Faculty of Agricultural Science, Nutritional Sciences and Environmental Management Professorship of Agricultural Production Economics of the Justus-Liebig University Giessen Socio-economic Characteristics and Land Allocation on Tidal Swampland Agriculture in Indonesia Inaugural-dissertation to obtain the doctoral degree (Dr.agr.) in the Faculty of Agricultural Science, Nutritional Sciences, and Environmental Management Justus Liebig University Giessen Submitted by Ahmad Yousuf Kurniawan from Indonesia Giessen, 2024 With permission from the Faculty of Agricultural Sciences, Nutritional Sciences, and Environmental Management Justus-Liebig university of Giessen Examination committee: 1. Referee: Prof. Dr. Joachim Aurbacher 2. Referee: Prof. Dr. Bernd Honermeier Examiner: Prof. Dr. Martin Petrick Examiner: Prof. Dr. Christian Herzig Chairman: Prof. Dr. Jan Siemens Date of Examination: 20.12.2023 ACKNOWLEDGEMENTS First of all, I am grateful to Almighty Allah SWT who blessing me to complete this research. I am most grateful to my advisor, Prof. Dr. Joachim Aurbacher, for his continuous guidance and suggestions for my research and also kindly support throughout the time of my study. I am thankful to my co-advisor, Prof. Dr. Bernd Honermeier for his valuable comments and suggestions and also to Prof. Dr. Martin Petrick and Prof. Dr. Christian Herzig for their precious comments and suggestions in my thesis. I am also indebted to all the lecturers, staffs and technicians in the Faculty of Agricultural Science, Nutritional Sciences and Environmental Management, and also the International Studies Centre (ISC) and the Promotion staffs for their help towards the successful completion of my study. This research and my study in JLU were supported by DAAD. The support to continue my study from Department of Agribusiness, Faculty of Agriculture, Lambung Mangkurat University in South Kalimantan, Indonesia is also appreciated. I am heartfelt thank Prof. Dr. S. Bauer for his helping in my early years of studying in Germany. I am heartfelt thank to fellow friends in DAAD Program in Agricultural Economics and Related Sciences in JLU for their help and warm kindness, also all Indonesian students and diasporas in Giessen-Frankfurt for sharing their joyful moments. Most importantly, I would like to thank my father and mother, my foster father and mother, my brothers, and also my father and mother in law for their continuing prays and supports. Finally, I would like to devote my special gratitude to my dear wife, Ir. Hesti Wijayanti, Ph.D, who always gives me love, and strength with her constant support and also my daughters, Danish Yusfia Rahmi and Diandra Yusfia Rizki who always cheers me with happiness during my study. Ahmad Yousuf Kurniawan 2024 TABLE OF CONTENTS i Table of Contents Table of Contents i List of Tables iv List of Figures vi Abbreviations vii 1 INTRODUCTION 1 1.1 Background 1 1.2 Problem statement 3 1.3 Research objectives 7 1.4 Significance of the study 7 2 SWAMPLAND FOR AGRICULTURE IN INDONESIA 9 2.1 Swampland distribution 9 2.2 Swampland characteristics and problems 11 2.3 Swampland used for agriculture in Indonesia 12 2.4 Tidal Swampland 15 2.4.1 The nature of tidal swampland 15 2.4.2 Tidal swampland for agriculture 17 2.4.3 Agricultural practice in tidal swampland ecosystem 19 2.5 Swampland for agriculture: some countries’ experience 22 2.5.1 Papua New Guinea 22 2.5.2 West Africa 23 2.5.3 Rwanda 24 2.5.4 Vietnam 25 2.5.5 Thailand 26 2.5.6 India 27 2.5.7 USA 28 2.5.8 Lesson to be learned 29 3 RESEARCH DESIGN 30 3.1 Location of the study area 30 3.2 Rationale for choosing the study area 30 3.3 Data collection 31 3.3.1 Primary data collection and sampling design 32 ii TABLES OF CONTENTS 3.3.2 Secondary data collection 33 3.4 Data analysis 33 3.4.1 Descriptive analysis 33 3.4.2 Household modelling analysis 33 4 REVIEW OF STUDY AREA CONDITIONS 35 4.1 Geography and climatology 35 4.1.1 Geography 35 4.1.2 Temperature and climate 37 4.2 Population and labor 38 4.3 Agricultural Production 40 4.4 Institutions, social services, and infrastructures 43 4.4.1 Agricultural extension service 44 4.4.2 Education and health facilities 44 4.4.3 Transportation and irrigation facilities 45 5 SOCIO-ECONOMIC ANALYSIS OF FARM-HOUSEHOLDS IN TIDAL SWAMPLAND 46 5.1 Household characteristics of farm households surveyed 46 5.1.1 Household size 46 5.1.2 Dependency ratio 47 5.1.3 Age of the head of household 47 5.1.4 Ethnicity and origin 47 5.1.5 Education and farming experience 48 5.1.6 Farmer affiliation 49 5.1.7 Farmer mobility 50 5.2 Land resources 51 5.2.1 Land holding status 51 5.2.2 Soil fertility 53 5.3 Household agricultural production 54 5.3.1 Crops production 55 5.3.2 Perennial crops 57 5.3.3 Livestock and fishery 58 5.3.4 Plantation 59 TABLE OF CONTENTS iii 5.4 Household Income and Expenditures 59 5.4.1 Household income 59 5.4.2 Farm household expenditure 67 6 MATHEMATICAL MODELLING OF A TIDAL SWAMPLAND FARM HOUSEHOLD 68 6.1 Mathematical programming model 68 6.2 Model description 69 6.2.1 Objective function 72 6.2.2 Resource Constraints 73 6.2.3 Prediction of simulated crops price 76 6.3 The modelling results 77 6.3.1 The model’s base solution (optimum) result 78 6.3.2 Model validation 79 6.3.3 Gross margin simulation with crop price fluctuations 80 6.4 Discussion of the modelling results 84 6.4.1 The model’s base solution (optimum) result 84 6.4.2 Gross margin simulation with crop price fluctuations 88 6.5 Model limitations 89 7 CONCLUSIONS AND RECOMMENDATIONS 91 7.1 Conclusions 91 7.1.1 Policy recommendation 91 7.1.2 Recommendations for further research 92 8 SUMMARY 93 8.1 Research background and framework 93 8.2 Research findings from the field study 94 8.3 Research findings from the modelling results 95 ABSTRACT 96 ZUSAMMENFASSUNG 97 REFERENCES 99 APPENDICES 113 iv LIST OF TABLES List of Tables Table 2.1 : The swampland area in the major islands of Indonesia ……… 10 Table 4.1 : Land use in Barito Kuala …………………………………….. 42 Table 5.1 : Household characteristics of the 200 surveyed household …... 46 Table 5.2 : Education and farming experience of the sample farm households …………………………………………………… 48 Table 5.3 : Selected farmer relation of the sample farm households …….. 49 Table 5.4 : Number of the farmer who has activities outside their home village ………………………………………………………… 50 Table 5.5 : Farmer appraisal on soil fertility ……………………………… 54 Table 5.6 : The average of rice productivity (ton/ha) …………………….. 55 Table 5.7 : Labor needed to cultivate paddy rice (man-hours/ha) ………… 56 Table 5.8 : Revenue Cost Ratio (RCR) of rice …………………………… 56 Table 5.9 : The average number of productive trees per household ……… 57 Table 5.10 : The BCR, NPV, and IRR of perennial crops in swampland area during a 16 years period with the interest rate of 10 %...... 58 Table 5.11 : The average number of livestock per household ……………… 58 Table 5.12 : The land area of smallholding plantation in Barito Kuala ……. 59 Table 5.13 : The average of main crops yields and its price in the different typology ………………………………………………………. 60 Table 5.14 : The farm production cost of the main corps (IDR per ha) ……. 62 Table 5.15 : The income from other crops (palawija) (in million IDR) ……. 62 Table 5.16 : The income from livestock (in million IDR) …………………. 63 Table 5.17 : The annual off-farm and non-farm inside the farmer home village (in million IDR) ………………………………………. 63 Table 5.18 : Average and share income of the farm household samples …… 64 Table 5.19 : Gini ratios across tidal swamp typology ……………………… 66 Table 5.20 : The share of different cost item to the household total expenditure (%) ………………………………………………. 68 Table 6.1 : The list of endogenous and exogenous variables in the mathematical programming model ……………………….. 71 Table 6.2 : Table price of rice, orange, and coconut (Million IDR/ton) in South Kalimantan …………………………………………….. 76 Table 6.3 : The actual output price correlation of the main crops and perennial crops in South Kalimantan …………………………. 77 Table 6.4 : Base solution (optimum) level ………………………………... 78 Table 6.5 : Base solution of labor use in various swamps typologies …… 79 Table 6.6 : Model results and their PAD …………………………………. 80 Table 6.7 : The statistical parameters of simulated crops price LIST OF TABLES v (in million IDR/ton) …………………………………………... 81 Table 6.8 : The simulated price correlation of the main crops and Perennial crops in South Kalimantan …………………….. 81 Table 6.9 : The statistical parameters of simulated gross margin (in million IDR) for different typology ………………………. 83 Table 6.10 : The percentage of fluctuation of crops prices and gross margin ………………………………………………………... 84 vi LIST OF FIGURES List of Figures Figure 1.1 : Rice production and consumption in Indonesia, 1990-2017 … 3 Figure 1.2 : Rice import in Indonesia, 1990-2020 ………………………… 4 Figure 2.1 : Swampland distribution in Indonesia ………………………… 9 Figure 2.2 : Swampland typology based on water prevailing in the field …. 17 Figure 2.3 : One-way water system ……………………………………….. 20 Figure 2.4 : The sorjan system ……………………………………………. 22 Figure 3.1 : Map of the study area ………………………………………… 31 Figure 3.2 : Data collection ……………………………………………….. 32 Figure 4.1 Land use in South Kalimantan, 2011 ………………………… 36 Figure 4.2 : Rainfall in South Kalimantan, 1978-2012 …………………… 38 Figure 4.3 : Population structure by age group in South Kalimantan, 2012 39 Figure 4.4 : The percentage of agricultural area by type in South Kalimantan, 2011 ……………………………………… 40 Figure 4.5 : The harvested area of main and secondary crops, 2011 ……… 41 Figure 4.6 : Production trend of selected crops in South Kalimantan …….. 41 Figure 4.7 : Productivity of selected crops in South Kalimantan …………. 42 Figure 4.8 : Rice production in Barito Kuala 1995-2012 …………………. 43 Figure 4.9 : Secondary crops production in Barito Kuala 1995-2012 …….. 43 Figure 5.1 : The distribution of landholding in the study area ……………. 52 Figure 5.2 : The average size of landownership and landholding of the respondents ………………………………………………. 53 Figure 5.3 : The Lorentz curve of household income inequality in different tidal swamp typology ………………………………. 66 Figure 5.4 : Household expenditure by cost items ………………………… 67 Figure 6.1 : Analytical framework of the mathematical model …………… 70 Figure 6.2 : Steps of determination the simulated prediction crops price …. 77 Figure 6.3 : Histograms of price change of the crops …………………….. 82 Figure 6.4 : Gross margin of farmers for each typology of swampland …… 83 ABBREVIATION vii Abbreviations ADB Asian Development Bank AIAT Assessment Institute for Agricultural Technology BCR Benefit-Cost Ratio CBS Central Bureau of Statistic (Badan Pusat Statistik) BULOG National Logistics Agency (Badan Urusan Logistik) CIA Central Intelligent Agency of USA DEA Data Envelopment Analysis FAO Food and Agriculture Organization FFS Farmer Field School GDP Gross Domestic Product GIZ Deutsche Gesselshaft für Internationale Zusammenarbeit GoI Government of Indonesia HYV High Yielding Variety of rice. IAARD Indonesia Agency for Agricultural Research and Development IDR Indonesian Rupiah Inpara Inbreed Swampland Rice IRR Internal Rate of Return IRRI International Rice Research Institute IR-42 Inbreed Rice released by IRRI in 1970s ISARI Indonesian Swampland Agriculture Research Institute (Balittra) MoA Ministry of Agriculture MRP Mega Rice Project NPV Net Present Value OECD Organization for Economic Co-operation and Development PNG Papua New Guinea RCR Revenue-Cost Ratio RIDS Rice Indonesia Data Service RJPPP Long Term Agriculture Development Plan (Rencana Jangka Panjang Pembangunan Pertanian) UN United Nations UNDP United Nation Development Project USDA United State Department of Agriculture VEPA Vietnam Environmental Protection Agency WB World Bank INTRODUCTION 1 1 INTRODUCTION 1.1 Background Indonesia is an archipelagic country in Southeast Asia located between the Indian and Pacific oceans. This archipelago has 13,700 islands that expand 5,100 kilometers east to west and 1,931 kilometers north to south (MAPZONE, 2003). The land area is 1.91 million km2, the marine territory is 3.26 km2 (OECD, 2012a), and the marine exclusive economic zone (EEZ) is 2.9 million km2 (CBS, 2012a). The population was 237.64 million in 2010, with a population density of 131.18 people per km2 and a 1.49 % growth rate (1990-2010) (CBS, 2012a). Meanwhile the population increased to 270.20 million in the next ten year (2020), with a population density of 141 people per km2 and a 1.25 % growth rate (CBS, 2020). The Dutch East India Company (1602-1800), the Netherlands East Indies (1800- 1942), and Japan colonized the Indonesian archipelago (1942-1945). Indonesia declared independence after Japan surrendered in 1945, but it took four years for the Netherlands to agree to a transfer of sovereignty in 1949, following some violence and negotiations mediated by the United Nations (UN) (RICKLEFS, 2001). The economy had achieved remarkable rapid growth, macroeconomic stability, and steadily declining poverty by the mid-1960s. Between 1966 and 1996, the average growth rate of the Gross National Product (GNP) per capita was more than 5 %, and poverty fell from 60 % to 11 %. (DARYANTO, 1999). Indonesian GNP reached $2,530 in 2010 and is expected to reach $4,140 by 2021 (MACROTRENDS, 2022a). During the 1997-1998 economic crisis, Indonesia experienced a 13.7 % economic contraction, the highest inflation rate at 78.1 percent, and an increase in unemployment to 17.1 % and poverty from 17.1 % in 1996 to 24.1 % in 1999. (DARYANTO, 1999, SURYAHADI et al., 2012). Economic growth has gradually increased since the regime changed and several improvements in politics and economic policies were implemented. Between 2000 and 2010, the average real GDP growth rate was 5.2 %, with a 4.0 % increase in real GDP per capita (ELIAS & NOONE, 2011). Furthermore, it increased to 4.6 % between 2011 and 2020 (MACROTRENDS, 2022b). According to UNDP (2010), unemployment fell from 11 % in 2005 to slightly more than 8 % in 2009. It was reduced by nearly half to 4.28 % in 2020 (MACROTRENDS, 2022c). While poverty rates remain high, they are gradually decreasing. Meanwhile, the Indonesian Human Development Index (HDI) increased by an average of 1.4 % per year between 2002 and 2008. And it continued to rise at 0.76 % annual rate between 2010 and 2021 (CBS, 2021). Indonesia recently surpassed China, Japan, and South Korea to become East Asia's fourth-largest economy (ELIAS & NOONE, 2011). It is one of the world's emerging 2 INTRODUCTION market economies and a member of the G-20 major economies. In 2011, the rate of Gross Domestic Product (GDP) growth was 6.2 % (CIA, 2016; OECD, 2012b), and GDP per capita was US$ 3,643. Industry contributes the most to GDP (47.2 %), followed by services (38.1 %) and agriculture (14.7 %) (CBS, 2012). The GDP was US$ 3,873 in 2020 (MACROTRENDS, 2022b), however, services became the biggest contributor to GDP (42.18 %), followed by industries (40.48 %), and agriculture (13.7 %) (CBS, 2021). Furthermore, Indonesia achieved poverty reduction in 2019 by lowering the poverty rate by more than half since 1999, to 9.4 %. (WORLD BANK, 2020). Agricultural progress in Indonesia from 1966 to 1996 was a success story. The implementation of Green Revolution technology has significantly increased productivity. According to AKIYAMA (2004), agricultural growth was 3.7 %, with land productivity accounting for 90 % of the increase. This outstanding performance significantly contributed to the achievement of Indonesian development goals, including food security, low and stable prices, employment opportunities, and foreign earnings/savings (DARYANTO, 1999). Food production was impacted by the 1997 and 1998 economic crises and the El Nino weather pattern. El Nino caused widespread crop failure and crop delays. In 1997, rice production fell by 4 %, and in 1998, it fell by 8 %. This decline was caused in part by an increase in food imports and the conversion of secondary food crop use from livestock feed to human consumption (DARYANTO, 1999). Despite the rapid development of industry and services, agriculture remains an important part of the economy. Even though contributes only for 13.6 % of total GDP, agriculture employs 38.9 % of the labor force (CIA, 2016). Furthermore, SURYAHADI & HADIWIDJAYA (2011) discovered that agriculture contributes to poverty reduction in rural areas. This is significant because nearly half of the population lives in rural areas (ELIAS & NOONE, 2011). Rice is one of the agricultural commodities that has become a source of concern in Indonesia. Rice is a commodity with not only an economic but also a political and social dimension (MASTUR et al., 2022). If this product is in short supply, it will have an impact on the social-economic and political stability of the community (SIDIK, 2004). Environmentally, Indonesia is severely impacted by heavy monsoon rain during the wet season and relatively little rain during the other periods, making it difficult to cultivate other staple crops. Food security and national rice self-sufficiency remain top priorities for the Indonesian government. According to CBS (2012), Indonesia has the 7th highest per capita rice consumption rate in the world, at 133 kilograms per person. The Indonesian government also estimates that rice accounts for roughly half of its INTRODUCTION 3 people's daily calorie and protein requirements, respectively. Domestic rice supply could represent food security in this country of 270.2 million people (CBS, 2021). However, according to the USDA (2012) and STATISTA (2022) total rice consumption has risen faster than total rice production in recent years (1990 – 2013), as the growth rate of national rice area and yield has lowered, and then the consumption has slightly decreased since 2014, but appears to be on the rise in the next following years (Figure 1.1). As a result, providing enough rice as a staple food remains a major challenge in Indonesia. Figure 1.1 Rice production and consumption in Indonesia, 1990-2017 Source: USDA (2012) and STATISTA (2022) In 2013, the government implemented the long-term agricultural development plan (RJPPP) for 2013-2045. Based on the identification of the picture and challenges of Indonesian agriculture up to 2045, the concept provides a clear and comprehensive direction for agriculture and related sector development, allowing agricultural problems to be anticipated (RIDS, 2012). However, increasing farmers' prosperity by relying solely on rice as a source of income is challenging due to high input costs, low output costs, an inefficient trade system, and environmental issues. As a result, farmers' enthusiasm for rice cultivation has declined. Moreover, the low productivity of rice farming has triggered farmers to seek for alternative farming schemes that can increase their income (ADAM et al., 2013). High yielding intercropped alternative crops with rice should be introduced as an additional source of income (WILDAYANA et al., 2016). 1.2 Problem statement Rice is one of Asia's most important cereal crops. According to WU et al. (2010), Asia's paddy rice fields accounted for more than 90 % of the total global rice 0.00 10000.00 20000.00 30000.00 40000.00 50000.00 60000.00 M ill ed R ic e ( 10 00 t o n ) Year Production Consumption 4 INTRODUCTION cultivated area. The major rice-producing countries in Asia accounted for more than half of the world's population. Rice demand will rise over time, as DOBERMANN (2012) estimated that an additional 1 billion people will require 100 million tons of rice. Indonesia, the world's fourth most populous country after China, India, and the United States (MASTUR et al., 2022), is more vulnerable to rice shortages since people rely on rice for calories and protein. Rice production is insufficient to meet local demand. Between 2002 and 2007, the average rice consumption was 27.83 million metric tons, with annual consumption per capita of 127.67 kg (MUTTAQIN & MARTIANTO, 2009). It increased to 134 kg per capita per year between 2009 and 2013 (FAOSTAT, 2007). From 2014 to 2021, it decreased to only 127 kg per capita per year (STATISTA, 2022). Since 1992, Indonesia has imported rice, with the average amount increasing year after year. Indonesia imported 1478.35 million tons of rice per year between 1980 and 1999 (USDA, 2012). Between 1990 and 2020, imports fluctuated dramatically (Figure 1.2). Imports have been increased in recent years to compensate for the decline in rice production caused by El Nino. Climate change causes more frequent occurrences of abiotic stresses for rice such as drought, flood, salinity, and more frequent pests and diseases attack (SASMITA & NUGRAHA, 2020). This condition, however, has caused concern in the global rice market. According to CBS (2012b), imports reached 2.75 million metric tons in 2011. Over the last ten years, Indonesia has continued to require imports, although in varying amount (CBS, 2021). Figure 1.2 Rice import in Indonesia, 1990-2020. Sources: FAOstat (2007) and CBS (2021) Furthermore, while population growth continues, rice productivity remains stagnant. It has only increased from approximately 4.38 tons per hectare in 1993 to 4.98 tons per hectare in 2011 (CBS, 2012b). It was only slightly increased to 5.11 - 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 4,000,000 4,500,000 5,000,000 1 9 90 1 9 91 1 9 92 1 9 93 1 9 94 1 9 95 1 9 96 1 9 97 1 9 98 1 9 99 2 0 00 2 0 01 2 0 02 2 0 03 2 0 04 2 0 05 2 0 06 2 0 07 2 0 08 2 0 09 2 0 10 2 0 11 2 0 12 2 0 13 2 0 14 2 0 15 2 0 16 2 0 17 2 0 18 2 0 19 2 0 20 Im p o rt ( to n n es ) Year INTRODUCTION 5 tons per hectare in 2020 (CBS, 2021). This slowdown in productivity improvement (see Table 1.1) is due to a number of factors, including the near completion of modern variety spread, declining fertilizer marginal productivity, a less favorable price environment, and a reduction in irrigation investment (PASANDARAN & ZULIASRI, 2001). As a result, increasing production is one solution to the problem. Increasing rice production, however, presents some challenges. DOBERMANN (2012) noted that some megatrends are already emerging in the rice sector, such as land scarcity and rising input costs, necessitating an increase in productivity to improve labor, water, fertilizer, and energy efficiency. Table 1.1 The percentage change (%) of harvested area, yields, and productivity of paddy (wetland and dryland), 1970-2000 Region Item 1970– 1975 1975– 1980 1980– 1985 1985– 1990 1990– 1995 1995– 2000 Java Harvested Area (%) 1.58 0.53 2.1 0.44 0.22 0.96 Yield (%) 3.42 6.06 5.63 2.33 0.71 0.70 Productivity (%) 1.81 5.50 3.45 1.88 0.49 -0.26 Outside Java Harvested Area (%) 1.33 1.93 1.71 2.02 3.23 -0.34 Yield (%) 3.96 5.45 5.68 3.98 3.7 0.67 Productivity (%) 2.60 3.45 3.91 1.93 0.46 1.01 Indonesia Harvested Area (%) 1.47 1.17 1.92 1.18 1.72 0.29 Yield (%) 3.63 5.83 5.65 2.97 1.94 0.69 Productivity (%) 2.13 4.6 3.66 1.76 0.22 0.27 Sources: PASANDARAN & ZULIASRI (2001) Another issue in Indonesia is land scarcity caused by land conversion from agricultural to non-agricultural use. Massive land conversion has occurred. It has a significant impact on food production, particularly in paddy fields (FIRMAN, 2000). More than 106,000 hectares of land were estimated to have been converted in 1991- 1993, including 58,000 hectares (54.7 %) of residential areas, 16,452 hectares (15.5 %) of industrial land, 5,210 hectares (4.9 %) of offices, and 26,774 hectares (25.3 %) of other urban land uses (MINISTRY OF ENVIRONMENT, 1997). During this decade, land conversion has reached approximately 100,000 ha per year. As reported in SASMITA & NUGRAHA (2020), the average arable land conversion to non-agricultural use was 96,512 ha per year. Approximately 79.3 % of that has 6 INTRODUCTION occurred on Java Island (APRIYANA, 2011), which produces 60 % of national rice (HUSSAIN et al., 2006). This condition is also intensified by land erosion and sedimentation (LUKAS, 2014). However, due to high population and food demand pressures, swampland utilization has emerged as an alternative method of developing agriculture and plantations. Indonesia has approximately 6 million hectares of tropical swamplands that can be used for agriculture (NOOR, 2004). Among them, 657,546 hectares were cultivated (SETIOBUDI & FAGI, 2009). South Kalimantan has a total area of 17,828 hectares, with acid-sulfate soil accounting for approximately 80 % of it. If managed properly, this area has economic value (AMALI et al., 2003). It can also be used for rice cultivation, which will help improve national rice production. Furthermore, Indonesia has the potential to play a significant role as a global rice supplier. The use of tidal swampland is an alternative method of increasing rice production and farmer income while also strengthening farmers' household economies and food security (ALIHAMSYAH, 2004). However, one of the major issues in rice cultivation is low productivity, which may be due to inefficient input use and land degradation. It was also due to the lack of some important factors, such as water management, better seeds, fertilizers, and farmers’ education and training (ADAM et al., 2013). As a result, research on swampland for farming, particularly its marginal characteristics, was frequently published (VAN DEN EELAART, 1981, DENT & PONS, 1995, NOOR, 2007). Farmers must adapt to address this issue. Previous research has looked into the adaptation strategies used by farmers and other stakeholders to deal with swampland farming. MEGAWATY et al. (2012) and NOOR & SOSIAWAN (2020) proposed a water management technique. DARSANI & ANNISA (2018) optimized sulfated soil treatment in swampland areas. SAIDY & AZIZ (2009) investigated the Indonesian government's proposed strategy for dealing with sea-level rise in tidal swampland. In addition to studies on the characteristics of tidal swampland and management strategies, the study of socio-economic aspects of swampland farming is intriguing and should be pursued further due to its contribution to agricultural productivity. In other words, how farmers adapt to increased income is an interesting issue that needs to be investigated. So far, research on the socio-economic analysis of swampland in Indonesia has concentrated on macro-management (JOOSTEN & CLARKE, 2002). Meanwhile, socio-economic studies of farmers' adaptation to swampland conditions, particularly tidal swampland conditions, are still in the early stages. INTRODUCTION 7 1.3 Research objectives Using the afore mentioned cases as the research foundation, the overall objective of this study is to investigate the socio-economic and land allocation strategies used by farmers to cope with marginal conditions in the swampland area of South Kalimantan, Indonesia. This can be accomplished by focusing on the following specific objectives: 1. To describe the socio-economic characteristics of a swampland area in South Kalimantan, Indonesia. 2. To determine a model of farmer households which has optimum gross margin under restricted resources in the tidal swampland area of South Kalimantan, Indonesia. 3. To simulate the fluctuation of gross margin as a result of crop price fluctuations in the tidal swampland area of South Kalimantan, Indonesia. 1.4 Significance of the study Rice is one of the agricultural commodities that has become a source of concern in Indonesia. Rice is a commodity with both economic and political and social implications. Furthermore, while population growth continues, rice productivity remains stagnant. Another issue in Indonesia is land scarcity caused by land conversion from agricultural to non-agricultural use, particularly on Java Island, where 60 % of the rice is produced (SASMITA & NUGRAHA, 2020) Due to the high population and food demand, swampland utilization is an alternative way to develop agriculture and plantations. Indonesia has a lot of tropical swamplands that can be used for farming. However, increasing farmers' prosperity by relying solely on rice as a source of income is difficult due to high input prices, low output prices, an inefficient trade system, and environmental issues. As a result, farmers are less interested in growing rice as their primary crop. As a consequence, high-economic alternative crops intercropped with rice should be introduced as an alternative source of income (PERRY, 1985). Furthermore, it is important to understand how land typology affects farm household land allocation. Therefore, policies can be tailored to the complexities of swamp farm households. Thus, this study was carried out in order to contribute to this issue and provide recommendations for managing land in the tidal swampland area. This study also investigates the impact of crop price changes on farmer households' income. 8 INTRODUCTION Moreover, this research was conducted on the Indonesian outlier of Kalimantan Island, which has the most marginal land. Because the majority of the development is concentrated on Java Island, this will contribute to further agricultural development as well as an effort to disperse national development. Rice production will be ensured by the establishment of a food production center outside of arable land on Java and Bali Island. SWAMPLAND FOR AGRICULTURE IN INDONESIA 9 2 SWAMPLAND FOR AGRICULTURE IN INDONESIA This chapter discusses swampland in general, including the definition of swampland agriculture, swamp characteristics, swamp utilization for agriculture, and a variety of socio-economic and environmental issues. Following that, a discussion of tidal swampland is presented, covering definition and characteristics, agricultural use, and agricultural farming practice. 2.1 Swampland distribution A swamp is a forested wetland that is often located along large rivers and is critically dependent on natural water level fluxes (KEDDY, 2010, HUGHES, 2003). Several large swamps can be found along major rivers such as the Amazon, Mississippi, and Congo (KEDDY et al., 2009), as well as on the shores of large lakes (WILCOX et al., 2007). Water can be saltwater, brackish water, or freshwater. According to the USGS (United States Geological Survey), swamps are a type of forested lowland, spongy land that is generally saturated with water and covered with trees and aquatic vegetation that can tolerate periodic inundation. Figure 2.1. Swampland distribution in Indonesia Source: SUSANTO (2003) Swampland is a lowland area that is waterlogged all or nearly all of the year (SUBAGYO, 2006). The majority of the swamps are formed by a large river, and they are critically dependent on natural water level fluctuations, such as sea tide and rainfall. 10 SWAMPLAND FOR AGRICULTURE IN INDONESIA According to VAN DEN EELAART (2014), swampland accounts for roughly 36 % of Indonesia's total coastal area. Agriculture accounts for approximately 15 % of the total land area. The majority of this reclaimed land is used for rice farming in the provinces of South Sumatera, Jambi, West Kalimantan, and South Kalimantan. Other undeveloped swamp areas can be found in Papua New Guinea (Irian Jaya). Figure 2.1 depicts the distribution of swamps across the Indonesian archipelago. Several surveys, including NEDECO/EUROCONSULT-BIEC (1984), SUBAGYO et al. (1990), NUGROHO et al. (1991), and PUSLITTANAK (2000) have been conducted to estimate the total of swampland areas in Indonesia based on swamp classification and coverage to present a detailed estimation. According to their findings, the swampland area of four major Indonesian islands reached 33.41 million hectares, with 13.28 million hectares of lowland (monotonous) swampland and 20.13 million hectares of tidal swampland. Tidal swampland is divided into five land typologies based on soil formation: 10.90 million ha of peatland; 2.07 million ha of potential swampland; 4.34 million ha of potential acid sulfate; 2.37 million ha of actual acid sulfate; and 0.44 million ha of brackish. The Ministry of Public Works and Housing reported the latest data in Table 2.1. As has been shown, the largest swampland is on Kalimantan (Borneo) Island, which contains 35.06 % of the total swampland. Thus, the focus of this research is on Kalimantan Island. Because swamp characteristics vary across the islands, the following discussion will focus on swampland in South Kalimantan. The Government of Indonesia (GoI) has developed 1.80 million hectares of swampland, 49.44 % of which is in Kalimantan (Table 2.1). Meanwhile, the public and private sectors have developed over 2.4 million hectares (MAAS, 2003, NOOR & JUMBERI, 2005). More than 4.2 million hectares of swampland have been developed for agriculture. Table 2.1 The swampland area in the major islands of Indonesia Major Islands Total Swampland (million ha) Total developed for agricultural purposes by government (million ha) Tidal Monotonous Total Tidal Monotonous Total Sumatra 6.60 2.78 9.37 0.69 0.11 0.80 Kalimantan 8.13 3.58 11.71 0.69 0.19 0.89 Sulawesi 1.15 0.64 1.80 0.07 0.01 0.08 Papua 4.22 6.31 10.52 - 0.02 0.02 Total 20.10 13.30 33.40 1.46 0.34 1.80 Source: MINISTRY OF PUBLIC WORK (2009) SWAMPLAND FOR AGRICULTURE IN INDONESIA 11 In 2009, the swamp developed for agriculture accounted for 52.11 % of total agricultural wetland of 8.06 million hectares, or 16.62 % of the 25.27 million hectares of total agricultural land in Indonesia (MINISTRY OF AGRICULTURE, 2014). Thus, swamp agriculture is essential to Indonesian agriculture. 2.2 Swampland characteristics and problems Swamp ecosystems have poor characteristics and are vulnerable to natural change (drought, fire, and flooding) as well as management failure (reclamation, opening, and improper cultivation). Swampland characteristics vary across Kalimantan. Some swamp areas have a peat layer with varying thicknesses and peat maturity, others have tidal problems, and still others have acidity problems. The soil itself has a number of issues related to its formation. Swamp peatland is prone to irreversible drying, subsidence, and nutrient deficiency. The dried peat has a hydrophobic surface (unable to bind the water and nutrients optimally). Acid sulfate soils can be found in some areas. This soil is distinguished by its low pH and the presence of a sulfuric horizon with overlying sulfide materials, primarily pyrite (FeS2) 1 (DENT & PONS, 1995, SHAMSHUDDIN et al., 2014). The soil becomes acidic (pH level 2-3) when the pyrite layer oxidizes, and the saturation of iron (Fe2+) and aluminum (Al3+) increases (NOOR & JUMBERI, 2005). This noxious water could seep into the drainage system and end up in the river. The soil is dominated by the low activity of clay, which has a weak structure and is prone to erosion (LEIWAKABESSY, 1989). If proper precautions are not taken, swamp opening will result in over draining. Because of highly acidic conditions (pH decrease to 2-3), nutrient deficiency, and an increase in Al3+, Fe2+, H2S, CO2, and organic acids, soil fertility will decrease when the groundwater level below the pyrite layer and the peat becomes irreversibly dried and hydrophobic (NOOR & JUMBERI, 2005). However, by adding lime or basalt, replenishing organic matter, and managing water tables to increase soil pH, the land can still be used productively for rice and other crops (SHAMSHUDDIN et al., 2014). Based on these facts, swamp reclamation, land preparation, and farming techniques should be conducted properly. 1 Pyrite (FeS2) or iron sulfide is a sulfide mineral that is found below the top soil in swampland. It is formed by marine sedimentation a thousand years ago in brackish water that contains saturated sulfate compound (SO4). If it is exposed to the air (O2) and oxidized, it will create hydrogen sulfate which causes higher acidity on soil and water (AGUS & SUBIKSA, 2008). 12 SWAMPLAND FOR AGRICULTURE IN INDONESIA 2.3 Swampland used for agriculture in Indonesia Swamp agriculture has a long history among Indonesian farmers (see Appendix 1). Archeological evidence suggests that in the 13th century, local people in the Pawan basin, West Kalimantan, opened swamp areas for agriculture and settlement (HARYONO, 2012). The Bugis have used swampland for agriculture since the early twentieth century. They have many years of experience reclaiming lowland areas and dealing with the related soil and water management issues. The Bugisse, followed by the Banjarese and Malays, have reclaimed approximately 2 million swamplands along the eastern coast of Sumatera and along the western and southern parts of Kalimantan using traditional techniques (SURYADI & MOERWANTO, 2013). They produced rice at a rate of about 0.8-1 ton per hectare (MAAS, 2003). In the 1680s, the first scientific swampland exploration discovered peat in Sumatera (NOOR, 2012). In 1895, detailed exploration was carried out in eastern Sumatera. It was then followed by surveys in the 1930s and 1950s on Kalimantan's western and southern coasts, as well as Sumatera's eastern coast. These surveys only looked at the ecology, flora and fauna, and other characteristics to make comparisons to the subtropical swamp (MAAS, 2003). In South Kalimantan, the Dutch East Indies government started the first large-scale swamp reclamation for agriculture and settlement in 1920. They dredged two canals that connect Kapuas Murung and the Barito River in 1936. They relocated people from Java to swamp areas in Kalimantan a year later to expand their colony program, distribute the population, and grow rubber and coconuts (NOOR & SARWANI, 2013). This effort was continued after Indonesia's independence in 1945. The Government of Indonesia (GoI) divided the history of swampland reclamation into three periods: (1) the 1945-1960s; (2) the 1969-1995s; and (3) the 1995-2000s. The first era began with canal dredging to improve accessibility in Sumatera and Kalimantan. Three major canals (anjir) were dredged and widened in Kalimantan to connect two major rivers, the Barito and the Kapuas Murung. The community then dug up the sub-canals (handil) and cleared the land for agriculture. The sub- canals were 2-3 meters wide, 0.5-2.0 meters deep, and 2-3 kilometers long, with a distance of 200-300 meters between them (NOOR, 2012). The second period (1969-1995) was marked by the launch of the Tidal Rice Field Reclamation Project (P4S)2, despite widespread skepticism about its success 2 P4S (Proyek Pembukaan Persawahan Pasang Surut) was a project to open peat swampland for agricultural purposes. The project was held on 1969 - 1984 under the World Bank (IBRD) sponsor. Major universities (IPB, UGM, and ITB) were involved to survey and design the SWAMPLAND FOR AGRICULTURE IN INDONESIA 13 (SUBAGYO et al., 1996, NOOR, 2012). However, the government has recognized swampland as a potential agricultural resource since 1968. Initially, experts questioned this potential because of constraints including hydrology, thick peat, soil acidity, and low soil consistency, which resulted in soil subsidence, soil nutrient depletion caused by tidal movement, seawater intrusion, and inaccessibility, as summarized by NOTOHADIPRAWIRO (1994). However, inspired by the Bugis people's long experience and Thailand and Vietnam's success in opening the Delta Mekong, the government continued the reclamation. The P4S project was created to address a 2-million-ton rice deficit by reopening 5.25 million hectares of land. To support the project, the government also launched the transmigration program3. Until 1995, the government had reclaimed 1.18 million hectares of swampland and local communities had reclaimed 3.0 million hectares (NOOR, 2004). Some swamp areas have been developed into cities and regencies, with agriculture serving as the major sector. The last phase (1996-2000s) was marked by the launch of the Mega Rice Project (MRP) in 1996. This project aimed to restore rice self-sufficiency by opening up more than 1 million hectares of peat swamp in Central Kalimantan Province. Since 1992, Indonesia has been a rice importer after achieving rice self-sufficiency in 1984. Rice imports risen rapidly from 0.6 million tons in 1994 to 1.8 million tons in 1995. The project began without a proper environmental impact assessment to determine the capacity of swamp peatland for rice production and to review the peatland conversion plan for the type of infrastructure development (HECKER, 2005). Around 13,500 migrant households from populated islands have been settled to work on this agricultural project. This project included the construction of 917 km of primary and secondary canals, as well as 11,839 km of tertiary ditches, which connected the peat dome to the sea (NOOR & SARWANI, 2013). The canal, however, agriculture area layout (VAN DEN EELAART, 1981; SUBAGYO et al., 1996). During Pelita I (1969– 1974), 32,000 hectares of swampland have been surveyed and 60 % of those have been converted into agricultural based settlement areas. 3 Transmigration project is the resettlement project to redistribute families from the crowded island (Java, Bali, and Madura) to the sparsely populated islands of Sumatera, Kalimantan, Sulawesi, and Papua. The program was first initiated by the Dutch East Indies government in 1905 as part of the colonization to reduce population pressure in Java Island and provide cheap plantation workforce in sparsely islands (FEARNSIDE, 1997; HOLDEN et al., 1995). The project reached its peak under Soeharto’s leadership (1968-1998), when 3,264,902 families have been trans-located to outlying and sparsely populated islands, including 279,580 families to Kalimantan (TIRTOSUDARMO, 2009; NUGROHO, 2013). The project was considered as the largest people migration in the world (MARR, 1990). At the end of the Soeharto period, 130,667 families were translocated to outlying islands during 1999-2007 (TIRTOSUDARMO, 2009). 14 SWAMPLAND FOR AGRICULTURE IN INDONESIA was not constructed with sluice gates, which caused over-draining of the peatlands. Opening peat forests emits massive amounts of greenhouse gases due to their role in carbon storage (JAENICKE et al., 2008). During the dry season, the dried peat ignites peat forest fires since it is combustible. When El Nino hit Indonesia in 1997- 1998, the forest fires in Kalimantan reached 3.06 million hectares, the majority of which were peat forests (LIEW et al., 1998). As a result, a massive amount of smoke haze spread to neighboring countries. The drained peat in Kalimantan released 0.81 to 2.57 gigatons of carbon, which is equivalent to 13 to 14 % of the average annual global carbon emissions from fossil fuels (PEAT PORTAL, 2004). Peat depletion also causes massive soil acidification, which reduces land fertility. Many migrant farmers have left the area due to high farm costs and low productivity as a result of a lack of knowledge about swampland farming practices (MAAS, 2003). Approximately half of the first migrant households abandoned large areas of land. Because the tidal wave could not reach the rice plots during the dry season, the water level dropped rapidly. The seawater intrusion spread further inland. As a result, it is difficult for the indigenous people to obtain fresh water. Meanwhile, during the rainy season, this area experienced frequent flooding (NOOR & SARWANI, 2013). The remaining farmers left and began clearing new peat areas. Due to a lack of capital and a reliance on family labor, they used slash and burn land clearing, which reduced soil fertility. According to NOOR (2010), peat fire reduce land productivity as the peat degrades. Rice yields fell from 3.0–3.5 ton per hectare to 0.05–1.50 ton per hectare. As a result, these frequently burned lands lost their fertility and were eventually abandoned by farmers. According to MAAS (2003), while swamp conversion begins with hydrology, hydro-topography, and soil assessment, the farming technique used to adapt to such conditions is the most important factor in its success. Farmers will achieve the highest yield only in the early years if traditional farming practices4 are followed. Later, as organic matter depletes, soil fertility declines rapidly, and farmers tend to move to the new open area. As a result, many reclaimed swampland areas have been fallowed, making them vulnerable to forest fires. According to an integrated swamp development project (ISDP) report, nearly 60% of reclaimed swampland was fallowed between 1994 and 1999 (possibly 70% if the ex-MRP area was included) (MAAS, 2003). During the dry season, the fallow areas are vulnerable to forest fires. However, the government continues to see the swamp area as a potential land resource for increasing agricultural production, particularly rice production. The government cited a large undeveloped area, flat topography, water 4 Based on the usual farming technology developed in Java Island. This technology is not adaptive to swampland environment SWAMPLAND FOR AGRICULTURE IN INDONESIA 15 availability, and a low population as justifications for a large-scale swamp conversion. If properly rehabilitated, the ex-MRP area has the potential to serve as a national food basket. SUHARTANTO (2007) proposed that the area be rehabilitated for several reasons: (i) migrant farmers face severe poverty; (ii) there are massive government assets (2.5 trillion IDR) as well as 1.45 million hectares of fallowed land; (iii) environmental damage causes ecological, economic, and socio-cultural problems; and (iv) the potential for further sustainable development. Furthermore, many farmers rely on their land for a living. In 2009, approximately 10,000 people (2,600 families) lived in 14 settlement areas along the Kapuas River's bank (MEDRILZAM et al., 2017). As a result, the GoI issued Presidential Instruction (Inpres) No. 2/2007 to help accelerate swamp rehabilitation and revitalization. The ex-MRP area was divided into five zones by the GoI: Zone A (393,302 ha), Zone B (164,836 ha), Zone C (441,436 ha), Zone D (153,453 ha), and Zone E (424,269 ha). ISARI surveyed an area of 563,248 ha (in total) from Zones A, B, and D of the ex- MRP area in 2012. They discovered that 29 % of the surveyed area is suitable for rice, 19 % for secondary crops and vegetables, 20 % for perennial crops and plantation, and 32% for conservation area and limited types of plants. Large-scale swamp reclamation tents cause environmental and human disasters. Simultaneously, abandoning the reclaimed area without treatment introduces a new adversity. Thus, rather than creating new swamp forests, maximizing productivity and existing reclaimed swampland is the best choice (see Appendix 2). 2.4 Tidal Swampland Tidal swampland accounts for roughly 20.1 million hectares (12.31 %) of Indonesia's 162.4 million hectares of land resources (SURYADI, 2006, MINISTRY OF PUBLIC WORK, 2009). The following explanation goes into greater detail about tidal swampland. 2.4.1 The nature of tidal swampland Tidal swampland is a swampland area where water movement caused by regular tidal fluctuation influences the water in wells and canals (VAN DEN EELAART, 1981). The depth of the water is controlled by both tides and rainfall. Tidal swampland can be divided into four typologies based on the prevailing water levels in fields (hydro- topography) (see Figure 2.2) (NOORSYAMSI, et al., 1984, VAN GILST, 1992, WIDJAJA-ADHI, et al., 1992, WIDJAJA-ADHI & KARAMA, 1994): 16 SWAMPLAND FOR AGRICULTURE IN INDONESIA ● Type A area is directly affected by tidal movement and always flood. During the spring tide, the water depth can fluctuate by up to 2.5 meters in 24 hours. ● Type B area is directly affected by tidal movement but only floods during the spring tide. Figure 2.2. Swampland typology based on water prevailing in the field Source: Modified from SUYAMTO (2007) ● Type C area is indirectly affected by the tidal movement. Where the depth is less than 50 cm, the tide influences the ground-water table. Rainfall has a greater impact on the water table than tides. ● Type D area is not affected by high tide. The only source of surface water is rainfall. The ground-water table is more than 50 cm deep. Tidal swampland is classified as marginal land (HIDAYAT et al., 2010), which means that its potential productivity is limited by the high variability of physical, biological, and socio-economic factors (PARTOHARDJONO, 1993). The primary environmental issues with tidal swampland, according to FOLKERTSMA (1998), YANTI (2002) and NUGRAHA et al. (2016), are the highly complex nature of the soil characteristics, the uncontrolled hydrologic regime, and the high level of acidity and the toxicity of metal substances, such as aluminum, iron, and manganese. Tidal swampland is distinguished by high soil acidity (low pH) and the presence of a high concentration of pyrite, Al3+ and Fe3+, and quartz/sand. Tidal swamplands may also have deep organic layers (peat) that are more than 200 cm thick. As a result, these characteristics pose numerous constraints to agriculture due to the low and unbalanced nutrient status of plants (crops) required. Plants may also be poisoned by the soil. According to PRASETYO et al. (1990) the soil can be classified into two major soil types. The soils on the relatively low-lying waterlogged interior, i.e. the alluvial- marine plains and old riverbeds, are the first to be considered. These soils have a SWAMPLAND FOR AGRICULTURE IN INDONESIA 17 brown layer (20-60 cm) overlying a gray layer that is generally pyritic. Pyrite content can reach up to 8% iron sulfide (FeS2). They have a (silty) clayey texture, are high in organic matter (5-14 %), are poorly drained, half to nearly ripe, and are mottled in most cases. The acidity (pH) level is between 3 and 4. Most of these soils are covered by a thin peat layer (10-20 cm). Second, higher ground soils, such as river levees and coastal ridges, differ from lower ground soils. These soils are similar to the first, but they are nearly ripe. They contain 4 % and 6.5 % organic matter, respectively, and have a low pyrite content (FeS2< 1.5 %). Because the pH level is around 5-6, the soil reaction is slightly acidic to neutral. At greater depths (> 125 cm), these soils are overlain by a gray pyritic subsoil. These soils were classified as tropaquepts by the Soil Taxonomy. The deeper the peat layer, the greater the need for a proper drainage system to increase soil pH (ANWARHAN, 1981). 2.4.2 Tidal swampland for agriculture Unlike the traditional farming practice on Java Island, which employs gravity full- irrigated systems, farming in a tidal ecosystem deals primarily with less arable land, tidal hydrology, and acid soil. The hydrology of different areas varies greatly and can change over time. The difference between high and low tides in the secondary channel, according to ANWAR & MAWARDI (2011), ranges from 42 to 204 cm depending on the typology, season, and distance from the primary riverbank. As a result of their distinct characteristics, farmers must implement an appropriate farming system in tidal ecosystems (YANTI, 2002). Even though the productivity is lower than in irrigated areas, reclaimed swampland is still promising for crop production in the future. The managed swamp area was estimated to be approximately 1,044,695 hectares in 2005 (SYAUKAT, 2011). The crops that can be grown in this area are limited. To produce an adequate yield, selected crops should be tolerant of adverse soil conditions such as high salinity and acidity. Concerning the sulfide layer, DENT & PONS (1995) proposed that farmers keep the soil moist to limit oxidation and manage the hydrology to ensure better acid leaching. Tidal swamplands are suitable for wet rice fields and other selected crops such as coconuts, oranges, and several secondary crops planted in the dikes due to the availability of abundant water, especially during the wet season, the hydrology properties, and the flat topographical setting (DJAMHARI, 2002). DARSANI et al. (2020) proposed that approximately 1.05 million hectares of swampland could be cultivated for rice with one year of planting and productivity of 4-5 ton per hectare. 18 SWAMPLAND FOR AGRICULTURE IN INDONESIA However, the total contribution of swampland to the national rice production is only 4-5 million ton per year (DARSANI et al., 2020). VAN DEN EELAART (2014) proposed swampland as an alternative for rice cultivation for several reasons. To begin with, it avoids the potential limitation of increasing gravity-irrigated coverage. Most gravity-irrigated rice fields are concentrated on Java Island, where land conversions occurred quickly. Second, the expansion of gravity-irrigated areas outside Java (e.g. South Sulawesi and North Sumatra provinces) for increasing rice production is considered as having no significant prospects at a competitive cost due to a number of issues, including the excessive soil problems that many irrigation projects have experienced, the limitation of physio-geography landscape, and the hydrology and topography of the adjoining rivers and basins, which are not suitable or only have very limited for irrigation. Moreover, the costs of maintenance and investments to increase rice production should be considered. As a result, in areas other than Java, the swamp scheme application may be more profitable than the gravity-irrigated one. Furthermore, physical conditions have a significant impact on rice yields. Most of the developed swampland is found in tidal river sections along the sweet water. In general, only one harvest results in a year with yields of 1.5-2.5 tons/ha. Besides that, the results are significantly lower than those of gravity-irrigated land. The land is then abandoned under stagnant water conditions with high acidity. Historically, the best yields have been found near rivers, where frequent flooding with fresh water may occur during high tide (VAN DEN EELAART, 2014). In the long-term, the rice yield productivity may be lower (less than 3 tons/ha) due to several factors, including: (1) inadequate water management at the primary system level; (2) several reclaimed areas are not mature enough due to a lack of water arrangement; (3) a lack of appropriate credit facilities; (4) a lack of infrastructure and post-harvest facilities and management; (5) severe pest and disease attacks; and (6) labor shortages (SUPRIYANTO et al., 2010). The Telang project (South Sumatera, Indonesia) achieved 6 tons/ha of optimum yields using improved rice varieties despite frequent tidal flooding. However, non- tidal flooded areas that made some improvements, such as installing water control structures, building more tertiary drains, and upgrading existing canals, obtained the same results. These systems most likely produced two crops per season. Even though pump irrigation has yet to be used in the area, the introduction of mechanized land preparation could be a significant factor (VAN DEN EELAART, 2014). SWAMPLAND FOR AGRICULTURE IN INDONESIA 19 According to VAN DEN EELAART (2014) and SUPRIYANTO et al. (2010), several factors influence land fertility, including (i) tidal flooding (tidal irrigation) ability, (ii) land levels related to the average water level due to the neap tide, and (iii) land levels related to the average water level due to the spring tide. These aforementioned factors are primarily concerned with water supply and percolation rates. Furthermore, water supply and percolation rates influence the severity of toxic and acidic conditions. The worse conditions, such as higher acid and toxic levels, were caused by insufficient water supply and percolation quantities. In addition, primary infrastructure maintenance and on-farm water management contribute to yields. The presence of high acidity soil layers in acid sulfate soils, on the other hand, had no significant effect. Fisheries are dependent on the quality of water in rivers and canals. The ecosystems that feed the swamps are dominated by freshwater fishing. Canals near large rivers, with their primary catchment in the uplands, are suitable for freshwater shrimp farming. Near the coast, brackish water fisheries such as shrimp, milkfish, and crab farming are feasible. However, due to high acidity, shrimp farming was not feasible in areas with acid-sulfated soils. Otherwise, blackfish that can survive in extremely acidic water for a few months were a better choice. It should be noted that the benefits to fisheries are limited in rivers with most of their catchment area in peat domes (ombrogenous peat soil) (VAN DEN EELAART, 2014). However, fish farming necessitates massive investment to maintain a consistent freshwater supply, dam construction, and feed. 2.4.3 Agricultural practice in tidal swampland ecosystem The main factor in tidal swampland farming is water control. A water arrangement can also be used to leach toxic substances, reduce pyrite oxidation, and land subsidence, and keep saltwater out of the field. Given these facts, one-way water control is appropriate for tidal swampland, particularly typologies A and B (SARAGIH, 2013). The secondary and tertiary channels in the system (Figure 2.3) are protected by one- way flap gates. Water enters the irrigation channel through the inlet flap gates while the outlet flap gates are closed. At low tide, water thrust automatically closes the inlet flap gates, while the outlet flap gates open and water flows out (SARAGIH, 2013). This system optimally supports water circulation and toxic leaching. Agriculture in swamp areas differs from agriculture in gravity-irrigated areas. To manage swamp areas, tide hydrology and the presence of pyrite must be properly considered. Agricultural practice, particularly rice cultivation, exhibits several characteristics. In swamp areas, local rice varieties are generally preferred because 20 SWAMPLAND FOR AGRICULTURE IN INDONESIA they can withstand deep flooding and acidity (high adaptability), require fewer inputs (easy to cultivate by the farmers), and are more profitable (SUMAWINATA, 1992, KHAIRULLAH, 2020). These varieties include bayar putih, bayar kuning, and lemo, which take 7-9 months from seedling to harvest. They have long, slim grains with a delicious flavor and aroma (SUMAWINATA, 1992, KHAIRULLAH et al., 2013). These varieties are the most popular in South and Central Kalimantan, so their prices are higher than others (KHAIRULLAH, 2020). Figure 2.3. One-way water system Source: SARAGIH (2013). Weeds and grass are cut with a scythe-like tool (tajak) or a long knife (parang) when the water level is higher than 30 cm. The grass cuttings are then decomposed by dispersing them throughout the plots and submerging them for 15-20 days. The rotten grasses are piled into a mound and left half-submerged for 15-20 days. The piles are then turned over and stored for the next ten days. Finally, the decomposed grasses are distributed across the plots. To deal with the seasonal hydrology situation, multiple seedling stages are used. Using this method, farmers require only 5 kg of seeds for a hectare of rice plot. The first seedbeds (teradakan) are usually prepared in October in normal years, or in November in El Nino years. Seedbeds are classified into two types: those prepared on dry soils with dry seeds and those prepared on the raft with pre-germinated field bunds. The first method is used in relatively high locations, such as dykes or paddy field bunds. Following the removal of the grass, 60-70 seeds are placed in a hole and covered with soil and ash. Tetujah (goat’s hoof) is used to make the holes. The holes are separated by about 10 cm and left for 40 days. The other technique is SWAMPLAND FOR AGRICULTURE IN INDONESIA 21 performed on a mud-covered floating raft. Pre-germinated seeds are sown on the mud surface and allowed to germinate for 15 days. The seedlings are then transferred to the second seedbed (ampakan) in one corner of each rice plot. Typically, the planting space is 20-30 cm. They are kept there for 40 days to help the seeds grow and multiply. On the periphery of each paddy plot, the third seedling (lacakan) is prepared. The seedlings are kept there for 60 days before being planted (tanam) on the others. Two lacakan are normally placed 30 x 30 cm apart in a single hole. Planting should be completed by February in the type A area and by March through April in the types B and C areas. Harvesting occurs in August and September after about 4-5 months of development. As a consequence of climate change, some steps (e.g., lacakan) should be overlapped to match the water table level. As a result, the amount of seeds needed per hectare will rise (SUMAWINATA, 1992, KHAIRULLAH et al., 2013). Farmers can also cultivate it twice a year by intercropping it with high-yielding rice varieties (HYVs). This arrangement is known as sawit dupa (once seedling–twice harvesting) in the local language. Local and HYV seeds are planted (typically in October) (KHAIRULLAH et al., 2013). The HYVs seeds are planted in the rice plot while the local variety seeds are still in the second and third seedling stages. After 90 days, the HYVs are harvested, and the local variety is planted on the plot. The water level in the plots is controlled by simple water gates built in a primary ditch by piling up tree trunks or branches. Similar but smaller water gates are built in the secondary ditch, which is dug at an angle to the main ditch. When the rainy season begins in early November, the water gates are kept open. The acids produced during the dry season, according to SUMAWINATA (1992), can be leached out and drained off through the main ditch. The water gates are closed in late December, and the fields are submerged. Farmers prepare the land in submerged fields by cutting the grass and successively transplanting seedlings. The final transplant is scheduled for March. The water gates will be shut down until June. As a result, the plots are submerged from the time the land is prepared until the end of the vegetative phase. Thus, the plots are submerged from the land preparation to the end of the vegetative phase. In addition, microchannels should be dug in the plot in swampland type B. SARWANI (2003) found that the microchannels accelerate the acidity leaching since the plots are only flooded on the spring tide. Raised beds for vegetables, fruits, and perennial crops such as coconut and orange can be established in a rice plot. This system is known locally as the sorjan system (see Figure 2.4). These raised beds can also be used to control the excess mud that enters the plot every year (HIDAYAT, 2010). In the rice plot, the mud layer should be about 30 cm thick. The raised bed measures 2-3 meters wide by 0.5-0.75 meters tall. To compensate for the high cost, the farmer typically constructs one raised-bed 22 SWAMPLAND FOR AGRICULTURE IN INDONESIA per year until he has 5-8 raised-beds in a hectare. Raised beds are built during the dry season and then left to leach to reduce soil acidity. To improve soil fertility, lime and organic matter are added. Rainfall will remove the acidity matter at the beginning of the wet season. One month before planting, holes are dug with a 5- meter space between them in orange farming. The size of the hole varies depending on the soil type and layer beneath. The topsoil and organic fertilizer mixture is then inserted and left for a week. The grafted seed is then planted. Farmers can plant 200-250 orange trees per hectare of land, depending on the swamp type. Every year, rice straw is spread over muddy soil and manure to maintain the raised bed. Figure 2.4. The sorjan system Source: Own description based on the field survey (2014) 2.5 Swampland for agriculture: some countries’ experience Agriculture on swampland dates to the prehistoric era. Archaeologists discovered the transition of farming from highland to drained wetland gardens in Kuk Swamp, Papua New Guinea (BOURKE, 2009). In another part of the world, Maya farmers on the Yucatán peninsula intensively modified swamps to make a living (BEACH, et al., 2009). The reclamation of the swamps continued in accordance with the increase in population. The advance of technology accelerated the conversion. Around half of the world's wetlands have been reclaimed for agricultural, settlement, industrial, and urban purposes (VERHOEVEN & SETTER, 2010). The reclamation has negative consequences because the environment and biodiversity status have deteriorated significantly. Reclamation, on the other hand, has positive outcomes such as improved economic conditions, reduced poverty, increased food production, and accelerated rural development. The following sub chapters describe some countries’ experiences with swampland reclamation. 2.5.1 Papua New Guinea In Papua New Guinea (PNG), the swamp forest is located on the northern lowland, between the central range to the south and the Pacific Ocean to the north (WIKRAMANAYAKE et al., 2002). According to archeological research, the Kuk Swamp area has been used for taro starch cultivation since 10,000 years. An SWAMPLAND FOR AGRICULTURE IN INDONESIA 23 arrangement of small island beds was built 7,000 years ago to breed water-tolerant plants such as banana and swamp gardens (BOURKE, 2009). According to WATSON (1965) and BALLARD (2001), the introduction of sweet potatoes was critical in a dramatic change from highlands to drained wetland farming three centuries before the first contact in the 1930s. Based on the archeological findings, the indigenous people constructed a network of drainage ditches, boundary markers, roads, and pig tracks. The spatial constraint was accommodated by these structures (BALLARD, 2001). This was followed by a movement from the highlands to the lowlands (MAY, 2004). In the 2000s, nearly half of the rural population lived in lowland areas (ALLEN & BOURKE, 2009). Because they are staple foods for the locals, sweet potato, banana, cassava, and swamp taro dominate lowland food production. According to a 2000 survey, sweet potatoes provide approximately 66 % for staple foods (BOURKE et al., 2009). Because the environment in PNG is ideal for mixed species planting, it is widely used. This method is quite efficient, resulting in higher total yields and lower labor input (ALLEN & BOURKE, 2009). The introduction of new cultivars has a significant impact on both the shortening of the fallow period and the lengthening of the cropping period. Agricultural production rises sharply when soil maintenance techniques (e.g., composting, managed tree fallow, crop rotation, and erosion control) are used (BOURKE, 2001). The agriculture sector employs 85 % of the population and accounts for 25 % of the country's GDP (ADB, 2015). 2.5.2 West Africa West Africa has 1.2 million hectares of mangrove swamps, with 200,000 hectares cleared for rice cultivation in Nigeria, Guinea Bissau, Gambia, Guinea, Senegal, and Sierra Leone over the last century. This region contributed 10% of total regional rice production (AGYEN-SAMPONG, 1994). Domesticated rice varieties from Niger and its neighbors were improved in order to increase yield (JOHNY et al., 1981). Nigeria is currently the largest rice producer on the subcontinent, accounting for 40% of total rice production (NWAOBIALA, 2010, FAOSTAT, 2007). In Nigeria, swamp and rain-fed lowland rice production accounts for half of total rice production (TASHIKALMA, 2014). Furthermore, swamp agriculture serves as not only a primary source of food but also as an employer and source of labor. WARDA (West Africa Rice Development Association), the current AfricaRice, has been developing the lowlands as part of the West-Africa Development Project since the early 1970s. They have released new adaptable varieties that have been 24 SWAMPLAND FOR AGRICULTURE IN INDONESIA continuously integrated with international rice research. They also work on rice trade improvement and development hubs across African countries as representatives of primary rice-growing environments and various market opportunities (AFRICARICE, 2015). Because it is in a tidal area, the farming outcome is dependent on salt-free periods, which affect the length of rice growth. In the favorable season, the yield reaches 2 tons per hectare, which is higher than 1 ton per hectare in the non-swamp area (AGYEN-SAMPONG, 1994). Physical constraints (e.g., Al and Fe toxicity, phosphorus and nitrogen deficiency, salinity, brown spot infestation, and acidity), biological constraints (e.g., low yield and susceptibility to pest and environmental stresses), and socio-economic constraints (e.g., limited labor, accessibility, extension and education, and credit) all limit production (AGYEN-SAMPONG, 1994, ONIAH et al., 2008). In this case, the adaptive rice variety should be introduced along with a suitable extension service. The yield cycle should be shortened to compensate for the decrease in rainfall. This condition, on the other hand, is similar to the tidal swamps located in South Kalimantan, Indonesia. 2.5.3 Rwanda Rwanda's economy is heavily dependent on agriculture, which accounts for 34 % of national GDP and 70 % of exports. Agriculture employs more than 80 % of the population (MUHINDA, 2013). In 2010, the national poverty line poverty headcount ratio was 44.9 % of the population (WORLD BANK DATA, 2015). Based on these facts, agriculture will be the backbone of poverty reduction and economic development by 2020 (WORLD BANK, 2011). Swamp and marshland in Rwanda were converted into a highly productive rice scheme. This World Bank program to improve to reduce poverty through agricultural development. The swamp was drained by diverting small rivers into two peripheral canals from which small-holder plots were gravity-irrigated (SEEBOERGER, 2014). In 2013, approximately 23,683 hectares of marshland were developed for agriculture, with a projected increase to 65,000 hectares by 2017 (MUHINDA, 2013). They learned from this experience that: (1) 10-20% of the swamps and marshland should be left for ecological buffering and nature; (2) the slope of the straightened main river-bed should have some stable speed-breakers/drop-structures; (3) intensive production necessitates a careful biological and mineral-fertilizer based soil fertility strategy; and (4) the silt-load of the rivers after heavy storms should be SWAMPLAND FOR AGRICULTURE IN INDONESIA 25 used systematically to re-fertilize the plots (5) The environmental impact of greenhouse gases in irrigated rice should be closely monitored; (6) the schemes can also be used for conventional intermittent irrigation rather than long-term submersion of rice plots. (7) A contour line-oriented plot layout can cut land- leveling work in half. (8) The majority of the work could be done by hand, e.g., cash-for-work; (9) According to local poverty-ranking-tools, the allocation of the many new plots is best for the poor (BERNHARD MEIER ZU BIESSEN, 2015, ESIRU, 2014, SEEBOERGER, 2014). Because the project is still in its early stages, no side effects of swamp reclamation have been observed. According to MUHINDA (2013), one of the challenges in the agricultural sector is a lack of private sector investment. Private and public sector financial institutions are hesitant to enter the credit market. Another challenge is the lack of farmer skills in modern and sustainable farming. 2.5.4 Vietnam Since 1968, Vietnam has recognized wetlands for their significant contribution to socio-economic development. They have contributed to the agricultural transition from rice importers to rice exporters from 1976 to 2003. Tourism and fishing are two other roles (VEPA, 2005). The Mekong Delta and the Red River Delta are Vietnam's most important swamps. The Mekong Delta was one of Southeast Asia's first civilizations (TORELL & SALAMANCA, 2003). In that order, the deltas cover 3.9 million hectares and 302,318 hectares (76 % is estuarine wetland) (VEPA, 2005). However, artificial wetlands (rice plots, fishponds) have increased, resulting in a decrease in natural wetlands, according to the MINISTRY OF NATURAL RESOURCES AND ENVIRONMENT (2001). Rice cultivation expanded from 2 million hectares in 1976 to 3.82 million hectares in 2004. According to preliminary statistics, 50% of reclaimed wetlands in Vietnam were used for crops, 25% for aquaculture, and 10% for artificial lakes and reservoirs (VEPA, 2005). For decades, paddy rice cultivation has dominated the Mekong Delta. Because of the harsh environment, only a small population lived there at first. Canals were dredged in the 1930s and 1940s to drain swamp areas, make them available, and connect them to cities and neighboring districts (NI et al., 2003). Rice cultivation began in the 1960s with traditional dwarf rice, and multiple rice cropping was introduced in the 1970s. Nguyen Chi Thahn classified swamps in 1993 based on the body of water, geomorphology, and hydrologic characteristics. In 2004, the Ministry of 26 SWAMPLAND FOR AGRICULTURE IN INDONESIA Agriculture adopted this classification as the agricultural development standard (VEPA, 2005). This expansion caused the construction of additional canals, the relocation of people, and the improvement of infrastructure (NI et al., 2003). As a result, agriculture occupied approximately 83 % of the Mekong Delta, utilizing an extensive network of canals, irrigation, drainage, and village connectors. Rice cultivation, on the other hand, is the area's primary source of agricultural household income (TORELL & SALAMANCA, 2003). Despite increased agricultural production, wetlands in Vietnam face significant challenges, such as population growth, low policy implementation, a lack of integration among intersectoral parties, a lack of research and development, agrochemical pollution, and natural disaster impacts (VEPA, 2005). Low yields are caused by environmental issues, and poverty increases among small farmers as they lose their land due to debt. Direct assistance for landless farmers has failed, as they have lost both their money and their land. As a result, direct aid should be accompanied with education, technical assistance, and short-term financial assistance (NI et al., 2003). 2.5.5 Thailand Thailand has 3,660,000 hectares of wetland (7.5 % of its total land area) (TRISURAT, 2006). Peat swamps cover 56,475 hectares of the total (YOSHINO et al., 2010). For generations, the wetlands have supported the Thai people's livelihood. During the 19th and 20th centuries, the government reclaimed freshwater wetlands and mangrove forests. Because of the importance of wetlands, the country ratified Ramsar Convention5 in 1998 in order to develop a national policy and action plan for wetland management (TRISURAT, 2006). From the 1960s to the early 1980s, large amounts of new land, including swampland, were made available for farming. Agriculture was the main driver of the economy at the time, accounting for 70 % of employment. Rural poverty fell from 60 % in the 1960s to 10 % in the 2000s (LETURQUE & WIGGINS, 2010) as a result of high agricultural growth since the mid-1980s (KASEM & THAPA, 2012). According to YOSHINO et al. (2010) estimated that 45 % of the tropical swamps in Malaysia and Thailand's peninsula have been converted into industrial forests, built-up areas, and agricultural lands. 5 Ramsar Convention is an intergovernmental treaty to provide the framework for national action and international cooperation for the conservation and wise use of wetlands and their resources. Currently, 169 parties are involved to protect 214.94 million hectare of wetlands in 2,231 Ramsar sites (Ramsar.org). SWAMPLAND FOR AGRICULTURE IN INDONESIA 27 Wetland rice cultivation is common in agricultural areas, particularly along the Chao Phraya River. This swamp area has helped Thailand's economy transition from a rice importer to a well-known Southeast Asian rice exporter. Rice exports stimulated economic development and integration into the global economy (FLAHERTY et al., 1999). The government aided rice, cassava, and sugarcane production by providing credit, inorganic fertilizers, and pesticides, as well as guaranteeing minimum prices (KASEM & THAPA, 2012). However, in other places, rapid growth is offset by the depletion of environmental and natural resources. As a result, the Thai government promotes for sustainable agriculture. Crop diversification with high-value crops is encouraged, and so are the livestock and fisheries. Chemical fertilizers and pesticides have been reduced by raising awareness about balanced chemical use, discouraging the use of inorganic fertilizers, and reducing inorganic fertilizer imports. In addition, the government promotes organic fertilizer and healthy food (KASEM & THAPA, 2012). Other issues in Thailand include the effects of forest conversion to agriculture on the river basin (WILK et al., 2001) and the transition from agriculture to industry (KASEM & THAPA, 2012). 2.5.6 India India's total wetland area is approximately 58.3 million hectares, with rice plots accounting for 71% of this total (BASSI et al., 2014). Most inland wetlands are directly or indirectly dependent on major rivers such as the Ganga, Brahmaputra, Narmada, and others. Himalayan wetlands, coastal wetlands, and reservoirs are the other types of wetlands (PRASAD et al., 2002). They are important in a variety of sectors, including food baskets (agriculture and fishing), water storage and supply, wildlife habitat, environmental buffer, tourism, and fuel (BASSI et al., 2014). Development and population growth, agriculture, deforestation, and over-irrigation all contribute to wetlands degradation in India (PRASAD et al., 2002). In Punjab, for example, most wetlands have degraded due to inadequate ecological restoration (LADHAR, 2002). Meanwhile, due to over drainage, artificial wetlands for agriculture are prone to drought (MABWOGA & THUKRAL, 2014). In the 1950s, natural wetlands covered approximately 60,000 ha in Punjab. The wetlands now only cover about 15,000 ha. The area of agricultural wetlands has increased from 6,500 ha in the 1960s to 8,000 ha recently (LADHAR, 2002). Recently, approximately 84 % of the wetlands have been cultivated, with the remaining 5.7 % being forested (JERATH et al., 1995; LADHAR, 2002). The lesson learned is that short-term drainage of low land will have an adverse effect on the environment. Modern agriculture that drains the water flow to the 28 SWAMPLAND FOR AGRICULTURE IN INDONESIA surrounding area endangers agricultural sustainability. Monoculture leads to excessive use of chemical fertilizer and pesticides (LADHAR, 2002). This agricultural run-off, as well as erosion silt, harms the local flora and fauna. To address the issue, a legal framework and policy support are required, and a qualified organization should be involved. The first policy support for wetland conservation was enacted by the government in 2000. Reduced agricultural run-off containing pesticides and fertilizers is part of the policy (BASSI et al., 2014). In addition, several rules were issued and gradually improved to meet the Ramsar Convention. However, because these rules were not tailored to local rights, the community and local stakeholders were not heavily involved (BASSI et al., 2014). 2.5.7 USA During the period of European settlement in the 1600s, the United States had nearly 90 million hectares of wetland (DAHL & ALLORD, 1997). There were only 41.8 million hectares left in the mid-1980s (DAHL et al., 1991, ZEDLER, 1996). This acreage represented only 47% of the wetland status in the 1780s. This area is still deteriorating, owing primarily to extensive and massive agriculture in the drained wetland (DAHL, 1990). DAHL et al. (1991) estimated that agriculture was responsible for 54% of the loss. The large scale of wetland conversion occurred between 1800 and 1860 because of high population growth and massive migration to the lower area, which resulted in a high demand for land. This conversion was accelerated by technological advances, such as steam-powered canal dredging and mechanical farming tools (DAHL & ALLORD, 1997). One of the most notable tidal swamps in the United States is reported in Maryland, where a significant change has occurred. Before European settlement, it was estimated that there were 485,622 hectares of tidal swamp, but this was reduced to 50 %. It happened as a result of deep-water habitat conversion, saltwater and freshwater impoundments, ditching, and a lack of government regulation (TINER & BURKE, 1995, HARRISON et al., 2004). Based on the facts, the government should prioritize wetlands mitigation. Preserving wetland habitat, replacing wetland losses, and mitigating upland losses are all part of the process (ZEDLER, 1996). However, there is still considerable disagreement about which should be prioritized, and which technique is more efficient than others. SWAMPLAND FOR AGRICULTURE IN INDONESIA 29 2.5.8 Lesson to be learned Humans have a long history of dealing with swampy environments. Many cultures developed around swamp agriculture. Swampland is still used as a resource for economic development in many countries today. The swamp is commonly exploited in developing countries. They recognize that the swamp contributes to agricultural development, reducing rural poverty and unemployment. Several countries, including Thailand, Indonesia, Vietnam, and India, have had success in reclaiming swamps, despite the fact that there are still many environmental problems. They have paid more attention in the last decade to restoring nature and implementing sustainable swamp agriculture. Their works are examples of how they attempted to address environmental issues. Their efforts can be served as a model for Rwanda and Papua New Guinea. On the other hand, developed countries' priorities and values for swamp areas have shifted from exploitation to preservation and mitigation. They are aware that the rapid reduction in swamp area may have an adverse effect on the ecology. Furthermore, uncontrolled exploitation will harm the economy in the long run. The bitterness of large-scale swamp reclamation has been felt in the United States. Swamp preservation is a difficult task that must be done on a continuous basis. All stakeholders must work together to solve numerous interconnected problems with limited funds and resources. The challenges for swamp rehabilitation are choices and priorities. An intervention is a factor that will affect others simultaneously. Swamps can be beneficial to food production if they are managed properly. However, its reclamation and use must be carefully considered. The reclamation should reconsider the soil, hydrology, and plants cultivated. The proportion of converted and forested swamp should be maintained. Humans have a long history of dealing with swamp agriculture. They have local knowledge which could be used to guide future development. Crop selection is also important. Crops should be economically valuable and resistant to harsh environmental conditions (acidity, salinity, dryness, inundation). Because swamp agriculture can only use a limited amount of land, crop selection and land use are critical issues. 30 STUDY AREAS AND RESEARCH DESIGN 3 RESEARCH DESIGN 3.1 Location of the study area The research was carried out on Kalimantan Island, which has extensive tidal swampland rice farming. According to WÖSTEN et al. (2008), Kalimantan has approximately 6 million hectares of peatland, primarily tidal swampland. The swamp areas are located in the southern part of Kalimantan, specifically in the provinces of South Kalimantan and Central Kalimantan. Based on the BPS report (2016), the planted area of tidal swampland for rice in South Kalimantan was 166,324 ha, which produced 602,709.19 tons of rice (NINGSIH et al., 2020). South Kalimantan is the island's smallest province. It is considered the island's gateway due to its strategic location to other parts of Indonesia, particularly the main island (Java). This province has a large tidal swamp area that is used for farming (AGRICULTURAL BUREAU OF SOUTH KALIMANTAN, 2012). As a result, the research focuses on a tidal swamp area in South Kalimantan (the green area on the map in Figure 3.1). Furthermore, the study concentrated on the regencies of Barito Kuala and Tanah Laut. 3.2 Rationale for choosing the study area South Kalimantan Province was chosen for this study. In 2009, the swampland area in this province was approximately 553,551 ha; those are spread out in Barito Kuala (239,830 ha), Tapin (102,322 ha), Tanah Laut (80,467 ha), and Banjar (52,592 ha) (KHAIRULLAH et al., 2021a). However, from the survey in 2019, there was a change in the distribution: Barito Kuala, Banjar, Tanah Laut, and Tapin account for 226,899, 74,273, 56,430, and 37,295 ha, respectively (KHAIRULLAH et al., 2021a). As a result, Barito Kuala and Tanah Laut regencies was chosen as the primary focus due to some reasons: - Barito Kuala has extensive tidal swampland farming, whereas Tanah Laut has extensive type A swamp farming. - Tidal swampland has been used since the 1900s (SUTIKNO & NOOR,1997) and became widely available in the 1970s (WÖSTEN et al., 2008). As a result, farmers in this area have experience of the challenges associated with cultivating rice in a specific environment. - Tidal swampland accounted for 196,419 hectares, or approximately 29.60 % of rice fields in South Kalimantan. It is distributed to 54.29 % in Barito Kuala, 18.37 % in Banjar, 12.92 % in Tapin, and 14.14 % in other regencies (CBS OF SOUTH KALIMANTAN, 2009). STUDY AREAS AND RESEARCH DESIGN 31 - The area has become the intention for agricultural development in order to increase the production through the implementation of government programs run by the Indonesian regency, regional, and central governments. Figure 3.1. Map of the study area 3.3 Data collection The method of data collection is critical for obtaining accurate and valid data. As a result, the sampling procedure is a critical step in data collection. The sampling is performed to verify a representative image of rice farming in tidal swampland. The data was collected from both primary and secondary sources. From October 2013 to February 2014, a field survey was conducted in the two regencies. The data collection procedures are depicted in Figure 3.2. Source: http://bjn.wikipedia.org Source: National Spatial Planning Study Zone: 1 Barito Kuala (inner area) 2 Tanah Laut (coastal area) 1 2 32 STUDY AREAS AND RESEARCH DESIGN Figure 3.2. Data collection Source: Own depiction 3.3.1 Primary data collection and sampling design The sample farmers were chosen using a multistage sampling procedure, as shown in Figure 3.2. First, two regions with large swampland areas were purposefully selected (i.e. Barito Kuala and Tanah Laut). WIDJAJA-ADHI et al. (1992; 1994) proposed a swampland typology based on flooding, which was also taken into account in sampling. Rice is primarily grown in swampland types A (deep flooding), B (medium flooding), and C (shallow flooding), with a small amount grown in swampland type D (not affected by flooding). Two villages were chosen from each region to represent those types of swamplands. In the third stage, some steps were taken to ensure a high level of representativeness. Initially, a sampling frame of farmers was created with the help of extension officers, village chiefs, and other stakeholders. The information was gathered through deep interviews, focus group discussions (FGD), and structured questionnaire interviews during the field survey (see Appendix 4). The questionnaire interview included 200 household heads from each of the three swamp typologies: 72 from typology A, 64 from typology B, and 64 from typology C. According to ethnicity, 129 household heads were locals and 71 were transmigrants. They were primarily questioned about their detailed activities for a single crop calendar year (August 2012 to July 2013). Their activities from the previous year and the two years before were also documented. STUDY AREAS AND RESEARCH DESIGN 33 A standardized questionnaire was used to interview farmers. They were asked about household general characteristics, farming experience, information on land ownership and land use, household assets, employment including non-farm activities, farm management, farm distance from home, cropping patterns at the present and five years before, physical cultural system, cost and returns of rice farming, prices of inputs and outputs, membership in the farmer group and cooperative, information sources, access to credit, role of the government for farming management, problems of rice farming, etc. Appendix 5 contains the complete questionnaire. 3.3.2 Secondary data collection Secondary data was gathered from non-governmental organizations' literature and reports, administrative offices including the Ministry of Agriculture, research centers, statistical yearbooks, and annual reports. 3.4 Data analysis The data was analyzed using two techniques: descriptive analysis and household modeling analysis. 3.4.1 Descriptive analysis The primary goal of the descriptive analysis is to gain in-depth knowledge of the study area and household socio-economic circumstances. Instead of going straight to the analytical results, this approach describes the state of the targeted respondents. Descriptive analysis was used in this study to present the social- demographic and economic characteristics of the surveyed farm households. Both were the primary parameters for further empirical investigation. The MS-Excel package was used to calculate statistical parameters such as the percentage, mean, standard deviation, and F-test. The standard deviation was applied to assess the difference within the tidal swampland typology. Meanwhile, the F-test was used to analyze the difference among the tidal swampland typologies A, B, and C. The analysis describes the household characteristics, farm characteristics, and land-use system. Farm management, as well as a few farming technologies applied, are also clearly explained. 3.4.2 Household modelling analysis To evaluate land allocation in swamp agriculture, a household modeling based on linear programming was developed. The objective function is to calculate the gross 34 STUDY AREAS AND RESEARCH DESIGN margin. According to HAZELL & NORTON (1986), the linear programming model can be written as follows: 𝑚𝑎𝑥𝑍 = ∑ 𝑐𝑗𝑋𝑗 𝑛 𝑗=1 (3.1) In such a way that ∑ 𝑎𝑖𝑗𝑋𝑗 𝑛 𝑗=1 ≤ 𝑏𝑖 , 𝑎𝑙𝑙 𝑖 = 1 𝑡𝑜 𝑚 (3.2) and 𝑋𝑗 ≥ 0, 𝑎𝑙𝑙 𝑗 = 1 𝑡𝑜 𝑛 (3.3) In this study, the objective of this mathematical modelling is to maximize Z, that is, gross margin. A detailed explanation of it would be explained further in Chapter 6. However, maximization of Z (gross margin) should not violate any of the fixed resource constraints expressed by equation 3.2 (i.e., farm household activities in tidal swampland area). The resource constraints also do not have any negative activity levels, as expressed by equation 3.3. These resource constraints include land availability, labor capacity, capital capacity, and home consumption in a certain season (dry and rainy season). Furthermore, the details of these resource constraints will be explained in Chapter 6. This research attempted to predict future crop prices based on stochastic numbers in the simulated crop price prediction by using average crop price data from the previous few years and the Cholesky Decomposition method. The GAMS solver runs the average prices only once. Appendix 6 contains the GAMS code that was used in this model. The simulated crop prices were then used to compute a simulated gross margin. This procedure will be thoroughly explained in Chapter 6. GAMS ver. 25.1, STATA 14, and MS-Excel 2010 were used for the household modeling and crop price simulation. FIELD STUDY FINDING 35 4 REVIEW OF STUDY AREA CONDITIONS This chapter presents the condition of South Kalimantan with a focus on the Barito Kuala and Tanah Laut regencies. The purpose of this chapter is to describe the study area's geographic, climate, and socio-economic conditions, including population and labor, agricultural production, institutions, social services, and infrastructures. 4.1 Geography and climatology The study was conducted on Kalimantan (Borneo) Island. Because the island is so close to the equator, it has a tropical climate with high humidity. Kalimantan has a total land area of 737,188 km2 (GAVEAU et al., 2014). Geographically, the island is shared by Indonesia, Malaysia, and Brunei. Indonesian territory (known as Kalimantan) accounts for 72.36 % of the island's area (GAVEAU et al., 2014) and 69.5 % of its population (CBS, 2014). West Kalimantan, North Kalimantan, East Kalimantan, Central Kalimantan, and South Kalimantan are the five provinces of Kalimantan (CBS, 2013). The swamp covers 11.37 million hectares and is spread across all provinces (MINISTRY OF PUBLIC WORK, 2009). South Kalimantan has approximately 250,000 ha of tidal swampland agriculture potential. Only 177,148 ha (70.86 %) of swampland has been reclaimed, with 133,702 ha used for agriculture (AR-RIZA et al., 1997). Barito Kuala, Banjar, Tanah Laut, Tapin, Kotabaru, and Banjarmasin are the six regencies in which they are located. The Barito Kuala and Banjar regencies have the most tidal swampland in South Kalimantan, accounting for 67.28 % and 17.03 %, respectively (AGRICULTURAL BUREAU OF SOUTH KALIMANTAN, 2012). 4.1.1 Geography South Kalimantan is located between 114o19’13”–116o33’28” East Longitude and 1o21’49 –4o10’14” South Latitude. In terms of area, it is the smallest province on Kalimantan Island. It covers 37,530.52 km2, or 6.98 % of Kalimantan Island and 1.9% of total Indonesian territory. The territory comprises 42.99 % of the forest, 22.14 % of savanna (including fallowed land), 17.55 % of agriculture, and 11.63 % of plantation (CBS OF SOUTH KALIMANTAN, 2012) (Figure 4.1). 36 FIELD STUDY FINDINGS Figure 4.1. Land use in South Kalimantan, 2011 Source: CBS OF SOUTH KALIMANTAN (2012) South Kalimantan's geography is divided into two major geographical areas: the lowland, which includes swamps and peatlands, and the mountainous area. The majority of the land (74.81 %) is flat, with a slope ranging from 0 to 15%. South Kalimantan is divided into eleven regencies (Barito Kuala, Tanah Laut, Kotabaru, Banjar, Tapin, Hulu Sungai Selatan, Hulu Sungai Tengah, Hulu Sungai Utara, Tabalong, Tanah Bumbu, and Balangan), as well as two cities (Banjarmasin and Banjarbaru) (CBS OF SOUTH KALIMANTAN, 2012). The research was carried out in the regencies of Barito Kuala and Tanah Laut, which have a large swampland area with distinct typologies. Barito Kuala is located in the western part of the province of South Kalimantan, between 114o20’50” E until 114o50’18” E and 2o29’50”–3o30’18” S. With a total area of 2,996.96 km2, the region accounts for approximately 7.99 % of the total area of South Kalimantan. Its topography is mostly flat lowland with a 0.20 % slope and an elevation of 0.20-3.00 meters above sea level (CBS OF BARITO KUALA, 2014). The majority of the district is tidal swampland with a thin peat layer in a few places. The dominant soil type is alluvial6, which covers approximately 60-64 % of the total area, with the remaining part being organosol7 (CBS OF BARITO KUALA, 2014). 6 Alluvial soil is formed by the sedimentation of sand and mud from the basin river. The area is the best for tidal agriculture (CBS OF BARITO KUALA, 2014). 7 Organosol or well-known as peat soil, is formed from decaying plant fibric which is waterlogged for a long period of time. The acidity is relatively high (CBS OF BARITO KUALA, 2014). Village 1.59% Industry 0.07% Mining 1.12% Agriculture 17.55% Plantation 11.63% Savana 22.14% Forest 42.99% Water 1.21% Other 1.70% FIELD STUDY FINDING 37 The Barito Kuala Regency is located on the banks of the Barito River. Tamban, Serapat, and Talaran are the three main canals (anjir) that connect the Barito and Kapuas rivers. Tidal movement, rainfall, and the state of land use on the river's banks and headwat