

The propagules collected in Thuwal were characterized by a low nutrient

concentration compared to the seedlings (C = 32.61±0.74, N = 0.86±0.10, P = 0.035±0.001,

Fe = 0.0006±0.0001 mmol g DW-1), indicating extreme iron depletion relative to the leaves

(930 C: 25 N: 1 P: 0.02 Fe, n=5).

All seedlings survived the first experiment, but three seedlings died in the EDTA

treatment and one in the Fe + EDTA treatment in the second experiment. There were

significant differences between treatments in both experiments in terms of seedling height,

number of leaves and nodes after three months. Seedling growth differed significantly with

treatments in the first experiment (R2=0.62, F=49.75, P<0.0001), with the treatment receiving iron reaching the largest size, even exceeding that in the treatment receiving complete nutrients (Fig. 3a). We also found significant differences in seedling growth in the second experiment (R2=0.68, F =47.21, P<0.0001), with the seedlings fertilized with iron alone reaching significantly higher heights than those receiving EDTA or Fe+EDTA and those under control conditions (Fig. 3b). Hence, iron addition consistently led to a fast growth rate in both experiments (Tukey HSD post hoc test, P < 0.05., Table 3).

Comparable significant results were found in the number of nodes produced during the first experiment (R2=0.73, F =82.1, P<0.0001), where all treatments involving

iron addition produced significantly more nodes than the control. Node production in the

second experiment was slightly higher in the EDTA and Fe+EDTA treatments (R2=0.57, F 116

=29.10, P<0.0001) than those receiving iron alone and the control (Fig. 3a, b and Table 3,

Tukey HSD post hoc test, P < 0.05).

The number of leaves produced also differed significantly among treatments but was significantly higher in the treatment receiving nitrogen in the first experiment

(R2=0.71, F =73.92, P<0.0001) and the treatment receiving Fe+EDTA in the second experiment (R2=0.62, F =36.98, P<0.0001) (Fig. 3a, b and Table 3, Tukey HSD post hoc test,

P < 0.05).

Root development (as g DW seedling-1) increased under iron and phosphorous addition at the end of the first experiment (R2=0.62, F =3.50, P=0.0008, Tukey HSD post hoc test, P < 0.05 Fig. 4a and Table 3). Root development was significantly higher in the iron treatment in both experiments (R2=0.59, F =9.38, P<0.0001, Tukey HSD post hoc test, P <

0.05 Fig. 4b and Table 3).

Iron addition led to higher leaf Chl a concentration in the first experiment (R2=0.76,

F =7.68, P<0.0001, Tukey HSD post hoc test, P < 0.05, Fig. 5a), whereas leaf Chl a concentration in the second experiment was higher in both EDTA and FeEDTA treatments than in the iron treatment alone and in the control (R2=0.85, F =30.24, P<0.0001, Tukey

HSD post hoc test, P < 0.05, Fig. 5 b).

117

Figure 3: Mean (± SE) for height, number of nodes and leaves of mangrove seedlings over time under control and different nutrient addition treatments across experiments testing for (A) N, P and Fe addition and (B) components of response to iron addition (Fe vs. EDTA). 118

Table 3: Mean (± SE) growth rate (cm seedling-1 day-1), leaf and node production rate

(number of leaves and nodes seedling-1 day-1), and root development (g DW seedling-1day-

1) of Red Sea mangrove seedlings under different nutrient addition treatments. The slopes

of the fitted linear regressions are between seedling height and time. Treatments linked

with the same letter did not differ significantly among themselves (Tukey HSD multipile

comparision post-hoc test, P > 0.05).

Treatment Height Leaf Nodes Root production production development 1st Control 0.213±0.016bc 0.052±0.004b 0.026±0.002ab 0.004±0.002a experiment P 0.126±0.018c 0.022±0.003c 0.011±0.002c 0.005±0.003a N 0.160±0.020bc 0.090±0.008a 0.017±0.002b -0.005±0.004a Fe 0.272±0.020a 0.064±0.003b 0.032±0.002a 0.012±0.004a Fe+P+N 0.270±0.018ab 0.072±0.003ab 0.035±0.001a 0.006±0.002a Mixed 0.268±0.019b 0.069±0.003b 0.035±0.001a -0.003±0.004a Nutrients 2nd Control 0.204±0.021bc 0.040±0.004a 0.020±0.002a 0.014±0.002ab experiment Fe 0.249±0.023a 0.044±0.004a 0.022±0.002a 0.018±0.004a EDTA 0.193±0.032c 0.052±0.007a 0.025±0.004a 0.013±0.003ab FeEDTA 0.242±0.027b 0.052±0.005a 0.026±0.003a 0.004±0.004b

119

Figure 4: Mean (± SE) for root development of mangrove seedlings over time under control and different nutrient addition treatments across experiments testing for (A) N, P and Fe addition and (B) components of response to iron addition (Fe vs. EDTA). 120

Figure 5: Mean (± SE) for leaf Chl a concentration over time of fully developed mangrove apical leaf under control and different nutrient addition treatments across experiments testing for (A) N, P and Fe addition and (B) components of response to iron addition (Fe vs.

EDTA). Bars showing different letters within a sampling event identify significantly different treatments (P < 0.05), as indicated by post Tukey’s HSD multiple comparison test. 121

Nutrient addition resulted in significant differences in carbon, nitrogen and

phosphorus concentrations in the leaves (R2=0.89 and F =18.70; R2=0.93 and F= 30.95; and

R2=0.78 and F =7.84, respectively, P<0.0001), with treatments receiving mixed nutrients

exhibiting higher leaf nitrogen and phosphorous concentrations. The iron concentration

differed significantly between treatments and time (R2=0.53, F= 2.39 and P=0.0137,

Supplementary Materials Table S2) in the first experiment, but it was significantly higher in

the plants receiving only iron compared with the EDTA, Fe + EDTA and control treatments

in the second experiment (R2=0.38, F= 4.18 and P=0.0188, Supplementary Materials Table

S3). The leaf carbon concentration declined sharply when leaf nitrogen, phosphorous and

iron concentrations reached levels below limiting thresholds (N < 2 mmol N g DW-1, P <

0.04 mmol P g DW-1, and Fe < 0.001 mmol Fe g DW-1 (Fig. 6).

The C:N ratio increased over time in plants receiving P, followed by the control and

Fe treatments (Fig. 7). The C:P ratio was relatively high, indicative of phosphorous limitation in the control, iron, nitrogen and mixed nutrient treatments. The C:Fe ratio was high and decreased in the control, mixed nutrient and nitrogen treatments, whereas treatment with Fe, P and Fe+P+N exhibited an increase followed by a subsequent decrease

(Fig. 7). The N:P ratio showed no significant changes in the control and iron treatments, although the N:P ratio increased over time in the treatments receiving nitrogen and mixed nutrients. Both N:Fe and P:Fe ratios decreased over time in the control, mixed nutrient, nitrogen and phosphorous treatments and increased and then decreased subsequently in the Fe+P+N and Fe treatments (Fig. 7). 122

Figure 6: The relationship between carbon and nutrient concentration (mmol nutrient g-1

DW) in fully developed apical leaves of Avicennia marina of seedlings. The solid line represents the reciprocal regression equation fitted across data derived from all treatments combined (mmol C = 34.071879 - 7.4611856*Recip mmol N), (mmol C = 123

36.763021 - 0.2475941*Recip mmol P) and (mmol C = 26.25834 + 0.0034299*Recip mmol

Fe). 124

Figure 7: Mean (± SE) nutrients stoichiometric ratios over time of fully developed apical leaves receiving different nutrient additions (n=3 ). The dashed line represents the value for propagules. 125

Discussion

1. Nutrient concentration and stoichiometric ratios

Mangrove stands in the Central Red Sea are characterized by low leaf nutrient concentrations and low carbon-to-nutrient stoichiometric ratios, with an overall mean of

1918 C: 36 N:1 P: 0.5 Fe, indicative of severe nutrient depletion, particularly P and Fe, across stands and pointing to the likelihood of nutrient limitation of Central Red Sea mangroves. However, there were important differences among stands, with mangroves receiving nutrient inputs from industrial and urban sources at Petro Rabigh showing the highest concentrations of nitrogen, phosphorus and iron across stands. Fe deficiency was particularly acute in the propagules collected to initiate the seedling experiments, which had P:Fe ratios that were 10 to 20 fold higher than those of leaves, indicative of extreme Fe deficiency in these propagules. Unfortunately, C: N: P: Fe ratios for balanced mangrove growth have not yet been defined. Nevertheless, C: N: P ratios for nutrient sufficient A. marina growth can be inferred from those in the fertilized treatments reported by Alongi

(Alongi 2011), which correspond to 346 C: 14 N: 1 P, similar to the treatments receiving P in our experiments, which had the lowest C:N:P ratio. Even the control plants, which did not receive any nutrient additions, in the stand examined by Alongi (Alongi 2011) had a

C:N:P ratio of 711:30:1, well below the ratios found in leaves of Central Red Sea mangroves.

The stoichiometric ratios in A. marina seedlings receiving Fe+N+P in our experiments can be considered to be nutrient-sufficient plants. These ratios are quite similar to those for nutrient-sufficient macroalgae and N-fertilized Avicennia marina plants in Northern

Queensland, Australia, but are higher than those for nutrient-sufficient seagrass and phytoplankton (Table 4). 126

Table 4: Atomic stoichiometric ratios across marine primary producers compared to Avicennia marina.

Redfield ratio C: N: P: Fe Reference (Atomic) Macroalgae (median values) 800 C:49 N: 1 P (Duarte 1992) Phytoplankton 106 C:16 N:1 P (Redfield 1963) Marine Macroalgae and Seagrasses 550 C:30 N:1 P (Atkinson and Smith 1983) Seagrass 474 C:24 N:1 P (Duarte 1990) Avicennia marina leaves in in north Queensland 732 C:52 N:1 P (Alongi 2011) Australia (receiving N additions, 50 mmol N m d ) −2 Avicenni−1 marina leaves in in north Queensland 346 C:14 N:1 P (Alongi 2011) Australia (receiving P additions, 10 mmol p m d ) −2 Kenya−1 Avicennia marina green leaves 1927 C:59 N:1 P (Ochieng and Erftemeijer 2002) Kenya Avicennia marina Senescent leaves 3869 C:52 N:1 P (Ochieng and Erftemeijer 2002) Terrestrial plants leaves 1212 C:28 N:1 P (McGroddy, Daufresne et al. 2004) Terrestrial plants litter 3007 C:45 N:1 P (McGroddy, Daufresne et al. 2004) Red Sea Avicennia marina leaves 790 C:50 N:1 P:0.1 This study (experimentally, Fe+N+P additions at 125 day) Fe Red Sea Avicennia marina leaves 1918 C:36 N:1 P: 0.5 This study (wild stands) Fe Red Sea Avicennia marina propagules 930 C: 25 N: 1 P: This study (Thuwal Island) 0.02 Fe

2. Nutrient requirements and critical concentrations

The experimental results allow a first approximation of the critical level of nutrient concentrations required to support nutrient-sufficient growth of Avicennia marina seedlings. These were defined as the nutrient concentrations below which carbon concentrations declined steeply (cf. Fig. 6). The inferred thresholds are 2 mmol N g DW-1, 127

0.04 mmol P g DW-1, and 0.001 mmol Fe g DW-1. These critical thresholds are higher than

the thresholds of nitrogen and phosphorus reported for seagrass (N < 1.2 mmol N g-1, P <

0.06 mmol P g-1 , recalculated from (Duarte 1990)).

Our results provide compelling evidence that the growth of Avicennia marina seedlings

in the oligotrophic Central Red Sea requires environmental Fe inputs, as iron addition

alone led to a growth response and an increase in Chl a concentrations comparable with

those in the treatments also receiving nitrogen and phosphorous. Although our experiment

showed iron to be the main limiting factor for seedling growth, nitrogen addition

significantly increased the number of leaves while addition of phosphorous and

particularly iron led to increased root development. Yet, the experimental treatment that

combined N, P and Fe addition did not increase growth beyond that achieved by Fe

fertilization alone, substantiating Liebig’s law of the minimum, with iron as the limiting

nutrient. The results from the second experiment confirmed that it was indeed iron, and

not the ligand added in the first experiment (EDTA), which elicited the observed growth

responses.

The finding of iron-limited growth of Central Red Sea seedlings is consistent with

(Alongi 2010), who reported high iron requirements in the transition stage from seedling

to sapling in five mangrove species growing in north Queensland, Australia. Iron limitation

in mangrove growth in the Central Red Sea is expected, as iron limitation is characteristic

of marine coastal habitats dominated by biogenic carbonate sediments, which are typically

iron depleted (Duarte, Martín et al. 1995). Moreover, the lack of riverine inputs, which are the main vector of iron to coastal marine sediments (Poulton and Raiswell 2002, Raiswell

2006), also renders the Red Sea prone to iron limitation. Iron-limited seagrass growth has 128 also been reported in the Mexican Caribbean (Duarte, Martín et al. 1995), where the karstic landscape lacks riverine input and where mangroves are also stunted. Similar to our findings in the Red Sea, iron addition stimulated leaf Chl a concentration in Caribbean seagrasses receiving iron (Duarte, Martín et al. 1995).

3. Nutrient inputs to the Red Sea

In the absence of rivers and negligible precipitation, such as in the Red Sea, nutrient inputs from the Indian Ocean (Murray and Johns 1997, Johns and Sofianos 2012, Churchill,

Bower et al. 2014, Raitsos, Yi et al. 2015), sub-surface mixing (Triantafyllou, Yao et al. 2014,

Yao, Hoteit et al. 2014), and atmospheric sources are likely to be the most significant path for new nutrient input (Brindley, Osipov et al. 2015), including nitrogen fixation and inputs of phosphorus and iron with dust deposition (Aerts and Chapin 1999, Schroth, Crusius et al.

2009). The Red Sea receives one of the highest dust inputs of all the oceans (Jish Prakash,

Stenchikov et al. 2015), and mangrove canopies may intercept dust during periods of dust storms, thereby increasing iron supply above that corresponding to passive dust deposition. The solubility of atmospheric iron varies between 0.01% to 80% depending on acid deposition, with 95% of atmospheric iron derived from desert dust and the rest from combustion processes (Mahowald, Engelstaedter et al. 2009). Although data from the study area are lacking, seabird guano e.g. brown booby birds, Red-billed tropic-birds and Bridled terns may also be an important source of nutrients to Red Sea mangroves, which support abundant bird communities (Evans 1987). The results presented also suggest anthropogenic inputs to be significant in mangrove stands near developed areas, as exemplified by the mangroves near Petro Rabigh, which exhibited elevated concentrations of iron as well as nitrogen and phosphorous. 129

A range of mangrove fertilization experiments were previously conducted, mostly focused on nitrogen and phosphorous addition, but those exploring iron limitation have been scarce. The outcome of these experiments depend on location, with some mangrove stands shown to be limited by phosphorous (Feller 1995, Koch and Snedaker 1997, Feller,

McKee et al. 2003, Lovelock, Feller et al. 2004, Lovelock, Feller et al. 2006) and others to be limited by nitrogen (Feller, Whigham et al. 2003, Naidoo 2009). Prior to the present study, iron limitation in mangrove growth had been reported, on the basis of experimental addition, only in mangroves in north Queensland, Australia (Alongi 2010). The finding that mangroves in the Central Red Sea are nutrient-limited, as evidenced by low nutrient concentrations in their leaves and very low iron levels in the propagules they form, allows us to begin to understand the functioning and drivers of these important components of the

Red Sea ecosystem. Nutrient-limited plants are likely to re-absorb significant amounts of nutrients before shedding their leaves (Reef, Feller et al. 2010), suggesting that nutrient- limited mangroves may act as a sink rather than a source of nutrients to the ecosystem.

Moreover, the decomposition rate of plant litter decreases with increasing C:N ratios

(Enriquez, Duarte et al. 1993). Nutrient-limited Central Red Sea mangroves are thus likely to produce relatively recalcitrant litter, which may act as a carbon sink in mangrove sediments. Lastly, our results show that nutrient-deficient mangroves are likely to produce nutrient-poor propagules. The resulting seedlings require additional nutrient inputs to support high growth and survival. The absence of significant nutrient inputs may, thus, act as a bottle-neck in the stability of the population (Selvam 2007). Nutrient-deficient propagules may also occur in other nutrient-limited mangrove stands. This suggestion requires further testing. Increased anthropogenic nutrient emissions in the Central Red Sea 130

are likely to lead to increased mangrove growth and improved nutrient status with

consequences for the ecosystem. Nutrient limitation may also affect the capacity of

mangroves to respond to other stresses affecting the Red Sea, such as the high temperature

and lack of fresh water supply, as nutrient limitation would impose limits to the capacity of

the plants to synthesize proteins and other molecules involved in building resilience to

these stresses. Hence, nutrient limitation likely renders the plants more vulnerable to the

extreme environmental conditions in the Central Red Sea. However, responses to multiple

stresses have not yet been investigated as we lack information on the thermal thresholds of

Avicennia marina in the Central Red Sea as well as on the consequences of a low fresh water

supply for the performance of the plants. These responses, along with the implications of

nutrient limitation in modulating the resistance to stress, should be targeted by future

research.

Conclusion.

In conclusion, nutrient concentrations in Avicennia marina leaves point to the

presence of nutrient-deficient mangrove stands in the Central Red Sea, with particularly

low concentrations of phosphorus and iron. Moreover, experimental nutrient addition

confirms iron to be the primary limiting nutrient for Avicennia marina seedlings in the

Central Red Sea. The stunted nature of Red Sea mangroves, with heights typically below 2

m, is, therefore, likely the result of acute iron limitation, although only long-term

fertilization experiments can confirm this suggestion. The iron-limited nature of Red Sea

Avicennia marina stands is consistent with the biogenic nature of the sediments in the Red

Sea, dominated by carbonates, and the lack of riverine sources of iron. We submit that dust deposition may be the most important source of iron to undisturbed mangrove stands, 131 although anthropogenic nutrient inputs, which are still highly localized along the Red Sea, can also locally elevate iron and phosphorus concentrations in the receiving stands.

Acknowledgements

We thank KAUST workshops, Coastal and Marine Resources Core Labs, and Analytical core lab for help with sampling and analyses and the Presidency of Meteorology and

Environment (PME) for providing weather data. We also thank Joao Curdia for help with the nutrient addition experiment and Vincent Saderne for providing data on light levels under mangrove canopies in the region. We also thank Virginia Unkefer for reviewing the manuscript and Heno Hwang for his illustration of the nursery experiment. The research reported in this paper was supported by King Abdullah University of Science and

Technology.

Conflict of Interest Statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Author Contributions statement

HA, CD and XI designed the study, HA carried out the experimental and analytical measurements, HA, CD and XI conducted the statistical analysis and HA, CD and XI wrote the manuscript.

132

References

Aerts, R. and F. S. Chapin (1999). "The mineral nutrition of wild plants revisited: a re- evaluation of processes and patterns." Advances in ecological research 30: 1-67.

Al-Farawati, R. (2011). "Spatial and Seasonal Distribution of Total Dissolved Copper and Nickel in the Surface Coastal Waters of Rabigh, Eastern Red Sea, Saudi Arabia." Journal of King Abdulaziz University: Earth Sciences 22(1).

Almahasheer, H., A. Aljowair, C. M. Duarte and X. Irigoien (2016). "Decadal stability of Red Sea mangroves." Estuarine, Coastal and Shelf Science 169: 164-172.

Alongi, D. (2011). "Early growth responses of mangroves to different rates of nitrogen and phosphorus supply." Journal of Experimental Marine Biology and Ecology 397(2): 85-93.

Alongi, D. M. (2010). "Dissolved iron supply limits early growth of estuarine mangroves." Ecology 91(11): 3229-3241.

Atkinson, M. and S. Smith (1983). "C: N: P ratios of benthic marine plants1." Limnology and Oceanography 28(3): 568-574.

Balk, M., J. A. Keuskamp and H. J. Laanbroek (2015). "Potential Activity, Size, and Structure of Sulfate-Reducing Microbial Communities in an Exposed, Grazed and a Sheltered, Non- Grazed Mangrove Stand at the Red Sea Coast." Frontiers in microbiology 6.

Ball, M., W. Chow and J. Anderson (1987). "Salinity-induced potassium deficiency causes loss of functional photosystem II in leaves of the grey mangrove, Avicennia marina, through depletion of the atrazine-binding polypeptide." Functional Plant Biology 14(3): 351-361.

Banguera-Hinestroza, E., W. Eikrem, H. Mansour, I. Solberg, J. Cúrdia, K. Holtermann, B. Edvardsen and S. Kaartvedt (2016). "Seasonality and toxin production of Pyrodinium bahamense in a Red Sea lagoon." Harmful Algae 55: 163-171.

Bovell, O. (2011). "Guyana mangrove nursery manual." Guyana Mangrove Restoration Project, Georgetown, Guyana.

Brindley, H., S. Osipov, R. Bantges, A. Smirnov, J. Banks, R. Levy, P. Jish Prakash and G. Stenchikov (2015). "An assessment of the quality of aerosol retrievals over the Red Sea and evaluation of the climatological cloud‐free dust direct radiative effect in the region." Journal of Geophysical Research: Atmospheres 120(20).

Bruckner, A. (2011). Khaled bin Sultan Living Oceans Foundation habitat mapping and characterization of coral reefs of the Saudi Arabian Red Sea: 2006–2009.Final Report Part I Panoramic Press Phoenix. 133

Churchill, J. H., A. S. Bower, D. C. McCorkle and Y. Abualnaja (2014). "The transport of nutrient-rich Indian Ocean water through the Red Sea and into coastal reef systems." Journal of Marine Research 72(3): 165-181.

Clarke, A. and L. Johns (2002). "Mangrove nurseries: Construction, propagation and planting: Fisheries Guidelines." Department of Primary Industries, Queensland, Fish Habitat Guideline FHG 4.

Clough, B. (1984). "Growth and salt balance of the mangroves Avicennia marina (Forsk.) Vierh. and Rhizophora stylosa Griff. in relation to salinity." Functional Plant Biology 11(5): 419-430.

Connor, D. (1969). "Growth of grey mangrove (Avicennia marina) in nutrient culture." Biotropica: 36-40.

Douabul, A. and A. M. Haddad (1970). "The Red Sea and Yemen’s Red Sea Environments." Hassell and Assoc., AMSAT and UNOPS: 1-16.

Duarte, C. M. (1990). "Seagrass nutrient content." Marine ecology progress series. Oldendorf 6(2): 201-207.

Duarte, C. M. (1992). "Nutrient concentration of aquatic plants: patterns across species." Limnology and Oceanography 37(4): 882-889.

Duarte, C. M., O. Geertz-Hansen, U. Thampanya, J. Terrados, M. D. Fortes, L. Kamp-Nielsen, J. Borum and S. Boromthanarath (1998). "Relationship between sediment conditions and mangrove Rhizophora apiculata seedling growth and nutrient status." Marine Ecology Progress Series (MEPS) 175: 277-283.

Duarte, C. M., M. Martín and G. Margarita (1995). "Evidence of iron deficiency in seagrasses growing above carbonate sediments." Limnology and Oceanography 40(6): 1153-1158.

Edwards, F. J. (1987). "Climate and oceanography." The Red Sea: 45-68.

Enriquez, S., C. M. Duarte and K. Sand-Jensen (1993). "Patterns in decomposition rates among photosynthetic organisms: the importance of detritus C: N: P content." Oecologia 94(4): 457-471.

Eshel, G. and N. H. Naik (1997). "Climatological Coastal Jet Collision, Intermediate Water Formation, and the General Circulation of the Red Sea*." Journal of Physical oceanography 27(7): 1233-1257.

Evans, P. G. (1987). "Sea birds of the Red Sea." Red Sea: Key environments (Edwards, AJ and Head, SM eds.): 315-338.

Feller, I. C. (1995). "Effects of nutrient enrichment on growth and herbivory of dwarf red mangrove (Rhizophora mangle)." Ecological monographs: 477-505. 134

Feller, I. C., K. L. McKee, D. F. Whigham and J. P. O'Neill (2003). "Nitrogen vs. phosphorus limitation across an ecotonal gradient in a mangrove forest." Biogeochemistry 62(2): 145- 175.

Feller, I. C., D. F. Whigham, K. L. McKee and C. E. Lovelock (2003). "Nitrogen limitation of growth and nutrient dynamics in a disturbed mangrove forest, Indian River Lagoon, Florida." Oecologia 134(3): 405-414.

Gheith, A. M. and M. A. Abou-ouf (1996). "Textural characteristics, mineralogy and fauna in the shore zone sediments at Rabigh and Sharm Al-Kharrar, eastern Red Sea, Saudi Arabia." Marine Scienes-Ceased lssuerg 17(1): 1-2.

Hart, J. R. (2000). "Ethylenediaminetetraacetic acid and related chelating agents." Ullmann's Encyclopedia of Industrial Chemistry.

Ismael, A. A. (2015). Phytoplankton of the Red Sea. The Red Sea, Springer: 567-583.

Jish Prakash, P., G. Stenchikov, S. Kalenderski, S. Osipov and H. Bangalath (2015). "The impact of dust storms on the Arabian Peninsula and the Red Sea." Atmospheric Chemistry and Physics 15(1): 199-222.

Johns, W. E. and S. S. Sofianos (2012). "Atmospherically forced exchange through the Bab el Mandeb Strait." Journal of Physical Oceanography 42(7): 1143-1157.

Koch, M. S. (1997). "Rhizophora mangle L. Seedling Development into the Sapling Stage across Resource and Stress Gradients in Subtropical Florida1." Biotropica 29(4): 427-439.

Koch, M. S. and S. C. Snedaker (1997). "Factors influencing Rhizophora mangle L. seedling development in Everglades carbonate soils." Aquatic Botany 59(1): 87-98.

Krauss, K. W., C. E. Lovelock, K. L. McKee, L. López-Hoffman, S. M. Ewe and W. P. Sousa (2008). "Environmental drivers in mangrove establishment and early development: a review." Aquatic Botany 89(2): 105-127.

Lovelock, C., I. C. Feller, K. McKee, B. Engelbrecht and M. Ball (2004). "The effect of nutrient enrichment on growth, photosynthesis and hydraulic conductance of dwarf mangroves in Panama." Functional Ecology 18(1): 25-33.

Lovelock, C. E., I. C. Feller, M. C. Ball, B. M. Engelbrecht and M. L. Ewe (2006). "Differences in plant function in phosphorus‐and nitrogen‐limited mangrove ecosystems." New Phytologist 172(3): 514-522.

Lü, X. T., G. T. Freschet, D. F. Flynn and X. G. Han (2012). "Plasticity in leaf and stem nutrient resorption proficiency potentially reinforces plant–soil feedbacks and microscale heterogeneity in a semi‐arid grassland." Journal of Ecology 100(1): 144-150. 135

Mahowald, N. M., S. Engelstaedter, C. Luo, A. Sealy, P. Artaxo, C. Benitez-Nelson, S. Bonnet, Y. Chen, P. Y. Chuang and D. D. Cohen (2009). "Atmospheric iron deposition: Global distribution, variability, and human perturbations*." Annual Review of Marine Science 1: 245-278.

Mandura, A. (1997). "A mangrove stand under sewage pollution stress: Red Sea." Mangroves and Salt marshes 1(4): 255-262.

Mandura, A., S. Saifullah and A. Khafaji (1987). "Mangrove Ecosystem of Southern Red Sea Coast of Saudi Arabia." Proc.Saudi Biol.Soc 10: 165-193.

McGroddy, M. E., T. Daufresne and L. O. Hedin (2004). "Scaling of C: N: P stoichiometry in forests worldwide: implications of terrestrial Redfield-type ratios." Ecology 85(9): 2390- 2401.

McKee, K. L. (1995). "Interspecific variation in growth, biomass partitioning, and defensive characteristics of neotropical mangrove seedlings: response to light and nutrient availability." American Journal of Botany: 299-307.

Murray, S. P. and W. Johns (1997). "Direct observations of seasonal exchange through the Bab el Mandab Strait." Geophysical Research Letters 24(21): 2557-2560.

Naidoo, G. (2009). "Differential effects of nitrogen and phosphorus enrichment on growth of dwarf< i> Avicennia marina mangroves." Aquatic Botany 90(2): 184-190.

Ochieng, C. A. and P. L. Erftemeijer (2002). "Phenology, litterfall and nutrient resorption in Avicennia marina (Forssk.) Vierh in Gazi Bay, Kenya." Trees 16(2-3): 167-171.

Pearman, J., S. Kürten, Y. Sarma, B. Jones and S. Carvalho (2016). "Biodiversity patterns of plankton assemblages at the extremes of the Red Sea." FEMS microbiology ecology.

Polidoro, B. A., K. E. Carpenter, L. Collins, N. C. Duke, A. M. Ellison, J. C. Ellison, E. J. Farnsworth, E. S. Fernando, K. Kathiresan, N. E. Koedam, S. R. Livingstone, T. Miyagi, G. E. Moore, V. Ngoc Nam, J. E. Ong, J. H. Primavera, S. G. Salmo, J. C. Sanciangco, S. Sukardjo, Y. Wang and J. W. Yong (2010). "The loss of species: mangrove extinction risk and geographic areas of global concern." PLoS One 5(4): e10095.

Poulton, S. and R. Raiswell (2002). "The low-temperature geochemical cycle of iron: from continental fluxes to marine sediment deposition." American Journal of Science 302(9): 774-805.

Raiswell, R. (2006). "Towards a global highly reactive iron cycle." Journal of Geochemical Exploration 88(1): 436-439.

Raitsos, D. E., Y. Pradhan, R. J. Brewin, G. Stenchikov and I. Hoteit (2013). "Remote sensing the phytoplankton seasonal succession of the Red Sea." PLoS One 8(6): e64909. 136

Raitsos, D. E., X. Yi, T. Platt, M. F. Racault, R. J. Brewin, Y. Pradhan, V. P. Papadopoulos, S. Sathyendranath and I. Hoteit (2015). "Monsoon oscillations regulate fertility of the Red Sea." Geophysical Research Letters 42(3): 855-862.

Redfield, A. C. (1963). "The influence of organisms on the composition of sea-water." The sea: 26-77.

Reef, R., I. C. Feller and C. E. Lovelock (2010). "Nutrition of mangroves." Tree Physiology 30(9): 1148-1160.

Rushdi, A. I. (2015). Calcite and Aragonite Saturation States of the Red Sea and Biogeochemical Impacts of Excess Carbon Dioxide. The Red Sea, Springer: 267-279.

Salt, D. E., R. Smith and I. Raskin (1998). "Phytoremediation." Annual review of plant biology 49(1): 643-668.

Sato, G., S. Negassi and A. Z. Tahiri (2011). "The only elements required by plants that are deficient in seawater are nitrogen, phosphorous and iron." Cytotechnology 63(2): 201-204.

Schroth, A. W., J. Crusius, E. R. Sholkovitz and B. C. Bostick (2009). "Iron solubility driven by speciation in dust sources to the ocean." Nature Geoscience 2(5): 337-340.

Selvam, V., ed (2007). "Trees and shrubs of the Maldives." RAP Publication(12).

Steiner, A. A. and H. van Winden (1970). "Recipe for ferric salts of ethylenediaminetetraacetic acid." Plant physiology 46(6): 862.

Sultan, S., F. Ahmad and N. Elghribi (1996). "Sea level variability in the central Red Sea." Oceanologica acta 19(5).

Talley, L. D. (2002). "Salinity patterns in the ocean." Encyclopedia of global change. Volume: the earth system: physical and chemical dimensions of global environmental change (eds MacCracken MC, Perry JS): 629-640.

Thompson, L. R., C. Field, T. Romanuk, D. Ngugi, R. Siam, H. Dorry and U. Stingl (2013). "Patterns of ecological specialization among microbial populations in the Red Sea and diverse oligotrophic marine environments." Ecology and evolution 3(6): 1780-1797.

Triantafyllou, G., F. Yao, G. Petihakis, K. Tsiaras, D. Raitsos and I. Hoteit (2014). "Exploring the Red Sea seasonal ecosystem functioning using a three‐dimensional biophysical model." Journal of Geophysical Research: Oceans 119(3): 1791-1811.

Wellburn, A. R. (1994). "The spectral determination of chlorophylls a and b, as well as total carotenoids, using various solvents with spectrophotometers of different resolution." Journal of plant physiology 144(3): 307-313. 137

Yao, F., I. Hoteit, L. J. Pratt, A. S. Bower, P. Zhai, A. Köhl and G. Gopalakrishnan (2014). "Seasonal overturning circulation in the Red Sea: 1. Model validation and summer circulation." Journal of Geophysical Research: Oceans 119(4): 2238-2262.

Zimmermann, C. F., C. W. Keefe and J. Bashe (1997). Method 440.0: Determination of Carbon and Nitrogen in Sediments and Particulates of Estuarine/Coastal Waters Using Elemental Analysis, United States Environmental Protection Agency, Office of Research and Development, National Exposure Research Laboratory.

138

Supplementary Materials Table S 1: Data for the weather conditions during the experiment. Data were provided by the Presidency of Meteorology and Environment (PME)

Relative Air Temp Pressure

Humidity

Year

-

Min Max level level level Mean Month Wind Speed Rainfall total Rainfall Min AD Max AD Mean AD Min DB 12 Min DB Max DB 10 Min station Min station Max station station Max Mean station Mean station Mean level sea Mean Vapor Pressure Mean Vapor J-14 7 24.9 29.4 20.5 33 15 56 88 16 1011.4 1013.3 1017.3 1004.2 15 17.4 F-14 8 25.4 30.3 21.1 37 17.1 53 93 12 1009.7 1011.6 1015.5 1003.7 0 16.7 M-14 8 26.7 31.7 22.2 35 17.6 53 91 11 1008.3 1010.2 1016.4 1003.2 777.7 18.4 A-14 8 30.1 35.4 25.5 41 19 48 87 14 1006.7 1008.7 1011.7 1000.2 777.7 20 M-14 7 30.6 36.1 25.9 42.6 22.5 51 93 12 1005.2 1007.1 1011 1000.2 777.7 21.7 J-14 8 32.1 37.8 27 48 24 52 94 5 1002.4 1004.3 1007 997.7 0 23.7 J-14 8 33.9 39.4 29.1 45 27 46 87 14 1002.3 1004.2 1006.4 997.5 6 23.4 A-14 7 34 39 29.7 42.1 26 50 89 14 1001.3 1003.2 1005.2 997.8 0 26 S-14 8 32 36.5 28.5 39 26.5 58 90 29 1003.9 1005.8 1007.7 999 0 26.8 O-14 7 30.1 34.9 26.2 38 23.4 61 98 22 1008 1009.9 1011.7 1004 0 25.5 N-14 6 27.6 32.9 23.3 37.4 19.2 53 88 17 1010.2 1012.1 1013.9 1005.5 31 19.2 D-14 6 26.8 31.3 22.4 34 18 57 93 21 1012 1013.9 1015.5 1007.9 0 19.6 J-15 8 23.7 28.4 19.4 35 13.4 51 87 7 1013.8 1015.8 1019.9 1007.2 0 15.1 F-15 8 25.2 30 20.9 35 16.7 57 88 16 1010.4 1012.3 1015.6 1005 0 18 M-15 8 27.4 32.1 23.4 40.6 20.8 55 87 13 1008.7 1010.6 1013.1 1002.5 0 19.2 A-15 9 27.7 32 24.1 37 20.4 49 86 20 1008.1 1010 1016.4 1002.8 777.7 17.9 M-15 7 31.5 36.1 27.5 41.1 25 49 85 10 1005.1 1007 1009.4 1001.1 0 21.7 J-15 8 31.6 36.7 26.7 40 23.4 50 86 11 1003.2 1005.1 1007.8 999.3 0 22.2 J-15 8 33.5 39.4 27.8 43 25.5 45 91 9 1002 1003.9 1006.3 998.4 0 21.8 139

Table S 2: Mean (± SE) nutrient concentration (mmol g DW-1) in fully developed apical leaves (n=3) of Avicennia marina seedlings grown under different experimental nutrient addition treatments. R2 and F value correspond to an ANOVA testing for significant differences between treatments over time. * P between 0.01 and 0.05, ** P < 0.01. Treatments linked with the same letter did not differ significantly among themselves (Tuckey HSD multiple comparison post-hoc test, P > 0.05).

Time Treatment C (mmolg-1) N (mmolg-1) P (mmolg-1) Fe ( µmol g-1) 63 day Control 31.63±0.56ab 2.32±0.07cde 0.042±0.006bcd 0.9±0.12b (Jun) Fe 32.07±0.52ab 2.79±0.32bc 0.061±0.015ab 3.3±0.87a Fe+P+N 31.68±0.44ab 2.73±0.17bc 0.048±0.004abcd 1.6±0.5ab Mixed Nutrients 33.09±0.18a 3.93±0.11a 0.072±0.008a 1.1±0.13ab N 32.12±0.75ab 3.40±0.22ab 0.057±0.002ab 0.7±0.14ab P 32.46±0.53a 1.69±0.04efg 0.053±0.005abc 1.3±0.17ab 86 day Control 28.87±0.67abcd 1.47±0.06fg 0.027±0.004cd 1.1±0.0035ab (Jul) Fe 32.58±0.92a 2.24±0.05cdef 0.048±0.003abcd 1.1±0.048ab Fe+P+N 31.52±1.05ab 2.66±0.24bc 0.039±0.002bcd 0.7±0.056b Mixed Nutrients 31.68±0.28ab 2.70±0.09bc 0.044±0.002abcd 1.3±0.25ab N 31.66±0.10ab 2.56±0.37bcd 0.033±0.005bcd 1.1±0.42ab P 30.18±0.40abc 1.24±0.05gh 0.034±0.004bcd 0.9±0.18b Control 27.96±0.74abcd 1.13±0.03gh 0.023±0.002d 2.7±1.3ab Fe 27.86±1.78abcd 1.15±0.01gh 0.024±0.002d 1.3±0.16ab 125 day Fe+P+N 26.17±0.73cd 1.67±0.10efg 0.034±0.003bcd 1.7±0.19ab (Aug) Mixed Nutrients 26.91±1.35bcd 1.73±0.16defg 0.028±0.004cd 1.3±0.095ab N 23.86±1.40d 1.59±0.07efg 0.021±0.002d 1.3±0.3ab P 15.83±2.28e 0.44±0.08h 0.021±0.003d 1.6±0.11ab R2 0.89 0.93 0.78 0.53 F Ratio 18.70** 30.95** 7.84** 2.39* 140

Table S 3: Mean (± SE) for nutrients concentrations (mmol g DW-1) in the experimental fully developed apical leaves of

Avicennia marina seedlings comparing Iron with EDTA (n=3). R2 and F value correspond to an ANOVA testing for significant differences between treatments. * P between 0.01 and 0.05, ** P < 0.01. Treatments linked with the same letter did not differ significantly among themselves (Tuckey HSD multiple comparison post-hoc test, P > 0.05).

Time Treatment N Rows Fe (mmolg-1) Control 6 0.0019±0.0001ab EDTA 6 0.0014±0.0002b 130 day (Jul) Fe 6 0.0022±0.0002a FeEDTA 6 0.0017±0.0001ab R2 0.38 F Ratio 4.18*

141

Chapter Four

Nutrient reabsorption and flux of Avicennia marina in an ultra-

oligotrophic environment

Hanan Almahasheer1,2,, Carlos M. Duarte1 and Xabier Irigoien1

1 King Abdullah University of Science and Technology (KAUST), Red Sea Research Center, Thuwal 23955-6900, Kingdom of Saudi Arabia

2 Biology Department, University of Dammam (UOD), Dammam 31441-1982, Kingdom of Saudi Arabia

142

Abstract

Mangroves in the Red Sea live in an oligotrophic sea without permanent freshwater

inputs. Understanding the mechanisms to cope with nutrient limitation is, therefore,

important to understand their distribution and nutrient dynamics in the coastal ecosystem.

We measured nutrients (N, P, and Fe) as a function of age in Avicennia marina leaves from

the Central Red Sea to estimate reabsorption rates in ten sites from four different

mangrove stands in the Central of the Red Sea. We found that the concentration of N and P

but not Fe declined with age. Total content also declined in the older leaves with

reabsorption estimates of 69%, 72% and 35% for N, P, and Fe respectively. The results

suggest that mangroves trees in the Red Sea are likely to be a nutrient sink instead than a

source to the surrounding ecosystem.

Keywords: leaves, Nitrogen, Phosphorous, Iron, limitation, Retention, Mangroves, Red Sea.

143

Introduction:

Nutrient’s reabsorption is a key internal strategy for plants to cope with nutrient limitation in challenging environments. The withdrawal of nutrients from senesced leaves allows plants to reuse them again (Aerts and Chapin 1999) and to reproduce structures

and new leaves (Chapin III and Van Cleve 2000, Ochieng and Erftemeijer 2002), reducing

the competition and nutrient uptake (Killingbeck 1996). Also, at the ecosystem level, it

influences the elemental nutrients cycling (Aerts and Chapin 1999). Plants, in general,

enhance the growth rate in the shoot compared to the root as a response to nutrient

enrichment (Lovelock, Ball et al. 2009), proven by the increase of the root biomass in

mangrove seedlings when the nutrients or light are low (McKee 1995). Although

mangroves are rich in carbon (Donato, Kauffman et al. 2011), they are often limited in

nutrients (Koch and Snedaker 1997), which can result in dwarfing (Naidoo 2009).

Nutrient limitation in mangrove seedlings and trees has generally been studied in

terms of nitrogen and phosphorus limitation (Boto and Wellington 1983, Naidoo 1987,

Feller 1995, Koch and Snedaker 1997, Feller, Whigham et al. 1999, Feller, McKee et al.

2003, Feller, Whigham et al. 2003, Lovelock, Feller et al. 2004, Lovelock, Feller et al. 2006,

Lovelock, Feller et al. 2007, Naidoo 2009, Alongi 2011). However, iron has also been found

to be limiting (Alongi 2010) (Almahasheer et al., submitted) and it has been suggested that

the low availability of the three elements may be the reason for their absence in several

coastlines (Sato, Negassi et al. 2011).

Consequently, reabsorption mechanisms been extensively studied in terrestrial and aquatic ecosystems see reviews in (Killingbeck 1996, Hemminga, Marba et al. 1999). A

recent review by (Reef, Feller et al. 2010) reported reabsorption values in different 144

mangrove species, were the nitrogen reabsorption ranged from 69% to <5%. N and P

reabsorption in Avicennia marina has been reported a limited number of times (Rao,

Woitchik et al. 1994, Ochieng and Erftemeijer 2002, Alongi, Clough et al. 2005, Lovelock,

Feller et al. 2007, Lovelock, Sorrell et al. 2010, Zhou, Wei et al. 2010, Lovelock, Feller et al.

2011, Wei, Liu et al. 2015) but to our knowledge there is not a study assessing N, P, and Fe at the same time.

The aim of this paper was to estimate leaf production, the flux of nutrients and to estimate the rate of the nutrient reabsorption in Avicennia marina leaves growing in a strong oligotrophic environment such the Red Sea, with high salinity, low rainfall and low nutrient supply (Mandura 1997). We are also categorizing the completeness of the reabsorption process by looking at the maximal withdrawal of nutrients that Avicennia can

retain from senescent leaves and compare it to other results of Avicennia where possible.

Methods

1. Sampling across mangrove stands

Leaves were collected from monospecific stands of Avicennia marina in ten sites within

four different mangrove locations in the Central Red Sea: mangrove stands in Thuwal

Island, Khor Alkharar are far from human disturbances, Economic City lagoon is currently under development, whereas stands in Petro Rabigh are close to petrochemical activities

(Fig.1). Further information’s about the study locations and environmental factors controlling them are described in (Almahasheer, submitted). A total of 91 leaves were

collected in March 2015. In each location, a total of two to three long axillary branches in

different sites were randomly selected. The collected leaves were gently washed with sea 145 water and numbered in the field based on their location in the axillary shoot starting from the first leaf near the meristem to the last attached senescent leaf.

146

Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was

produced with ArcMap Version 10.2. Background map credits: the World Administrative

Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company.

2. Estimating density and flux

To calculate the flux of nutrients in mangroves leaves (mg element m -2 year -1) we (a)

estimated the number of leaves produced m -2 year -1 as the following:

Number of leaves produced m -2 year -1 = Number of meristems m-2 × Number of internodes

meristem -1 y-1 × 2.

The number of meristems m-2 were estimated in the field by measuring the girth of the

trees and the number of meristems that each tree produced in one or couple of branches,

using a line transect method. The transects had a corridor of 3 meters wide and 5 to 15

meters long in two locations (n= 74 trees for Khor Alkarrar and n =144 tree for Thuwal

Island). To estimate the total number of meristems we used the relation between

meristems and tree girth (see results). The number of internodes meristem-1 y-1 for the

same two locations above, where obtained from Almahasheer et al (submitted) (9.56 node

y-1 for Khor Alkharrar and 9.72 y-1 for Thuwal Island), and multiplied by 2 ( rates of annual

node production =half of leaf production) (Duke and Pinzon 1992). This estimate was

multiplied by the mean nutrient content of the each element in the leaves to get the flux of

each nutrient (mg element m -2 year -1).

3. Chemical analysis

Each leaf was photographed individually weight remained and finally grinded constant.and For then Carbon dried and at nitrogen60 ˚C in theconcentration, oven until thewe

weighted approx. 2 mg of each powdered leaf to the nearest 0.001 mg with an ultra-micro 147

balance in a pre-combusted aluminum capsule, then the concentration of C and N was

determined by combustion and thermal conductivity using a FLASH 2000 CHNS Analyzer

(Zimmermann, Keefe et al. 1997). Iron and phosphorus concentrations were determined

after digesting approx. 0.50 mg of the leaf with 6ml concentrated HNO3 and 2ml of H2O2 in

(Spalla, Baffi et al. 2009). The samples wereDigi PREP let to digestion cool then systems diluted forto 40ml2 hours using at Mili95 ˚C-Q water to be subsequently analyzed by

Inductively Coupled Plasma-Optical Emission Spectrometry (Varian Inc. model 720-ES).

The quality control is represented as the percent recovery. For Carbon and Nitrogen, we used a standard reference material (BBOT) and a calibration standard (Sulfanilamide) to obtain an average recovery of 101.1% for nitrogen and 100.7% for carbon n=30. For Iron and Phosphorous we used two different standards (Inorganic Ventures and PerkinElmer’s

Pure Plus), duplicated the samples and spiked them with the standards every 20 samples to have a recovery of 104% Fe and 98% P n=25. Additionally, we used an SRM: standard reference material from NIST with a recovery of 74% for Apple leaves and 80% for Peach leaves n =5. Finally, we removed any results below the detection limits of the ICP.

4. Statistical analysis

The nutrient content (mg leaf-1) was calculated as the product of the nutrients concentrations (mg g DW-1) and the total dry mass (g DW leaf-1) of the leaves (Lin and

Wang 2001). We also calculated the reabsorption in (mg leaf-1) from the fully developed leaf with max dry weight and last senescent leaf for leaves were photography showed that there was not any significant loss due to grazing or other causes (Lin and Wang 2001,

Lovelock, Feller et al. 2004). The age of the leaves was estimated using the production of

the nodes (Erickson and Michelini 1957, Duarte, Marba et al. 1994) in the same locations of 148

the Central of the Red Sea, where Avicennia marina produced 9.58 node y-1, resulting in an

estimated Plastocron interval of PI= 38 (Almahasheer et al, submitted).

Statistical analyses, including descriptive statistics, linear regression analyses of age vs. element, general linear models to test differences among stands, and Tukey HSD posthoc test to assess pairwise differences were carried out using JMP.

Results

1. Nutrient content, concentration, and reabsorption

The average nitrogen concentration was significantly higher in Petro Rabigh

compared to the other areas, whereas, the average nitrogen content did not differ among

locations. Similarly, averages of phosphorous content and concentration did not differ

significantly among stand. However, the average of iron content and concentration was

higher in both Petro Rabigh and Economic City compared to Alkarrar, while Thuwal

showed intermediate values (Table 1 and 2).

Table 1: Mean (± SE) for nutrients concentration (mg g DW-1) in Avicennia marina leaves from four different locations in the Central Red Sea, and the results from ANOVA and Tukey

HSD post hoc tests for differences in nutrient concentrations among locations. * = 0.05 > P

> 0.01. ** = P < 0.01. Locations sharing the same superscript letters do not differ among themselves in nutrient concentration for a particular element.

Location Alkarrar Petro Economic Thuwal F Ratio Rabigh City No. Rows 15 27 30 19 Location N 10.62±0.88ab 13.46±1.08a 10.68±0.69ab 9.39±0.60b 4.064** P 0.59±0.05a 0.84±0.11a 0.65±0.07a 0.68±0.06a 1.581ns 149

Fe 0.26±0.03b 0.56±0.06a 0.49±0.06a 0.44±0.03ab 4.264**

Table 2: Mean (± SE) for nutrients content (mg leaf-1) in Avicennia marina leaves from four

different locations in the Central Red Sea, and the results from ANOVA and Tukey HSD

posthoc tests for differences among in nutrient content among locations. * = 0.05 > P >

0.01. ** = P < 0.01. Locations sharing the same superscript letters do not differ among

themselves in nutrient content for a particular element.

Location Alkarrar Petro Economic Thuwal F Ratio Rabigh City No. Rows 15 27 30 19 Location N 1.96±0.29a 2.86±0.34a 2.65±0.29a 1.93±0.28a 2.112ns P 0.11±0.01a 0.15±0.01a 0.15±0.02a 0.13±0.02a 1.431ns Fe 0.04±0.01b 0.13±0.02a 0.12±0.01a 0.09±0.01ab 4.951**

The nitrogen and phosphorous concentration in the leaves declined linearly with

leaf age showing the dilution or leaching effect of nutrients as the leaves mature and

enlarge. However, iron concentration was stable or increased with leave age (Fig. 2). The regression intercept and slope for the overall elemental concentrations, along with R2 and F

ratio of ANOVA testing differences of elements are presented in Table 3. The slopes of

regression between locations were negative for nitrogen and phosphorous in all location

and positive in Petro Rabigh and Economic City for iron.

150

Figure 2: Accumulation rate of nutrients concentration (Difference in mg g DW-1 to the

value of the first leaf) with the age of Avicennia marina leaves in the Central Red Sea. One tree each site. The fitted lines present nitrogen in blue, phosphorous in red, and green for iron.

151

Table 3: Intercept (± SE) and Slope (± SE) for nutrients concentration (mg g DW-1 to the initial value of the first leaf) in Avicennia marina leaves in the Central Red Sea. The slopes are per tree (one tree each site).

Element Location /Site Intercept Intercept Slope Slope R2 F ratio Error Error /P value N Alkarrar 1st island 1.263 0.149 -0.001 0.001 0.30 3.03 ns N Alkarrar 2nd island 1.171 0.085 -0.003 0.001 0.89 33.67** N Economic city 1st island 1.239 0.138 -0.001 0.001 0.22 2.25 ns N Economic city 2nd island 0.939 0.081 -0.002 0.000 0.70 23.46** N Economic city 3rd island 1.365 0.129 -0.003 0.001 0.80 23.86** N Petro Rabigh Far - pipe 1.103 0.269 -0.001 0.002 0.06 0.25 ns N Petro Rabigh few meters - pipe 1.001 0.313 -0.001 0.002 0.04 0.31 ns N Petro Rabigh under pipe 0.939 0.084 -0.002 0.000 0.65 18.88** N Thuwal Fringe 1.339 0.180 -0.002 0.001 0.40 4.61 ns N Thuwal land 1.124 0.095 -0.001 0.000 0.58 11.19** P Alkarrar 1st island 0.940 0.057 -0.002 0.000 0.89 58.52** P Alkarrar 2nd island 1.371 0.310 -0.002 0.002 0.19 0.95 ns P Economic city 1st island 1.503 0.175 -0.002 0.001 0.49 7.60* P Economic city 2nd island 0.797 0.100 -0.002 0.000 0.65 18.66** P Economic city 3rd island 1.166 0.092 -0.003 0.000 0.85 33.93** P Petro Rabigh Far - pipe 0.954 0.133 -0.004 0.001 0.81 17.46* P Petro Rabigh few meters - pipe 0.870 0.115 -0.003 0.001 0.70 16.24** P Petro Rabigh under pipe 0.802 0.094 -0.002 0.000 0.72 25.30** P Thuwal Fringe 0.980 0.058 -0.002 0.000 0.92 77.01** P Thuwal land 1.083 0.084 -0.002 0.000 0.70 18.97** Fe Alkarrar 1st island 0.950 0.133 -0.001 0.001 0.11 0.90 ns Fe Alkarrar 2nd island 0.827 0.235 0.001 0.002 0.15 0.68 ns Fe Economic city 1st island 0.811 0.134 0.000 0.001 0.00 0.01 ns Fe Economic city 2nd island 1.415 0.575 0.008 0.002 0.50 10.07** Fe Economic city 3rd island 0.851 0.157 0.000 0.001 0.01 0.08 ns Fe Petro Rabigh Far - pipe 0.757 0.231 -0.002 0.002 0.34 2.11 ns Fe Petro Rabigh few meters - pipe 1.187 0.343 0.001 0.002 0.02 0.12 ns Fe Petro Rabigh under pipe 0.970 0.597 0.013 0.002 0.72 26.27** Fe Thuwal Fringe 0.838 0.127 -0.001 0.001 0.29 2.90 ns Fe Thuwal land 0.755 0.146 0.000 0.001 0.00 0.01 ns

152

However, the nutrient contents were highly variable between sites with a general

trend to decline in leaves around 200 days old in most of the sites inferring a retrieval of

nutrients from old leaves before shedding them (Fig. 3). The regression intercept and slope

for the overall elemental content, along with R2 and F ratio of ANOVA testing differences of

elements are presented in (Table 4). N and P reabsorption was detected in all examined

sites whereas Fe reabsorption was only detected in nine out of the ten sites. Because we

did not find a significant difference in the reabsorption efficiency between locations for N, P

and Fe (Table 5), we calculated the grand mean of reabsorption to be 69% N, 72% P and

35% Fe. Reabsorption efficiency was not related to the maximum nutrient concentration

and therefore independent of the nutrient status of the plants. However, reabsorption in N

and P correlated significantly (rN-P=0.82, P =0.0037), whereas reabsorption in Fe and P

correlated but the probability was not significant (rFe-P=0.58, P =0.0787), and reabsorption in Fe and N did not correlate (rN-Fe =0.37, P =0.2899).

However, the N: P ratio showed a general trend to increase from values of 10 in young leaves to 20 at leaves 200 days old and then to around 30 in senescent leaves (Fig.

4).

153

Figure 3: Accumulation rate of nutrients content (difference in mg leaf-1 to the value of the first leaf) with the age of Avicennia marina leaves in the Central Red Sea. One tree each site.

The fitted lines present nitrogen in blue, phosphorous in red, and green for iron.

154

Table 4: Intercept (± SE) and Slope (± SE) for content (mg leaf-1 to the initial value of the first leaf) in Avicennia marina leaves in the Central Red Sea. The slopes are per tree (one tree each site).

Element Location /Site Intercept Intercept Slope Slope R2 F ratio Error Error /P value N Alkarrar 1st island 2.228 0.590 -0.001 0.003 0.03 0.20 ns N Alkarrar 2nd island 2.886 1.699 0.001 0.011 0.00 0.02 ns N Economic city 1st island 3.177 2.008 0.021 0.010 0.35 4.24 ns N Economic city 2nd island 1.399 0.464 0.004 0.002 0.30 4.34 ns N Economic city 3rd island 2.411 0.838 -0.003 0.004 0.07 0.47 ns N Petro Rabigh Far - pipe 2.926 2.705 0.024 0.018 0.30 1.67 ns N Petro Rabigh few meters - pipe 1.198 1.552 0.009 0.009 0.13 1.04 ns N Petro Rabigh under pipe 0.933 0.262 0.001 0.001 0.16 1.85 ns N Thuwal Fringe 3.621 2.225 0.008 0.010 0.08 0.57 ns N Thuwal land 1.448 1.609 0.019 0.007 0.47 7.17* P Alkarrar 1st island 1.519 0.221 -0.003 0.001 0.49 6.66* P Alkarrar 2nd island 3.685 3.009 0.010 0.020 0.06 0.27 ns P Economic city 1st island 3.885 1.999 0.015 0.010 0.22 2.22 ns P Economic city 2nd island 1.156 0.275 0.001 0.001 0.09 1.02 ns P Economic city 3rd island 1.989 0.716 -0.002 0.004 0.05 0.34 ns P Petro Rabigh Far - pipe 2.281 0.963 0.000 0.006 0.00 0.01ns P Petro Rabigh few meters - pipe 1.021 0.383 0.001 0.002 0.01 0.06 ns P Petro Rabigh under pipe 0.848 0.161 -0.001 0.001 0.11 1.22 ns P Thuwal Fringe 2.295 0.904 0.000 0.004 0.00 0.00 ns P Thuwal land 1.348 1.414 0.016 0.006 0.46 6.75 ns Fe Alkarrar 1st island 1.619 0.410 0.000 0.002 0.00 0.02 ns Fe Alkarrar 2nd island 1.452 1.531 0.022 0.010 0.54 4.64 ns Fe Economic city 1st island 0.815 1.170 0.026 0.006 0.71 19.23** Fe Economic city 2nd island 0.152 3.359 0.071 0.014 0.72 26.12** Fe Economic city 3rd island 1.431 0.741 0.003 0.004 0.09 0.63 ns Fe Petro Rabigh Far - pipe 1.001 0.872 0.010 0.006 0.42 2.93 ns Fe Petro Rabigh few meters - pipe 1.442 1.507 0.017 0.009 0.35 3.84 ns Fe Petro Rabigh under pipe -1.800 2.935 0.055 0.012 0.67 20.40** Fe Thuwal Fringe 1.768 1.103 0.007 0.005 0.23 2.10 ns Fe Thuwal land -0.379 0.567 0.025 0.002 0.93 104.37**

155

Table 5: Mean (± SE) for nutrients resorption in Avicennia marina leaves from four different locations in the Central Red Sea and the results from ANOVA to tests for differences between locations. * = 0.05 > P > 0.01. ** = P < 0.01.

Location Alkarrar Petro Economic Thuwal Grand F Ratio Rabigh City mean No.Rows 2 3 3 2 10 Location N 71.80±0.76 64.86±12.61 69.00±7.94 72.83±1.06 69.08±3.99 0.14ns P 70.01±2.02 70.96±4.04 74.02±4.90 74.96±4.79 72.49±1.93 0.26ns Fe 37.14±13.18 38.03±10.69 46.81±8.52 12.36±34.26 35.35±7.67 0.52ns

Figure 4: Accumulation rate of N: P ratio vs. time in Avicennia marina leaves in the Central

Red Sea. One tree per site. Based on elemental content. The line is a smoothed line.

156

2. Leaf production and nutrient flux

The trees girth and number of meristems were significantly related with R2 =0.91,

F=378.88 and P<0.0001 and R2 =0.93, F=582.15 and P<0.0001 in Khor Alkarrar and

Thuwal Island respectively (Fig. 5a and b).

In Khor Alkarrar Avicennia marina produced 7871.5 leaf-1 m-2y-1 and in Thuwal

7345.2 leaf-1 m-2 y-1 with an overall average of 7608.3 leaf-1 m-2 y-1. The estimated nutrients

flux (Table. 6) ranged from 15000 to 22000 mg m-2 y-1 for nitrogen, 800 to 1100 mg m-2 y-1

for phosphorous and 300 to 1000 mg m-2 y-1 for iron.

157

1000

600

400 300

200

100 A 70 50

30

20

10

7

5

3

2

Log(meristems) = 0.3196785 + 1.6363156*Log(girth) 1

0.7 2 3 4 5 6 20 30 50 1 10 100

Gi th

600 B 400 300

200

100

60

40

30

20

10

6

4

3

2

Log(meristems) = 1.1529447 + 1.7421698*Log(girth) 1

0.7 0.4 0.6 2 3 4 5 6 20 30 40 1 10

Gi th

Figure 5: Regression analysis for the Avicenna marina girth and total number of meristems

produced in (a) Khor Alkharar and (b) Thuwal Island.

158

Table 6: Mean of nutrients fluxes (mg element m -2 year -1) in Avicennia marina leaves from four different locations in the Central Red Sea

Location Alkarrar Petro Rabigh Economic City Thuwal N 14912 21760 20162 14684 P 837 1141 1141 989 Fe 304 989 913 685

Discussion

Our results of N and P reabsorption are in the range of previous estimates of nutrient reabsorption in Avicennia marina (Table. 7), which are in the higher range of all other aquatic and terrestrial plants (Hemminga, Marba et al. 1999, Chapin III and Van Cleve

2000, Reef, Feller et al. 2010). However, to our knowledge, iron reabsorption has not been previously estimated in mangroves. We propose that Avicennia in the Central of the Red

illustratedSea have aby lower(Almahasheer, iron reabsorption in review). efficiencyWe suggest ≤ that 35% because making of themthe biogenic iron limited nature asof the sediments in the Red Sea, dominated by carbonates, and the lack of riverine sources of iron they do not acquire iron from the sediment, however, because plant leaves in general act as the root they lose and absorbed nutrients by their stomata (Aerts and Chapin 1999).

We hypothesise that dust deposition maybe the most important source of iron, that mainly

0.01%because to soil 80% iron due solubility to atmospheric is ∼0.1% acids while (Mahowald, the atmospheric Engelstaedter iron solubility et al. 2009) varies. Then between when the leaves obtain Iron, it difficult for them to keep it/ reabsorb it, because many metals cannot easily pass through the plasma membrane of their large atomic weight unless 159 passed as metal chelates. And that potentially might be due to the high mobility of N and P in the phloem compared to an intermediate mobility of Fe in plants (White 2012).

Table 7: Comparison of nutrients resorption in Avicennia marina leaves from different locations worldwide.

location N-RE P-RE Fe-RE Ref Kenya 69% (Rao, Woitchik et al. 1994) Kenya 68% 61% (Ochieng and Erftemeijer 2002) China 57%-68% 49%-62% (Wei, Liu et al. 2015) New Zealand 20%-60% 10%-60% (Lovelock, Feller et al. Vegetated & fertilized 2007) North coast of western 63%-71% 43%-72% (Alongi, Clough et al. 2005) Australia Exmouth Gulf 19%-44% -1%-60% (Lovelock, Feller et al. north-west of Western 2011) Australia(before and after Cyclone Pancho) New Zealand 60% 60% (Lovelock, Sorrell et al. (fertilized with 2010) either N or P or Unfertilized) China 56% 31% (Zhou, Wei et al. 2010) Red Sea 69% 72% 37% Our study

The changes in N and P concentration with leave age indicate that there is a dilution effect where an increased leaf mass can be sustained with the same amount of N and P.

However this is not observed with iron, where concentration remains stable or increase, suggesting that iron was the most limiting factor in building up the leave biomass. This 160

result confirms experimental work with seedlings suggesting that iron is the primary

limiting factor for mangroves growth in the Central Red Sea (Almahasheer submitted).

Aerts and Chapin (Aerts and Chapin 1999) proposed that when the foliar N: P mass

ratio <14 the plant is N limited, whereas ratios > 16 would indicate P limitation. This ratio

has been used to show N limitation (Wei, Liu et al. 2015) and situations where the N:P ratio

of mature leaves was higher than the one of young and senescent leaves (Zhou, Wei et al.

2010). Our results show an increase of N: P ratio with age suggesting an additional

phosphorous limitation as the leaves mature. However, N: P ratio should only be used to

assess which nutrient limits the biomass production at the vegetation level and only when

factors other than N or P are unlikely to be limiting (Güsewell and Koerselman 2002).

Considering the reabsorption levels, we suggest that mangrove trees in the Red Sea

act as nutrients sink not as a source to the adjacent ecosystem.

However, the leaves shedding imply a significant turnover of nutrients, where the nutrients collected by the roots are brought back to the top layer of the sediments. If we extrapolate the estimated fluxes to the area covered by mangroves in the Red Sea, around

135 km2, the total flux per year is about 2414 metric tons of nitrogen, 139 tons of

phosphorous and 98 tons of iron. This turnover of nutrients can play a significant role in an

oligotrophic ecosystem without rivers and where the annual budget of nutrients input from

the Indian Ocean is considered to be null or even negative (Bethoux 1988, Souvermezoglou,

Metzl et al. 1989).

References

Aerts, R. and F. S. Chapin (1999). "The mineral nutrition of wild plants revisited: a re- evaluation of processes and patterns." Advances in ecological research 30: 1-67. 161

Alongi, D. (2011). "Early growth responses of mangroves to different rates of nitrogen and phosphorus supply." Journal of Experimental Marine Biology and Ecology 397(2): 85-93.

Alongi, D., B. Clough and A. Robertson (2005). "Nutrient-use efficiency in arid-zone forests of the mangroves Rhizophora stylosa and Avicennia marina." Aquatic botany 82(2): 121- 131.

Alongi, D. M. (2010). "Dissolved iron supply limits early growth of estuarine mangroves." Ecology 91(11): 3229-3241.

Bethoux, J. (1988). "Red Sea geochemical budgets and exchanges with the Indian Ocean." Marine Chemistry 24(1): 83-92.

Boto, K. K. and J. J. Wellington (1983). "Phosphorus and nitrogen nutritional status of a northern Australian mangrove forest." Marine Ecology Progress Series-pages: 11: 63-69.

Chapin III, F. S. and K. Van Cleve (2000). Approaches to studying nutrient uptake, use and loss in plants. Plant physiological ecology, Springer: 185-207.

Donato, D. C., J. B. Kauffman, D. Murdiyarso, S. Kurnianto, M. Stidham and M. Kanninen (2011). "Mangroves among the most carbon-rich forests in the tropics." Nature Geoscience 4(5): 293-297.

Duarte, C. M., N. Marba, N. Agawin, J. Cebrian, S. Enriquez, M. D. Fortes, M. E. Gallegos, M. Merino, B. Olesen, K. Sandjensen, J. Uri and J. Vermaat (1994). "Reconstruction of seagrass dynamics: age determinations and associated tools for the seagrass ecologist." Marine Ecology Progress Series 107(1-2): 195-209.

Duke, N. C. and Z. S. M. Pinzon (1992). "Aging Rhizophora seedlings from leaf scar nodes: a technique for studying recruitment and growth in mangrove forests." Biotropica: 173-186.

Erickson, R. O. and F. J. Michelini (1957). "The plastochron index." American Journal of Botany: 297-305.

Feller, I. C. (1995). "Effects of nutrient enrichment on growth and herbivory of dwarf red mangrove (Rhizophora mangle)." Ecological monographs: 477-505.

Feller, I. C., K. L. McKee, D. F. Whigham and J. P. O'Neill (2003). "Nitrogen vs. phosphorus limitation across an ecotonal gradient in a mangrove forest." Biogeochemistry 62(2): 145- 175.

Feller, I. C., D. F. Whigham, K. L. McKee and C. E. Lovelock (2003). "Nitrogen limitation of growth and nutrient dynamics in a disturbed mangrove forest, Indian River Lagoon, Florida." Oecologia 134(3): 405-414.

Feller, I. C., D. F. Whigham, J. P. O'Neill and K. L. McKee (1999). "Effects of nutrient enrichment on within-stand cycling in a mangrove forest." Ecology 80(7): 2193-2205. 162

Güsewell, S. and W. Koerselman (2002). "Variation in nitrogen and phosphorus concentrations of wetland plants." Perspectives in Plant Ecology, Evolution and Systematics 5(1): 37-61.

Hemminga, M. A., N. Marba and J. Stapel (1999). "Leaf nutrient resorption, leaf lifespan and the retention of nutrients in seagrass systems." Aquatic Botany 65(1-4): 141-158.

Killingbeck, K. T. (1996). "Nutrients in senesced leaves: keys to the search for potential resorption and resorption proficiency." Ecology 77(6): 1716-1727.

Koch, M. S. and S. C. Snedaker (1997). "Factors influencing Rhizophora mangle L. seedling development in Everglades carbonate soils." Aquatic Botany 59(1): 87-98.

Lin, P. and W.-q. Wang (2001). "Changes in the leaf composition, leaf mass and leaf area during leaf senescence in three species of mangroves." Ecological Engineering 16(3): 415- 424.

Lovelock, C., I. C. Feller, K. McKee, B. Engelbrecht and M. Ball (2004). "The effect of nutrient enrichment on growth, photosynthesis and hydraulic conductance of dwarf mangroves in Panama." Functional Ecology 18(1): 25-33.

Lovelock, C. E., M. C. Ball, K. C. Martin and I. C. Feller (2009). "Nutrient enrichment increases mortality of mangroves." PLoS One 4(5): e5600.

Lovelock, C. E., I. C. Feller, M. F. Adame, R. Reef, H. M. Penrose, L. Wei and M. C. Ball (2011). "Intense storms and the delivery of materials that relieve nutrient limitations in mangroves of an arid zone estuary." Functional Plant Biology 38(6): 514-522.

Lovelock, C. E., I. C. Feller, M. C. Ball, B. M. Engelbrecht and M. L. Ewe (2006). "Differences in plant function in phosphorus‐and nitrogen‐limited mangrove ecosystems." New Phytologist 172(3): 514-522.

Lovelock, C. E., I. C. Feller, J. Ellis, A. M. Schwarz, N. Hancock, P. Nichols and B. Sorrell (2007). "Mangrove growth in New Zealand estuaries: the role of nutrient enrichment at sites with contrasting rates of sedimentation." Oecologia 153(3): 633-641.

Lovelock, C. E., B. K. Sorrell, N. Hancock, Q. Hua and A. Swales (2010). "Mangrove Forest and Soil Development on a Rapidly Accreting Shore in New Zealand." Ecosystems 13(3): 437- 451.

Mahowald, N. M., S. Engelstaedter, C. Luo, A. Sealy, P. Artaxo, C. Benitez-Nelson, S. Bonnet, Y. Chen, P. Y. Chuang and D. D. Cohen (2009). "Atmospheric iron deposition: Global distribution, variability, and human perturbations*." Annual Review of Marine Science 1: 245-278.

Mandura, A. (1997). "A mangrove stand under sewage pollution stress: Red Sea." Mangroves and Salt marshes 1(4): 255-262. 163

McKee, K. L. (1995). "Interspecific variation in growth, biomass partitioning, and defensive characteristics of neotropical mangrove seedlings: response to light and nutrient availability." American Journal of Botany: 299-307.

Naidoo, G. (1987). "Effects of salinity and nitrogen on growth and water relations in the mangrove, Avicennia marina (Forsk.) Vierh." New Phytologist 107(2): 317-325.

Naidoo, G. (2009). "Differential effects of nitrogen and phosphorus enrichment on growth of dwarf Avicennia marina mangroves." Aquatic Botany 90(2): 184-190.

Ochieng, C. A. and P. L. Erftemeijer (2002). "Phenology, litterfall and nutrient resorption in Avicennia marina (Forssk.) Vierh in Gazi Bay, Kenya." Trees 16(2-3): 167-171.

Rao, R., A. Woitchik, L. Goeyens, A. Van Riet, J. Kazungu and F. Dehairs (1994). "Carbon, nitrogen contents and stable carbon isotope abundance in mangrove leaves from an east African coastal lagoon (Kenya)." Aquatic Botany 47(2): 175-183.

Reef, R., I. C. Feller and C. E. Lovelock (2010). "Nutrition of mangroves." Tree Physiology 30(9): 1148-1160.

Sato, G., S. Negassi and A. Z. Tahiri (2011). "The only elements required by plants that are deficient in seawater are nitrogen, phosphorous and iron." Cytotechnology 63(2): 201-204.

Souvermezoglou, E., N. Metzl and A. Poisson (1989). "Red Sea budgets of salinity, nutrients and carbon calculated in the Strait of Bab-El-Mandab during the summer and winter seasons." Journal of Marine Research 47(2): 441-456.

Spalla, S., C. Baffi, C. Barbante, C. Turretta, G. Cozzi, G. Beone and M. Bettinelli (2009). "Determination of rare earth elements in tomato plants by inductively coupled plasma mass spectrometry techniques." Rapid Communications in Mass Spectrometry 23(20): 3285-3292.

Wei, S. D., X. W. Liu, L. H. Zhang, H. Chen, H. Zhang, H. C. Zhou and Y. M. Lin (2015). "Seasonal changes of nutrient levels and nutrient resorption in Avicennia marina leaves in Yingluo Bay, China." Southern Forests 77(3): 237-242.

White, P. (2012). "Long-distance transport in the xylem and phloem." Marschner’s Mineral Nutrition of Higher Plants: 49-70.

Zhou, H.-C., S.-D. Wei, Q. Zeng, L.-H. Zhang, N. F.-y. Tam and Y.-M. Lin (2010). "Nutrient and caloric dynamics in Avicennia marina leaves at different developmental and decay stages in Zhangjiang River Estuary, China." Estuarine Coastal and Shelf Science 87(1): 21-26.

Zimmermann, C. F., C. W. Keefe and J. Bashe (1997). Method 440.0: Determination of Carbon and Nitrogen in Sediments and Particulates of Estuarine/Coastal Waters Using Elemental Analysis, United States Environmental Protection Agency, Office of Research and Development, National Exposure Research Laboratory. 164

Supplementary Materials

Figure S 1: Slope (± SE) for the regression analysis of leaves nutrients concentration (mg g

DW-1 to initial value of the first leaf).

-3 10 15 Alkarrar Petro Rabigh Economic City 10 Thuwal

5 Slope and st. error 0

-5 N P Fe Elements Concentration Figure S 2: Slope (± SE) for the regression analysis of leaves nutrients content (mg leaf-1 to initial value of the first leaf). 165

0.08 Alkarrar Petro Rabigh Economic City 0.06 Thuwal

0.04

0.02 Slope and st. error

0

-0.02 N P Fe Elements Content

166

Chapter Five

Heavy metal dynamics in Red Sea Mangrove leaves

Hanan Almahasheer1,2,, Carlos M. Duarte1 and Xabier Irigoien1

1 King Abdullah University of Science and Technology (KAUST), Red Sea Research Center, Thuwal 23955-6900, Kingdom of Saudi Arabia

2 Biology Department, University of Dammam (UOD), Dammam 31441-1982, Kingdom of Saudi Arabia

167

Abstract

The role of mangroves in heavy metal sequestration is still relatively unknown. On one side the woody parts of the mangrove retain heavy metals for long periods, but on the other leaves may return heavy metals collected by roots to the upper layers of the sediment. In this paper, we analyzed the concentrations of heavy metals as a function of age in

mangrove leaves to understand whether some heavy metals are retained by the plant and

to quantify the amounts shed with senescent leaves. Our results revealed that mangroves

appear to have a high tolerance of metals in contaminated sediments, and their leaves may

serve as a biological indicator as well as a source of contamination. Most metals

accumulated with age and were shed with senescent leaves, returning therefore to the

upper layers of the sediment. Only for copper, we could observe a reabsorption

mechanisms similar to the one described for nutrients.

Keywords: Heavy Metals, rate, Phytoremediation, Avicennia marina, Mangroves, Red Sea.

168

Introduction

Inorganic pollutants, such as heavy metals have both anthropogenic, and earth crust

origins and are not degradable (Lasat 2002, Nagajyoti, Lee et al. 2010, Moore, Kröger et al.

2011). Global development has increased the release of heavy metals into the ecosystems.

As an example over the last five decades 22.000 t, 939.000 t, 783.000 and 1.350.000 t of Cd,

Cu, Pb and Zn respectively have been released into different ecosystems (Singh, Labana et

al. 2003). These metals get concentrated as they move up in the food chain which

propagates the contamination and complicates the treatment (Wang, Chen et al. 2009).

However, plants usually absorb them from the soil along with nutrients

2014) and can, therefore, be used to phytoremediation the environment(Ovečka at and lowTakáč cost

(Robinson, Green et al. 2003). There are different mechanisms of plants phytoremediation.

Briefly, plants phytofiltrate the pollutant mostly metals from the aquatic systems using

their roots, phytoextract metals/organics from the soil and concentrate them in the

roots/shoot, phytostabilize the pollutants in soil, phytovolatilize elements from the soil

through the plant foliage and also plants and associated microbes phytodegrade the soil

organic pollutants (Garbisu and Alkorta 2001). Moreover, it has long been documented that plants growing in toxic environment cannot prevent taking them up, but they control it over three different strategies; accumulation mostly in above ground parts, exclusion where metals concentration in the shoot is very low compared to the soil, and indicator plants where the uptake to the shoot is regulated and reflects the soil levels (Baker 1981).

Most heavy metals are not toxic for plant, and some of them are named micronutrients and considered essential elements for plant growth (e.g. Co, Cu, Fe, Mn, Mo, Ni, Al, Rb, Ti, and

Zn) until they exceed a certain limit (Appenroth 2010, Kabata-Pendias 2010). Others are 169 non-essential for plant growth and toxic (e.g. Cd, Pb, U, Cr, Ag, Hg, Zr), also As and Se are metalloids but share the toxicity (Bothe 2011, Shahid, Ferrand et al. 2013).

Mangroves potentially phytoremediate coastal ecosystems as they have proven their efficient adaptation to contaminated coastal environments by accumulating the organic sediments combined with metals and nutrients thus reducing their distribution (Nath,

Birch et al. 2014). As a result of their roots structure, mangrove roots are very efficient at trapping and immobilizing high concentrations of metals compared to other parts of the plant (Wang, Qiu et al. 2013). Further, as they grow in salty water, they are the only plant that can be used for phytoremediation in coastal sediments. Interestingly, their leaves serve as metals bio-indicators (Murray 1985, Pinheiro, e Silva et al. 2012).

Nevertheless, through the shedding of their leaves mangroves could also potentially remobilize heavy metals that were buried deep in the sediment. And therefore, although in the long term the net effect would be a reduction of heavy metals in the sediment column because accumulation in the shoot and roots, the short time effects of mangroves could be to maintain high levels of heavy metals in the sediment surface. However, mangroves also have the capacity to reabsorb nutrients before shedding the leaves (Alongi, Clough et al.

2005, Zhou, Wei et al. 2010)Almahasheer, et al in prep) and if the same applied to heavy metals the reduction would also apply to the surface layers of the sediment. Overall, pollution in mangroves have been widely studied in recent years (Zhang, Xu et al. 2014), but to our knowledge the dynamics of heavy metals in mangrove leaves has not been investigated and it is, therefore, difficult to evaluate the role of mangroves in the heavy metals dynamics in the sediment. 170

Therefore, the aim of this paper is to study the concentrations of heavy metals rates as a

function of age in mangrove leaves. We test the hypothesis that mangroves retain heavy

metals before shedding leaves, as it does with nutrients (Alongi, Clough et al. 2005, Zhou,

Wei et al. 2010)Almahasheer, et al in prep), If this hypothesis is supported, it will imply

that both in the short and long term mangroves would contribute to reducing the load of

heavy metals in all sediment layers.

Methods

1. Study location and leaves sampling

Leaves were collected in March 2015 from four different locations in the Central Red

Sea: Thuwal Island and Khor Alkharar are locations far from human disturbances, whereas

Petro Rabigh is a location with petrochemical industry, and Economic-city is a new industrial city under development (Fig.1). Further information’s about the study locations and environmental factors controlling them are described in (Almahasheer, nursery paper). Leaves were collected, gently washed with sea water, and numbered in the field based on their location in the axillary shoot starting from the first leaf near the meristem to the last attached senescent leaf. To calculate the flux of elements in mangroves leaves (mg

element m -2 year -1) we estimated the number of leaves shed per year and square meter

using leaf production and tree density measurements (see Chapter 4)

171

Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative

Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company. 172

2. Chemical analysis and quality control

weight.All leaves We processed were photographed each leaf individually and then driedby homogenizing individually the at 60samples ˚C oven then till digesting constant

approx. 0.50 mg of the leaf with 6ml concentrated HNO3 and 2ml of H2O2 in Digi PREP

(Spalla, Baffi et al. 2009). The samples were left to

digestioncool and then systems diluted for to2 40mlhours Milli at 95-Q ˚Cto be subsequently analyzed with Inductively Coupled

Plasma-Optical Emission Spectrometry (ICP-OES). The quality control is presented as the

percent recovery. We used two different standards: Inorganic Ventures and PerkinElmer’s

Pure Plus, duplicated the samples and spiked them with the standards every 20 samples to

have a recovery of 108%. Additionally, we used an SRM: standard reference material from

NIST to have a recovery of 104% for Apple leaves and 126% for Peach leaves. The

reference material passed all elements present in the SRM except Se, As, V and Pb. Finally,

we removed any results below the detection limits of the ICP.

3. Statistical analysis

The element content (mg leaf DW-1) was calculated by multiplying the element concentrations (mg g DW-1) with the leaf dry mass (g leaf DW-1) (Lin and Wang 2001). The

age of the leaves was estimated using the production of the nodes in the same locations of

the Central of the Red Sea, were Avicennia marina produced 9.58 node y-1. By simple

calculation we estimated the plastocron interval to be PI= 38 (Erickson and Michelini 1957,

Duarte, Marba et al. 1994) and (Almahasheer et al, submitted). Statistical analyses,

including descriptive statistics, linear regression analyses of age vs. element, general linear

models to test differences among stands, and Tukey HSD posthoc test to assess pairwise

differences were carried out using JMP. 173

Results

Nine and eleven out of the twenty two heavy metals that were analyzed showed significant differences in concentration and contents between locations (Table 1 and 2).

Generally, the concentration of these metals was significantly higher in Petro Rabigh followed by Economic City then Thuwal Island and finally Alkarrar, whereas, the elemental content was higher in Economic City followed by Petro Rabigh then Thuwal Island and finally Alkarrar. (Tukey HSD post hoc test, P < 0.05). Details of the raw data for both elements content and concentration are provided in the supplementary materials (Tables

S1 and S2).

174

Table 1: Mean ± St. Error for the element concentrations (mg g DW-1). Results of ANOVA by location and sites for each elements showing F and significant differences at **P < 0.01; *P between 0.01 and 0.05. Letters indicate significant differences of Tukey test HSD post hock between locations for each element separately.

Location Alkarrar Economic City Petro Rabigh Thuwal F ratio N Rows 15 30 27 19 location Ag 0.01235±0.00215b 0.02794±0.00414a 0.03109±0.00370a 0.02304±0.00234ab 3.94* Al 0.20778±0.02328b 0.37307±0.04527a 0.41386±0.04088a 0.33916±0.02444ab 3.82* As 0.00216±0.00033ab 0.00215±0.00026b 0.00329±0.00028a 0.00273±0.00034ab 3.65* Be 0.00002±0.00000a 0.00007±0.00003a 0.00005±0.00002a 0.00005±0.00004a 0.46ns Cd 0.00021±0.00002a 0.00017±0.00003a 0.00011±0.00002a 0.00020±0.00004a 1.87ns Co 0.00016±0.00008a 0.00033±0.00007a 0.00034±0.00005a 0.00020±0.0006a 1.72ns Cr 0.00087±0.00010a 0.00157±0.00025a 0.00123±0.00010a 0.00119±0.00009a 2.30ns Cu 0.00218±0.00015a 0.00290±0.00024a 0.00306±0.00033a 0.00316±0.00076a 0.86ns Fe 0.26202±0.03370b 0.48587±0.06011a 0.55512±0.05563a 0.43934±0.02949ab 4.26** Li 0.00045±0.00002a 0.00055±0.00003a 0.00053±0.00003a 0.00045±0.00003a 2.40ns Mn 0.02324±0.00115b 0.02812±0.00229b 0.10157±0.01109a 0.02397±0.00191b 34.36** Mo 0.00282±0.00022a 0.00192±0.00029ab 0.00145±0.00024b 0.00178±0.00026ab 3.61* Nb 0.00009±0.00005a 0.00022±0.00006a 0.00037±0.00009a 0.00021±0.00007a 2.15ns Ni 0.00181±0.00033a 0.00212±0.00022a 0.00151±0.00014a 0.00141±0.00012a 2.80* Pb 0.00866±0.00081a 0.00872±0.00096a 0.00790±0.00065a 0.00844±0.00103a 0.19ns Rb 0.05785±0.00187a 0.05976±0.00136a 0.05480±0.00149a 0.05848±0.00233a 1.84ns Se 0.00743±0.00111a 0.00591±0.00070a 0.00530±0.00067a 0.00719±0.00111a 1.33ns Sr 0.05785±0.00187a 0.05976±0.00136a 0.05480±0.00149a 0.05848±0.00233a 1.84ns Ti 0.00850±0.00114b 0.01805±0.00233a 0.01942±0.00215a 0.01621±0.00125ab 4.17** V 0.01645±0.00127ab 0.01469±0.00133b 0.02072±0.00169a 0.02127±0.00152a 4.74** Zn 0.01606±0.00529a 0.01034±0.00062ab 0.01689±0.00158a 0.00654±0.00050b 5.49** Zr 0.00071±0.00007a 0.00100±0.00009a 0.00084±0.00007a 0.00083±0.00006a 2.45ns 175

Table 2: Mean ± St. Error of the element for content (mg leaf-1), Results of ANOVA by location and sites for each elements showing F and significant differences at **P < 0.01; *P between 0.01 and 0.05. Letters indicate significant differences of Tukey test HSD post hock between locations for each element separately.

Location Alkarrar Economic city Petro Rabigh Thuwal F ratio N Rows 15 30 27 19 location Ag 0.00190±0.00033b 0.00648±0.00087a 0.00748±0.00124a 0.00477±0.00085ab 4.86** Al 0.03540±0.00416b 0.08839±0.01063a 0.09646±0.01352a 0.06825±0.00981ab 4.56** As 0.00042±0.00008a 0.00059±0.00009 a 0.00075±0.00009 a 0.00058±0.00011a 1.65ns Be 0.00000±0.00000a 0.00003±0.00001a 0.00001±0.00001a 0.00002±0.00002a 0.60ns Cd 0.00004±0.00001a 0.00006±0.00001a 0.00003±0.00001a 0.00005±0.00002a 0.86ns Co 0.00002±0.00001a 0.00008±0.00002a 0.00009±0.00002a 0.00005±0.00002a 1.92ns Cr 0.00015±0.00002b 0.00042±0.00009a 0.00030±0.00004ab 0.00026±0.00005ab 2.70ns Cu 0.00040±0.00005b 0.00070±0.00007a 0.00059±0.00005ab 0.00054±0.00009ab 2.87* Fe 0.04362±0.00538b 0.11621±0.01414a 0.12953±0.01808a 0.09011±0.01336ab 4.95** Li 0.00008±0.00001b 0.00014±0.00001a 0.00012±0.00001ab 0.00009±0.00001ab 3.27* Mn 0.00442±0.00058b 0.00724±0.00081b 0.02502±0.00373a 0.00565±0.00111b 18.42** Mo 0.00050±0.00006a 0.00064±0.00012a 0.00037±0.00008a 0.00035±0.00006a 2.08ns Nb 0.00001±0.00001a 0.00005±0.00001a 0.00010±0.00003a 0.00007±0.00003a 2.46ns Ni 0.00031±0.00005b 0.00055±0.00008a 0.00036±0.00005ab 0.00030±0.00005ab 3.50* Pb 0.00160±0.00024a 0.00226±0.00036a 0.00172±0.00021a 0.00178±0.00033a 0.96ns Rb 0.01039±0.00097a 0.01572±0.00157a 0.01244±0.00124a 0.01238±0.00184a 2.14ns Se 0.00152±0.00032a 0.00158±0.00027a 0.00119±0.00021a 0.00137±0.00029a 0.48ns Sr 0.01039±0.00097a 0.01572±0.00157a 0.01244±0.00124a 0.01238±0.00184a 2.14ns Ti 0.00143±0.00020b 0.00426±0.00052a 0.00454±0.00067a 0.00322±0.00046ab 5.08** V 0.00303±0.00040a 0.00405±0.00053a 0.00519±0.00069a 0.00511±0.00090a 1.90ns Zn 0.00230±0.0004ab 0.00280±0.00036ab 0.00378±0.00054a 0.00132±0.00021b 5.52** Zr 0.00012±0.00001b 0.00025±0.00003a 0.00019±0.00002ab 0.00018±0.00004ab 3.09* 176

The concentration remained constant with age for most heavy metals except for

Vanadium, Cadmium and Niobium that increased and Copper that decreased (Fig. 2).

Therefore the content of metals increased significantly with leaf age P<0.01 (Fig. 3).

There were significant differences in the slopes (accumulation rates) between sites

(ANOVA test). The detailed slopes calculated per tree are provided in supplementary materials (Tables S3 and S4). Correlations between different metal concentrations are presented in (Table 3). Silver, aluminum, iron, titanium and partially lithium show a strong correlation.

The fluxes per m-2 are provided in table S5. An extrapolation of the measured fluxes to the area covered by mangroves in the Red Sea (135 km2), results as an example in 50 kg y-1 for cadmium, 570 kg y-1 for copper, 74080 kg y-1 for Aluminium, and 97440 kg y-1 for

Iron.

177

Figure 2: The increase or decrease of metal elements with the age of Avicennia marina leaves in the Central Red Sea. The slopes of the fitted linear regressions provide an estimate of metal accumulation rate (units mg metal gDW-1 day-1). The red line is the fitted linear regressions. The elements are ordered based on their goodness of the fit.

178

0.007 0.08 1.2 0.0014

0.006 0.0012 1 0.06 0.005 0.001

0.8 0.004 0.0008 0.04 0.003 0.6 0.0006

0.002 0.0004 0.02 0.4

0.001 0.0002

0.2 0 0 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 001 0 50 100 150 200 250 300 350 400 0 0 0002

0.002 0.0014

0.007 0.0012 0.015

0.0015 0.006 0.001

0.005 0.0008 0.01 0.001 0.004

0.0006

0.003

0.0004 0.0005 0.002 0.005 0.0002 0.001

0 0 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 50 0 0 0002 0 001 0 100 150 200 250 300 350 400

1.6 0.25

0.0012 0.005 1.4

0.2 0.001 1.2 0.004

1 0.0008 0.15 0.003

0.8 0.0006 0.002 0.1 0.6

0.0004

0.4 0.001 0.05 0.0002 0.2 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 50 0 0 0 0 100 150 200 250 300 350 400

0.03

0.09 0.0016 0.025 0.005 0.08 0.0014

0.07 0.0012 0.02 0.004 0.06 0.001 0.015 0.003 0.05 0.0008

0.04 0.0006 0.01 0.002

0.03 0.0004

0.005 0.001 0.0002 0.02

0 0.01 0 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 0002 0 50 100 150 200 250 300 350 400 0

0.06 0.035 0.09 0.02

0.08 0.03 0.05

0.07 0.015 0.025

0.06 0.04 0.02

0.05 0.01 0.03 0.015

0.04

0.01 0.02 0.005 0.03

0.005 0.02 0.01 0 0 0.01

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 0 0 005

0.09

0.08 0.002

0.07

0.06 0.0015

0.05

0.04 0.001

0.03

0.02 0.0005

0.01

0 50 100 150 200 250 300 350 400 0 50 0 0 100 150 200 250 300 350 400

Figure 3: The increase of metal elements with the age of Avicennia marina leaves in the

Central Red Sea. The slopes of the fitted linear regressions provide an estimate of metal accumulation rate (units mg metal leaf-1 day-1). The red line is the fitted linear regressions.

The elements are ordered alphabetically. 179

0.0004 0.025 0.25 0.002 0.00035

0.02 0.2 0.0003 0.0015

0.00025 0.015 0.15 0.0002 0.001

0.01 0.00015 0.1

0.0005 0.0001 0.005

0.00005 0.05

0 0 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 0 00005

0.0025 0.0005

0.0004 0.002 0.0015 0.0004

0.0003 0.0015 0.0003

0.001

0.0002 0.001 0.0002

0.0001 0.0001 0.0005 0.0005

0 0 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0

0.07 0.4 0.0003 0.0025

0.06 0.35 0.00025 0.002

0.3 0.05

0.0002 0.25 0.0015 0.04

0.2 0.00015 0.03 0.001

0.15 0.0001 0.02 0.0005 0.1

0.00005 0.01 0.05 0

0 50 100 150 200 250 300 350 400 50 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 0 0 100 150 200 250 300 350 400 0

0.008 0.002 0.0005 0.035 0.007

0.03 0.0004 0.006 0.0015

0.005 0.025

0.0003

0.004 0.02 0.001 0.003 0.0002 0.015

0.002

0.01 0.0001 0.0005 0.001

0.005 0 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 50 0 0 001 0 0 100 150 200 250 300 350 400

0.006 0.035 0.016

0.005 0.014 0.03 0.015

0.012 0.004 0.025

0.01 0.02 0.003 0.01 0.008

0.015 0.002 0.006

0.01 0.005 0.001 0.004

0.005 0.002 0

0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 50 0 0 0 0 100 150 200 250 300 350 400

0.014 0.0009

0.012 0.0008

0.0007 0.01

0.0006

0.008 0.0005

0.006 0.0004

0.0003 0.004

0.0002

0.002 0.0001

0 50 100 150 200 250 300 350 400 50 0 0 0 100 150 200 250 300 350 400

Table 3: Correlation for elements concentration mg g DW-1 and probability (** p<0.01, * P between 0.01 and 0.05, and no star P>0.05).

180

Discussion

Although there are doubts about whether concentrations in roots and shoots are higher or similar to those found in sediments (Alongi, Clough et al. 2003, MacFarlane, Pulkownik

et al. 2003, MacFarlane, Koller et al. 2007), in the long term mangroves act as heavy metals

sinks by taking them in the shoot and root tissues. Therefore mangroves can be used for

coastal phytoremediation. However, when considering mangroves for phytoremediation

the short term effects of leaves shedding should be taken into consideration. Table. 4,

presents heavy metals concentrations in mangrove leaves in different species and

environments. The variability suggests that as found in this and other studies (Saenger and

McConchie 2004, Naidoo, Hiralal et al. 2014), the concentration in the leaves reflect the

concentration in the sediment, and therefore leaves can be used a bio-indicators. Further,

our study relating the heavy metals concentration to leave age indicates that for most 181

metals the content increases with the leaf growth (stable concentration), with no

reabsorption before shedding. This implies that in the short term mangroves remobilize

heavy metals that were deep in the sediment by up taking them with their roots system and

bringing them back to the surface of the sediment through leaves shedding.

Some metals present a different behavior. Copper concentration in the leaves decreases

with age, suggesting that the A. marina reabsorbs copper the same way it does with other

nutrients (Usman, Alkredaa et al. 2013). A similar effect for copper concentrations has been

observed in other mangrove species were concentrations were higher in the small leaves

than in the mature leaves (Saenger and McConchie 2004, Pinheiro, e Silva et al. 2012). In

plants, copper is essential in various metabolic processes from photosynthesis to lignin

synthesis but in high concentrations can be toxic and therefore is under tight regulation

(Pilon, Abdel-Ghany et al. 2006). The fact that is being reabsorbed by the plant suggests

that it might be a limiting factor in the Central Red Sea area. On the other, hand vanadium

and cadmium concentrations in the leaves increase with age. Whether vanadium is

essential for plants is under debate, but there is evidence that at high concentrations it has deleterious effects (Imtiaz, Rizwan et al. 2015). Cadmium is considered a non-essential metal with toxic effects even at low concentrations (Gallego, Pena et al. 2012). For both the increase in concentration suggests that A. marina has the capacity to detoxify by transferring the two elements to leaves that will be shed.

The observed concentrations of heavy metals in the leaves of A. marina in the Central

Red Sea are generally in the lower range of values observed around the planet (Table. 4), including the southern Red Sea (Usman, Alkredaa et al. 2013) which suggest that mangroves in the area are not under severe heavy metal pollution stress. 182

Table 4: Comparing different mangrove species and locations around the globe, the

selected results are about the leaves only and the unit -1 and mg kg-1), species

abbreviation: (A.i, Acanthus ilicifolius ;A.c, Aegiceras corniculatums are in (μg ;g A.a, Avicennia alba; A.o,

Avicennia officinalis; A.m, Avicennia marina ;B.c, Bruguiera cylindrical ;B.g, Bruguiera

gymnorhiza ;C.d, Ceriops decandra ;C.t, Ceriops tagal ;E.a, Excoecaria agallocha ;H.f,

Heritiera fomes ;H.t, Hibiscus tiliaceus ;K.c, Kandelia candel ;L.r, Lumitzera racemose ;R.a,

Rhizophora apiculate ;R.m, Rhizophora mucronata ;R.s, Rhizophora stylosa ;R, Rhizophora

mangle ;S.c, Sonneratia caseolaris ).

Ref spp plant A B C M N part Site Ag Al Cd Cr Cu Fe Li Mn Ni Pb Rb Se Sr Ti V Zn Zr s e o o b

our A.m leaves 12. 207. 2. 0. 0.2 0. 0. 2.1 262. 0. 23.2 2. 0. 1.8 8.6 57. 7. 57. 16. 16. 0. 8.5 study 35 78 16 02 1 16 87 8 02 45 4 82 09 1 6 85 43 85 45 06 71 27. 373. 2. 0. 0.1 0. 1. 485. 0. 28.1 1. 0. 2.1 8.7 59. 5. 59. 18. 14. 10. 2.9 1 94 07 15 07 7 33 57 87 55 2 92 22 2 2 76 91 76 05 69 34 31. 413. 3. 0. 0.1 0. 1. 3.0 555. 0. 101. 1. 0. 1.5 54. 5. 54. 19. 20. 16. 0. 7.9 09 86 29 05 1 34 23 6 12 53 57 45 37 1 8 3 8 42 72 89 84 23. 339. 2. 0. 0. 1. 3.1 439. 0. 23.9 1. 0. 1.4 8.4 58. 7. 58. 16. 21. 6.5 0. 0.2 04 16 73 05 2 19 6 34 45 7 78 21 1 4 48 19 48 21 27 4 83 propagul 1.2 26.1 2. 0. 0.3 0. 0. 4.0 33.6 0. 1. 0.7 5.3 8. 5.3 1.4 0.3 6.0 0. 3.61 0 7.9 es 8 1 82 37 5 3 7 4 7 15 34 8 4 45 4 7 3 1 53 A.m leaves (Usma n, 1.0 9. 35 29. Alkred 2.3 4 3 6.6 5 aa et al. Red Sea Red 2013) A.o leaves 14. 225. 23. 10 87.8 (Agora 78 3 21 7.8 moort R.a 10. 103. 167. 12. 16.

hy, 25 8 4 23 8 Chen R.m 19. 140. 12. 40. 391 et al. 9 2 61 3 2008) E.a 8.1 246. 27. 76. 91.9 2 7 35 6 B.c 17. 169. 219. 17. 11

46 2 4 39 6.9 C.d 95. 181. 235. 11. 9.3 05 1 6 82 A.c 13. 189. 12. 12. 51.1 39 5 91 8 A.i 13. 105. 15. 67. India 293 79 7 99 5 (Birch, A.m leaves 1. 0.1 0. 0. 0.6 6.1 15 1.3 11 Nath et 8 1 09 42 2 al. 0.0 0. 2. 1 20 66 3.2 4.6 19 2015) 2 25 7 0. 0.0 0. 0. 0.4 0.3 12 23 14 38 1 05 28 5 6 Ref spp plant A B C M N Ag Al Cd Cr Cu Fe Li Mn Ni Pb Rb Se Sr Ti V Zn Zr part s e o o b (Nath, A.m leaves Birch 0. 0.0 0. 0. 0.6 0.7 10 119 39 14 et al. 48 1 07 49 1 4 Australia 2014) (Pinhe R Green 0. 1.8 365. iro, e Silva et 42 3 35 al.

Senescen 2012) 0. 0.7 302. t

Brazil 44 9 1

A.o, young (Chow A.m, dhury, A.a, 0. 1. 1. 13. 106. 206. 1.5 1.3 21. 0.4 Favas B.g, 05 19 19 39 8 21 5 2 88 et al. C.t,

2015) R.a, India 183

A.c,E. mature a, L.r, 0. 0.3 0. 1. 12. 91.6 357. 1.3 1.6 21. H.f, 1 7 65 15 44 6 76 8 5 16 S.c

B.g leaves 0.1 0. 12 0.5 1.7 12 (Wang, 4 6 Qiu et A.i 0.0 3 17 2.1 1.6 69 al. 9 2013) K. c 0.1 3. 8 0.5 0.6 21 2 3 A.m 0.1 0. 10 0.2 ND 12 4 2

A.c 0.0 1. 12 1.3 0.9 17 9 8 S.c 0.1

China 1 8 0.7 0.7 40 5 (Sadiq A.m leaves and Zaidi 107. 1.8 0. 1. 4.4 129. 17.8 0. 2.0 6.8 42. 1.2 11. 1.7 1994) 16 9 82 38 2 41 7 92 8 9 59 4 38 Arabian Arabian Gulf

A.m young 16. 35. (Saeng leaves 2.9 er and 01 6 9 McCon ≤0. chie A.c 12. 43. 2004) 3.4 01 1 7 ≤0. A.m 20. 6.8 3 01 9 ≤0. A.c 20. 5.1 2.4 01 7 ≤0. H.t 23. 8.3 3.6 01 5 ≤0. E.a 13. 38. 4.3 01 3 5 ≤0. B.g 16. 7.6 2.5 01 4 ≤0. A.m 0.0 18. 34. 2.4 1 3 6

A.c 20. 38. 4.7 01 4 9 ≤0. R.s 19. 29. 1.6 01 6 2 ≤0. E.a 16. 9.2 2.8 01 7 ≤0. C.t 0.0 13. 20. 4 2 4 1

A.m 3.7 1 11 01 ≤0. R.s 2.3 1 5.7 01 ≤0. A.m old 10. 25. leaves 3.2 01 9 2 ≤0. A.c 30. 8.4 4 01 4 ≤0. A.m 17. 4.7 3.8 01 4 ≤0. A.c 13. 3.6 3.6 01 7 ≤0. H.t 24. 7.5 4.2 01 2 ≤0. E.a 10. 34. 5 01 5 3 B.g ≤0. 16. 7.2 3.2 01 8 A.m ≤0. 12. 22. 2.7 01 2 2 A.c ≤0. 10. 28. 6.1 01 7 4 R.s 0.0≤0. 13. 2.8 22 4 3 E.a 0.0 Australia 8.7 4.2 9.5 4 184

C.t 0.0 11. 21. 3.6 2 5 8 A.m 3.2 1 9.6 01 R.s ≤0. 2.1 1 4.4 01 A.m young 99.6 6. <0.≤0. <0 3.7 <0 1.6 0.9 <0. 12.

leaves 4 12 05 .1 3 .1 8 5 1 9 5. <0. <0 <0 <0. 13. 286 4.4 1.7 0.9 2 05 .1 .1 1 1 4. <0. <0 <0 <0. 11. 417 2.8 1.8 1 8 05 .1 .1 1 2 7. <0. <0 <0 <0. 12. 286 6.6 1.6 1.3 8 05 .1 .1 1 6 11 <0. <0 <0 <0. 256 6 2.1 1.3 8.9 .3 05 .1 .1 1 6. <0. <0 <0 <0. 131 4 1.6 1.1 9.6 1 05 .1 .1 1 9. <0. <0 <0 <0. 10. 293 3.5 1.8 1 2 05 .1 .1 1 3 <0. <0 <0 <0. 10. 127 6 3 1.5 1 05 .1 .1 1 4 4. <0. <0 <0 <0. 10. 69 3 1.5 1 9 05 .1 .1 1 8 <0. <0 <0 <0. 12. 87 4 2.7 1.6 0 05 .1 .1 1 6 3. <0. <0 <0 <0. 88 2.4 1.4 0.9 7.9 8 05 .1 .1 1 leaf litter <0. 32. 68. 4.6 01 3 7 A.m leaves 9.8 8.7 14. 1.3 13 (Parva 9 6 44 resh, 0.7 12. 14. 6.9 17.

Abedi 4 05 98 6 75 et al. 1.0 9.3 7.0 8.2 19.

2011) 9 3 5 4 75 0.7 10. 5.6 7.7 20.

9 66 1 6 74 0.7 9.2 6.7 7.8 20.

9 9 3 4 07 0.8 9.4 20. Iran 6.2 8.8 8 5 69

Conclusions

- Heavy metal concentrations in mangrove leaves in the Central Red Sea present

significant differences between impacted and non-impacted sites.

- Values are in the lower range of those observed around the globe suggesting a

limited impact. 185

- Most metals concentrations remain constant with leaf age. Copper decreases

suggesting reabsorption whereas vanadium and cadmium increase suggesting a

detoxifying mechanism.

- Because concentration remains constant with age, the content increases with the

growth of the leaf, and therefore mangroves partially remove heavy metals from the

deep sediment (roots system) to the surface of the sediment when shedding the

leaves.

Acknowledgements

The research reported in this paper was supported by King Abdullah University of Science

and Technology. We thank the Costal and Marine Resources core lab and Analytical core lab

for their help.

Author Contributions statement

H.A., C.M.D and X.I designed the study, H.A. carried out the field and lab work, H.A., C.M.D, and X.I. did the statistical analysis and H.A., C.M.D and X.I. wrote the manuscript. Authors declare no conflict of interest.

References

Agoramoorthy, G., F.-A. Chen and M. J. Hsu (2008). "Threat of heavy metal pollution in halophytic and mangrove plants of Tamil Nadu, India." Environmental Pollution 155(2): 320-326.

Alongi, D., B. Clough and A. Robertson (2005). "Nutrient-use efficiency in arid-zone forests of the mangroves Rhizophora stylosa and Avicennia marina." Aquatic botany 82(2): 121- 131.

Alongi, D. M., B. F. Clough, P. Dixon and F. Tirendi (2003). "Nutrient partitioning and storage in arid-zone forests of the mangroves Rhizophora stylosa and Avicennia marina." Trees 17(1): 51-60. 186

Appenroth, K.-J. (2010). "What are “heavy metals” in Plant Sciences?" Acta Physiologiae Plantarum 32(4): 615-619.

Baker, A. J. (1981). "Accumulators and excluders‐strategies in the response of plants to heavy metals." Journal of plant nutrition 3(1-4): 643-654.

Birch, G., B. Nath and P. Chaudhuri (2015). "Effectiveness of remediation of metal- contaminated mangrove sediments (Sydney estuary, Australia)." Environmental Science and Pollution Research 22(8): 6185-6197.

Bothe, H. (2011). Plants in heavy metal soils. Detoxification of heavy metals, Springer: 35- 57.

Chowdhury, R., P. J. Favas, J. Pratas, M. Jonathan, P. S. Ganesh and S. K. Sarkar (2015). "Accumulation of trace metals by mangrove plants in Indian Sundarban Wetland: Prospects for Phytoremediation." International journal of phytoremediation 17(9): 885-894.

Duarte, C. M., N. Marba, N. Agawin, J. Cebrian, S. Enriquez, M. D. Fortes, M. E. Gallegos, M. Merino, B. Olesen, K. Sandjensen, J. Uri and J. Vermaat (1994). "Reconstruction of seagrass dynamics: age determinations and associated tools for the seagrass ecologist." Marine Ecology Progress Series 107(1-2): 195-209.

Erickson, R. O. and F. J. Michelini (1957). "The plastochron index." American Journal of Botany: 297-305.

Gallego, S. M., L. B. Pena, R. A. Barcia, C. E. Azpilicueta, M. F. Iannone, E. P. Rosales, M. S. Zawoznik, M. D. Groppa and M. P. Benavides (2012). "Unravelling cadmium toxicity and tolerance in plants: insight into regulatory mechanisms." Environmental and Experimental Botany 83: 33-46.

Garbisu, C. and I. Alkorta (2001). "Phytoextraction: a cost-effective plant-based technology for the removal of metals from the environment." Bioresource technology 77(3): 229-236.

Imtiaz, M., M. S. Rizwan, S. Xiong, H. Li, M. Ashraf, S. M. Shahzad, M. Shahzad, M. Rizwan and S. Tu (2015). "Vanadium, recent advancements and research prospects: A review." Environment international 80: 79-88.

Kabata-Pendias, A. (2010). Trace elements in soils and plants, CRC press.

Lasat, M. M. (2002). "Phytoextraction of toxic metals." Journal of environmental quality 31(1): 109-120.

Lin, P. and W.-q. Wang (2001). "Changes in the leaf composition, leaf mass and leaf area during leaf senescence in three species of mangroves." Ecological Engineering 16(3): 415- 424. 187

MacFarlane, G., A. Pulkownik and M. Burchett (2003). "Accumulation and distribution of heavy metals in the grey mangrove,< i> Avicennia marina(Forsk.) Vierh.: biological indication potential." Environmental Pollution 123(1): 139-151.

MacFarlane, G. R., C. E. Koller and S. P. Blomberg (2007). "Accumulation and partitioning of heavy metals in mangroves: a synthesis of field-based studies." Chemosphere 69(9): 1454- 1464.

Moore, M. T., R. Kröger and C. R. Jackson (2011). "The role of aquatic ecosystems in the elimination of pollutants." Ecological Impacts of Toxic Chemicals: 225-237.

Murray, F. (1985). "Cycling of fluoride in a mangrove community near a fluoride emission source." Journal of applied ecology: 277-285.

Nagajyoti, P., K. Lee and T. Sreekanth (2010). "Heavy metals, occurrence and toxicity for plants: a review." Environmental Chemistry Letters 8(3): 199-216.

Naidoo, G., T. Hiralal and Y. Naidoo (2014). "Ecophysiological responses of the mangrove Avicennia marina to trace metal contamination." Flora - Morphology, Distribution, Functional Ecology of Plants 209(1): 63-72.

Nath, B., G. Birch and P. Chaudhuri (2014). "Assessment of sediment quality in Avicennia marina-dominated embayments of Sydney Estuary: the potential use of pneumatophores (aerial roots) as a bio-indicator of trace metal contamination." Sci Total Environ 472: 1010- 1022.

eavy metal toxicity stress in plants: biological and biotechnological tools." Biotechnology advances 32(1): 73-86. Ovečka, M. and T. Takáč (2014). "Managing h Parvaresh, H., Z. Abedi, P. Farshchi, M. Karami, N. Khorasani and A. Karbassi (2011). "Bioavailability and concentration of heavy metals in the sediments and leaves of grey mangrove, Avicennia marina (Forsk.) Vierh, in Sirik Azini creek, Iran." Biological trace element research 143(2): 1121-1130.

Pilon, M., S. E. Abdel-Ghany, C. M. Cohu, K. A. Gogolin and H. Ye (2006). "Copper cofactor delivery in plant cells." Current opinion in plant biology 9(3): 256-263.

Pinheiro, M. A. A., P. P. G. e Silva, L. F. de Almeida Duarte, A. A. Almeida and F. P. Zanotto (2012). "Accumulation of six metals in the mangrove crab Ucides cordatus (Crustacea: Ucididae) and its food source, the red mangrove Rhizophora mangle (Angiosperma: Rhizophoraceae)." Ecotoxicology and environmental safety 81: 114-121.

Robinson, B., S. Green, T. Mills, B. Clothier, L. Fung, S. Hurst, V. Snow, I. McIvor, S. Sivakumaran and C. van den Dijssel (2003). "Assessment of phytoremediation as best management practice for degraded environments." Environmental Management using Soil- Plant Systems (eds. LD Currie, RB Stewart and CWN Anderson). Occasional Report(16). 188

Sadiq, M. and T. Zaidi (1994). "Sediment composition and metal concentrations in mangrove leaves from the Saudi coast of the Arabian Gulf." Science of the total environment 155(1): 1-8.

Saenger, P. and D. McConchie (2004). "Heavy metals in mangroves: methodology, monitoring and management." Envis Forest Bulletin 4: 52.

Shahid, M., E. Ferrand, E. Schreck and C. Dumat (2013). Behavior and impact of zirconium in the soil–plant system: plant uptake and phytotoxicity. Reviews of Environmental Contamination and Toxicology Volume 221, Springer: 107-127.

Singh, O., S. Labana, G. Pandey, R. Budhiraja and R. Jain (2003). "Phytoremediation: an overview of metallic ion decontamination from soil." Applied Microbiology and Biotechnology 61(5-6): 405-412.

Spalla, S., C. Baffi, C. Barbante, C. Turretta, G. Cozzi, G. Beone and M. Bettinelli (2009). "Determination of rare earth elements in tomato plants by inductively coupled plasma mass spectrometry techniques." Rapid Communications in Mass Spectrometry 23(20): 3285-3292.

Usman, A. R., R. S. Alkredaa and M. Al-Wabel (2013). "Heavy metal contamination in sediments and mangroves from the coast of Red Sea: Avicennia marina as potential metal bioaccumulator." Ecotoxicology and environmental safety 97: 263-270.

Wang, L. K., J. P. Chen, Y.-T. Hung and N. K. Shammas (2009). Heavy metals in the environment, CRC Press.

Wang, Y., Q. Qiu, G. Xin, Z. Yang, J. Zheng, Z. Ye and S. Li (2013). "Heavy metal contamination in a vulnerable mangrove swamp in South China." Environmental monitoring and assessment 185(7): 5775-5787.

Zhang, Z.-W., X.-R. Xu, Y.-X. Sun, S. Yu, Y.-S. Chen and J.-X. Peng (2014). "Heavy metal and organic contaminants in mangrove ecosystems of China: a review." Environmental Science and Pollution Research 21(20): 11938-11950.

Zhou, H.-C., S.-D. Wei, Q. Zeng, L.-H. Zhang, N. F.-y. Tam and Y.-M. Lin (2010). "Nutrient and caloric dynamics in Avicennia marina leaves at different developmental and decay stages in Zhangjiang River Estuary, China." Estuarine Coastal and Shelf Science 87(1): 21-26.

189

Supplementary Materials Table S 1: Raw data for heavy metals content in Avicennia marina leaves

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) (days) Alkarrar 38 0.08 2.E-03 2.E- 2.E-04 4.E- 2.E-05 1.E-05 1.E-04 2.E- 3.E- 4.E- 1.E- 3.E-04 0.E+00 1.E- 4.E-04 6.E- 4.E-04 6.E- 9.E- 1.E- 7.E- 1.E- 1st island 02 06 04 02 05 03 04 03 03 04 03 03 04 76 0.13 2.E-03 4.E- 2.E-05 2.E- 2.E-05 3.E-05 1.E-04 3.E- 6.E- 7.E- 2.E- 4.E-04 0.E+00 2.E- 1.E-03 7.E- 4.E-04 7.E- 2.E- 1.E- 4.E- 9.E- 02 06 04 02 05 03 04 03 03 03 03 03 05 114 0.2 4.E-03 5.E- 4.E-04 9.E- 3.E-05 2.E-05 1.E-04 5.E- 7.E- 9.E- 5.E- 5.E-04 0.E+00 4.E- 2.E-03 1.E- 2.E-03 1.E- 2.E- 2.E- 3.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 152 0.22 4.E-03 7.E- 6.E-04 3.E- 7.E-05 0.E+00 2.E-04 5.E- 8.E- 1.E- 5.E- 7.E-04 0.E+00 6.E- 2.E-03 1.E- 0.E+00 1.E- 3.E- 3.E- 3.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 190 0.17 2.E-03 3.E- 3.E-05 2.E- 5.E-05 0.E+00 2.E-04 5.E- 4.E- 8.E- 4.E- 5.E-04 0.E+00 7.E- 1.E-03 1.E- 2.E-03 1.E- 1.E- 3.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 228 0.15 2.E-03 4.E- 4.E-04 9.E- 2.E-05 2.E-04 2.E-04 5.E- 4.E- 8.E- 4.E- 4.E-04 0.E+00 7.E- 2.E-03 1.E- 9.E-04 1.E- 1.E- 3.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 266 0.17 2.E-03 3.E- 6.E-04 3.E- 3.E-05 3.E-07 1.E-04 3.E- 4.E- 7.E- 3.E- 6.E-04 3.E-05 4.E- 2.E-03 1.E- 2.E-03 1.E- 1.E- 3.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 304 0.17 4.E-03 6.E- 5.E-04 3.E- 4.E-05 4.E-05 2.E-04 2.E- 7.E- 1.E- 5.E- 3.E-04 9.E-05 5.E- 2.E-03 1.E- 2.E-03 1.E- 3.E- 4.E- 1.E- 1.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 343 0.11 2.E-03 3.E- 2.E-04 3.E- 2.E-05 2.E-05 9.E-05 2.E- 4.E- 6.E- 3.E- 3.E-04 6.E-05 2.E- 4.E-04 7.E- 1.E-03 7.E- 1.E- 3.E- 1.E- 9.E- 02 06 04 02 05 03 04 03 03 03 03 03 05 Alkarrar 38 0.05 6.E-04 6.E- 0.E+00 4.E- 3.E-06 0.E+00 3.E-05 1.E- 6.E- 2.E- 8.E- 1.E-04 0.E+00 3.E- 6.E-04 2.E- 2.E-05 2.E- 2.E- 7.E- 5.E- 3.E- 2nd 03 07 04 03 05 04 05 03 03 04 04 04 05 island 76 0.19 1.E-03 2.E- 7.E-04 2.E- 4.E-05 2.E-05 2.E-04 5.E- 2.E- 7.E- 4.E- 1.E-04 0.E+00 2.E- 2.E-03 9.E- 9.E-04 9.E- 6.E- 2.E- 2.E- 1.E- 02 06 04 02 05 03 04 03 03 04 03 03 04 114 0.3 0.E+00 3.E- 6.E-04 3.E- 5.E-05 0.E+00 9.E-05 7.E- 3.E- 1.E- 7.E- 5.E-04 0.E+00 1.E- 2.E-03 2.E- 3.E-03 2.E- 8.E- 4.E- 2.E- 2.E- 02 06 04 02 04 03 04 02 02 04 03 03 04 152 0.29 8.E-04 4.E- 1.E-03 1.E- 7.E-05 3.E-05 1.E-04 6.E- 5.E- 1.E- 7.E- 8.E-04 0.E+00 1.E- 1.E-03 2.E- 4.E-03 2.E- 2.E- 5.E- 2.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 190 0.31 2.E-03 3.E- 4.E-04 2.E- 9.E-05 0.E+00 1.E-04 6.E- 3.E- 1.E- 9.E- 1.E-03 0.E+00 2.E- 4.E-03 1.E- 4.E-03 1.E- 8.E- 6.E- 2.E- 1.E- 02 06 04 02 04 03 04 02 02 04 03 03 04 228 0.21 0.E+00 3.E- 5.E-04 2.E- 7.E-05 3.E-05 3.E-04 2.E- 4.E- 9.E- 6.E- 8.E-04 3.E-05 3.E- 2.E-03 1.E- 1.E-03 1.E- 1.E- 5.E- 1.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 Petro 38 0.1 8.E-04 1.E- 2.E-04 6.E- 4.E-07 0.E+00 0.E+00 4.E- 1.E- 5.E- 2.E- 3.E-05 0.E+00 1.E- 8.E-04 3.E- 8.E-04 3.E- 4.E- 6.E- 2.E- 3.E- Rabigh 02 07 04 02 05 03 04 03 03 04 04 03 05 under 38 0.1 0.E+00 9.E- 1.E-04 6.E- 0.E+00 2.E-05 6.E-05 4.E- 1.E- 5.E- 2.E- 3.E-05 0.E+00 5.E- 5.E-04 4.E- 1.E-03 4.E- 3.E- 7.E- 1.E- 4.E- pipe 03 07 04 02 05 03 05 03 03 04 04 03 05 76 0.14 8.E-04 2.E- 2.E-04 1.E- 5.E-06 0.E+00 6.E-05 7.E- 3.E- 6.E- 4.E- 0.E+00 0.E+00 5.E- 7.E-04 6.E- 0.E+00 6.E- 8.E- 1.E- 2.E- 6.E- 02 06 04 02 05 03 05 03 03 04 03 03 05 114 0.12 3.E-03 5.E- 4.E-04 2.E- 1.E-05 5.E-05 1.E-04 6.E- 6.E- 6.E- 4.E- 2.E-05 0.E+00 2.E- 1.E-03 6.E- 3.E-04 6.E- 2.E- 7.E- 2.E- 1.E- 02 06 04 02 05 03 04 03 03 03 04 03 04 152 0.12 2.E-03 3.E- 3.E-04 2.E- 5.E-06 0.E+00 9.E-05 5.E- 4.E- 5.E- 4.E- 2.E-05 0.E+00 9.E- 9.E-04 8.E- 8.E-04 8.E- 1.E- 9.E- 2.E- 8.E- 02 06 04 02 05 03 05 03 03 03 04 03 05 190 0.29 5.E-03 7.E- 7.E-04 2.E- 1.E-05 7.E-05 3.E-04 9.E- 9.E- 1.E- 1.E- 1.E-03 0.E+00 3.E- 3.E-03 2.E- 2.E-03 2.E- 3.E- 4.E- 4.E- 2.E- 02 05 04 02 04 02 04 02 02 03 03 03 04 228 0.29 9.E-03 1.E- 8.E-04 7.E- 1.E-05 1.E-04 4.E-04 1.E- 1.E- 1.E- 2.E- 4.E-04 0.E+00 2.E- 0.E+00 2.E- 3.E-03 2.E- 5.E- 5.E- 5.E- 2.E- 01 06 03 01 04 02 04 02 02 03 03 03 04 266 0.2 5.E-03 7.E- 7.E-04 5.E- 2.E-05 4.E-05 2.E-04 6.E- 1.E- 9.E- 1.E- 6.E-04 4.E-05 3.E- 2.E-03 1.E- 3.E-04 1.E- 4.E- 4.E- 3.E- 2.E- 02 06 04 01 05 02 04 02 02 03 03 03 04 190

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) (days) Petro 304 0.36 1.E-02 1.E- 4.E-04 9.E- 3.E-05 6.E-05 4.E-04 1.E- 2.E- 1.E- 2.E- 1.E-03 8.E-05 5.E- 2.E-03 2.E- 2.E-03 2.E- 6.E- 8.E- 6.E- 3.E- Rabigh 01 06 03 01 04 02 04 02 02 03 03 03 04 under 343 0.44 3.E-02 3.E- 1.E-03 1.E- 4.E-05 2.E-04 8.E-04 1.E- 4.E- 3.E- 4.E- 7.E-04 2.E-05 8.E- 3.E-03 3.E- 4.E-03 3.E- 1.E- 1.E- 7.E- 5.E- pipe 01 05 03 01 04 02 04 02 02 02 02 03 04 381 0.28 1.E-02 2.E- 1.E-03 7.E- 5.E-05 2.E-04 5.E-04 6.E- 2.E- 2.E- 2.E- 1.E-03 8.E-05 5.E- 3.E-03 2.E- 0.E+00 2.E- 8.E- 6.E- 1.E- 3.E- 01 06 04 01 04 02 04 02 02 03 03 02 04 381 0.28 9.E-03 1.E- 1.E-03 7.E- 3.E-05 1.E-04 3.E-04 4.E- 1.E- 1.E- 2.E- 1.E-03 5.E-05 3.E- 3.E-03 2.E- 2.E-04 2.E- 5.E- 6.E- 1.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 02 04 Petro 38 0.08 2.E-03 3.E- 1.E-04 2.E- 4.E-06 4.E-05 1.E-04 2.E- 4.E- 4.E- 8.E- 2.E-04 1.E-05 1.E- 1.E-03 5.E- 6.E-04 5.E- 1.E- 2.E- 1.E- 9.E- Rabigh 02 06 04 02 05 03 04 03 03 03 03 03 05 few 38 0.08 2.E-03 3.E- 2.E-04 2.E- 6.E-06 4.E-05 9.E-05 2.E- 4.E- 3.E- 8.E- 2.E-04 7.E-06 8.E- 8.E-04 5.E- 4.E-04 5.E- 1.E- 2.E- 2.E- 7.E- meters 02 06 04 02 05 03 05 03 03 03 03 03 05 from pipe 76 0.15 5.E-03 6.E- 3.E-04 4.E- 2.E-05 5.E-05 2.E-04 2.E- 7.E- 6.E- 2.E- 0.E+00 1.E-04 2.E- 7.E-04 8.E- 2.E-04 8.E- 2.E- 4.E- 2.E- 1.E- 02 06 04 02 05 02 04 03 03 03 03 03 04 114 0.24 9.E-03 1.E- 8.E-04 5.E- 3.E-05 1.E-04 3.E-04 6.E- 2.E- 2.E- 4.E- 1.E-04 3.E-04 5.E- 4.E-03 1.E- 2.E-03 1.E- 6.E- 8.E- 4.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 152 0.23 2.E-02 2.E- 9.E-04 7.E- 5.E-05 2.E-04 4.E-04 5.E- 2.E- 2.E- 3.E- 3.E-05 3.E-04 7.E- 2.E-03 1.E- 2.E-03 1.E- 9.E- 7.E- 3.E- 3.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 190 0.36 2.E-02 3.E- 1.E-03 1.E- 5.E-05 3.E-04 7.E-04 8.E- 4.E- 3.E- 6.E- 8.E-05 4.E-04 1.E- 3.E-03 2.E- 6.E-04 2.E- 1.E- 1.E- 6.E- 4.E- 01 05 04 01 04 02 03 02 02 02 02 03 04 228 0.39 1.E-02 2.E- 1.E-03 2.E- 2.E-04 4.E-04 7.E-04 9.E- 2.E- 2.E- 5.E- 9.E-04 5.E-04 7.E- 3.E-03 2.E- 2.E-03 2.E- 7.E- 1.E- 5.E- 5.E- 01 04 04 01 04 02 04 02 02 03 02 03 04 266 0.33 1.E-02 1.E- 2.E-03 4.E- 8.E-05 4.E-05 5.E-04 3.E- 2.E- 2.E- 4.E- 4.E-04 2.E-04 5.E- 2.E-03 1.E- 1.E-03 1.E- 5.E- 1.E- 5.E- 2.E- 01 05 04 01 04 02 04 02 02 03 02 03 04 266 0.32 9.E-03 1.E- 5.E-04 2.E- 5.E-05 1.E-04 3.E-04 3.E- 2.E- 2.E- 4.E- 8.E-05 2.E-04 6.E- 3.E-03 2.E- 2.E-03 2.E- 6.E- 1.E- 5.E- 2.E- 01 05 04 01 04 02 04 02 02 03 02 03 04 Petro 38 0.03 3.E-03 3.E- 2.E-04 2.E- 0.E+00 0.E+00 8.E-05 3.E- 5.E- 4.E- 3.E- 2.E-05 0.E+00 9.E- 3.E-04 2.E- 2.E-04 2.E- 2.E- 3.E- 9.E- 6.E- Rabigh 02 06 04 02 05 03 05 03 03 03 04 04 05 Far from 76 0.17 5.E-03 6.E- 5.E-04 3.E- 2.E-06 3.E-05 2.E-04 6.E- 9.E- 7.E- 2.E- 3.E-04 5.E-05 1.E- 5.E-04 9.E- 9.E-04 9.E- 3.E- 3.E- 2.E- 1.E- the pipe 02 06 04 02 05 02 04 03 03 03 03 03 04 114 0.22 4.E-03 7.E- 1.E-03 4.E- 3.E-05 4.E-05 2.E-04 7.E- 9.E- 9.E- 3.E- 3.E-04 4.E-05 2.E- 2.E-03 1.E- 1.E-03 1.E- 3.E- 4.E- 2.E- 1.E- 02 06 04 02 05 02 04 02 02 03 03 03 04 152 0.3 1.E-02 1.E- 9.E-04 5.E- 3.E-05 2.E-04 4.E-04 8.E- 2.E- 1.E- 6.E- 7.E-04 4.E-05 5.E- 1.E-03 2.E- 1.E-03 2.E- 6.E- 6.E- 3.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 190 0.22 4.E-03 6.E- 1.E-03 5.E- 3.E-05 3.E-05 2.E-04 8.E- 9.E- 9.E- 4.E- 5.E-04 1.E-04 2.E- 1.E-03 1.E- 2.E-03 1.E- 3.E- 6.E- 1.E- 1.E- 02 06 04 02 05 02 04 02 02 03 03 03 04 228 0.24 6.E-03 1.E- 1.E-03 6.E- 3.E-05 2.E-05 3.E-04 5.E- 2.E- 2.E- 6.E- 9.E-05 2.E-04 5.E- 3.E-03 1.E- 0.E+00 1.E- 5.E- 7.E- 1.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 Economic 38 0.05 1.E-03 2.E- 0.E+00 9.E- 7.E-06 1.E-05 7.E-05 2.E- 3.E- 3.E- 1.E- 5.E-06 0.E+00 1.E- 5.E-04 2.E- 3.E-04 2.E- 9.E- 3.E- 5.E- 5.E- city 1st 02 07 04 02 05 03 04 03 03 04 04 04 05 island 38 0.05 1.E-03 2.E- 3.E-04 1.E- 5.E-06 8.E-06 8.E-05 2.E- 2.E- 3.E- 9.E- 3.E-05 0.E+00 7.E- 5.E-04 3.E- 8.E-04 3.E- 9.E- 4.E- 4.E- 6.E- 02 06 04 02 05 04 05 03 03 04 04 04 05 76 0.25 4.E-03 6.E- 6.E-04 3.E- 4.E-05 0.E+00 3.E-04 8.E- 7.E- 1.E- 5.E- 4.E-04 0.E+00 4.E- 2.E-03 2.E- 1.E-03 2.E- 3.E- 3.E- 3.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 114 0.29 6.E-03 6.E- 8.E-04 6.E- 5.E-05 5.E-04 5.E-04 8.E- 8.E- 1.E- 5.E- 8.E-04 0.E+00 1.E- 3.E-03 2.E- 2.E-03 2.E- 3.E- 3.E- 2.E- 2.E- 02 06 04 02 04 03 03 02 02 03 03 03 04

191

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) (days) Economic 152 0.43 3.E-03 1.E- 1.E-03 4.E- 4.E-04 4.E-04 2.E-03 2.E- 2.E- 3.E- 1.E- 2.E-03 2.E-04 2.E- 6.E-03 3.E- 2.E-03 3.E- 6.E- 5.E- 4.E- 8.E- city 1st 01 04 03 01 04 02 03 02 02 03 03 03 04 island 190 0.42 1.E-02 2.E- 1.E-03 8.E- 1.E-04 5.E-05 6.E-04 1.E- 2.E- 2.E- 9.E- 2.E-03 0.E+00 1.E- 2.E-03 3.E- 5.E-03 3.E- 7.E- 5.E- 3.E- 4.E- 01 05 03 01 04 03 03 02 02 03 03 03 04 228 0.47 1.E-02 2.E- 4.E-04 4.E- 1.E-04 2.E-04 6.E-04 1.E- 2.E- 2.E- 1.E- 8.E-04 0.E+00 1.E- 4.E-03 3.E- 3.E-03 3.E- 7.E- 5.E- 3.E- 4.E- 01 05 03 01 04 02 03 02 02 03 03 03 04 266 0.49 1.E-02 2.E- 1.E-03 2.E- 1.E-04 1.E-04 6.E-04 2.E- 3.E- 3.E- 1.E- 2.E-03 0.E+00 1.E- 1.E-03 3.E- 2.E-03 3.E- 1.E- 7.E- 4.E- 4.E- 01 05 03 01 04 02 03 02 02 02 03 03 04 304 0.44 1.E-02 2.E- 2.E-03 1.E- 7.E-05 0.E+00 6.E-04 1.E- 2.E- 2.E- 7.E- 2.E-03 0.E+00 9.E- 3.E-03 3.E- 4.E-03 3.E- 7.E- 6.E- 4.E- 3.E- 01 05 03 01 04 03 04 02 02 03 03 03 04 304 0.49 1.E-02 1.E- 1.E-03 9.E- 1.E-04 3.E-05 5.E-04 8.E- 1.E- 3.E- 8.E- 2.E-03 0.E+00 5.E- 4.E-03 3.E- 5.E-03 3.E- 5.E- 7.E- 3.E- 3.E- 01 06 04 01 04 03 04 02 02 03 03 03 04 Economic 38 0.06 0.E+00 5.E- 5.E-05 5.E- 0.E+00 0.E+00 3.E-05 4.E- 5.E- 3.E- 8.E- 6.E-06 0.E+00 8.E- 3.E-04 3.E- 1.E-04 3.E- 2.E- 3.E- 7.E- 4.E- city 2nd 03 07 04 03 05 04 05 03 03 04 04 04 05 38 0.07 1.E-04 6.E- 4.E-05 7.E- 0.E+00 6.E-07 3.E-05 4.E- 5.E- 3.E- 8.E- 4.E-05 0.E+00 8.E- 8.E-04 3.E- 1.E-05 3.E- 2.E- 4.E- 8.E- 4.E- 03 07 04 03 05 04 05 03 03 04 04 04 05 76 0.18 1.E-03 2.E- 3.E-04 2.E- 1.E-05 0.E+00 1.E-04 1.E- 2.E- 7.E- 2.E- 1.E-04 0.E+00 3.E- 2.E-03 9.E- 1.E-03 9.E- 7.E- 8.E- 2.E- 9.E- 02 06 03 02 05 03 05 03 03 04 04 03 05 114 0.22 2.E-03 3.E- 5.E-04 2.E- 0.E+00 0.E+00 1.E-04 9.E- 3.E- 8.E- 3.E- 3.E-05 0.E+00 2.E- 8.E-04 1.E- 1.E-03 1.E- 1.E- 1.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 152 0.24 5.E-03 5.E- 4.E-04 3.E- 1.E-05 0.E+00 2.E-04 8.E- 6.E- 9.E- 5.E- 3.E-05 0.E+00 2.E- 2.E-03 1.E- 0.E+00 1.E- 2.E- 2.E- 3.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 190 0.18 4.E-03 5.E- 2.E-04 4.E- 1.E-06 4.E-05 2.E-04 5.E- 7.E- 1.E- 4.E- 1.E-04 0.E+00 3.E- 1.E-03 9.E- 4.E-04 9.E- 3.E- 1.E- 2.E- 1.E- 02 06 04 02 04 03 04 03 03 03 03 03 04 228 0.26 5.E-03 6.E- 6.E-04 5.E- 2.E-05 0.E+00 2.E-04 7.E- 8.E- 1.E- 7.E- 1.E-03 0.E+00 3.E- 2.E-03 1.E- 2.E-03 1.E- 3.E- 3.E- 4.E- 1.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 266 0.38 5.E-03 9.E- 1.E-03 3.E- 3.E-05 4.E-05 3.E-04 8.E- 1.E- 2.E- 1.E- 5.E-04 4.E-05 3.E- 7.E-03 2.E- 0.E+00 2.E- 5.E- 7.E- 6.E- 3.E- 02 06 04 01 04 02 04 02 02 03 03 03 04 304 0.44 3.E-03 9.E- 0.E+00 1.E- 1.E-04 9.E-05 4.E-04 9.E- 1.E- 2.E- 1.E- 2.E-03 9.E-05 6.E- 9.E-04 2.E- 3.E-03 2.E- 4.E- 8.E- 6.E- 3.E- 02 04 04 01 04 02 04 02 02 03 03 03 04 343 0.47 7.E-03 1.E- 2.E-03 4.E- 1.E-04 6.E-05 4.E-04 1.E- 2.E- 2.E- 2.E- 1.E-03 7.E-05 9.E- 5.E-03 3.E- 1.E-06 3.E- 6.E- 1.E- 8.E- 4.E- 01 05 03 01 04 02 04 02 02 03 02 03 04 381 0.37 4.E-03 7.E- 1.E-03 2.E- 8.E-05 5.E-05 3.E-04 7.E- 9.E- 1.E- 2.E- 1.E-03 1.E-04 7.E- 4.E-03 2.E- 5.E-03 2.E- 3.E- 1.E- 6.E- 3.E- 02 05 04 02 04 02 04 02 02 03 02 03 04 381 0.34 3.E-03 6.E- 1.E-03 6.E- 5.E-05 1.E-04 3.E-04 5.E- 8.E- 2.E- 1.E- 1.E-03 0.E+00 3.E- 2.E-03 2.E- 2.E-03 2.E- 3.E- 8.E- 5.E- 2.E- 02 06 04 02 04 02 04 02 02 03 03 03 04 Economic 38 0.07 5.E-03 6.E- 2.E-04 3.E- 1.E-05 5.E-05 2.E-04 2.E- 8.E- 6.E- 3.E- 2.E-04 3.E-05 3.E- 8.E-04 5.E- 5.E-04 5.E- 3.E- 1.E- 7.E- 1.E- city 3rd 02 06 04 02 05 03 04 03 03 03 03 04 04 island 76 0.11 5.E-03 7.E- 2.E-04 3.E- 1.E-05 9.E-05 2.E-04 2.E- 9.E- 7.E- 5.E- 1.E-04 5.E-05 3.E- 4.E-04 8.E- 1.E-03 8.E- 3.E- 3.E- 9.E- 2.E- 02 06 04 02 05 03 04 03 03 03 03 04 04 114 0.17 1.E-02 2.E- 2.E-04 6.E- 3.E-05 1.E-04 4.E-04 4.E- 2.E- 1.E- 7.E- 6.E-04 1.E-04 6.E- 0.E+00 1.E- 4.E-04 1.E- 8.E- 3.E- 2.E- 3.E- 01 06 04 01 04 03 04 02 02 03 03 03 04 152 0.17 6.E-03 9.E- 0.E+00 3.E- 2.E-05 8.E-05 3.E-04 3.E- 1.E- 1.E- 6.E- 5.E-04 8.E-05 4.E- 1.E-03 1.E- 1.E-03 1.E- 4.E- 4.E- 1.E- 2.E- 02 06 04 01 04 03 04 02 02 03 03 03 04 190 0.27 2.E-02 2.E- 6.E-04 9.E- 5.E-05 2.E-04 2.E-03 9.E- 3.E- 2.E- 1.E- 1.E-04 3.E-04 1.E- 8.E-03 2.E- 2.E-03 2.E- 1.E- 6.E- 5.E- 5.E- 01 06 04 01 04 02 03 02 02 02 03 03 04

192

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) (days) Economic 228 0.21 9.E-03 1.E- 1.E-04 5.E- 3.E-05 1.E-04 4.E-04 3.E- 2.E- 1.E- 8.E- 1.E-05 1.E-04 6.E- 2.E-03 1.E- 2.E-03 1.E- 6.E- 5.E- 1.E- 3.E- city 3rd 01 06 04 01 04 03 04 02 02 03 03 03 04 island 266 0.22 9.E-03 1.E- 9.E-04 1.E- 3.E-05 8.E-05 3.E-04 3.E- 2.E- 1.E- 1.E- 0.E+00 2.E-04 6.E- 1.E-03 1.E- 5.E-04 1.E- 6.E- 6.E- 1.E- 3.E- 01 05 04 01 04 02 04 02 02 03 03 03 04 304 0.11 8.E-03 9.E- 4.E-04 4.E- 1.E-05 6.E-05 3.E-04 3.E- 1.E- 1.E- 6.E- 8.E-05 8.E-05 4.E- 1.E-03 8.E- 7.E-04 8.E- 5.E- 3.E- 8.E- 2.E- 02 06 04 01 04 03 04 03 03 03 03 04 04 Thuwal 38 0.04 1.E-03 2.E- 1.E-04 1.E- 6.E-06 1.E-05 6.E-05 1.E- 2.E- 3.E- 9.E- 6.E-05 0.E+00 8.E- 6.E-04 3.E- 6.E-04 3.E- 9.E- 4.E- 4.E- 4.E- land 02 06 04 02 05 04 05 03 03 04 04 04 05 76 0.08 3.E-03 3.E- 3.E-04 2.E- 1.E-05 2.E-05 1.E-04 3.E- 4.E- 5.E- 2.E- 2.E-04 0.E+00 8.E- 4.E-05 6.E- 7.E-04 6.E- 1.E- 1.E- 8.E- 8.E- 02 06 04 02 05 03 05 03 03 03 03 04 05 114 0.16 4.E-03 5.E- 3.E-04 2.E- 3.E-05 1.E-05 2.E-04 5.E- 7.E- 7.E- 4.E- 5.E-04 0.E+00 4.E- 1.E-03 1.E- 2.E-03 1.E- 3.E- 3.E- 1.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 152 0.19 4.E-03 7.E- 0.E+00 2.E- 5.E-05 2.E-05 2.E-04 6.E- 9.E- 9.E- 5.E- 6.E-04 0.E+00 3.E- 2.E-03 1.E- 2.E-04 1.E- 3.E- 4.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 190 0.26 4.E-03 6.E- 8.E-04 2.E- 4.E-05 5.E-05 3.E-04 7.E- 9.E- 1.E- 8.E- 5.E-04 4.E-05 5.E- 3.E-03 1.E- 1.E-03 1.E- 3.E- 6.E- 2.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 228 0.23 4.E-03 6.E- 5.E-04 2.E- 6.E-05 5.E-05 2.E-04 7.E- 8.E- 8.E- 6.E- 5.E-04 0.E+00 3.E- 9.E-04 1.E- 6.E-04 1.E- 3.E- 5.E- 2.E- 2.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 266 0.29 7.E-03 1.E- 1.E-03 3.E- 6.E-05 0.E+00 3.E-04 8.E- 1.E- 1.E- 1.E- 5.E-04 6.E-05 4.E- 2.E-03 2.E- 1.E-03 2.E- 5.E- 6.E- 2.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 304 0.41 1.E-02 1.E- 4.E-04 3.E- 4.E-04 4.E-04 7.E-04 2.E- 2.E- 2.E- 2.E- 1.E-03 4.E-04 7.E- 3.E-03 3.E- 2.E-03 3.E- 6.E- 1.E- 3.E- 7.E- 01 04 03 01 04 02 04 02 02 03 02 03 04 343 0.5 1.E-02 1.E- 2.E-03 5.E- 2.E-04 3.E-05 8.E-04 1.E- 2.E- 2.E- 2.E- 2.E-04 4.E-04 8.E- 6.E-03 3.E- 6.E-03 3.E- 7.E- 2.E- 4.E- 4.E- 01 06 03 01 04 02 04 02 02 03 02 03 04 343 0.29 1.E-02 2.E- 6.E-04 7.E- 4.E-05 1.E-04 5.E-04 6.E- 2.E- 2.E- 1.E- 8.E-05 2.E-04 5.E- 9.E-04 2.E- 2.E-03 2.E- 8.E- 8.E- 1.E- 3.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 Thuwal 38 0.03 1.E-03 2.E- 3.E-05 7.E- 0.E+00 5.E-06 4.E-05 5.E- 2.E- 2.E- 6.E- 0.E+00 0.E+00 5.E- 4.E-04 2.E- 6.E-04 2.E- 8.E- 3.E- 3.E- 3.E- Fringe 02 07 04 02 05 04 05 03 03 04 04 04 05 76 0.09 2.E-03 3.E- 2.E-04 1.E- 5.E-06 8.E-06 7.E-05 2.E- 3.E- 3.E- 1.E- 2.E-04 0.E+00 9.E- 4.E-04 5.E- 9.E-05 5.E- 1.E- 1.E- 5.E- 8.E- 02 06 04 02 05 03 05 03 03 03 03 04 05 114 0.12 4.E-04 3.E- 2.E-04 1.E- 1.E-05 0.E+00 1.E-04 2.E- 3.E- 3.E- 1.E- 4.E-04 0.E+00 9.E- 1.E-03 5.E- 7.E-04 5.E- 1.E- 2.E- 5.E- 7.E- 02 06 04 02 05 03 05 03 03 03 03 04 05 152 0.16 4.E-03 7.E- 4.E-04 3.E- 3.E-05 2.E-05 2.E-04 3.E- 8.E- 8.E- 3.E- 4.E-04 1.E-05 2.E- 8.E-04 8.E- 1.E-03 8.E- 3.E- 2.E- 9.E- 1.E- 02 06 04 02 05 03 04 03 03 03 03 04 04 190 0.24 7.E-03 9.E- 3.E-04 3.E- 1.E-05 0.E+00 2.E-04 5.E- 1.E- 9.E- 5.E- 6.E-04 2.E-06 3.E- 4.E-03 1.E- 2.E-03 1.E- 4.E- 5.E- 1.E- 2.E- 02 06 04 01 05 03 04 02 02 03 03 03 04 228 0.27 7.E-03 9.E- 1.E-03 3.E- 4.E-05 2.E-05 3.E-04 4.E- 1.E- 9.E- 5.E- 5.E-04 4.E-05 3.E- 1.E-03 1.E- 3.E-03 1.E- 4.E- 6.E- 1.E- 2.E- 02 06 04 01 05 03 04 02 02 03 03 03 04 266 0.24 3.E-03 5.E- 6.E-04 3.E- 3.E-05 9.E-05 2.E-04 4.E- 7.E- 8.E- 4.E- 0.E+00 0.E+00 3.E- 3.E-03 1.E- 4.E-04 1.E- 2.E- 7.E- 1.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 304 0.24 2.E-03 5.E- 1.E-03 2.E- 5.E-05 4.E-05 2.E-04 3.E- 7.E- 9.E- 4.E- 3.E-04 1.E-04 3.E- 3.E-03 1.E- 7.E-04 1.E- 2.E- 6.E- 6.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 04 04 343 0.17 2.E-03 4.E- 9.E-04 6.E- 1.E-05 0.E+00 1.E-04 3.E- 6.E- 7.E- 3.E- 3.E-05 9.E-05 1.E- 1.E-03 1.E- 1.E-03 1.E- 2.E- 5.E- 8.E- 1.E- 02 07 04 02 05 03 04 02 02 03 03 04 04

193

Table S 2: Raw data for heavy metals concentration in Avicennia marina leaves

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) (days) Alkarrar 38 0.08 2.E-03 2.E- 2.E-04 4.E- 2.E-05 1.E-05 1.E-04 2.E- 3.E- 4.E- 1.E- 3.E-04 0.E+00 1.E- 4.E-04 6.E- 4.E-04 6.E- 9.E- 1.E- 7.E- 1.E- 1st island 02 06 04 02 05 03 04 03 03 04 03 03 04 76 0.13 2.E-03 4.E- 2.E-05 2.E- 2.E-05 3.E-05 1.E-04 3.E- 6.E- 7.E- 2.E- 4.E-04 0.E+00 2.E- 1.E-03 7.E- 4.E-04 7.E- 2.E- 1.E- 4.E- 9.E- 02 06 04 02 05 03 04 03 03 03 03 03 05 114 0.2 4.E-03 5.E- 4.E-04 9.E- 3.E-05 2.E-05 1.E-04 5.E- 7.E- 9.E- 5.E- 5.E-04 0.E+00 4.E- 2.E-03 1.E- 2.E-03 1.E- 2.E- 2.E- 3.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 152 0.22 4.E-03 7.E- 6.E-04 3.E- 7.E-05 0.E+00 2.E-04 5.E- 8.E- 1.E- 5.E- 7.E-04 0.E+00 6.E- 2.E-03 1.E- 0.E+00 1.E- 3.E- 3.E- 3.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 190 0.17 2.E-03 3.E- 3.E-05 2.E- 5.E-05 0.E+00 2.E-04 5.E- 4.E- 8.E- 4.E- 5.E-04 0.E+00 7.E- 1.E-03 1.E- 2.E-03 1.E- 1.E- 3.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 228 0.15 2.E-03 4.E- 4.E-04 9.E- 2.E-05 2.E-04 2.E-04 5.E- 4.E- 8.E- 4.E- 4.E-04 0.E+00 7.E- 2.E-03 1.E- 9.E-04 1.E- 1.E- 3.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 266 0.17 2.E-03 3.E- 6.E-04 3.E- 3.E-05 3.E-07 1.E-04 3.E- 4.E- 7.E- 3.E- 6.E-04 3.E-05 4.E- 2.E-03 1.E- 2.E-03 1.E- 1.E- 3.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 304 0.17 4.E-03 6.E- 5.E-04 3.E- 4.E-05 4.E-05 2.E-04 2.E- 7.E- 1.E- 5.E- 3.E-04 9.E-05 5.E- 2.E-03 1.E- 2.E-03 1.E- 3.E- 4.E- 1.E- 1.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 343 0.11 2.E-03 3.E- 2.E-04 3.E- 2.E-05 2.E-05 9.E-05 2.E- 4.E- 6.E- 3.E- 3.E-04 6.E-05 2.E- 4.E-04 7.E- 1.E-03 7.E- 1.E- 3.E- 1.E- 9.E- 02 06 04 02 05 03 04 03 03 03 03 03 05 Alkarrar 38 0.05 6.E-04 6.E- 0.E+00 4.E- 3.E-06 0.E+00 3.E-05 1.E- 6.E- 2.E- 8.E- 1.E-04 0.E+00 3.E- 6.E-04 2.E- 2.E-05 2.E- 2.E- 7.E- 5.E- 3.E- 2nd 03 07 04 03 05 04 05 03 03 04 04 04 05 island 76 0.19 1.E-03 2.E- 7.E-04 2.E- 4.E-05 2.E-05 2.E-04 5.E- 2.E- 7.E- 4.E- 1.E-04 0.E+00 2.E- 2.E-03 9.E- 9.E-04 9.E- 6.E- 2.E- 2.E- 1.E- 02 06 04 02 05 03 04 03 03 04 03 03 04 114 0.3 0.E+00 3.E- 6.E-04 3.E- 5.E-05 0.E+00 9.E-05 7.E- 3.E- 1.E- 7.E- 5.E-04 0.E+00 1.E- 2.E-03 2.E- 3.E-03 2.E- 8.E- 4.E- 2.E- 2.E- 02 06 04 02 04 03 04 02 02 04 03 03 04 152 0.29 8.E-04 4.E- 1.E-03 1.E- 7.E-05 3.E-05 1.E-04 6.E- 5.E- 1.E- 7.E- 8.E-04 0.E+00 1.E- 1.E-03 2.E- 4.E-03 2.E- 2.E- 5.E- 2.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 190 0.31 2.E-03 3.E- 4.E-04 2.E- 9.E-05 0.E+00 1.E-04 6.E- 3.E- 1.E- 9.E- 1.E-03 0.E+00 2.E- 4.E-03 1.E- 4.E-03 1.E- 8.E- 6.E- 2.E- 1.E- 02 06 04 02 04 03 04 02 02 04 03 03 04 228 0.21 0.E+00 3.E- 5.E-04 2.E- 7.E-05 3.E-05 3.E-04 2.E- 4.E- 9.E- 6.E- 8.E-04 3.E-05 3.E- 2.E-03 1.E- 1.E-03 1.E- 1.E- 5.E- 1.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 Petro 38 0.1 8.E-04 1.E- 2.E-04 6.E- 4.E-07 0.E+00 0.E+00 4.E- 1.E- 5.E- 2.E- 3.E-05 0.E+00 1.E- 8.E-04 3.E- 8.E-04 3.E- 4.E- 6.E- 2.E- 3.E- Rabigh 02 07 04 02 05 03 04 03 03 04 04 03 05 under 38 0.1 0.E+00 9.E- 1.E-04 6.E- 0.E+00 2.E-05 6.E-05 4.E- 1.E- 5.E- 2.E- 3.E-05 0.E+00 5.E- 5.E-04 4.E- 1.E-03 4.E- 3.E- 7.E- 1.E- 4.E- pipe 03 07 04 02 05 03 05 03 03 04 04 03 05 76 0.14 8.E-04 2.E- 2.E-04 1.E- 5.E-06 0.E+00 6.E-05 7.E- 3.E- 6.E- 4.E- 0.E+00 0.E+00 5.E- 7.E-04 6.E- 0.E+00 6.E- 8.E- 1.E- 2.E- 6.E- 02 06 04 02 05 03 05 03 03 04 03 03 05 114 0.12 3.E-03 5.E- 4.E-04 2.E- 1.E-05 5.E-05 1.E-04 6.E- 6.E- 6.E- 4.E- 2.E-05 0.E+00 2.E- 1.E-03 6.E- 3.E-04 6.E- 2.E- 7.E- 2.E- 1.E- 02 06 04 02 05 03 04 03 03 03 04 03 04 152 0.12 2.E-03 3.E- 3.E-04 2.E- 5.E-06 0.E+00 9.E-05 5.E- 4.E- 5.E- 4.E- 2.E-05 0.E+00 9.E- 9.E-04 8.E- 8.E-04 8.E- 1.E- 9.E- 2.E- 8.E- 02 06 04 02 05 03 05 03 03 03 04 03 05 190 0.29 5.E-03 7.E- 7.E-04 2.E- 1.E-05 7.E-05 3.E-04 9.E- 9.E- 1.E- 1.E- 1.E-03 0.E+00 3.E- 3.E-03 2.E- 2.E-03 2.E- 3.E- 4.E- 4.E- 2.E- 02 05 04 02 04 02 04 02 02 03 03 03 04 228 0.29 9.E-03 1.E- 8.E-04 7.E- 1.E-05 1.E-04 4.E-04 1.E- 1.E- 1.E- 2.E- 4.E-04 0.E+00 2.E- 0.E+00 2.E- 3.E-03 2.E- 5.E- 5.E- 5.E- 2.E- 01 06 03 01 04 02 04 02 02 03 03 03 04 266 0.2 5.E-03 7.E- 7.E-04 5.E- 2.E-05 4.E-05 2.E-04 6.E- 1.E- 9.E- 1.E- 6.E-04 4.E-05 3.E- 2.E-03 1.E- 3.E-04 1.E- 4.E- 4.E- 3.E- 2.E- 02 06 04 01 05 02 04 02 02 03 03 03 04

194

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) (days) Petro 304 0.36 1.E-02 1.E- 4.E-04 9.E- 3.E-05 6.E-05 4.E-04 1.E- 2.E- 1.E- 2.E- 1.E-03 8.E-05 5.E- 2.E-03 2.E- 2.E-03 2.E- 6.E- 8.E- 6.E- 3.E- Rabigh 01 06 03 01 04 02 04 02 02 03 03 03 04 under 343 0.44 3.E-02 3.E- 1.E-03 1.E- 4.E-05 2.E-04 8.E-04 1.E- 4.E- 3.E- 4.E- 7.E-04 2.E-05 8.E- 3.E-03 3.E- 4.E-03 3.E- 1.E- 1.E- 7.E- 5.E- pipe 01 05 03 01 04 02 04 02 02 02 02 03 04 381 0.28 1.E-02 2.E- 1.E-03 7.E- 5.E-05 2.E-04 5.E-04 6.E- 2.E- 2.E- 2.E- 1.E-03 8.E-05 5.E- 3.E-03 2.E- 0.E+00 2.E- 8.E- 6.E- 1.E- 3.E- 01 06 04 01 04 02 04 02 02 03 03 02 04 381 0.28 9.E-03 1.E- 1.E-03 7.E- 3.E-05 1.E-04 3.E-04 4.E- 1.E- 1.E- 2.E- 1.E-03 5.E-05 3.E- 3.E-03 2.E- 2.E-04 2.E- 5.E- 6.E- 1.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 02 04 Petro 38 0.08 2.E-03 3.E- 1.E-04 2.E- 4.E-06 4.E-05 1.E-04 2.E- 4.E- 4.E- 8.E- 2.E-04 1.E-05 1.E- 1.E-03 5.E- 6.E-04 5.E- 1.E- 2.E- 1.E- 9.E- Rabigh 02 06 04 02 05 03 04 03 03 03 03 03 05 few 38 0.08 2.E-03 3.E- 2.E-04 2.E- 6.E-06 4.E-05 9.E-05 2.E- 4.E- 3.E- 8.E- 2.E-04 7.E-06 8.E- 8.E-04 5.E- 4.E-04 5.E- 1.E- 2.E- 2.E- 7.E- meters 02 06 04 02 05 03 05 03 03 03 03 03 05 from 76 0.15 5.E-03 6.E- 3.E-04 4.E- 2.E-05 5.E-05 2.E-04 2.E- 7.E- 6.E- 2.E- 0.E+00 1.E-04 2.E- 7.E-04 8.E- 2.E-04 8.E- 2.E- 4.E- 2.E- 1.E- pipe 02 06 04 02 05 02 04 03 03 03 03 03 04 114 0.24 9.E-03 1.E- 8.E-04 5.E- 3.E-05 1.E-04 3.E-04 6.E- 2.E- 2.E- 4.E- 1.E-04 3.E-04 5.E- 4.E-03 1.E- 2.E-03 1.E- 6.E- 8.E- 4.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 152 0.23 2.E-02 2.E- 9.E-04 7.E- 5.E-05 2.E-04 4.E-04 5.E- 2.E- 2.E- 3.E- 3.E-05 3.E-04 7.E- 2.E-03 1.E- 2.E-03 1.E- 9.E- 7.E- 3.E- 3.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 190 0.36 2.E-02 3.E- 1.E-03 1.E- 5.E-05 3.E-04 7.E-04 8.E- 4.E- 3.E- 6.E- 8.E-05 4.E-04 1.E- 3.E-03 2.E- 6.E-04 2.E- 1.E- 1.E- 6.E- 4.E- 01 05 04 01 04 02 03 02 02 02 02 03 04 228 0.39 1.E-02 2.E- 1.E-03 2.E- 2.E-04 4.E-04 7.E-04 9.E- 2.E- 2.E- 5.E- 9.E-04 5.E-04 7.E- 3.E-03 2.E- 2.E-03 2.E- 7.E- 1.E- 5.E- 5.E- 01 04 04 01 04 02 04 02 02 03 02 03 04 266 0.33 1.E-02 1.E- 2.E-03 4.E- 8.E-05 4.E-05 5.E-04 3.E- 2.E- 2.E- 4.E- 4.E-04 2.E-04 5.E- 2.E-03 1.E- 1.E-03 1.E- 5.E- 1.E- 5.E- 2.E- 01 05 04 01 04 02 04 02 02 03 02 03 04 266 0.32 9.E-03 1.E- 5.E-04 2.E- 5.E-05 1.E-04 3.E-04 3.E- 2.E- 2.E- 4.E- 8.E-05 2.E-04 6.E- 3.E-03 2.E- 2.E-03 2.E- 6.E- 1.E- 5.E- 2.E- 01 05 04 01 04 02 04 02 02 03 02 03 04 Petro 38 0.03 3.E-03 3.E- 2.E-04 2.E- 0.E+00 0.E+00 8.E-05 3.E- 5.E- 4.E- 3.E- 2.E-05 0.E+00 9.E- 3.E-04 2.E- 2.E-04 2.E- 2.E- 3.E- 9.E- 6.E- Rabigh 02 06 04 02 05 03 05 03 03 03 04 04 05 Far from 76 0.17 5.E-03 6.E- 5.E-04 3.E- 2.E-06 3.E-05 2.E-04 6.E- 9.E- 7.E- 2.E- 3.E-04 5.E-05 1.E- 5.E-04 9.E- 9.E-04 9.E- 3.E- 3.E- 2.E- 1.E- the pipe 02 06 04 02 05 02 04 03 03 03 03 03 04 114 0.22 4.E-03 7.E- 1.E-03 4.E- 3.E-05 4.E-05 2.E-04 7.E- 9.E- 9.E- 3.E- 3.E-04 4.E-05 2.E- 2.E-03 1.E- 1.E-03 1.E- 3.E- 4.E- 2.E- 1.E- 02 06 04 02 05 02 04 02 02 03 03 03 04 152 0.3 1.E-02 1.E- 9.E-04 5.E- 3.E-05 2.E-04 4.E-04 8.E- 2.E- 1.E- 6.E- 7.E-04 4.E-05 5.E- 1.E-03 2.E- 1.E-03 2.E- 6.E- 6.E- 3.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 190 0.22 4.E-03 6.E- 1.E-03 5.E- 3.E-05 3.E-05 2.E-04 8.E- 9.E- 9.E- 4.E- 5.E-04 1.E-04 2.E- 1.E-03 1.E- 2.E-03 1.E- 3.E- 6.E- 1.E- 1.E- 02 06 04 02 05 02 04 02 02 03 03 03 04 228 0.24 6.E-03 1.E- 1.E-03 6.E- 3.E-05 2.E-05 3.E-04 5.E- 2.E- 2.E- 6.E- 9.E-05 2.E-04 5.E- 3.E-03 1.E- 0.E+00 1.E- 5.E- 7.E- 1.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 Economic 38 0.05 1.E-03 2.E- 0.E+00 9.E- 7.E-06 1.E-05 7.E-05 2.E- 3.E- 3.E- 1.E- 5.E-06 0.E+00 1.E- 5.E-04 2.E- 3.E-04 2.E- 9.E- 3.E- 5.E- 5.E- city 1st 02 07 04 02 05 03 04 03 03 04 04 04 05 island 38 0.05 1.E-03 2.E- 3.E-04 1.E- 5.E-06 8.E-06 8.E-05 2.E- 2.E- 3.E- 9.E- 3.E-05 0.E+00 7.E- 5.E-04 3.E- 8.E-04 3.E- 9.E- 4.E- 4.E- 6.E- 02 06 04 02 05 04 05 03 03 04 04 04 05 76 0.25 4.E-03 6.E- 6.E-04 3.E- 4.E-05 0.E+00 3.E-04 8.E- 7.E- 1.E- 5.E- 4.E-04 0.E+00 4.E- 2.E-03 2.E- 1.E-03 2.E- 3.E- 3.E- 3.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 114 0.29 6.E-03 6.E- 8.E-04 6.E- 5.E-05 5.E-04 5.E-04 8.E- 8.E- 1.E- 5.E- 8.E-04 0.E+00 1.E- 3.E-03 2.E- 2.E-03 2.E- 3.E- 3.E- 2.E- 2.E- 02 06 04 02 04 03 03 02 02 03 03 03 04

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) 195

(days)

Economic 152 0.43 3.E-03 1.E- 1.E-03 4.E- 4.E-04 4.E-04 2.E-03 2.E- 2.E- 3.E- 1.E- 2.E-03 2.E-04 2.E- 6.E-03 3.E- 2.E-03 3.E- 6.E- 5.E- 4.E- 8.E- city 1st 01 04 03 01 04 02 03 02 02 03 03 03 04 island 190 0.42 1.E-02 2.E- 1.E-03 8.E- 1.E-04 5.E-05 6.E-04 1.E- 2.E- 2.E- 9.E- 2.E-03 0.E+00 1.E- 2.E-03 3.E- 5.E-03 3.E- 7.E- 5.E- 3.E- 4.E- 01 05 03 01 04 03 03 02 02 03 03 03 04 228 0.47 1.E-02 2.E- 4.E-04 4.E- 1.E-04 2.E-04 6.E-04 1.E- 2.E- 2.E- 1.E- 8.E-04 0.E+00 1.E- 4.E-03 3.E- 3.E-03 3.E- 7.E- 5.E- 3.E- 4.E- 01 05 03 01 04 02 03 02 02 03 03 03 04 266 0.49 1.E-02 2.E- 1.E-03 2.E- 1.E-04 1.E-04 6.E-04 2.E- 3.E- 3.E- 1.E- 2.E-03 0.E+00 1.E- 1.E-03 3.E- 2.E-03 3.E- 1.E- 7.E- 4.E- 4.E- 01 05 03 01 04 02 03 02 02 02 03 03 04 304 0.44 1.E-02 2.E- 2.E-03 1.E- 7.E-05 0.E+00 6.E-04 1.E- 2.E- 2.E- 7.E- 2.E-03 0.E+00 9.E- 3.E-03 3.E- 4.E-03 3.E- 7.E- 6.E- 4.E- 3.E- 01 05 03 01 04 03 04 02 02 03 03 03 04 304 0.49 1.E-02 1.E- 1.E-03 9.E- 1.E-04 3.E-05 5.E-04 8.E- 1.E- 3.E- 8.E- 2.E-03 0.E+00 5.E- 4.E-03 3.E- 5.E-03 3.E- 5.E- 7.E- 3.E- 3.E- 01 06 04 01 04 03 04 02 02 03 03 03 04 Economic 38 0.06 0.E+00 5.E- 5.E-05 5.E- 0.E+00 0.E+00 3.E-05 4.E- 5.E- 3.E- 8.E- 6.E-06 0.E+00 8.E- 3.E-04 3.E- 1.E-04 3.E- 2.E- 3.E- 7.E- 4.E- city 2nd 03 07 04 03 05 04 05 03 03 04 04 04 05 island 38 0.07 1.E-04 6.E- 4.E-05 7.E- 0.E+00 6.E-07 3.E-05 4.E- 5.E- 3.E- 8.E- 4.E-05 0.E+00 8.E- 8.E-04 3.E- 1.E-05 3.E- 2.E- 4.E- 8.E- 4.E- 03 07 04 03 05 04 05 03 03 04 04 04 05 76 0.18 1.E-03 2.E- 3.E-04 2.E- 1.E-05 0.E+00 1.E-04 1.E- 2.E- 7.E- 2.E- 1.E-04 0.E+00 3.E- 2.E-03 9.E- 1.E-03 9.E- 7.E- 8.E- 2.E- 9.E- 02 06 03 02 05 03 05 03 03 04 04 03 05 114 0.22 2.E-03 3.E- 5.E-04 2.E- 0.E+00 0.E+00 1.E-04 9.E- 3.E- 8.E- 3.E- 3.E-05 0.E+00 2.E- 8.E-04 1.E- 1.E-03 1.E- 1.E- 1.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 152 0.24 5.E-03 5.E- 4.E-04 3.E- 1.E-05 0.E+00 2.E-04 8.E- 6.E- 9.E- 5.E- 3.E-05 0.E+00 2.E- 2.E-03 1.E- 0.E+00 1.E- 2.E- 2.E- 3.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 190 0.18 4.E-03 5.E- 2.E-04 4.E- 1.E-06 4.E-05 2.E-04 5.E- 7.E- 1.E- 4.E- 1.E-04 0.E+00 3.E- 1.E-03 9.E- 4.E-04 9.E- 3.E- 1.E- 2.E- 1.E- 02 06 04 02 04 03 04 03 03 03 03 03 04 228 0.26 5.E-03 6.E- 6.E-04 5.E- 2.E-05 0.E+00 2.E-04 7.E- 8.E- 1.E- 7.E- 1.E-03 0.E+00 3.E- 2.E-03 1.E- 2.E-03 1.E- 3.E- 3.E- 4.E- 1.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 266 0.38 5.E-03 9.E- 1.E-03 3.E- 3.E-05 4.E-05 3.E-04 8.E- 1.E- 2.E- 1.E- 5.E-04 4.E-05 3.E- 7.E-03 2.E- 0.E+00 2.E- 5.E- 7.E- 6.E- 3.E- 02 06 04 01 04 02 04 02 02 03 03 03 04 304 0.44 3.E-03 9.E- 0.E+00 1.E- 1.E-04 9.E-05 4.E-04 9.E- 1.E- 2.E- 1.E- 2.E-03 9.E-05 6.E- 9.E-04 2.E- 3.E-03 2.E- 4.E- 8.E- 6.E- 3.E- 02 04 04 01 04 02 04 02 02 03 03 03 04 343 0.47 7.E-03 1.E- 2.E-03 4.E- 1.E-04 6.E-05 4.E-04 1.E- 2.E- 2.E- 2.E- 1.E-03 7.E-05 9.E- 5.E-03 3.E- 1.E-06 3.E- 6.E- 1.E- 8.E- 4.E- 01 05 03 01 04 02 04 02 02 03 02 03 04 381 0.37 4.E-03 7.E- 1.E-03 2.E- 8.E-05 5.E-05 3.E-04 7.E- 9.E- 1.E- 2.E- 1.E-03 1.E-04 7.E- 4.E-03 2.E- 5.E-03 2.E- 3.E- 1.E- 6.E- 3.E- 02 05 04 02 04 02 04 02 02 03 02 03 04 381 0.34 3.E-03 6.E- 1.E-03 6.E- 5.E-05 1.E-04 3.E-04 5.E- 8.E- 2.E- 1.E- 1.E-03 0.E+00 3.E- 2.E-03 2.E- 2.E-03 2.E- 3.E- 8.E- 5.E- 2.E- 02 06 04 02 04 02 04 02 02 03 03 03 04 Economic 38 0.07 5.E-03 6.E- 2.E-04 3.E- 1.E-05 5.E-05 2.E-04 2.E- 8.E- 6.E- 3.E- 2.E-04 3.E-05 3.E- 8.E-04 5.E- 5.E-04 5.E- 3.E- 1.E- 7.E- 1.E- city 3rd 02 06 04 02 05 03 04 03 03 03 03 04 04 76 0.11 5.E-03 7.E- 2.E-04 3.E- 1.E-05 9.E-05 2.E-04 2.E- 9.E- 7.E- 5.E- 1.E-04 5.E-05 3.E- 4.E-04 8.E- 1.E-03 8.E- 3.E- 3.E- 9.E- 2.E- 02 06 04 02 05 03 04 03 03 03 03 04 04 114 0.17 1.E-02 2.E- 2.E-04 6.E- 3.E-05 1.E-04 4.E-04 4.E- 2.E- 1.E- 7.E- 6.E-04 1.E-04 6.E- 0.E+00 1.E- 4.E-04 1.E- 8.E- 3.E- 2.E- 3.E- 01 06 04 01 04 03 04 02 02 03 03 03 04 152 0.17 6.E-03 9.E- 0.E+00 3.E- 2.E-05 8.E-05 3.E-04 3.E- 1.E- 1.E- 6.E- 5.E-04 8.E-05 4.E- 1.E-03 1.E- 1.E-03 1.E- 4.E- 4.E- 1.E- 2.E- 02 06 04 01 04 03 04 02 02 03 03 03 04 190 0.27 2.E-02 2.E- 6.E-04 9.E- 5.E-05 2.E-04 2.E-03 9.E- 3.E- 2.E- 1.E- 1.E-04 3.E-04 1.E- 8.E-03 2.E- 2.E-03 2.E- 1.E- 6.E- 5.E- 5.E- 01 06 04 01 04 02 03 02 02 02 03 03 04

Location leaf DW Ag Al As Be Cd Co Cr Cu Fe Li Mn Mo Nb Ni Pb Rb Se Sr Ti V Zn Zr and Site age (g) (days) 196

Economic 228 0.21 9.E-03 1.E- 1.E-04 5.E- 3.E-05 1.E-04 4.E-04 3.E- 2.E- 1.E- 8.E- 1.E-05 1.E-04 6.E- 2.E-03 1.E- 2.E-03 1.E- 6.E- 5.E- 1.E- 3.E- city 3rd 01 06 04 01 04 03 04 02 02 03 03 03 04 266 0.22 9.E-03 1.E- 9.E-04 1.E- 3.E-05 8.E-05 3.E-04 3.E- 2.E- 1.E- 1.E- 0.E+00 2.E-04 6.E- 1.E-03 1.E- 5.E-04 1.E- 6.E- 6.E- 1.E- 3.E- 01 05 04 01 04 02 04 02 02 03 03 03 04 304 0.11 8.E-03 9.E- 4.E-04 4.E- 1.E-05 6.E-05 3.E-04 3.E- 1.E- 1.E- 6.E- 8.E-05 8.E-05 4.E- 1.E-03 8.E- 7.E-04 8.E- 5.E- 3.E- 8.E- 2.E- 02 06 04 01 04 03 04 03 03 03 03 04 04 Thuwal 38 0.04 1.E-03 2.E- 1.E-04 1.E- 6.E-06 1.E-05 6.E-05 1.E- 2.E- 3.E- 9.E- 6.E-05 0.E+00 8.E- 6.E-04 3.E- 6.E-04 3.E- 9.E- 4.E- 4.E- 4.E- land 02 06 04 02 05 04 05 03 03 04 04 04 05 76 0.08 3.E-03 3.E- 3.E-04 2.E- 1.E-05 2.E-05 1.E-04 3.E- 4.E- 5.E- 2.E- 2.E-04 0.E+00 8.E- 4.E-05 6.E- 7.E-04 6.E- 1.E- 1.E- 8.E- 8.E- 02 06 04 02 05 03 05 03 03 03 03 04 05 114 0.16 4.E-03 5.E- 3.E-04 2.E- 3.E-05 1.E-05 2.E-04 5.E- 7.E- 7.E- 4.E- 5.E-04 0.E+00 4.E- 1.E-03 1.E- 2.E-03 1.E- 3.E- 3.E- 1.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 152 0.19 4.E-03 7.E- 0.E+00 2.E- 5.E-05 2.E-05 2.E-04 6.E- 9.E- 9.E- 5.E- 6.E-04 0.E+00 3.E- 2.E-03 1.E- 2.E-04 1.E- 3.E- 4.E- 2.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 190 0.26 4.E-03 6.E- 8.E-04 2.E- 4.E-05 5.E-05 3.E-04 7.E- 9.E- 1.E- 8.E- 5.E-04 4.E-05 5.E- 3.E-03 1.E- 1.E-03 1.E- 3.E- 6.E- 2.E- 2.E- 02 06 04 02 04 03 04 02 02 03 03 03 04 228 0.23 4.E-03 6.E- 5.E-04 2.E- 6.E-05 5.E-05 2.E-04 7.E- 8.E- 8.E- 6.E- 5.E-04 0.E+00 3.E- 9.E-04 1.E- 6.E-04 1.E- 3.E- 5.E- 2.E- 2.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 266 0.29 7.E-03 1.E- 1.E-03 3.E- 6.E-05 0.E+00 3.E-04 8.E- 1.E- 1.E- 1.E- 5.E-04 6.E-05 4.E- 2.E-03 2.E- 1.E-03 2.E- 5.E- 6.E- 2.E- 2.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 304 0.41 1.E-02 1.E- 4.E-04 3.E- 4.E-04 4.E-04 7.E-04 2.E- 2.E- 2.E- 2.E- 1.E-03 4.E-04 7.E- 3.E-03 3.E- 2.E-03 3.E- 6.E- 1.E- 3.E- 7.E- 01 04 03 01 04 02 04 02 02 03 02 03 04 343 0.5 1.E-02 1.E- 2.E-03 5.E- 2.E-04 3.E-05 8.E-04 1.E- 2.E- 2.E- 2.E- 2.E-04 4.E-04 8.E- 6.E-03 3.E- 6.E-03 3.E- 7.E- 2.E- 4.E- 4.E- 01 06 03 01 04 02 04 02 02 03 02 03 04 343 0.29 1.E-02 2.E- 6.E-04 7.E- 4.E-05 1.E-04 5.E-04 6.E- 2.E- 2.E- 1.E- 8.E-05 2.E-04 5.E- 9.E-04 2.E- 2.E-03 2.E- 8.E- 8.E- 1.E- 3.E- 01 06 04 01 04 02 04 02 02 03 03 03 04 Thuwal 38 0.03 1.E-03 2.E- 3.E-05 7.E- 0.E+00 5.E-06 4.E-05 5.E- 2.E- 2.E- 6.E- 0.E+00 0.E+00 5.E- 4.E-04 2.E- 6.E-04 2.E- 8.E- 3.E- 3.E- 3.E- Fringe 02 07 04 02 05 04 05 03 03 04 04 04 05 76 0.09 2.E-03 3.E- 2.E-04 1.E- 5.E-06 8.E-06 7.E-05 2.E- 3.E- 3.E- 1.E- 2.E-04 0.E+00 9.E- 4.E-04 5.E- 9.E-05 5.E- 1.E- 1.E- 5.E- 8.E- 02 06 04 02 05 03 05 03 03 03 03 04 05 114 0.12 4.E-04 3.E- 2.E-04 1.E- 1.E-05 0.E+00 1.E-04 2.E- 3.E- 3.E- 1.E- 4.E-04 0.E+00 9.E- 1.E-03 5.E- 7.E-04 5.E- 1.E- 2.E- 5.E- 7.E- 02 06 04 02 05 03 05 03 03 03 03 04 05 152 0.16 4.E-03 7.E- 4.E-04 3.E- 3.E-05 2.E-05 2.E-04 3.E- 8.E- 8.E- 3.E- 4.E-04 1.E-05 2.E- 8.E-04 8.E- 1.E-03 8.E- 3.E- 2.E- 9.E- 1.E- 02 06 04 02 05 03 04 03 03 03 03 04 04 190 0.24 7.E-03 9.E- 3.E-04 3.E- 1.E-05 0.E+00 2.E-04 5.E- 1.E- 9.E- 5.E- 6.E-04 2.E-06 3.E- 4.E-03 1.E- 2.E-03 1.E- 4.E- 5.E- 1.E- 2.E- 02 06 04 01 05 03 04 02 02 03 03 03 04 228 0.27 7.E-03 9.E- 1.E-03 3.E- 4.E-05 2.E-05 3.E-04 4.E- 1.E- 9.E- 5.E- 5.E-04 4.E-05 3.E- 1.E-03 1.E- 3.E-03 1.E- 4.E- 6.E- 1.E- 2.E- 02 06 04 01 05 03 04 02 02 03 03 03 04 266 0.24 3.E-03 5.E- 6.E-04 3.E- 3.E-05 9.E-05 2.E-04 4.E- 7.E- 8.E- 4.E- 0.E+00 0.E+00 3.E- 3.E-03 1.E- 4.E-04 1.E- 2.E- 7.E- 1.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 03 04 304 0.24 2.E-03 5.E- 1.E-03 2.E- 5.E-05 4.E-05 2.E-04 3.E- 7.E- 9.E- 4.E- 3.E-04 1.E-04 3.E- 3.E-03 1.E- 7.E-04 1.E- 2.E- 6.E- 6.E- 1.E- 02 06 04 02 05 03 04 02 02 03 03 04 04 343 0.17 2.E-03 4.E- 9.E-04 6.E- 1.E-05 0.E+00 1.E-04 3.E- 6.E- 7.E- 3.E- 3.E-05 9.E-05 1.E- 1.E-03 1.E- 1.E-03 1.E- 2.E- 5.E- 8.E- 1.E- 02 07 04 02 05 03 04 02 02 03 03 04 04

Table S 3: The detailed slopes calculated per tree for heavy metals content 197

Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD test Alkarrar 1st island 3.E-03 8.E-04 -1.E-06 4.E-06 0.01 0.1 0.7726 B Alkarrar 2nd island 8.E-04 7.E-04 -8.E-07 5.E-06 0.01 0.0 0.8839 B Economic city 1st island 4.E-04 2.E-03 4.E-05 1.E-05 0.70 18.5 0.0026* A Economic city 2nd island 6.E-04 9.E-04 1.E-05 4.E-06 0.55 12.3 0.0057* B Economic city 3rd island 7.E-03 4.E-03 1.E-05 2.E-05 0.06 0.4 0.5678 A Petro Rabigh far from the pipe 3.E-03 3.E-03 2.E-05 2.E-05 0.14 0.6 0.4732 AB Petro Rabigh few meters from pipe 3.E-03 4.E-03 4.E-05 2.E-05 0.32 3.3 0.114 A Petro Rabigh under pipe -2.E-03 3.E-03 4.E-05 1.E-05 0.53 11.4 0.0071* AB Thuwal Fringe 2.E-03 2.E-03 5.E-06 8.E-06 0.06 0.4 0.5377 B Thuwal land -1.E-03 1.E-03 4.E-05 5.E-06 0.85 44.2 0.0002* AB Alkarrar 1st island 4.E-02 1.E-02 2.E-05 6.E-05 0.01 0.1 0.7594 C Alkarrar 2nd island 1.E-02 9.E-03 1.E-04 6.E-05 0.44 3.2 0.1498 C Economic city 1st island 2.E-02 2.E-02 5.E-04 1.E-04 0.71 19.6 0.0022* AB Economic city 2nd island 5.E-03 1.E-02 2.E-04 4.E-05 0.73 27.6 0.0004* C Economic city 3rd island 9.E-02 5.E-02 2.E-04 2.E-04 0.08 0.5 0.496 AB Petro Rabigh far from the pipe 3.E-02 3.E-02 3.E-04 2.E-04 0.39 2.6 0.184 ABC Petro Rabigh few meters from pipe 5.E-02 5.E-02 5.E-04 3.E-04 0.35 3.7 0.0953 A Petro Rabigh under pipe -2.E-02 3.E-02 5.E-04 1.E-04 0.67 20.2 0.0012* BC Thuwal Fringe 3.E-02 2.E-02 1.E-04 9.E-05 0.17 1.5 0.2636 C Thuwal land -4.E-03 1.E-02 4.E-04 5.E-05 0.91 77.9 <.0001* BC Alkarrar 1st island 2.E-04 2.E-04 7.E-07 7.E-07 0.10 0.8 0.4002 B Alkarrar 2nd island 3.E-04 4.E-04 2.E-06 3.E-06 0.09 0.4 0.5565 AB Economic city 1st island 2.E-04 2.E-04 4.E-06 1.E-06 0.64 14.1 0.0056* AB Economic city 2nd island -4.E-05 3.E-04 3.E-06 1.E-06 0.43 7.4 0.0214* AB Economic city 3rd island 6.E-05 2.E-04 2.E-06 1.E-06 0.25 2.0 0.204 B Petro Rabigh far from the pipe 2.E-04 3.E-04 5.E-06 2.E-06 0.71 9.9 0.0348* A Petro Rabigh few meters from pipe 1.E-04 2.E-04 4.E-06 1.E-06 0.56 8.8 0.0207* A Petro Rabigh under pipe -2.E-05 1.E-04 3.E-06 5.E-07 0.81 42.2 <.0001* AB Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD 198

test Thuwal Fringe -1.E-04 2.E-04 4.E-06 9.E-07 0.72 17.9 0.0039* AB Thuwal land -5.E-05 3.E-04 3.E-06 1.E-06 0.47 7.2 0.0276* AB Alkarrar 1st island 5.E-06 2.E-06 -3.E-09 1.E-08 0.02 0.1 0.7535 A Alkarrar 2nd island 2.E-06 1.E-06 2.E-09 7.E-09 0.01 0.1 0.8316 A Economic city 1st island 5.E-05 8.E-05 1.E-08 4.E-07 0.00 0.0 0.9778 A Economic city 2nd island -6.E-06 2.E-05 1.E-07 6.E-08 0.19 2.4 0.1554 A Economic city 3rd island 3.E-06 2.E-06 1.E-08 1.E-08 0.19 1.4 0.2848 A Petro Rabigh far from the pipe 2.E-06 3.E-07 2.E-08 2.E-09 0.95 73.7 0.0010* A Petro Rabigh few meters from pipe -2.E-05 4.E-05 3.E-07 2.E-07 0.22 2.0 0.2004 A Petro Rabigh under pipe 1.E-06 3.E-06 3.E-08 1.E-08 0.28 4.0 0.0742 A Thuwal Fringe 1.E-06 7.E-07 3.E-09 3.E-09 0.08 0.6 0.4467 A Thuwal land -3.E-05 7.E-05 3.E-07 3.E-07 0.11 1.0 0.3476 A Alkarrar 1st island 3.E-05 1.E-05 2.E-08 6.E-08 0.02 0.1 0.7169 AB Alkarrar 2nd island 3.E-06 2.E-05 4.E-07 1.E-07 0.77 13.1 0.0223* AB Economic city 1st island 5.E-05 7.E-05 3.E-07 4.E-07 0.09 0.8 0.4065 A Economic city 2nd island -2.E-05 2.E-05 3.E-07 8.E-08 0.53 11.1 0.0075* AB Economic city 3rd island 2.E-05 1.E-05 4.E-08 6.E-08 0.06 0.4 0.5479 AB Petro Rabigh far from the pipe -5.E-06 7.E-06 2.E-07 5.E-08 0.80 15.6 0.0168* AB Petro Rabigh few meters from pipe -1.E-05 4.E-05 5.E-07 2.E-07 0.35 3.7 0.0945 AB Petro Rabigh under pipe -6.E-06 3.E-06 1.E-07 1.E-08 0.87 68.2 <.0001* B Thuwal Fringe 2.E-06 1.E-05 1.E-07 5.E-08 0.40 4.6 0.069 B Thuwal land -3.E-05 7.E-05 6.E-07 3.E-07 0.33 4.0 0.0818 AB Alkarrar 1st island 2.E-05 4.E-05 8.E-08 2.E-07 0.02 0.2 0.7046 AB Alkarrar 2nd island 2.E-06 1.E-05 8.E-08 9.E-08 0.15 0.7 0.4501 AB Economic city 1st island 2.E-04 1.E-04 -1.E-07 6.E-07 0.00 0.0 0.8663 AB Economic city 2nd island -2.E-05 1.E-05 3.E-07 5.E-08 0.69 22.7 0.0008* B Economic city 3rd island 1.E-04 4.E-05 -1.E-08 2.E-07 0.00 0.0 0.95 AB Petro Rabigh far from the pipe 2.E-05 6.E-05 2.E-07 4.E-07 0.06 0.2 0.6481 AB Petro Rabigh few meters from pipe 5.E-05 8.E-05 7.E-07 5.E-07 0.21 1.9 0.2154 A Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD 199

test Petro Rabigh under pipe -2.E-05 2.E-05 4.E-07 9.E-08 0.72 26.3 0.0004* AB Thuwal Fringe -9.E-08 2.E-05 1.E-07 9.E-08 0.15 1.3 0.2954 B Thuwal land -3.E-05 8.E-05 5.E-07 4.E-07 0.21 2.1 0.185 AB Alkarrar 1st island 1.E-04 4.E-05 6.E-08 2.E-07 0.01 0.1 0.7576 B Alkarrar 2nd island 8.E-06 8.E-05 1.E-06 5.E-07 0.47 3.5 0.1354 AB Economic city 1st island 4.E-04 4.E-04 2.E-06 2.E-06 0.07 0.6 0.4771 A Economic city 2nd island 2.E-05 3.E-05 9.E-07 1.E-07 0.82 46.9 <.0001* B Economic city 3rd island 3.E-04 5.E-04 9.E-07 2.E-06 0.02 0.1 0.7294 AB Petro Rabigh far from the pipe 1.E-04 7.E-05 9.E-07 5.E-07 0.47 3.5 0.1341 AB Petro Rabigh few meters from pipe 9.E-05 1.E-04 2.E-06 6.E-07 0.54 8.3 0.0238* AB Petro Rabigh under pipe -4.E-05 8.E-05 2.E-06 3.E-07 0.67 20.0 0.0012* AB Thuwal Fringe 9.E-05 6.E-05 3.E-07 3.E-07 0.18 1.5 0.2547 B Thuwal land -5.E-05 9.E-05 2.E-06 4.E-07 0.76 24.7 0.0011* AB Alkarrar 1st island 4.E-04 1.E-04 -3.E-07 5.E-07 0.06 0.4 0.5416 B Alkarrar 2nd island 4.E-04 2.E-04 7.E-07 2.E-06 0.04 0.2 0.6943 AB Economic city 1st island 4.E-04 3.E-04 3.E-06 1.E-06 0.36 4.5 0.066 A Economic city 2nd island 6.E-04 1.E-04 4.E-07 5.E-07 0.04 0.5 0.5138 AB Economic city 3rd island 3.E-04 2.E-04 6.E-07 1.E-06 0.05 0.3 0.5866 B Petro Rabigh far from the pipe 5.E-04 2.E-04 1.E-06 1.E-06 0.19 0.9 0.3939 AB Petro Rabigh few meters from pipe 2.E-04 2.E-04 1.E-06 1.E-06 0.21 1.9 0.2098 B Petro Rabigh under pipe 5.E-04 2.E-04 7.E-07 6.E-07 0.11 1.3 0.282 AB Thuwal Fringe 3.E-04 1.E-04 -1.E-07 5.E-07 0.01 0.1 0.7885 B Thuwal land 9.E-05 2.E-04 3.E-06 8.E-07 0.64 13.9 0.0058* AB Alkarrar 1st island 6.E-02 1.E-02 -8.E-06 7.E-05 0.00 0.0 0.9018 C Alkarrar 2nd island 9.E-03 1.E-02 1.E-04 7.E-05 0.54 4.6 0.0976 C Economic city 1st island 2.E-02 3.E-02 7.E-04 2.E-04 0.71 19.2 0.0023* AB Economic city 2nd island 7.E-04 2.E-02 3.E-04 6.E-05 0.72 26.1 0.0005* C Economic city 3rd island 1.E-01 6.E-02 2.E-04 3.E-04 0.09 0.6 0.4592 AB Petro Rabigh far from the pipe 5.E-02 4.E-02 5.E-04 3.E-04 0.42 2.9 0.1624 ABC Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD 200

test Petro Rabigh few meters from pipe 6.E-02 6.E-02 7.E-04 3.E-04 0.35 3.8 0.0908 A Petro Rabigh under pipe -2.E-02 4.E-02 7.E-04 1.E-04 0.67 20.4 0.0011* BC Thuwal Fringe 4.E-02 2.E-02 2.E-04 1.E-04 0.23 2.1 0.1905 C Thuwal land -9.E-03 1.E-02 6.E-04 6.E-05 0.93 104.4 <.0001* BC Alkarrar 1st island 7.E-05 2.E-05 4.E-08 7.E-08 0.03 0.3 0.6303 CD Alkarrar 2nd island 4.E-05 3.E-05 4.E-07 2.E-07 0.50 4.0 0.1153 BCD Economic city 1st island 4.E-05 3.E-05 8.E-07 2.E-07 0.73 22.0 0.0016* A Economic city 2nd island 3.E-05 2.E-05 4.E-07 7.E-08 0.75 30.5 0.0003* CD Economic city 3rd island 9.E-05 4.E-05 2.E-07 2.E-07 0.15 1.0 0.3457 ABC Petro Rabigh far from the pipe 3.E-05 2.E-05 5.E-07 2.E-07 0.73 10.7 0.0306* ABCD Petro Rabigh few meters from pipe 4.E-05 4.E-05 7.E-07 2.E-07 0.61 11.1 0.0125* AB Petro Rabigh under pipe 2.E-05 3.E-05 4.E-07 1.E-07 0.61 15.7 0.0027* CD Thuwal Fringe 3.E-05 2.E-05 2.E-07 7.E-08 0.54 8.2 0.0239* D Thuwal land -5.E-06 2.E-05 6.E-07 9.E-08 0.85 45.9 0.0001* BCD Alkarrar 1st island 3.E-03 8.E-04 5.E-06 4.E-06 0.17 1.4 0.2724 BC Alkarrar 2nd island 1.E-03 2.E-03 3.E-05 1.E-05 0.65 7.6 0.0513 BC Economic city 1st island 2.E-03 2.E-03 3.E-05 8.E-06 0.57 10.7 0.0114* BC Economic city 2nd island -1.E-03 1.E-03 4.E-05 5.E-06 0.89 84.9 <.0001* BC Economic city 3rd island 4.E-03 2.E-03 2.E-05 1.E-05 0.35 3.2 0.1255 BC Petro Rabigh far from the pipe 5.E-04 1.E-02 3.E-04 9.E-05 0.70 9.2 0.0389* A Petro Rabigh few meters from pipe 8.E-03 9.E-03 2.E-04 5.E-05 0.62 11.6 0.0115* A Petro Rabigh under pipe -3.E-03 4.E-03 8.E-05 2.E-05 0.74 28.3 0.0003* B Thuwal Fringe 8.E-04 8.E-04 1.E-05 4.E-06 0.59 10.0 0.0158* C Thuwal land -2.E-03 2.E-03 5.E-05 7.E-06 0.87 51.4 <.0001* BC Alkarrar 1st island 5.E-04 1.E-04 -1.E-07 5.E-07 0.01 0.1 0.7879 B Alkarrar 2nd island -4.E-05 2.E-04 5.E-06 1.E-06 0.83 18.9 0.0122* AB Economic city 1st island 1.E-04 4.E-04 6.E-06 2.E-06 0.54 9.2 0.0161* A Economic city 2nd island -3.E-04 2.E-04 4.E-06 8.E-07 0.71 24.2 0.0006* B Economic city 3rd island 4.E-04 2.E-04 -1.E-06 9.E-07 0.23 1.8 0.225 B Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD 201

test Petro Rabigh far from the pipe 2.E-04 3.E-04 9.E-07 2.E-06 0.07 0.3 0.6205 B Petro Rabigh few meters from pipe 3.E-05 2.E-04 1.E-06 1.E-06 0.16 1.4 0.2792 B Petro Rabigh under pipe -1.E-04 2.E-04 3.E-06 7.E-07 0.65 19.0 0.0014* B Thuwal Fringe 3.E-04 2.E-04 -2.E-07 8.E-07 0.00 0.0 0.8598 B Thuwal land 3.E-04 2.E-04 6.E-07 9.E-07 0.05 0.4 0.5447 B Alkarrar 1st island -3.E-05 2.E-05 3.E-07 8.E-08 0.59 10.2 0.0152* C Alkarrar 2nd island -1.E-05 1.E-05 1.E-07 7.E-08 0.43 3.0 0.1583 BC Economic city 1st island 2.E-05 4.E-05 -3.E-08 2.E-07 0.00 0.0 0.8592 C Economic city 2nd island -2.E-05 2.E-05 2.E-07 7.E-08 0.46 8.4 0.0157* C Economic city 3rd island 6.E-05 7.E-05 4.E-07 4.E-07 0.15 1.1 0.3436 B Petro Rabigh far from the pipe -5.E-05 4.E-05 9.E-07 3.E-07 0.76 13.0 0.0226* ABC Petro Rabigh few meters from pipe 5.E-05 1.E-04 1.E-06 5.E-07 0.39 4.4 0.0741 A Petro Rabigh under pipe -2.E-05 1.E-05 2.E-07 5.E-08 0.60 14.8 0.0032* C Thuwal Fringe -3.E-05 2.E-05 3.E-07 1.E-07 0.60 10.4 0.0146* BC Thuwal land -1.E-04 8.E-05 1.E-06 3.E-07 0.61 12.3 0.0080* BC Alkarrar 1st island 3.E-04 2.E-04 5.E-07 7.E-07 0.07 0.5 0.506 B Alkarrar 2nd island 1.E-05 4.E-05 1.E-06 3.E-07 0.72 10.4 0.0323* B Economic city 1st island 4.E-04 3.E-04 2.E-06 2.E-06 0.22 2.3 0.1677 A Economic city 2nd island -7.E-06 1.E-04 2.E-06 4.E-07 0.62 16.4 0.0023* B Economic city 3rd island 4.E-04 3.E-04 9.E-07 2.E-06 0.05 0.3 0.5857 AB Petro Rabigh far from the pipe 3.E-05 1.E-04 2.E-06 8.E-07 0.57 5.3 0.0823 AB Petro Rabigh few meters from pipe 1.E-04 2.E-04 2.E-06 1.E-06 0.44 5.6 0.0497* AB Petro Rabigh under pipe -5.E-06 8.E-05 1.E-06 3.E-07 0.62 16.1 0.0025* B Thuwal Fringe 8.E-05 7.E-05 6.E-07 3.E-07 0.31 3.1 0.1198 B Thuwal land 6.E-05 1.E-04 2.E-06 4.E-07 0.65 15.2 0.0046* B Alkarrar 1st island 1.E-03 5.E-04 6.E-07 2.E-06 0.01 0.1 0.8021 A Alkarrar 2nd island 9.E-04 1.E-03 8.E-06 8.E-06 0.23 1.2 0.3347 A Economic city 1st island 1.E-03 1.E-03 7.E-06 6.E-06 0.18 1.8 0.2168 A Economic city 2nd island 6.E-04 1.E-03 8.E-06 4.E-06 0.28 3.9 0.0772 A Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD 202

test Economic city 3rd island 7.E-04 2.E-03 7.E-06 1.E-05 0.08 0.5 0.4957 A Petro Rabigh far from the pipe -9.E-05 6.E-04 1.E-05 4.E-06 0.62 6.4 0.0645 A Petro Rabigh few meters from pipe 9.E-04 6.E-04 8.E-06 3.E-06 0.49 6.7 0.0363* A Petro Rabigh under pipe 3.E-04 5.E-04 6.E-06 2.E-06 0.50 10.1 0.0098* A Thuwal Fringe 4.E-04 7.E-04 7.E-06 3.E-06 0.36 3.9 0.0879 A Thuwal land 2.E-04 1.E-03 8.E-06 5.E-06 0.29 3.2 0.1097 A Alkarrar 1st island 8.E-03 2.E-03 6.E-06 7.E-06 0.08 0.6 0.4753 BC Alkarrar 2nd island 5.E-03 4.E-03 5.E-05 3.E-05 0.41 2.7 0.1736 ABC Economic city 1st island 5.E-03 3.E-03 9.E-05 2.E-05 0.78 28.0 0.0007* A Economic city 2nd island 3.E-03 2.E-03 6.E-05 7.E-06 0.87 65.7 <.0001* BC Economic city 3rd island 8.E-03 3.E-03 2.E-05 2.E-05 0.16 1.2 0.3236 BC Petro Rabigh far from the pipe 3.E-03 3.E-03 5.E-05 2.E-05 0.61 6.3 0.0654 BC Petro Rabigh few meters from pipe 5.E-03 2.E-03 5.E-05 1.E-05 0.74 19.8 0.0030* B Petro Rabigh under pipe 2.E-03 3.E-03 5.E-05 1.E-05 0.69 22.3 0.0008* BC Thuwal Fringe 2.E-03 2.E-03 3.E-05 8.E-06 0.73 18.7 0.0034* C Thuwal land -2.E-04 3.E-03 8.E-05 1.E-05 0.82 36.2 0.0003* BC Alkarrar 1st island 6.E-04 6.E-04 3.E-06 3.E-06 0.16 1.3 0.2933 A Alkarrar 2nd island 5.E-04 1.E-03 1.E-05 1.E-05 0.27 1.5 0.2935 A Economic city 1st island 2.E-04 7.E-04 1.E-05 3.E-06 0.64 14.4 0.0053* A Economic city 2nd island -2.E-04 8.E-04 7.E-06 3.E-06 0.31 4.5 0.0592 A Economic city 3rd island 8.E-04 6.E-04 1.E-06 3.E-06 0.02 0.1 0.7158 A Petro Rabigh far from the pipe 7.E-04 9.E-04 3.E-06 6.E-06 0.05 0.2 0.6706 A Petro Rabigh few meters from pipe 4.E-04 4.E-04 5.E-06 2.E-06 0.45 5.8 0.0464* A Petro Rabigh under pipe 7.E-04 8.E-04 3.E-06 3.E-06 0.06 0.6 0.448 A Thuwal Fringe 7.E-04 7.E-04 2.E-06 3.E-06 0.06 0.4 0.5336 A Thuwal land -8.E-05 9.E-04 8.E-06 4.E-06 0.35 4.3 0.0728 A Alkarrar 1st island 8.E-03 2.E-03 6.E-06 7.E-06 0.08 0.6 0.4753 BC Alkarrar 2nd island 5.E-03 4.E-03 5.E-05 3.E-05 0.41 2.7 0.1736 ABC Economic city 1st island 5.E-03 3.E-03 9.E-05 2.E-05 0.78 28.0 0.0007* A Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD 203

test Economic city 2nd island 3.E-03 2.E-03 6.E-05 7.E-06 0.87 65.7 <.0001* BC Economic city 3rd island 8.E-03 3.E-03 2.E-05 2.E-05 0.16 1.2 0.3236 BC Petro Rabigh far from the pipe 3.E-03 3.E-03 5.E-05 2.E-05 0.61 6.3 0.0654 BC Petro Rabigh few meters from pipe 5.E-03 2.E-03 5.E-05 1.E-05 0.74 19.8 0.0030* B Petro Rabigh under pipe 2.E-03 3.E-03 5.E-05 1.E-05 0.69 22.3 0.0008* BC Thuwal Fringe 2.E-03 2.E-03 3.E-05 8.E-06 0.73 18.7 0.0034* C Thuwal land -2.E-04 3.E-03 8.E-05 1.E-05 0.82 36.2 0.0003* BC Alkarrar 1st island 2.E-03 5.E-04 7.E-07 3.E-06 0.01 0.1 0.8021 C Alkarrar 2nd island 2.E-04 3.E-04 5.E-06 2.E-06 0.54 4.6 0.0977 C Economic city 1st island 8.E-04 1.E-03 2.E-05 6.E-06 0.69 18.1 0.0028* AB Economic city 2nd island 8.E-05 6.E-04 1.E-05 2.E-06 0.71 24.2 0.0006* C Economic city 3rd island 4.E-03 2.E-03 9.E-06 1.E-05 0.10 0.7 0.4476 AB Petro Rabigh far from the pipe 2.E-03 1.E-03 2.E-05 9.E-06 0.40 2.7 0.1779 ABC Petro Rabigh few meters from pipe 2.E-03 2.E-03 2.E-05 1.E-05 0.33 3.4 0.1068 A Petro Rabigh under pipe -1.E-03 1.E-03 2.E-05 6.E-06 0.64 17.7 0.0018* BC Thuwal Fringe 1.E-03 9.E-04 5.E-06 4.E-06 0.17 1.5 0.2674 C Thuwal land -2.E-04 6.E-04 2.E-05 3.E-06 0.88 60.3 <.0001* BC Alkarrar 1st island 1.E-03 5.E-04 7.E-06 2.E-06 0.61 11.0 0.0129* D Alkarrar 2nd island 3.E-04 9.E-04 3.E-05 6.E-06 0.81 17.4 0.0140* BCD Economic city 1st island 4.E-04 6.E-04 2.E-05 3.E-06 0.86 51.1 <.0001* BCD Economic city 2nd island -2.E-03 8.E-04 3.E-05 3.E-06 0.87 68.4 <.0001* BCD Economic city 3rd island 2.E-03 1.E-03 1.E-05 6.E-06 0.36 3.3 0.1179 BCD Petro Rabigh far from the pipe 8.E-05 9.E-04 3.E-05 6.E-06 0.88 30.0 0.0054* AB Petro Rabigh few meters from pipe 1.E-03 1.E-03 4.E-05 7.E-06 0.83 33.3 0.0007* A Petro Rabigh under pipe -6.E-04 9.E-04 2.E-05 4.E-06 0.77 32.6 0.0002* CD Thuwal Fringe -2.E-05 9.E-04 2.E-05 4.E-06 0.78 24.7 0.0016* BCD Thuwal land -2.E-03 2.E-03 4.E-05 7.E-06 0.80 32.9 0.0004* BC Alkarrar 1st island 6.E-03 8.E-04 -2.E-05 4.E-06 0.74 20.1 0.0029* ABCD Alkarrar 2nd island 1.E-03 7.E-04 3.E-06 5.E-06 0.10 0.4 0.548 CDE Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD 204

test Economic city 1st island 1.E-03 5.E-04 1.E-05 3.E-06 0.65 14.8 0.0049* ABCD Economic city 2nd island 3.E-04 6.E-04 2.E-05 2.E-06 0.81 43.9 <.0001* ABC Economic city 3rd island 1.E-03 1.E-03 1.E-06 6.E-06 0.01 0.0 0.8603 DE Petro Rabigh far from the pipe 2.E-03 8.E-04 7.E-07 5.E-06 0.00 0.0 0.9108 BCDE Petro Rabigh few meters from pipe 1.E-03 6.E-04 2.E-05 3.E-06 0.77 23.9 0.0018* AB Petro Rabigh under pipe -5.E-04 1.E-03 3.E-05 4.E-06 0.77 34.1 0.0002* A Thuwal Fringe 5.E-04 3.E-04 2.E-06 1.E-06 0.20 1.8 0.2249 E Thuwal land 3.E-04 5.E-04 7.E-06 2.E-06 0.63 13.5 0.0063* DE Alkarrar 1st island 1.E-04 1.E-05 -1.E-07 7.E-08 0.23 2.1 0.1889 B Alkarrar 2nd island 8.E-05 5.E-05 4.E-07 3.E-07 0.20 1.0 0.3699 AB Economic city 1st island 1.E-04 1.E-04 1.E-06 7.E-07 0.24 2.5 0.1501 A Economic city 2nd island 1.E-05 4.E-05 8.E-07 1.E-07 0.75 29.9 0.0003* B Economic city 3rd island 2.E-04 9.E-05 4.E-07 4.E-07 0.13 0.9 0.3832 AB Petro Rabigh far from the pipe 6.E-05 4.E-05 6.E-07 3.E-07 0.55 4.8 0.0932 AB Petro Rabigh few meters from pipe 8.E-05 8.E-05 1.E-06 5.E-07 0.44 5.6 0.0503 AB Petro Rabigh under pipe 1.E-05 5.E-05 7.E-07 2.E-07 0.60 14.9 0.0032* B Thuwal Fringe 6.E-05 3.E-05 3.E-07 1.E-07 0.40 4.7 0.0679 B Thuwal land -5.E-05 1.E-04 1.E-06 4.E-07 0.58 11.2 0.0102* AB

Table S 4: The detailed slopes calculated per tree for heavy metals concentration 205

Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey HSD test Alkarrar 1st island 2.E-02 4.E-03 -2.E-05 2.E-05 0.19 1.6 0.2468 CD Alkarrar 2nd island 1.E-02 3.E-03 -4.E-05 2.E-05 0.50 4.1 0.1143 D Economic city 1st island 2.E-02 5.E-03 2.E-05 3.E-05 0.07 0.6 0.4457 BCD Economic city 2nd island 6.E-03 4.E-03 2.E-05 2.E-05 0.17 2.1 0.1791 CD Economic city 3rd island 7.E-02 2.E-02 -4.E-05 8.E-05 0.04 0.3 0.6347 A Petro Rabigh far from the pipe 7.E-02 2.E-02 -2.E-04 1.E-04 0.47 3.5 0.1336 BCD Petro Rabigh few meters from pipe 3.E-02 1.E-02 3.E-05 7.E-05 0.03 0.2 0.6549 B Petro Rabigh under pipe 2.E-03 6.E-03 1.E-04 2.E-05 0.66 19.5 0.0013* BCD Thuwal Fringe 3.E-02 7.E-03 -5.E-05 3.E-05 0.22 2.0 0.2045 CD Thuwal land 3.E-02 7.E-03 -7.E-06 3.E-05 0.01 0.1 0.8269 BC Alkarrar 1st island 3.E-01 4.E-02 3.E-05 2.E-04 0.00 0.0 0.8637 CD Alkarrar 2nd island 1.E-01 2.E-02 7.E-05 2.E-04 0.06 0.2 0.6545 D Economic city 1st island 3.E-01 5.E-02 -8.E-06 3.E-04 0.00 0.0 0.9761 BCD Economic city 2nd island 1.E-01 3.E-02 3.E-04 1.E-04 0.40 6.8 0.0262* D Economic city 3rd island 8.E-01 1.E-01 -3.E-04 7.E-04 0.02 0.1 0.7245 A Petro Rabigh far from the pipe 8.E-01 3.E-01 -2.E-03 2.E-03 0.33 2.0 0.2332 BCD Petro Rabigh few meters from pipe 4.E-01 1.E-01 2.E-04 7.E-04 0.01 0.1 0.8048 B Petro Rabigh under pipe 1.E-01 5.E-02 1.E-03 2.E-04 0.71 24.8 0.0006* BCD Thuwal Fringe 4.E-01 6.E-02 -6.E-04 3.E-04 0.36 3.9 0.0883 BCD Thuwal land 4.E-01 8.E-02 -1.E-04 3.E-04 0.02 0.1 0.7248 BC Alkarrar 1st island 2.E-03 9.E-04 3.E-06 4.E-06 0.08 0.6 0.462 B Alkarrar 2nd island 2.E-03 2.E-03 5.E-06 1.E-05 0.06 0.2 0.649 AB Economic city 1st island 3.E-03 1.E-03 -2.E-06 6.E-06 0.02 0.1 0.7284 AB Economic city 2nd island 9.E-04 6.E-04 5.E-06 3.E-06 0.29 4.1 0.0705 B Economic city 3rd island 1.E-03 1.E-03 4.E-06 6.E-06 0.09 0.6 0.4597 B Petro Rabigh far from the pipe 5.E-03 2.E-03 -9.E-07 1.E-05 0.00 0.0 0.9313 A Petro Rabigh few meters from pipe 2.E-03 7.E-04 4.E-06 4.E-06 0.13 1.0 0.3404 AB Petro Rabigh under pipe 1.E-03 6.E-04 6.E-06 2.E-06 0.40 6.7 0.0266* AB Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey 206

HSD test Thuwal Fringe 4.E-04 8.E-04 1.E-05 4.E-06 0.66 13.8 0.0075* AB Thuwal land 3.E-03 9.E-04 -2.E-06 4.E-06 0.02 0.2 0.7065 AB Alkarrar 1st island 4.E-05 1.E-05 -4.E-08 6.E-08 0.06 0.5 0.5147 A Alkarrar 2nd island 1.E-05 3.E-06 -2.E-08 2.E-08 0.24 1.3 0.3244 A Economic city 1st island 1.E-04 2.E-04 -5.E-08 9.E-07 0.00 0.0 0.9606 A Economic city 2nd island -5.E-06 3.E-05 2.E-07 1.E-07 0.17 2.1 0.1784 A Economic city 3rd island 3.E-05 7.E-06 1.E-09 3.E-08 0.00 0.0 0.9663 A Petro Rabigh far from the pipe 6.E-05 2.E-05 -2.E-07 1.E-07 0.35 2.1 0.2173 A Petro Rabigh few meters from pipe -3.E-05 9.E-05 7.E-07 5.E-07 0.22 2.0 0.2048 A Petro Rabigh under pipe 1.E-05 1.E-05 6.E-08 4.E-08 0.18 2.2 0.1659 A Thuwal Fringe 2.E-05 2.E-06 -3.E-08 1.E-08 0.58 9.7 0.0171* A Thuwal land -4.E-05 2.E-04 6.E-07 7.E-07 0.09 0.8 0.3874 A Alkarrar 1st island 2.E-04 5.E-05 9.E-08 2.E-07 0.03 0.2 0.6825 ABC Alkarrar 2nd island 5.E-05 4.E-05 1.E-06 3.E-07 0.83 19.1 0.0120* ABC Economic city 1st island 2.E-04 2.E-04 3.E-07 8.E-07 0.01 0.1 0.7578 A Economic city 2nd island -4.E-05 4.E-05 7.E-07 2.E-07 0.58 13.5 0.0043* BC Economic city 3rd island 2.E-04 3.E-05 -1.E-07 2.E-07 0.06 0.4 0.5511 ABC Petro Rabigh far from the pipe -2.E-05 3.E-05 7.E-07 2.E-07 0.76 12.3 0.0246* ABC Petro Rabigh few meters from pipe 3.E-05 1.E-04 1.E-06 6.E-07 0.32 3.2 0.1155 ABC Petro Rabigh under pipe 3.E-06 2.E-05 3.E-07 7.E-08 0.67 20.5 0.0011* C Thuwal Fringe 4.E-05 5.E-05 4.E-07 2.E-07 0.29 2.8 0.136 ABC Thuwal land 1.E-04 2.E-04 9.E-07 7.E-07 0.18 1.7 0.2253 AB Alkarrar 1st island 2.E-04 3.E-04 4.E-07 1.E-06 0.01 0.1 0.7765 ABC Alkarrar 2nd island 1.E-05 6.E-05 3.E-07 4.E-07 0.13 0.6 0.4837 ABC Economic city 1st island 6.E-04 4.E-04 -1.E-06 2.E-06 0.04 0.3 0.5706 ABC Economic city 2nd island -4.E-05 5.E-05 6.E-07 2.E-07 0.55 12.1 0.0060* C Economic city 3rd island 8.E-04 9.E-05 -1.E-06 5.E-07 0.59 8.7 0.0257* A Petro Rabigh far from the pipe 1.E-04 2.E-04 6.E-07 1.E-06 0.05 0.2 0.6627 ABC Petro Rabigh few meters from pipe 5.E-04 2.E-04 3.E-07 1.E-06 0.01 0.0 0.8394 AB Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey 207

HSD test Petro Rabigh under pipe 9.E-06 9.E-05 1.E-06 4.E-07 0.50 10.0 0.0100* ABC Thuwal Fringe 9.E-05 9.E-05 1.E-07 4.E-07 0.01 0.1 0.8184 BC Thuwal land 2.E-04 2.E-04 5.E-07 9.E-07 0.04 0.3 0.596 ABC Alkarrar 1st island 1.E-03 2.E-04 -4.E-07 9.E-07 0.02 0.2 0.693 B Alkarrar 2nd island 4.E-04 5.E-04 2.E-06 3.E-06 0.12 0.6 0.497 B Economic city 1st island 2.E-03 8.E-04 -1.E-06 4.E-06 0.01 0.1 0.7667 AB Economic city 2nd island 5.E-04 9.E-05 1.E-06 4.E-07 0.45 8.2 0.0167* B Economic city 3rd island 2.E-03 2.E-03 7.E-07 8.E-06 0.00 0.0 0.934 A Petro Rabigh far from the pipe 2.E-03 5.E-04 -6.E-06 3.E-06 0.40 2.6 0.18 AB Petro Rabigh few meters from pipe 1.E-03 2.E-04 1.E-06 1.E-06 0.08 0.6 0.4491 AB Petro Rabigh under pipe 3.E-04 2.E-04 4.E-06 7.E-07 0.74 28.3 0.0003* B Thuwal Fringe 1.E-03 2.E-04 -1.E-06 7.E-07 0.34 3.6 0.1009 B Thuwal land 1.E-03 2.E-04 2.E-07 1.E-06 0.01 0.1 0.8224 AB Alkarrar 1st island 3.E-03 4.E-04 -3.E-06 2.E-06 0.30 3.0 0.1255 A Alkarrar 2nd island 3.E-03 4.E-04 -5.E-06 3.E-06 0.49 3.8 0.1216 A Economic city 1st island 4.E-03 3.E-04 -4.E-06 2.E-06 0.38 4.8 0.0597 A Economic city 2nd island 6.E-03 4.E-04 -1.E-05 2.E-06 0.86 60.8 <.0001* A Economic city 3rd island 2.E-03 6.E-04 1.E-07 3.E-06 0.00 0.0 0.9683 A Petro Rabigh far from the pipe 8.E-03 2.E-03 -3.E-05 1.E-05 0.57 5.4 0.0812 A Petro Rabigh few meters from pipe 2.E-03 4.E-04 -3.E-06 2.E-06 0.27 2.7 0.1474 A Petro Rabigh under pipe 5.E-03 3.E-04 -8.E-06 1.E-06 0.84 53.6 <.0001* A Thuwal Fringe 8.E-03 3.E-03 -3.E-05 1.E-05 0.32 3.3 0.1116 A Thuwal land 4.E-03 3.E-04 -3.E-06 1.E-06 0.44 6.3 0.0369* A Alkarrar 1st island 4.E-01 6.E-02 -2.E-04 3.E-04 0.11 0.9 0.3744 CD Alkarrar 2nd island 1.E-01 3.E-02 2.E-04 2.E-04 0.15 0.7 0.4555 D Economic city 1st island 4.E-01 7.E-02 3.E-05 3.E-04 0.00 0.0 0.9244 BCD Economic city 2nd island 1.E-01 4.E-02 6.E-04 2.E-04 0.50 10.1 0.0099* D Economic city 3rd island 1.E+00 2.E-01 -3.E-04 1.E-03 0.01 0.1 0.7881 A Petro Rabigh far from the pipe 1.E+00 3.E-01 -3.E-03 2.E-03 0.34 2.1 0.2204 BCD Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey 208

HSD test Petro Rabigh few meters from pipe 6.E-01 2.E-01 3.E-04 1.E-03 0.02 0.1 0.7426 B Petro Rabigh under pipe 1.E-01 7.E-02 2.E-03 3.E-04 0.72 26.3 0.0004* CD Thuwal Fringe 5.E-01 8.E-02 -6.E-04 4.E-04 0.29 2.9 0.1326 BCD Thuwal land 5.E-01 9.E-02 -3.E-05 4.E-04 0.00 0.0 0.9391 BC Alkarrar 1st island 5.E-04 4.E-05 -1.E-09 2.E-07 0.00 0.0 0.9961 B Alkarrar 2nd island 3.E-04 3.E-05 2.E-07 2.E-07 0.14 0.7 0.4597 B Economic city 1st island 5.E-04 6.E-05 -3.E-09 3.E-07 0.00 0.0 0.992 B Economic city 2nd island 5.E-04 4.E-05 -2.E-07 2.E-07 0.13 1.4 0.2586 B Economic city 3rd island 8.E-04 1.E-04 -1.E-07 5.E-07 0.01 0.0 0.8393 A Petro Rabigh far from the pipe 9.E-04 3.E-04 -2.E-06 2.E-06 0.23 1.2 0.33 B Petro Rabigh few meters from pipe 5.E-04 9.E-05 5.E-07 5.E-07 0.13 1.0 0.3419 AB Petro Rabigh under pipe 4.E-04 6.E-05 2.E-07 2.E-07 0.05 0.5 0.4959 B Thuwal Fringe 4.E-04 6.E-05 -3.E-07 3.E-07 0.14 1.2 0.3163 B Thuwal land 6.E-04 1.E-04 -4.E-07 4.E-07 0.08 0.7 0.4386 B Alkarrar 1st island 2.E-02 2.E-03 3.E-05 9.E-06 0.58 9.6 0.0174* E Alkarrar 2nd island 1.E-02 1.E-03 7.E-05 6.E-06 0.97 134.2 0.0003* CDE Economic city 1st island 2.E-02 2.E-03 -3.E-06 9.E-06 0.02 0.1 0.718 E Economic city 2nd island 8.E-03 1.E-03 8.E-05 5.E-06 0.96 273.1 <.0001* E Economic city 3rd island 4.E-02 5.E-03 3.E-05 3.E-05 0.19 1.4 0.277 CD Petro Rabigh far from the pipe 8.E-02 2.E-02 6.E-04 2.E-04 0.80 16.3 0.0156* A Petro Rabigh few meters from pipe 1.E-01 2.E-02 1.E-04 1.E-04 0.22 2.0 0.2043 B Petro Rabigh under pipe 1.E-02 4.E-03 2.E-04 2.E-05 0.93 143.6 <.0001* C Thuwal Fringe 1.E-02 2.E-03 1.E-05 8.E-06 0.18 1.6 0.2526 E Thuwal land 2.E-02 2.E-03 5.E-05 8.E-06 0.84 42.2 0.0002* DE Alkarrar 1st island 4.E-03 4.E-04 -3.E-06 2.E-06 0.22 2.0 0.1998 A Alkarrar 2nd island 1.E-03 9.E-04 1.E-05 6.E-06 0.42 2.9 0.1633 ABC Economic city 1st island 1.E-03 8.E-04 1.E-05 4.E-06 0.38 4.9 0.0576 AB Economic city 2nd island -3.E-04 6.E-04 9.E-06 2.E-06 0.61 15.4 0.0028* BC Economic city 3rd island 3.E-03 9.E-04 -1.E-05 5.E-06 0.49 5.7 0.0546 BC Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey 209

HSD test Petro Rabigh far from the pipe 1.E-03 9.E-04 2.E-07 6.E-06 0.00 0.0 0.9755 ABC Petro Rabigh few meters from pipe 1.E-03 7.E-04 -2.E-06 4.E-06 0.05 0.4 0.5511 C Petro Rabigh under pipe -3.E-04 6.E-04 1.E-05 3.E-06 0.60 14.9 0.0032* BC Thuwal Fringe 2.E-03 9.E-04 -4.E-06 4.E-06 0.13 1.1 0.3309 ABC Thuwal land 3.E-03 6.E-04 -6.E-06 3.E-06 0.38 4.9 0.0572 ABC Alkarrar 1st island -2.E-04 1.E-04 2.E-06 5.E-07 0.64 12.7 0.0092* BC Alkarrar 2nd island -5.E-05 5.E-05 6.E-07 3.E-07 0.43 3.0 0.1583 BC Economic city 1st island 6.E-05 9.E-05 -8.E-08 4.E-07 0.00 0.0 0.8592 C Economic city 2nd island -5.E-05 4.E-05 5.E-07 2.E-07 0.45 8.3 0.0165* C Economic city 3rd island 5.E-04 2.E-04 1.E-06 1.E-06 0.22 1.7 0.2376 A Petro Rabigh far from the pipe -2.E-04 2.E-04 4.E-06 1.E-06 0.71 9.8 0.0354* AB Petro Rabigh few meters from pipe 4.E-04 3.E-04 2.E-06 2.E-06 0.18 1.5 0.2624 A Petro Rabigh under pipe -6.E-05 4.E-05 6.E-07 2.E-07 0.60 14.9 0.0031* C Thuwal Fringe -2.E-04 1.E-04 2.E-06 5.E-07 0.60 10.6 0.0138* BC Thuwal land -3.E-04 2.E-04 3.E-06 7.E-07 0.67 16.2 0.0038* BC Alkarrar 1st island 2.E-03 8.E-04 3.E-06 4.E-06 0.10 0.8 0.4098 AB Alkarrar 2nd island 4.E-04 3.E-04 2.E-06 2.E-06 0.21 1.0 0.365 C Economic city 1st island 2.E-03 6.E-04 -4.E-07 3.E-06 0.00 0.0 0.9136 AB Economic city 2nd island 8.E-04 3.E-04 2.E-06 1.E-06 0.16 2.0 0.1912 C Economic city 3rd island 4.E-03 9.E-04 -2.E-06 5.E-06 0.02 0.1 0.7455 A Petro Rabigh far from the pipe 2.E-03 8.E-04 -3.E-06 5.E-06 0.06 0.2 0.6452 BC Petro Rabigh few meters from pipe 2.E-03 5.E-04 2.E-06 3.E-06 0.09 0.7 0.4351 BC Petro Rabigh under pipe 7.E-04 3.E-04 2.E-06 1.E-06 0.32 4.8 0.0539 C Thuwal Fringe 1.E-03 2.E-04 -9.E-07 8.E-07 0.15 1.2 0.3043 C Thuwal land 2.E-03 4.E-04 -2.E-06 2.E-06 0.11 1.0 0.3443 BC Alkarrar 1st island 8.E-03 2.E-03 2.E-06 1.E-05 0.00 0.0 0.864 A Alkarrar 2nd island 1.E-02 4.E-03 -9.E-06 3.E-05 0.03 0.1 0.7493 A Economic city 1st island 1.E-02 2.E-03 -1.E-05 1.E-05 0.20 2.0 0.1929 A Economic city 2nd island 8.E-03 3.E-03 -2.E-07 1.E-05 0.00 0.0 0.9885 A Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey 210

HSD test Economic city 3rd island 6.E-03 7.E-03 3.E-05 3.E-05 0.08 0.5 0.4892 A Petro Rabigh far from the pipe 6.E-03 4.E-03 9.E-06 3.E-05 0.03 0.1 0.7439 A Petro Rabigh few meters from pipe 1.E-02 2.E-03 -1.E-05 1.E-05 0.10 0.8 0.4002 A Petro Rabigh under pipe 6.E-03 2.E-03 5.E-06 8.E-06 0.04 0.4 0.5327 A Thuwal Fringe 8.E-03 3.E-03 8.E-06 1.E-05 0.05 0.4 0.5719 A Thuwal land 1.E-02 4.E-03 -1.E-05 2.E-05 0.06 0.5 0.5131 A Alkarrar 1st island 6.E-02 5.E-03 3.E-06 2.E-05 0.00 0.0 0.9052 AB Alkarrar 2nd island 5.E-02 4.E-03 3.E-05 2.E-05 0.26 1.4 0.2964 BCD Economic city 1st island 6.E-02 3.E-03 5.E-06 2.E-05 0.01 0.1 0.79 ABC Economic city 2nd island 5.E-02 3.E-03 3.E-05 1.E-05 0.48 9.2 0.0126* BCD Economic city 3rd island 7.E-02 4.E-03 -4.E-05 2.E-05 0.38 3.8 0.1009 A Petro Rabigh far from the pipe 6.E-02 3.E-03 -2.E-05 2.E-05 0.19 1.0 0.3828 BCD Petro Rabigh few meters from pipe 6.E-02 2.E-03 -6.E-05 1.E-05 0.86 44.2 0.0003* CD Petro Rabigh under pipe 4.E-02 4.E-03 6.E-05 2.E-05 0.56 12.5 0.0054* CD Thuwal Fringe 5.E-02 3.E-03 2.E-05 2.E-05 0.18 1.5 0.2589 D Thuwal land 7.E-02 5.E-03 -3.E-05 2.E-05 0.20 1.9 0.2012 A Alkarrar 1st island 4.E-03 3.E-03 2.E-05 1.E-05 0.24 2.2 0.1821 A Alkarrar 2nd island 2.E-03 4.E-03 4.E-05 3.E-05 0.33 1.9 0.2375 A Economic city 1st island 8.E-03 3.E-03 -2.E-06 1.E-05 0.00 0.0 0.8951 A Economic city 2nd island 1.E-03 2.E-03 1.E-05 9.E-06 0.16 1.9 0.1996 A Economic city 3rd island 8.E-03 2.E-03 -8.E-06 1.E-05 0.07 0.5 0.5195 A Petro Rabigh far from the pipe 7.E-03 4.E-03 -1.E-05 2.E-05 0.06 0.3 0.635 A Petro Rabigh few meters from pipe 5.E-03 2.E-03 -2.E-06 1.E-05 0.01 0.0 0.8387 A Petro Rabigh under pipe 7.E-03 2.E-03 -9.E-06 1.E-05 0.07 0.7 0.4152 A Thuwal Fringe 1.E-02 4.E-03 -2.E-05 2.E-05 0.15 1.3 0.2997 A Thuwal land 1.E-02 3.E-03 -2.E-05 1.E-05 0.20 2.0 0.1917 A Alkarrar 1st island 6.E-02 5.E-03 3.E-06 2.E-05 0.00 0.0 0.9052 AB Alkarrar 2nd island 5.E-02 4.E-03 3.E-05 2.E-05 0.26 1.4 0.2964 BCD Economic city 1st island 6.E-02 3.E-03 5.E-06 2.E-05 0.01 0.1 0.79 ABC Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey 211

HSD test Economic city 2nd island 5.E-02 3.E-03 3.E-05 1.E-05 0.48 9.2 0.0126* BCD Economic city 3rd island 7.E-02 4.E-03 -4.E-05 2.E-05 0.38 3.8 0.1009 A Petro Rabigh far from the pipe 6.E-02 3.E-03 -2.E-05 2.E-05 0.19 1.0 0.3828 BCD Petro Rabigh few meters from pipe 6.E-02 2.E-03 -6.E-05 1.E-05 0.86 44.2 0.0003* CD Petro Rabigh under pipe 4.E-02 4.E-03 6.E-05 2.E-05 0.56 12.5 0.0054* CD Thuwal Fringe 5.E-02 3.E-03 2.E-05 2.E-05 0.18 1.5 0.2589 D Thuwal land 7.E-02 5.E-03 -3.E-05 2.E-05 0.20 1.9 0.2012 A Alkarrar 1st island 1.E-02 2.E-03 9.E-07 9.E-06 0.00 0.0 0.923 CD Alkarrar 2nd island 3.E-03 1.E-03 7.E-06 8.E-06 0.16 0.8 0.4291 D Economic city 1st island 2.E-02 3.E-03 2.E-07 1.E-05 0.00 0.0 0.9861 BCD Economic city 2nd island 4.E-03 2.E-03 2.E-05 8.E-06 0.43 7.6 0.0204* D Economic city 3rd island 4.E-02 8.E-03 -9.E-06 4.E-05 0.01 0.1 0.8306 A Petro Rabigh far from the pipe 4.E-02 1.E-02 -1.E-04 9.E-05 0.34 2.1 0.2237 BCD Petro Rabigh few meters from pipe 2.E-02 7.E-03 1.E-05 4.E-05 0.01 0.1 0.7624 B Petro Rabigh under pipe 3.E-03 3.E-03 6.E-05 1.E-05 0.70 23.8 0.0006* BCD Thuwal Fringe 2.E-02 3.E-03 -3.E-05 2.E-05 0.35 3.8 0.0938 BCD Thuwal land 2.E-02 4.E-03 -8.E-06 2.E-05 0.03 0.2 0.6423 BC Alkarrar 1st island 9.E-03 2.E-03 4.E-05 1.E-05 0.64 12.3 0.0099* DE Alkarrar 2nd island 8.E-03 2.E-03 6.E-05 1.E-05 0.83 19.7 0.0114* CD Economic city 1st island 8.E-03 1.E-03 2.E-05 6.E-06 0.57 10.6 0.0116* F Economic city 2nd island 9.E-05 1.E-03 6.E-05 6.E-06 0.92 109.0 <.0001* F Economic city 3rd island 2.E-02 3.E-03 3.E-05 1.E-05 0.46 5.0 0.0663 BC Petro Rabigh far from the pipe 8.E-03 2.E-03 9.E-05 2.E-05 0.90 36.2 0.0038* AB Petro Rabigh few meters from pipe 2.E-02 2.E-03 3.E-05 1.E-05 0.54 8.4 0.0233* A Petro Rabigh under pipe 4.E-03 1.E-03 5.E-05 5.E-06 0.90 90.7 <.0001* EF Thuwal Fringe 8.E-03 1.E-03 7.E-05 7.E-06 0.93 96.2 <.0001* CD Thuwal land 1.E-02 1.E-03 6.E-05 5.E-06 0.93 109.3 <.0001* BC Alkarrar 1st island 5.E-02 1.E-02 -2.E-04 7.E-05 0.49 6.7 0.0359* A Alkarrar 2nd island 1.E-02 6.E-04 -2.E-05 4.E-06 0.88 29.5 0.0056* BC Location Site Intercept I.St.Error Slope S.St.Error R2 F ratio p Value Tukey 212

HSD test Economic city 1st island 1.E-02 9.E-04 -7.E-06 4.E-06 0.25 2.7 0.1387 C Economic city 2nd island 1.E-02 9.E-04 1.E-05 4.E-06 0.58 13.7 0.0041* BC Economic city 3rd island 1.E-02 3.E-03 -1.E-05 2.E-05 0.07 0.5 0.5245 ABC Petro Rabigh far from the pipe 3.E-02 5.E-03 -1.E-04 3.E-05 0.73 10.7 0.0307* BC Petro Rabigh few meters from pipe 2.E-02 2.E-03 -9.E-06 1.E-05 0.11 0.8 0.3925 ABC Petro Rabigh under pipe 1.E-02 5.E-03 5.E-05 2.E-05 0.37 5.7 0.0375* AB Thuwal Fringe 7.E-03 8.E-04 -1.E-05 4.E-06 0.53 8.0 0.0255* C Thuwal land 1.E-02 8.E-04 -1.E-05 3.E-06 0.65 14.8 0.0049* C Alkarrar 1st island 1.E-03 2.E-04 -2.E-06 9.E-07 0.30 3.0 0.1288 B Alkarrar 2nd island 7.E-04 7.E-05 -7.E-07 5.E-07 0.37 2.4 0.2001 B Economic city 1st island 1.E-03 2.E-04 -1.E-06 1.E-06 0.10 0.9 0.3832 B Economic city 2nd island 5.E-04 5.E-05 6.E-07 2.E-07 0.45 8.3 0.0163* B Economic city 3rd island 2.E-03 2.E-04 -6.E-07 1.E-06 0.04 0.2 0.6421 A Petro Rabigh far from the pipe 2.E-03 4.E-04 -5.E-06 3.E-06 0.36 2.3 0.2062 B Petro Rabigh few meters from pipe 1.E-03 2.E-04 -6.E-08 1.E-06 0.00 0.0 0.9561 B Petro Rabigh under pipe 4.E-04 1.E-04 1.E-06 4.E-07 0.44 7.9 0.0183* B Thuwal Fringe 9.E-04 8.E-05 -9.E-07 4.E-07 0.45 5.8 0.0466* B Thuwal land 8.E-04 2.E-04 6.E-07 1.E-06 0.04 0.3 0.6024 B

Table S 5: The detailed slopes calculated per tree for heavy metals content 213

flux (mg m-2 y-1) Average flux (mg m-2 y-1) flux (ton km2 y-1) Location Alkarrar Economic city Petro Rabigh Thuwal N Rows 15.00 30.00 27.00 19.00 Ag 14.46 49.30 56.91 36.29 39.24 5.30 Al 269.34 672.50 733.90 519.27 548.75 74.08 As 3.20 4.49 5.71 4.41 4.45 0.60 Be 0.00 0.23 0.08 0.15 0.11 0.02 Cd 0.30 0.46 0.23 0.38 0.34 0.05 Co 0.15 0.61 0.68 0.38 0.46 0.06 Cr 1.14 3.20 2.28 1.98 2.15 0.29 Cu 3.04 5.33 4.49 4.11 4.24 0.57 Fe 331.88 884.17 985.51 685.59 721.78 97.44 Li 0.61 1.07 0.91 0.68 0.82 0.11 Mn 33.63 55.08 190.36 42.99 80.52 10.87 Mo 3.80 4.87 2.82 2.66 3.54 0.48 Nb 0.08 0.38 0.76 0.53 0.44 0.06 Ni 2.36 4.18 2.74 2.28 2.89 0.39 Pb 12.17 17.19 13.09 13.54 14.00 1.89 Rb 79.05 119.60 94.65 94.19 96.87 13.08 Se 11.56 12.02 9.05 10.42 10.77 1.45 Sr 79.05 119.60 94.65 94.19 96.87 13.08 Ti 10.88 32.41 34.54 24.50 25.58 3.45 V 23.05 30.81 39.49 38.88 33.06 4.46 Zn 17.50 21.30 28.76 10.04 19.40 2.62 Zr 0.91 1.90 1.45 1.37 1.41 0.19 214

Chapter Six

Carbon sink capacity of Red Sea mangroves

Hanan Almahasheer1, 2, Oscar Serrano Gras3, 4, Carlos M. Duarte1, Pere Masque3, 4, 5 and

Xabier Irigoien1

1 King Abdullah University of Science and Technology (KAUST), Red Sea Research Center, Thuwal 23955-6900, Kingdom of Saudi Arabia

2 Biology Department, University of Dammam (UOD), Dammam 31441-1982, Kingdom of Saudi Arabia

3School of Natural Sciences & Centre for Marine Ecosystems Research, Faculty of Health, Engineering and Science, Edith Cowan University, 270 Joondalup Drive, Joondalup, 6027

4The UWA Oceans Institute, University of Western Australia, 35 Stirling Highway, Crawley 6009

5Universitat Autònoma de Barcelona, Departament de Física -Institut de Ciència i Tecnologia Ambientals, Spain

215

Abstract

Mangroves, Avicennia marina, occupy about 135 Km2 in the Red Sea and represent one of

the most important vegetated communities in this otherwise arid and oligotrophic region.

We assessed the belowground carbon sequestered in 10 sites within four locations in the

Saudi coast of the Central Red Sea, to find that, the sediment accretion rate is within the

global range of mangrove forests. However, both Corg density and stock in mangrove

sediment were the lowest globally with an average of 42.5 Mg ha-1 and 0.0044 g cm-3

respectively, resulting in a low carbon sequestration rate of 2.22 to 34 Corg g m-2 yr-1

based on 14C and 210Pb estimates respectively. This low rate might be due to the extreme

conditions in the Red Sea such as low rainfall, nutrient limitation and high temperature,

reducing the growth rates of themangroves and increasing the rate of respiration in the sediment.

Keywords: blue carbon, carbon sequestration, Corg, 14C, 210 Pb, 13C, and 15 N.

δ δ

216

Introduction

Mangrove forests contribute directly to the economic development of costal

societies and also provide important global ecosystem services, in particular, their role in

carbon sequestration (Suratman 2008). Although mangroves occupy only 0.5% of global

coastal ocean area (Alongi 2014), and 0.7% of the tropical forests of the world (Giri,

Ochieng et al. 2011), they account for about 1% of the carbon sequestered by the world’s

forests, and as coastal habitats they account for 14% of the carbon sequestered by the

global ocean (Alongi 2012).Also, mangroves occupy 0.2% of the marine vegetated habitats

surface, but contribute 50% of carbon burial in marine sediments (Duarte, Losada et al.

2013), which implies a significant influence on the global carbon budget (Jennerjahn and

Ittekkot 2002, Kristensen, Bouillon et al. 2008). In terms of CO2 budget mangroves provide

approximately 1.6 billion dollar a year by sequestering more than 25.5 M ton of C a year

(Polidoro, Carpenter et al. 2010). Their significance in carbon sequestering is mainly due to

their ability to trap and retain sediments (Ewel, Twilley et al. 1998), contributing to a high

sedimentation rate compared to non mangroves area (Kathiresan 2003). Further, the high

litter production and low rate of soil respiration result in high net ecosystem production,

and delivery of high inputs of carbon to the sediment (Komiyama, Ong et al. 2008). On the

other hand, their canopy has declined globally and the losses continue every year (Valiela,

Bowen et al. 2001). The deforestation effect contributes to increase the amount of Co2 in

the atmosphere, and is estimated to be the second largest anthropogenic source after fossil

fuel combustion (Van der Werf, Morton et al. 2009), contributing to an emission of 1 Pg CO2

annually when marine vegetated habitats are lost (Duarte, Losada et al. 2013) and in particular emissions of 0.02– (Donato,

0.12 Pg carbon per year for mangrove deforestation 217

Kauffman et al. 2011). However, the Red Sea is one region where mangroves are not only stable but also have slightly expanded by 12% over the last 41 years (Almahasheer,

Aljowair et al. 2016). Therefore, mangroves can be playing a significant role in carbon sequestration in an arid area that is otherwise devoid of vegetation.

However, the carbon sequestration rates estimated in other regions of the world cannot be directly applied to the Red Sea. Carbon capture/storage of plant biomass and sediments rely on differences in sediments flux, rate of primary production, carbon allocation to root biomass, nutrient content in plants influencing the decomposition rates of organic matter, efficiency in plant canopy, and the microbial and macro faunal community effect on sediment remineralization rates (Atwood, Connolly et al. 2015). And while mangrove species survive harsh saline conditions, salinity increases reduce the carbon assimilation capacity and growth (Naidoo 2006). The mangroves growth in the Red

Sea is limited, with dwarfed trees and limited growth (Almahasheer, submitted). Further high temperatures and lack of rivers suggest higher organic matter degradation and lower sediment influx. To our knowledge only one recent paper has estimated the carbon sequestration rates in arid area (Ezcurra, Ezcurra et al. 2016), and still this is in an area,

Baja California, where the ocean is highly productive. The Red Sea combines an extremely arid climate with an oligotrophic sea, a unique combination for mangroves.

Therefore, our aim in this paper is to assess the amount of carbon that Central Red

Sea mangroves sequestered, which to date has never been reported, by combining estimates of carbon density down to about 1 m in depth along the profile of the sediment with date estimates derived from 210Pb and 14C isotopes to estimate (a) the burial rate of organic carbon over time and (b) the stock of organic carbon contained within the top 218 meter of the sediment. Further, we use stable isotopes to estimate potential contribution of different additional sources of organic carbon such as mangroves, halophytes, macroalgae, segrasses and plankton. These results inform the potential of blue carbon strategies, adopted as one of the mitigation strategies of the Kingdom of Saudi Arabia, in mitigating

CO2 emissions through the restoration and conservation of mangroves, which remains a healthy and stable ecosystem in the Red Sea.

Methods

1. Study location

This study was conducted in 10 sites within four locations in the Saudi coast of the

Central Red Sea (Fig. 1). The area in this study is about 80 Km long, from Thuwal island were mangroves grow on a shallow soil of weathered coral (Balk, Keuskamp et al. 2015) to

Alkarrar lagoon lying on a coastal plain north west Rabigh. Rabigh is a city between the two locations, with many factories and petrochemical industry, whereas King Abdullah

Economic City, about 40km south of Rabigh, is a new city under construction (Gheith and

Abou-ouf 1996, Al-Farawati 2011). Thuwal Island and Alkarrar lagoon are relatively away from human disturbance whereas the impacts in Rabigh and Economic city are expected to be higher. 219

Figure 1: Location of the sampled Central Red Sea mangrove stands. The map was produced with ArcMap Version 10.2. Background map credits: the World Administrative

Divisions layer provided by Esri Data and Maps and DeLorme Publishing Company. 220

2. Collecting and processing the samples

Cores were obtained using manual percussion and rotation of PVC pipes of class 12

with an inner diameter of 62.6 mm. 29 cores were obtained in the A. marina rhizosphere

sediments (< 0.5 m water depth), with an average core depth of 1.7 m deep except for one

core that was less than 1 m. We had two types of cores, whole cores and port cores with 3

cm holes in diameter. Cores were hammered gently in order to reduce compression. The

length of core barrel inserted into the sediment and the length of retrieved mangrove

sediment were recorded in order to correct the core lengths for compression effects and all

variables studied here are referenced to the corrected, decompressed depths. The cores

were sealed at both ends and transported vertically to the laboratory to be processed the

next day. The sediment core was sliced using a ceramic knife in 1cm thick slices, to be

subsequently dried at 60 °C oven till constant weight, to determine the dry bulk density (g

cm3). The slices were then grounded in an agate mortar and subdivided for different

analysis.

Additionally, we collected leaves from the same locations to analyze the carbon flux to

the sediment through shedding. We calculated the flux of carbon due to mangroves leaves

(mg carbon m-2 year-1) based on a previous estimation of number of leaves shed per year

using leaf production and tree density measurements in a square meter. The carbon

content was analyzed using a FLASH 2000 CHNS Analyzer, see chapter 4 for method and percent of accuracy.

3. Analyzing the samples

The granulometry analysis was performed using a Mastersizer 2000-Malvern

Instruments at Centro De Estudios Avanzados De Blanes, Spain. Briefly, for the grain size 221

analysis the samples were first digested with H2O2 30% for 48 hours, rinsed with tap water, then oven dried at 60 °C for 48h. The dried samples were manually passed through a 1mm sieve and the <1mm fraction was analyzed with the MasterSizer. However, some particles

>1mm passed through the manual sieve and were detected in the machine, therefore we

excluded results >1mm and we normalized the results to 100% again as the MasterSizer

measures the % of particles based on the size class. All granulometry analysis samples

were weighted before and after digestion, drying and the actual values are the average of 3

analytical replicates, in addition to rerunning the samples until the residuals were < 1%.

Grain size classification and texture were categorized following Wentworth (Wentworth

1922) scale of sediment grad and class.

For organic carbon (Corg) and d13C analyses, 0.5 g of ground sample was acidified with 1

M HCl to remove inorganic carbon, centrifuged at 3500 revolutions per minute, for 5 min,

and the supernatant with acid residues was carefully removed by pipette, avoiding

resuspension. The sample was then washed with Milli-Q water, centrifuged and the

supernatant removed. The residual samples were re dried at 60 °C for 72 hours or until dry

and then encapsulated for Corg analysis using a Thermo Delta V Conflo III coupled to a

Costech 4010 at UH Hilo Analytical Laboratory, USA. The content of Corg was calculated for

the bulk acidified samples. Carbon isotope ratios are expressed as d values in parts per

thousand and relative to the Vienna Pee Dee Belemnite standard. In addition, we collected

living materials from the same locations such as mangroves (aerial roots, green and

senescent leaves, stem, buds, flowers), halophytes, macro-algae, seagrasses (rhizomes and

leaves) and suspended particulate matter (fraction retained on a

filter). The living material samples were dried and analyzed as described0.7 μm above, pore however, diameter 222

they were not acidified as they do not contain calcium carbonate in their tissues.

Different approaches were used for quality control. Duplicate samples were run every

20 samples. The percent recovery ranged from 92 to 100.8 % for %N and

. In addition, two quality control (QC) samples wereδ15N; run and alongside 99.8 to

the101.2% samples for %C to andverify δ13C accuracy and precision of isotope data. The QC samples were NIST

1547 (Peach Leaves) for no acidified samples and NIST 8704 (Buffalo River Sediment) for

the acidified samples. The c

= - arbon isotope data were normalized to USGS 40 (δ13C vs VPDB

26.4) and USGS 41 (δ13C vs VPDB = 37.6) and had an accuracy- of 0.2‰. Whereas, nitrogenvs Air = isotope47.6) and data had were an normalized accuracy ofto 0.2USGS‰. 40 Finally, (δ15N twovs Air standards = 4.5) and of USGSknown 41 values(δ15N

(acetanilide and glycine) were run alternatively every 12 samples in every batch. The

IsoSource software package was used (Phillips and Gregg 2003) to estimate the proportion of different marine plants and plankton to the organic carbon in the sediment, using a 1% increment and 0.1‰ tolerance.

4. Dating

The sediment cores were dated combining 210Pb (short term-last century, 100 yrs.), and

radiocarbon dating 14C (Long term-last millennia, 1000 yrs.).

The concentrations of 210Pb in the upper 20 to 30 cm were determined in the fraction

<125 um at the Universitat Autònoma de Barcelona (Spain) by alpha spectrometry through

the measurement of its granddaughter 210Po, assuming radioactive equilibrium between

both radionuclides. Briefly, aliquots of about 200 mg of each sample were spiked with a

known amount of 209Po and acid-digested in a microwave with a mixture of HNO3 and HF.

After digestion, boric acid was added to avoid the formation of insoluble fluorides. The 223

resulting solutions were evaporated and diluted to 100 mL 1 M HCl and polonium isotopes

were autoplated onto pure silver disks. Before deposition, acid ascorbic was added to

complex Fe. Polonium emissions were measured by alpha spectrometry using Passivated

Implanted Planar Silicon, PIPS detectors (CANBERRA, Mod. PD-450.18 A.M.) The

concentrations of excess 210Pb were calculated by subtraction of the supported 210Pb, which

was evaluated from the determination of 226Ra by ultralow background liquid scintillation

counting (LSC, Quantulus 1220) in selected samples from each core. The remaining

solutions after polonium deposition were evaporated to dryness and the volume was

adjusted to 10 mL HCl 0.5M in 20 mL low-diffusion counting vials. 10 mL of Mineral Oil

Liquid scintillation cocktail were added, and mixtures were stored for 3 weeks in a dark,

temperature controlled area so that the 226Ra progeny would reach equilibrium. Samples

were counted sequentially for 1 h each and during 15 cycles. The concentration of 226Ra

was determined from the quantification of the alpha emissions of its decay products in

equilibrium (222Rn, 218Po and 214Po). Analyses reagent blanks, replicates and a reference

material (IAEA - 315, marine sediments) were carried out for both 210Pb and 226Ra to assess for any contamination and to ensure reproducibility of the results.

The results are synthesized in terms of concentrations profiles of 210Pb and 226Ra for

each core, as well as the estimated sediment accumulation rates where possible.

The radiocarbon were dated for 2 depths per core following standard procedures (ISO

17025 and ISO 9001) at the AMS Direct Laboratory, USA using Radiocarbon accelerator

mass spectrometry, after an acid base acid treatment for the shells and some samples of

sediments when the shells were not available.. The raw radiocarbon date reported by the

laboratory was calibrated (Calib 7.1; Marine13 curve) and corrected for the marine 224

reservoir effect (i.e. subtracting Delta R value of 116 ± 11 years based on our sampling

location from the uncorrected radiocarbon ages, see the Marine Reservoir Correction

Database; http://www.calib.qub.ac.uk/marine/). The estimated radiocarbon age is

expressed after correction and calibrated to years before present (present taken as AD

2014).

Finally, in order to estimate carbon sequestration whenever we did not have 14C data for a core we assume that the core had the same sediment accretion rates as the average from the same study location (4 cases for 14C). For 210Pb we only had one to two cores per

location where the age could be estimated and therefore we used that value for all other

cores in same the location.

5. Statistical analysis

Statistics analyses, including descriptive statistics, general linear models to test differences among stands, and Tukey HSD posthoc test to assess pairwise differences were carried out using JMP.

Results

The distribution of sediment grain size in Petro Rabigh was significantly higher in clay,

silt and very fine sand which also increased with depth compared to the other locations

resulting in a loamy sand sediment. On the other hand, the medium and large sand sizes

were significantly higher in Thuwal, followed by Economic city and Khor Alkharrar

resulting in a coarse sand texture as more than 25% of their total weight had sediment grain size higher than 1mm (Fig. 2 and Table. 1, Tukey HSD post hoc test, P < 0.05).

225

Table 1: Mean (± SE) for % sediment grain sizes and texture from four different locations in the Central Red Sea. R2 and F value correspond to an ANOVA testing for significant differences between locations. * P between 0.01 and 0.05, ** P < 0.01. locations with different letters indicate significant differences (Tukey HSD multiple comparison post-hoc test, P < 0.05).

226

Figure 2: Sediment grain size fractions, expressed as a percentage of the total sediment dry from four different locations in the Central Red Sea. Sediments were classified as;

227

The 210Pb concentration profiles were determined along the upper 20 to 30 cm of

the following sediment cores TMF; Thuwal Island and EME; Economic city; KMC and KME;

Khor Alkharrar while RMF; Petro Rabigh. These sediment showed mixed profiles between

the upper 2 and 10 cm (Fig. 3). The supported Ra-226 concentrations were 5.6 ± 0.6, 7.7 ±

1.0, 10.7 ± 1.6, 10.9 ± 1.3, and 12.0 ± 0.7 in KMC, TMF, RMF, EME, and KME respectively.

The excess 210Pb concentrations at undisturbed mangrove sediments decreased down to 11, 12, 15, and 22 cm depth, allowing us to apply CF:CS model to apply (CFCS model : Constant Flux Constant Sedimentation rate by (Krishnaswamy, Lal et al. 1971,

Robbins, Edgington et al. 1978) to estimate a sediment accumulation rate for the last century, which were 1.47±0.44, 0.37±0.07, 0.24±0.10, 0.21±0.07, and 0.07±0.01 cm yr-1 in

EME, KMC, RMF, TMF, and KME respectively, to be used as an upper limit.

228

Figure 3: Concentration profiles of total and excess 210Pb in mangrove sediments in central Red Sea.

The Corg stock in the 10 cm surface sediment was significantly higher in Khor

Alkarrar compared to the rest. However, the high 210Pb sediment accretion rate in

Economic city resulted in a significantly higher carbon sequestration in Economic city 229

compared to all locations (Table. 2, Tukey HSD post hoc test, P < 0.05). Similarly, the Corg

Stock in the 1st m sediment was significantly higher in Khor Alkarrar compared to the rest.

Nevertheless, the 14C sediment accretion rate did not differ among location resulting in an equivalent carbon sequestration rate (Table. 3, Tukey HSD post hoc test, P > 0.05).

Table 2: Mean (± SE) of organic carbon in 10 cm stock sediment from four different locations in the Central Red Sea, along with estimation of 210Pb sediment accretion rate for the last 100–150 years and carbon sequestration rate. R2 and F value correspond to an

ANOVA testing for significant differences between locations. * P between 0.01 and 0.05, **

P < 0.01. locations with different letters indicate significant differences (Tukey HSD multiple comparison post-hoc test, P < 0.05).

location N Corg Stock-in 10 cm thick soil 210Pb Carbon sediment Sequestration accretion using 210Pb rate Corg kg m-2 Corg g m-2 mm yr-1 Corg g m-2 yr-1 Economic 8 0.53±0.06 531.00±58.34b 14.68±0.00a 78.00±8.59a city b Khor Alkarar 7 1.54±0.43a 1541.29±429.91a 2.18±0.33b 31.00±9.46b Petro Rabigh 6 0.38±0.15b 382.33±153.47b 2.42±0.00b 9.17±3.75b Thuwal 8 0.53±0.15b 529.63±146.14b 2.07±0.00b 11.25±3.03b Island R2 0.37 0.37 0.99 0.73 F ratio 5.04** 5.04** 1636.85** 22.08** All 29 0.74±0.14 743.72±139.58 5.65±1.06 34.00±6.30

230

Table 3: Mean (± SE) of organic carbon in 1m stock sediment from four different locations

in the Central Red Sea, along with estimation of 14C sediment accretion rate for the last

1000 years. Both estimations resulted a carbon sequestration rate. R2 and F value

correspond to an ANOVA testing for significant differences between locations. * P between

0.01 and 0.05, ** P < 0.01. locations with different letters indicate significant differences

(Tucky HSD multiple comparison post-hoc test, P < 0.05).

location N Corg Stock-in 1m thick soil 14C sediment Carbon accretion Sequestration rate using 14C Corg kg m-2 Corg g m-2 mm yr-1 Corg g m-2 yr-1 Economic 8 3.84±0.29b 3837.73±291.30b 0.53±0.12a 2.22±0.70a city Khor Alkarar 7 7.62±1.53a 7617.64±1530.31a 0.51±0.19a 4.34±2.46a Petro Rabigh 6 2.47±0.47b 2471.28±469.61b 0.42±0.08a 1.17±0.36a Thuwal 8 3.04±0.42b 3033.59±416.13b 0.34±0.08a 1.14±0.37a Island R2 0.48 0.48 0.05 0.14 F ratio 7.77** 7.77** 0.46ns 1.35ns All 29 4.25±0.53 4245.58±533.36 0.45±0.06 2.22±0.65

Sediment in Khor Alkharrar were characterized by a high sediment and Corg density along with high %Corg compared to the rest locations (Table. 4, Tukey HSD post hoc test, P <

0.05), mainly because %Corg in cores KME, KMB and KMD ranged from 5 to 25% in the

upper 25 cm, whereas the rest of cores in Khor Alkharrar as well as other locations were

less than 2.5% (Fig. 4.a). Moreover, Khor Alkharrar cores density ranged from 0 to 2.5 g cm-

3 along the 180 cm sediment profile while rest of cores in all locations ranged from 0.5 to

1.5 g cm-3 with a maximum 2 g cm-3 in a few cores (Fig. 4.b). Likewise, the Corg density in 231

Khor Alkharrar were higher compared to rest of locations as the surface sediment reaches

0.05 g cm-3 while the rest were lower than 0.02 g cm-3 along the sediment profile (Fig. 4.c).

and 13C values for sediments and different groups are presented in (Table. 4, 5 andδ15N Figure. 5).δ and 13C values of marine plants and sediment indicated that the source of the Corgδ15N in the sedimentδ could be attributed as 36% from mangroves, 25% from halophytes, 15% from macro algae, 14% from suspended particulate matter, and 11% from seagrasses (Table. 6).

Finally, the carbon flux estimated from Avicennia marina leaves was 622, 748, 861 and

710 g C m-2 y-1 in Alkarrar, Petro Rabigh, Economic city and Thuwal respectively. With an average of 735g C m-2 y-1 for the Central Red Sea.

Table 4: Mean (± SE) of sediment characteristics from four different locations in the Central

Red Sea. R2 and F value correspond to an ANOVA testing for significant differences between locations. * P between 0.01 and 0.05, ** P < 0.01. locations with different letters indicate

significant differences (Tukey HSD multiple comparison post-hoc test, P < 0.05).

Location N Sediment Corg %Corg Isotopes Density Density (g cm-3) (g cm-3) Economic city 268 1.00±0.01c 0.0039±0.0001b 0.41±0.01b -δ13C19.32±0.19a 2.00±0.00δ15N b Khor Alkarar 250 1.24±0.02a 0.0074±0.0005a 1.23±0.20a -20.05±0.20b 1.96±0.02b Petro Rabigh 200 1.09±0.02b 0.0025±0.0001c 0.24±0.02b -20.81±0.22c 4.13±0.00a Thuwal 152 1.15±0.02b 0.0032±0.0002bc 0.29±0.02b -18.83±0.22a 2.13±0.00b R2 0.08 0.16 0.05 0.04 0.45 F ratio 28.38** 53.49** 17.20** 14.97** 241.66** All 870 1.12±0.01 0.0044±0.0002 0.58±0.06 -19.79±0.11 2.50±0.00

232

Table 5: Mean (± SE) of marine plants and sediment in the Central Red Sea.

Group N δ15N δ13C Macro algae 54 1.6±0.1 -12.5±0.5 Halophytes 33 4.7±0.5 -24.7±1.0 Mangrove 117 1.7±0.2 -26.1±0.1 Suspended particulate 12 2.8±0.2 -11.3±1.6 matter Seagrass 111 0.8±0.2 -8.2±0.2 Sediment 870 2.5±0.0 -19.8±0.1

Table 6: The sources of organic carbon in sediment obtained using marine plant and

sediment δ15N and δ13C values. Group Mean Min Max 1 %ile 50 99 STD %ile %ile DEV Macroalgae 0.146 0 0.44 0 0.13 0.4 0.107 Halophytes 0.251 0.09 0.39 0.12 0.25 0.36 0.057 Mangrove 0.357 0.26 0.49 0.28 0.35 0.47 0.045 Suspended 0.139 0 0.42 0 0.12 0.39 0.102 particulate matter Seagrass 0.107 0 0.32 0 0.09 0.3 0.079

233

A

B

C

Figure 4: Vertical profile of the % Corg and sediment and carbon density g cm-3 in mangrove sediments in central Red Sea. 234

A

B

Figure 5:

Vertical profile of δ13C and δ15N in mangrove sediments in central Red Sea.

Discussion

The overall Corg density in our results, 0.0044 g cm-3 is lower than any other mangrove

area worldwide according to a global estimate of 154 mangrove sites in a latitudinal range

from 22.4°S in the Indian Ocean to 55.5°N in the northeastern Atlantic, were they found an

average value of 0.055 g cm-3 and a minimum value of 0.023 g cm-3 (Chmura, Anisfeld et al.

2003).

The accretion rate in Economic city was the highest 14.7 mm yr-1, however, the average

accretion rate in our results was 5.65 mm yr-1 based on 210Pb(last 100 years) and 0.45 mm 235

yr-1 based on 14C (last millennia). Both results are within the average rate of soil accretion

in mangrove forests as they usually range from 0.1 to 10.0 mm yr-1 (Alongi 2012). Yet, higher than a recent estimate by (Lovelock, Cahoon et al. 2015) were the sediment accumulation using 210Pb was 1.2 and 1.7 mm yr-1 in Moreton Bay and southeastern

Australia respectively.

The carbon sequestration rate based on 14C in our results (1.14 g m-2 yr-1 to 4.34g m-2 yr-

1) was lower than a recent assessment in Baja California, where they found carbon

sequestration to be around 256 g m-2 yr-1 for Avicennia mudflats (Ezcurra, Ezcurra et al.

2016) . Another study in Pohnpei Island, Micronesia estimated that carbon accumulation rate was 93 g m-2 yr-1in conjunction with a sedimentation rate of 2 mm yr-1 (Fujimoto,

Imaya et al. 1999).

The very low carbon sequestration rates in the mangrove sediments of the Red Sea are

most likely due to the extreme conditions. The lack of rivers, the oligotrophic nature of the

sea and the extremely arid conditions result in growth limitation and dwarfed trees.

Further, the lack of rivers implies that there is no an influx of riverine sediments that can

be trapped by the mangroves root system and all carbon stored needs to be of

autochthonous origin. This is reflected by the low contribution of the suspended particulate

matter to the sedimentary carbon. The sequestration rate is still significantly lower than

the annual flux of leaves from trees to the sediment, reflecting high respiration rates due to the high temperature of the region, and in particular of the water bathing the mangroves, with minimums of 25oC in winter and exceeding 33 oC in summer. However, it has to be

considered that mangroves are the only forests in the coastal regions around the Red Sea, 236 and even under these harsh conditions, the observed carbon sequestration are still similar to those observed in tropical forests (Laffoley and Grimsditch 2009).

References

Al-Farawati, R. (2011). "Spatial and Seasonal Distribution of Total Dissolved Copper and Nickel in the Surface Coastal Waters of Rabigh, Eastern Red Sea, Saudi Arabia." Journal of King Abdulaziz University: Earth Sciences 22(1).

Almahasheer, H., A. Aljowair, C. M. Duarte and X. Irigoien (2016). "Decadal stability of Red Sea mangroves." Estuarine, Coastal and Shelf Science 169: 164-172.

Alongi, D. M. (2012). "Carbon sequestration in mangrove forests." Carbon management 3(3): 313-322.

Alongi, D. M. (2014). "Carbon cycling and storage in mangrove forests." Annual review of marine science 6: 195-219.

Atwood, T. B., R. M. Connolly, E. G. Ritchie, C. E. Lovelock, M. R. Heithaus, G. C. Hays, J. W. Fourqurean and P. I. Macreadie (2015). "Predators help protect carbon stocks in blue carbon ecosystems." Nature Climate Change 5(12): 1038-1045.

Balk, M., J. A. Keuskamp and H. J. Laanbroek (2015). "Potential Activity, Size, and Structure of Sulfate-Reducing Microbial Communities in an Exposed, Grazed and a Sheltered, Non- Grazed Mangrove Stand at the Red Sea Coast." Frontiers in microbiology 6.

Chmura, G. L., S. C. Anisfeld, D. R. Cahoon and J. C. Lynch (2003). "Global carbon sequestration in tidal, saline wetland soils." Global biogeochemical cycles 17(4).

Donato, D. C., J. B. Kauffman, D. Murdiyarso, S. Kurnianto, M. Stidham and M. Kanninen (2011). "Mangroves among the most carbon-rich forests in the tropics." Nature Geoscience 4(5): 293-297.

Duarte, C. M., I. J. Losada, I. E. Hendriks, I. Mazarrasa and N. Marbà (2013). "The role of coastal plant communities for climate change mitigation and adaptation." Nature Climate Change 3(11): 961-968.

Ewel, K., R. Twilley and J. Ong (1998). "Different kinds of mangrove forests provide different goods and services." Global Ecology & Biogeography Letters 7(1): 83-94.

Ezcurra, P., E. Ezcurra, P. P. Garcillán, M. T. Costa and O. Aburto-Oropeza (2016). "Coastal landforms and accumulation of mangrove peat increase carbon sequestration and storage." Proceedings of the National Academy of Sciences 113(16): 4404-4409. 237

Fujimoto, K., A. Imaya, R. Tabuchi, S. Kuramoto, H. Utsugi and T. Murofushi (1999). "Belowground carbon storage of Micronesian mangrove forests." Ecological Research 14(4): 409-413.

Gheith, A. M. and M. A. Abou-ouf (1996). "Textural characteristics, mineralogy and fauna in the shore zone sediments at Rabigh and Sharm Al-Kharrar, eastern Red Sea, Saudi Arabia." Marine Scienes-Ceased lssuerg 17(1): 1-2.

Giri, C., E. Ochieng, L. L. Tieszen, Z. Zhu, A. Singh, T. Loveland, J. Masek and N. Duke (2011). "Status and distribution of mangrove forests of the world using earth observation satellite data." Global Ecology and Biogeography 20(1): 154-159.

Jennerjahn, T. C. and V. Ittekkot (2002). "Relevance of mangroves for the production and deposition of organic matter along tropical continental margins." Naturwissenschaften 89(1): 23-30.

Kathiresan, K. (2003). "How do mangrove forests induce sedimentation?" Revista de biologia tropical 51(2): 355-360.

Komiyama, A., J. E. Ong and S. Poungparn (2008). "Allometry, biomass, and productivity of mangrove forests: A review." Aquatic Botany 89(2): 128-137.

Krishnaswamy, S., D. Lal, J. Martin and M. Meybeck (1971). "Geochronology of lake sediments." Earth and Planetary Science Letters 11(1-5): 407-414.

Kristensen, E., S. Bouillon, T. Dittmar and C. Marchand (2008). "Organic carbon dynamics in mangrove ecosystems: a review." Aquatic Botany 89(2): 201-219.

Laffoley, D. and G. D. Grimsditch (2009). The management of natural coastal carbon sinks, Iucn.

Lovelock, C. E., D. R. Cahoon, D. A. Friess, G. R. Guntenspergen, K. W. Krauss, R. Reef, K. Rogers, M. L. Saunders, F. Sidik and A. Swales (2015). "The vulnerability of Indo-Pacific mangrove forests to sea-level rise." Nature.

Naidoo, G. (2006). "Factors contributing to dwarfing in the mangrove Avicennia marina." Annals of botany 97(6): 1095-1101.

Phillips, D. L. and J. W. Gregg (2003). "Source partitioning using stable isotopes: coping with too many sources." Oecologia 136(2): 261-269.

Polidoro, B. A., K. E. Carpenter, L. Collins, N. C. Duke, A. M. Ellison, J. C. Ellison, E. J. Farnsworth, E. S. Fernando, K. Kathiresan and N. E. Koedam (2010). "The loss of species: mangrove extinction risk and geographic areas of global concern." PLoS One 5(4): e10095. 238

Robbins, J. A., D. N. Edgington and A. L. W. Kemp (1978). "Comparative 210Pb, 137Cs, and pollen geochronologies of sediments from Lakes Ontario and Erie." Quaternary Research 10(2): 256-278.

Suratman, M. N. (2008). Carbon sequestration potential of mangroves in southeast Asia. Managing Forest Ecosystems: The Challenge of Climate Change, Springer: 297-315.

Valiela, I., J. L. Bowen and J. K. York (2001). "Mangrove forests: One of the world's threatened major tropical environments." Bioscience 51(10): 807-815.

Van der Werf, G. R., D. C. Morton, R. S. DeFries, J. G. Olivier, P. S. Kasibhatla, R. B. Jackson, G. J. Collatz and J. Randerson (2009). "CO2 emissions from forest loss." Nature Geoscience 2(11): 737-738.

Wentworth, C. K. (1922). "A scale of grade and class terms for clastic sediments." The Journal of Geology 30(5): 377-392.

239

General Discussion

Mangrove trees connect land and sea, with an enormous root system they

create a special ecosystem which provides food and shelter for many organisms,

protect coasts from erosion, stabilize the sediment and consequently prevent the offshore seagrass beds and coral reef from being smoothed and act as carbon sinks.

However, the lack of knowledge about their role led to increasing the annual rate of

tree losses, making them rank amongst the most threatened ecosystems globally.

Further, mangroves are understudied; investigations on mangrove ecosystems were

less than 500 paper y-1 before 2005 according to a search we did in Scopus (1846 to

2015). Globally, published papers on mangroves, seagrass, and salt marshes account for 11-14% of coastal habitats publications, compared to 60% of coral reefs (Duarte,

Dennison et al. 2008). In the Red Sea, mangroves are the only naturally growing trees on shorelines and the first to be described in scientific literature (Saenger 2002); nevertheless, they have been overlooked in the past. We searched the word “Mangrove

Red Sea” on the 26 March 2016 and found 129 and 479 papers about Red Sea mangroves, compared to 38,891 and 14,451 for mangrove (global) in Elsevier's Scopus and Thomson Reuter's Web of Science (WOS) respectively. This corresponds to a range of 0.3 to 3.3% of the total scientific literature about mangroves. In contrast, coral reef literature in the Red Sea ranged from 2.5 to 18.5% of the global literature in WOS and

Scopus respectively (Table. 1).

240

Table 1: Percent of mangrove and coral reef published papers in the RED Sea vs. global.

Ecosystem Papers % WOS mangrove* 38891 mangrove* red-sea 129 0.3 coral reef* 62165 coral reef* red-sea 1576 2.5 Scopus mangrove* 14451 mangrove* red-sea 479 3.3 coral reef* 21475 coral reef* red-sea 3977 18.5 * In databases search is for (mangrove and mangroves)

The Red Sea is spatially and temporally heterogeneous with salinity ranging from 35 to 41 ppt, temperature from 21 to34 °C and chl a from 0.5 to4.0 mg m3

(DiBattista, Roberts et al. 2015). The rainfall is mostly short showers, 10–15 mm y −1 over the sea and an average of 60 mm y in the onshore desert of the Red Sea, −1 whereas evaporation significantly exceeds precipitation and is between 1.4 and 2 m yr (Rasul, Stewart et al. 2015). Nutrient inputs to the Red Sea are dominated by −1 inputs from the Indian Ocean, leading to a gradient of oligotrophication toward the north (Ismael 2015), concurrent with an increased salinity due to high evaporation losses (Talley 2002), while temperature declines from south to north, ultimately imposing the northern biogeographical boundary of mangroves in the northern Red

Sea. Therefore, we aimed to provide the scientific underpinnings of the Red Sea mangrove ecosystems to fill the gaps of knowledge on their growth dynamics under extreme conditions, such as the Red Sea, with the specific target of evaluating selected ecosystem services such as carbon storage, sequestration, and phytoremediation 241

potential. Eventually, this information will shed light on the benefit of conserving

mangroves while maintaining a healthy ecosystem.

In the Red Sea, general information about mangrove distribution is scarce.

Hence, we used aerial imagery to determine changes in mangrove populations over 41

years, finding an increase of approximately 12%. Specifically, the total coverage of Red

Sea mangrove forests in 1972, 2000 and 2013 was 120, 132 and 135 km2, respectively.

The decrease of numbers over time due to urban development, aquaculture, and other impacts was compensated with afforestation projects, including those in Yanbu in

Saudi Arabia and Hirgigo in Eritrea. This stability in the Red Sea is in contrast with a sharp decline globally. Therefore, this result is a milestone in assessing the status of mangroves, not only in the Red Sea as an ecosystem, but also to help balance reports of the global status of mangroves and to understand local coverage and distribution.

Knowledge of the population dynamics is essential in predicting mangrove

forest development and recovery from disturbance (Thi Ha, Duarte et al. 2003).

Nonetheless, Avicennia marina has less than ten phenology reports worldwide. As a

result, we started monthly in situ measurements for new branches, buds, flowers and

propagules. Besides, we measured the height and internodal distance, as the

internodal length of Avicennia marina allows the study of the growth patterns and

reconstruction and calculation of the plastochrone interval, a key property to convert

biological into chronological time (Gill and Tomlinson 1971, Duke and Pinzon 1992).

We found the general population stunted at around 2 meters, with their node production genetically controlled and not varying between five locations with an average of 9.59 node y-1. This resulted in a plastochrone intervals of 38 days. The 242

internodal length varied significantly between locations, resulting in growth

differences possibly reflecting the environmental conditions of each location (i.e. the

limitation in nutrient supply and fresh water inputs). This result was validated by a

direct observation from seedlings growing in a nursery with PI=37.9 days and was

consistent with that inferred from the observation of annual sub-branch production

PI=39.20 days. This 38 days plastochrone interval provides a strong scale to convert

number of internodes into time elapsed, which can be used to develop a deep

understanding of the time needed for mangrove growth, production, and elemental

flux.

Furthermore, bud and flower production initiated at the time of the summer

solstice (20-23 June), when the temperature is highest, but peaked in the autumn,

when atmospheric humidity is highest. This relationship with temperature,

particularly suggests that this species must have a plasticity not only in the Red Sea,

where they grow in their northern limit of the Indo-Pacific with extreme conditions of high temperature but globally, as mangrove biomass and reproductive development are negatively related to latitude and positively related to temperature (Duke 1990,

Alongi 2002). Therefore, the lower growth rates in the Red Sea are unlikely to be attributable to temperature alone as temperatures are high in the region. Actually,

Avicennia marina has the widest distribution between mangroves from (Duke 1991,

Smith 2013, Saintilan, Wilson et al. 2014) tropical latitudes to the temperate zone of

southern Australia (Duke, Ball et al. 1998).

Hence, the reason for the limited growth might be attributed to other factors such as salinity, frequency of tidal influence, soil waterlogging, surface hydrology, soil 243 redox potential and nutrient limitation (Naidoo 2010). Although they can normally live in a saline habitat, both carbon assimilation capacity and growth are reduced as salinity increases (Naidoo 2006). However, in the southern area of the Red Sea, where salinity is above ocean average and temperature is higher than that in the central Red

Sea, but nutrient concentrations are much higher, mangrove trees are much taller than those in the central Red Sea (Mandura, Saifullah et al. 1987, Mandura 1997), pointing at nutrient availability as the primary factor limiting growth in the central Red Sea.

Most mangrove fertilization experiments focused on nitrogen and phosphorus additions, but those exploring iron limitation are scarce. As a result, we conducted a fertilization experiment (N, P and Fe and combinations) and we assessed nutrient status in collected propagules and naturally growing Avicennia marina leaves. Our results provide evidence that the growth of Avicennia marina seedlings in the Central

Red Sea requires environmental Fe inputs, as iron additions alone led to a growth response and increased in chlorophyll a concentration comparable to that in the treatments also receiving nitrogen and phosphorous. Although our experiment showed iron to be the main limiting factor for seedling growth, nitrogen addition significantly enhanced the number of leaves while both phosphorous, and particularly iron, led to increased root development. Yet, the experimental treatment with combined N, P and Fe did not enhance growth beyond that achieved by Fe fertilization alone, indicating iron to be the growth limiting factor according to Liebig’s law of the minimum. Whereas the second experiment aimed at separating the possible effects of iron and the chelant used in combination in the first experiment, which were found to enhance seedling significantly. The results from the second experiment conducted 244 confirmed that it was indeed iron, and not the ligand added in the first experiment

(EDTA), which elicited the growth responses observed. Furthermore, we found naturally growing mangroves in the central Red Sea characterized by low leaf nutrient concentrations (N < 1.5 %, P < 0.09 %, Fe < 0.06) and carbon to nutrient stoichiometric ratios, with an overall mean of 1918 C: 36 N: 1 P: 0.5 Fe, indicative of severe nutrient limitation, particularly P and Fe. Likewise, the propagules were characterized by a low nutrient concentration relative to leaves 930 C: 25 N: 1 P: 0.02 Fe. Particularly, Fe deficiency was severe in the propagules collected to initiate the experiment, which had

P: Fe ratios 10 to 20 fold higher than those of leaves, indicative of extreme Fe deficiency in these propagules. The stoichiometric ratios in A. marina seedlings receiving Fe+N+P additions in our experiments can be considered to represent those of nutrient-sufficient plants 790 C:50 N:1 P:0.1 Fe.

Then we aimed to further explore the mechanisms contributing to their capability in coping with this ultra-oligotrophic environment by estimating the reabsorption rates which were 69%, 72% and 35% for N, P, and Fe respectively.

Further, we estimated leaf production to be 7608 leaf-1 m-2 y-1 and consequently estimated the flux of nutrients to range from 15 to 22 g m-2 y-1 for nitrogen, 0.8 to 1.1 g m-2 y-1 for phosphorous and 0.3 to 1 g m-2 y-1 for iron. Therefore, the leaves shedding imply a significant turnover of nutrients, where the nutrients collected by the roots are brought back to the top layer of the sediments. If we extrapolate the estimated fluxes to the area covered by mangroves in the Red Sea, around 135 km2, the total flux per year is about 2414 metric tons of nitrogen, 139 tons of phosphorous and 98 tons of 245

iron per year. This turnover of nutrients can play a significant role in an oligotrophic

ecosystem.

On the other hand, industrial and urban development in the Saudi Coast of the

Red Sea is increasing the level of contaminants. As a consequence, we tested whether

mangroves can retain heavy metals before shedding leaves, as they did with nutrients,

which would increase their phytoremediation capabilities. Most metals concentrations

remain constant with leaf age. Copper decreases suggesting reabsorption whereas

vanadium and cadmium increase are indicating a detoxifying mechanism.

Furthermore, concentration remaining constant with age, the content increases with

the growth of the leaf, and therefore mangroves partially remove heavy metals from

the deep sediment (roots system) and bring them back to the surface of the sediment

when shedding the leaves. which implies that leaf shedding could potentially

remobilize heavy metals that were buried deep in the sediment. Therefore, although in

the long term the net effect would be a reduction of heavy metals in the sediment

column because accumulation in the shoot, roots, and deep sediment, in the short time

mangroves, could contribute to maintaining high levels of heavy metals in the

sediment surface. The same extrapolation we did for nutrients was applied to heavy

metals to estimate the flux in Red Sea mangrove forests, resulting in 50 kg y-1 for

cadmium, 570 kg y-1 for copper and 74080 kg y-1 for Aluminum.

The second ecosystem service we target was carbon storage and sequestration.

The Kingdom of Saudi Arabia is one of the main producers of oil in the world, and mangrove forests are the only coastal forests in the Kingdom. We assumed that in contrast with other mangrove forests, CO2 capture by mangroves in the Red Sea would 246

be extremely limited due to the arid conditions of this region. Indeed we found

sequestration rates to be low in comparison with other regions, with average values for the last hundred years of 34 g m-2 y-1, which expanded to the total mangroves covered area of the Red Sea would imply 4590 tons of carbon sequestered per year.

This represents 5 % percent of the flux due to leaves.

Figure 1: Flow chart summarizing results obtained from Central Red Sea mangroves. References

Alongi, D. M. (2002). "Present state and future of the world's mangrove forests." Environmental conservation 29(03): 331-349. 247

DiBattista, J. D., M. B. Roberts, J. Bouwmeester, B. W. Bowen, D. J. Coker, D. F. Lozano‐ Cortés, J. Howard Choat, M. R. Gaither, J. P. A. Hobbs and M. T. Khalil (2015). "A review of contemporary patterns of endemism for shallow water reef fauna in the Red Sea." Journal of Biogeography.

Duarte, C. M., W. C. Dennison, R. J. Orth and T. J. Carruthers (2008). "The charisma of coastal ecosystems: addressing the imbalance." Estuaries and coasts 31(2): 233-238.

Duke, N. (1990). "Phenological trends with latitude in the mangrove tree Avicennia marina." The Journal of Ecology: 113-133.

Duke, N. (1991). "A systematic revision of the mangrove genus Avicennia (Avicenniaceae) in Australasia*." Australian Systematic Botany 4(2): 299-324.

Duke, N., M. Ball and J. Ellison (1998). "Factors influencing biodiversity and distributional gradients in mangroves." Global Ecology & Biogeography Letters 7(1): 27-47.

Duke, N. C. and Z. S. M. Pinzon (1992). "Aging Rhizophora seedlings from leaf scar nodes: a technique for studying recruitment and growth in mangrove forests." Biotropica: 173-186.

Gill, A. M. and P. B. Tomlinson (1971). "Studies on the growth of red mangrove (Rhizophora mangle L.) 3. Phenology of the shoot." Biotropica: 109-124.

Ismael, A. A. (2015). Phytoplankton of the Red Sea. The Red Sea, Springer: 567-583.

Mandura, A. (1997). "A mangrove stand under sewage pollution stress: Red Sea." Mangroves and Salt marshes 1(4): 255-262.

Mandura, A., S. Saifullah and A. Khafaji (1987). "Mangrove Ecosystem of Southern Red Sea Coast of Saudi Arabia." Proc.Saudi Biol.Soc 10: 165-193.

Naidoo, G. (2006). "Factors contributing to dwarfing in the mangrove Avicennia marina." Annals of botany 97(6): 1095-1101.

Naidoo, G. (2010). "Ecophysiological differences between fringe and dwarf Avicennia marina mangroves." Trees 24(4): 667-673.

Rasul, N. M., I. C. Stewart and Z. A. Nawab (2015). Introduction to the Red Sea: Its Origin, Structure, and Environment. The Red Sea, Springer: 1-28.

Saenger, P. (2002). Physico-chemical Factors and Mangrove Performance. Mangrove Ecology, Silviculture and Conservation, Springer: 101-146.

Saintilan, N., N. C. Wilson, K. Rogers, A. Rajkaran and K. W. Krauss (2014). "Mangrove expansion and salt marsh decline at mangrove poleward limits." Global change biology 20(1): 147-157.

Smith, T. J. (2013). Forest Structure 248

Tropical Mangrove Ecosystems, American Geophysical Union: 101-136.

Talley, L. D. (2002). "Salinity patterns in the ocean." Encyclopedia of global change. Volume: the earth system: physical and chemical dimensions of global environmental change (eds MacCracken MC, Perry JS): 629-640.

Thi Ha, H., C. M. Duarte, N. H. Tri, J. Terrados and J. Borum (2003). "Growth and population dynamics during early stages of the mangrove Kandelia candel in Halong Bay, North Viet Nam." Estuarine, Coastal and Shelf Science 58(3): 435-444.