0
A scenario-based spatial model of
land use and land cover change in
the Langkloof valley, South Africa
A combined qualitative and spatial approach for investigating the drivers of a
complex socio-ecological landscape
Roos van der Deijl
November 29, 2019
Examiner
Dr. Peter Verburg, VU University Amsterdam
Second examiner Dr. Jeanne Nel, VU University Amsterdam
Co-assessor
Dr. Erik Cammeraat, University of Amsterdam
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
1
Research thesis (42 ECTS)
by Roos van der Deijl (student number 10470263)
for the partial fulfilment of the MSc. Earth Science,
track Future Planet Ecosystem Science,
at the Institute of Biodiversity and Ecosystem Dynamics,
University of Amsterdam,
September 2019
.The Langkloof, mapped as
a 3D terrain from satellite images
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
2
Table of Contents
ABSTRACT ... 4
LIST OF TABLES ... 5
LIST OF FIGURES ... 6
ACKNOWLEDGEMENTS ... 8
CHAPTER 1. RESEARCH CONTEXT AND DESIGN ... 9
1.1. Introduction ... 9
1.2. Research aim and questions ... 10
1.3. Research design ... 11
CHAPTER 2. A CONCEPTUAL MODEL OF THE DRIVERS OF LAND USE & LAND
COVER CHANGE IN THE LANGKLOOF ... 12
2.1. Introduction ... 12
2.1.1. The Langkloof: area overview ...12
2.1.2. Living Lands: collaborations working on living landscapes ...17
2.2. Methods ... 17
2.3. Results ... 18
2.3.1. Natural drivers ... 18
2.3.2. Socio-economic drivers ... 23
2.3.3. Political drivers ... 29
2.3.4. Innovation & collaboration ... 32
CHAPTER 3. A SPATIAL MODEL OF FUTURE LAND USE & LAND COVER CHANGE
IN THE LANGKLOOF ...35
3.1. Introduction ... 35
3.1.1. LUCC modelling ... 35
3.1.2. Scenarios for the future: from storylines to projections ... 36
3.2. Methods ... 37
3.2.1. Land use types ... 37
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
3
3.2.3. Spatial policies and restrictions ... 40
3.2.4. Land use type specific conversion settings... 40
3.2.5. Land cover requirements ... 43
3.2.6. Statistical analysis of location factors ... 44
3.2.7. Allocation procedure ... 47
3.2.8. Scenario storylines ... 49
3.2.9. Scenario translation from storyline to model parameters ... 51
3.2.10. Model validation ... 52
3.3. Results ... 54
3.3.1. Land cover requirements ... 54
3.3.2. Statistical analysis for the prediction of land use types by location characteristics ... 56
3.3.3. Allocation of land use and cover changes ... 59
3.3.4. Model validation ... 65
CHAPTER 4. DISCUSSION ... 67
4.1 Synthesis conceptual model ... 67
4.2 Synthesis spatial model ... 68
4.3 Conclusions and policy recommendations ... 70
REFERENCES AND DATA ... 72
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
4
Abstract
Land use and land cover change (LUCC) is now predominantly driven by human activities. It often leads to land degradation, desertification and deforestation, which are putting a major pressure on the Earth’s climate and ecosystems. Fortunately, more than 2 billion hectares of land worldwide are suitable for restoration. The implementation of landscape restoration is now recognized as a priority by the United Nations and many governments. It is therefore of crucial importance to increase knowledge on the landscape-specific conditions for restoration. This is not an easy task, as LUCC of a landscape is driven by complex interactions of social, economic, and political drivers. Spatial LUCC models provide the opportunity to unravel these complex relations and to visualise the patterns of future land use changes under different scenarios. This study therefore implements a LUCC model in the Langkloof valley in South Africa, where the organization Living Lands is working on socio-ecological landscape restoration. The study followed two aims. First, a conceptual model was developed by means of interviews and literature review to map the drivers of land use change in the Langkloof. Second, the Dyna-CLUE spatial model of future LUCC in the Langkloof was used to simulate land use change 20 years into the future. Next to a base model run, four scenarios were simulated. These were based on storylines that were developed from the conceptual model. The conceptual model showed a complex image with as the main threats the increasing droughts and extreme weather events, the invasion of alien species, and the commercialisation of the agricultural sector. The main opportunities include the diversification of commodities, the cultivation of honeybush, tourism, and the value-added industry. The spatial model results clearly show the threat of alien invasives, and a distinction in land use dynamics between the valley floor as opposed to the hillslopes and mountains. More research is needed to expand these modelling efforts with more thorough data. It is important that these results are communicated to policy makers so that informed decisions can be made about the future of the landscape.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
5
List of tables
Table 1 - Answers of landowner interviews.. ... 24
Table 2 – Land use types and their codes as used for the spatial model. ... 38
Table 3 - Mapping methods of the 1990 land cover map, specified per land use type. ... 39
Table 4 - Elasticity settings for each land use type ... 41
Table 5 - Land use types conversion matrix for the base model. ... 42
Table 6 - Calculation of land use requirements ... 43
Table 7 – Location factors and sources. ... 45
Table 8 - Hypotheses and sources for the regression models for each land use type. ... 46
Table 9 - Scenario settings based on the scenario storylines. *The calculation of the increases in demand is explained in the text. ... 51
Table 10 - Regression coefficients for the binary logistic regression models for each land use type and their predicting location factors. ... 58
Table A1 - Experts consulted in the process of conceptual model development……….……….82
Table A2 - Recode of 2014 map………..…...82
Table A3 - Conversion matrix scenario 1……….…83
Table A4 - Conversion matrix scenario 2……….…………...84
Table A5 - Conversion matrix scenario 3………...84
Table A6 - Conversion matrix scenario 4……….85
Table A7 - Conversion matrix validation model……….………85
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
6
List of figures
Figure 1 - Map of South Africa indicating the location of the Langkloof……….………13
Figure 2 - View of the Langkloof, looking North-West, with indications of the most important ecosystem services……….……….……….………...……….…..…...……13
Figure 3 - Apple orchards in the Langkloof.……….………....……….…..……14
Figure 4 - Map of the greater Langkloof area indicating different agricultural sub-communities, primary land use activities, and towns or villages………...………..……15
Figure 5 – Indigenous fynbos vegetation on the hills and pastures in the valley floor………..…16
Figure 6 - Land-use/land-cover change (LUCC) in the Kromme river catchment over the past century.………16
Figure 7 – Workflow of the research process…….……….………11
Figure 8 - Natural drivers of land use change in the Langkloof………...……….……19
Figure 9 - Temperatures in Joubertina (central Langkloof) ……….………20
Figure 10 – Rainfall in Joubertina (central Langkloof). ……….21
Figure 11 – Projected median change in the average seasonal rainfall (mm) over South Africa……….……….………..……21
Figure 12 - The black wattle plant……….……….….…….…22
Figure 13 - Socio-economic drivers of land use change in the Langkloof.….……….………23
Figure 14 - Export tonnage and prices of Honeybush tea in South Africa from 1999-2017...27
Figure 15 - Rand to US dollar exchange rate from February 1990 to June 2019………27
Figure 16 - Fuel price index, fertilizer price index, and Brent Crude Oil price index assumptions for South Africa from 2010 to 20126………..28
Figure 17 - Annual percentage change in the cost of labour and electricity………..….………29
Figure 18 - Political drivers of land use change in the Langkloof………...……..……30
Figure 19 - Protected areas in the Langkloof………..………31
Figure 20 - Conservation areas in the Langkloof………...……….31
Figure 21 - Knowledge & Innovation drivers of land use change in the Langkloof………33
Figure 22 - Overview of the information flow in the Dyna-CLUE model………..………37
Figure 23 - Predicted apple and pear production and area in South Africa……….……….…..44
Figure 24 - Flow chart of the allocation procedure of the Dyna-CLUE model……….…47
Figure 25 - Scenario storylines………..50
Figure 26 - Land cover requirements in hectares for each land use type over the 20 years in the base model………..…….54
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
7
Figure 27 - Land requirements per scenario for each land use type in the Langkloof, over all model run years 1-20 (2014 to 2033)………..………..……….55 Figure 28 - Land use areas in 2014 and land cover requirements for 2033 for the four scenario
runs.………..……….………..56 Figure 29 - LUCC maps of the Langkloof in 2014 and 2033 as produced by the Dyna-CLUE base model
run………...…..………..……….60 Figure 30 - Zoom-in on the South Eastern part of the Langkloof maps of 2014, 2023, and 2033 as produced by the Dyna-CLUE base model run.……….……….60 Figure 31 – Land areas in hectares in the year 2014 and the land cover requirements in 2033 for each land use type in the base model, categorized by valley floor and hillslopes & mountains.…………...………….……….61 Figure 32 - Percent change from 2014 to 2033 for all land use types in the valley floor and hills & mountains for the base model………..………...….……….………..61 Figure 33 – Map results of the Dyna-CLUE LUCC model of the Langkloof for the year 2033 for all 4
scenarios………..………63 Figure 34 - Percent change from 2014 to 2033 for each land use types in the valley floor and hills & mountains for scenario 1……….………..64 Figure 35 - Percent change from 2014 to 2033 for each land use types in the valley floor and hills & mountains for scenario 2………..……….64 Figure 36- Percent change from 2014 to 2033 for each land use types in the valley floor and hills & mountains for scenario 3………..………..……….……..64 Figure 37- Percent change from 2014 to 2033 for each land use types in the valley floor and hills & mountains for scenario 4………..……….………..………….……….64 Figure 38 – Fits for several sampling window sizes for the validation model……….…………...65 Figure 39 – Maps resulting from the model run with the initial year 1990 and the final year of
2014…………...……….…..66 Figure A1 – Conceptual model on the drivers of land use change in the Langkloof………..86 Figure A2 - Conceptual framework of manifestations and underlying drivers for agricultural land change. From: (van Vliet, de Groot, Rietveld, & Verburg (2015)………..87 Figure A3 - Driving Factors of LULC change in the Western Cape Province. From: Tizora, Roux, Mans, & Cooper (n.d.)……...……… ……...……….…….87
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
8
Acknowledgements
The process of this thesis has been a learning journey with many helpers along the way. The story started in my course ‘Analysis and Modelling Lab’, where I followed a module on the Dyna-CLUE land use model taught by Dr. Peter Verburg. The fun working out the puzzles of complex land use systems ignited my enthusiasm for land use modelling. Around the same time, I was introduced to the work of Commonland on holistic landscape restoration, which highly inspired me. So, when the time came to choose a research thesis topic, I inquired for possibilities to do my thesis work with Commonland, where Simon Moolenaar and Willemijn de Iongh referred me to their South African partner Living Lands. This brought me in contact with Maya Beukes, who directed me to the Langkloof team, and in particular Ancia Cornelius as my local supervisor. I asked Dr. Peter Verburg to be my supervisor and Dr. Erik Cammeraat to be my reader. Dr. Jeanne Nel acted as my unofficial supervisor especially on the conceptual modelling and understanding of the South African context, which is her home country. When the research proposal was accepted, I spent 4 months in the Langkloof to explore the landscape and get to know the work of Living Lands. These months have taught me a lot about the potential and challenges of the on-the-ground work of socio-ecological landscape restoration, as well as about the complexity of the Langkloof landscape. This experience was indispensable to the study’s conceptualisation of the landscape.
Many people supported me in this thesis process. First and foremost, I would like to thank Peter, Jeanne, and Ancia, who supervised me in their own expertise along the process. I was lucky to be supported by Peter’s expertise on land use modelling and his patience when the process was extended. Jeanne helped me through the first conceptual steps and constantly inspired me with new ideas and approaches. Ancia was and is my Langkloof buddy, always showing me the beauty of the landscape and explaining the social complexity. I would also like to thank Dr. Erik Cammeraat for his supervision and for reading my thesis.
I would like to thank the Living Lands team for hosting me in their office and home and teaching me about the Langkloof: thank you Liz Metcalfe, Matt Sephton, Roderick Juba, and Thelani Grant. Tiahnah Göbel; thank you for welcoming me into the Eastern Cape with a beautiful learning journey.
The spatial modelling would not have been possible without the wealth of data provided by the WRC project team, led by Dr. Julia Glenday. Julia and Faith Jumbi also took me along on a day of fieldwork, from which I learned a lot. I am very excited to see the results of their impressive participatory hydrological modelling project, and if my thesis can be of any use to their work that would make me very happy. Furthermore, I would like to thank all my interviewees for helping me build and evaluate my conceptual model and for giving me new insights.
I would also like to thank Sian Theron, Sameena Khan, and Eugenie Angus for the infamous Buiteklub nights and making my Langkloof time fun. The same thanks go to my housemate Jonathan Snyders, and my housemate and co-student Laetitia Vallée. Our common time in the Langkloof (long valley) must have been meant to be, since Vallée means ‘valley’ in French, and my last name ‘van der Deijl’ means ‘from the dune valley’ in Dutch.
During this thesis, I took on an extra challenge by starting a new job at Commonland Foundation. This job offered me a unique opportunity to work on my thesis while constantly staying in touch with the field of landscape restoration. At the same time, combining work and thesis is challenging and I could not have gotten through it without the constant support and encouragement of my parents and siblings. Furthermore, thank you Boris, for your never-ending positive spirits, and thanks to all my friends for being my cheerleaders.
Last but not least, I would like to thank my Romée Prijden and Giulia Sardano for reading my thesis and providing me with useful feedback.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
9
Chapter 1. Research context and design
“While natural forces dominated the appearance of the land’s surface for billions of years, humans are now recognized as the main driver shaping the environment in the modern world”
- Prestele et al., 2016
1.1. Introduction
Since the beginning of human society, we have moderated the Earth’s land to meet our needs. Land provides us with food, clothing, shelter, heat, transportation, recreation, aesthetic pleasure, social status, spiritual needs, and territorial sovereignty (Briassoulis, 2009). However, our growing world population is posing increasing demands on the Earth. Accordingly, while the most common drivers of land use change have always been of natural origin, since recent history humans are the dominant drivers of land use and cover change (LUCC) (Crutzen, 2006). This shift marks the beginning of a new geological epoch named the ‘Anthropocene’ (ibid.). The unprecedented human pressure on land resources often leads to a competition or trade-off between demand for land functions that provide food, water and energy on the one hand, and services that support and regulate all life cycles on Earth on the other hand (UNCCD, 2017). As a consequence, all agro-ecological systems and land cover types are showing severe degradation (Le, Nkonya, & Mirzabaev, 2016). Between 1998 and 2013, approximately 20% of cropland and 27% of rangeland worldwide declined in productivity, mainly as a result of land/water use and management practices (ibid.). Also natural land is degrading; only 25-40% of the Earth’s land surface can still be qualified as ‘natural’ (Ellis, 2011). Overall, land degradation covers 30% of global land area. About 3 billion people live in degraded areas, but the number of people affected by it is higher because of the dependence on ecosystem services from these areas (Le et al., 2016).
The specific consequences of LUCC for the Earth’s climate and ecosystems are diverse (Briassoulis, 2009; IPBES, 2016; Nelson, 2005; Ostberg, Schaphoff, Lucht, & Gerten, 2015). First, habitat loss as a result of land use change in terrestrial ecosystems has caused the largest impact on biodiversity throughout history (Pereira, Navarro, & Martins, 2012). Also, a number of land changes have been shown to contribute heavily to human CO2 emissions (Le Quéré et al., 2018), and non-CO2 greenhouse gas emissions to the atmosphere
(Tubiello et al., 2015). These changes include deforestation, afforestation, logging and forest degradation, shifting cultivation, and regrowth of forests following wood harvest or abandonment of agriculture (Le Quéré et al., 2018). In addition to these biochemical effects, LUCC causes some biophysical effects to the Earth’s surface which also affect global climatic cycles (Betts, Falloon, Goldewijk, & Ramankutty, 2007). One such effect is surface albedo, where a lower albedo can lead to a local warming due to decreased shortwave light reflections (ibid.). Also, the conversion of forest to cropland or pasture reduces the aerodynamic roughness of the landscape (Betts et al., 2007). This reduces evaporation, which leads to a decrease in the fluxes of moisture and latent heat from the surface to the atmosphere, which again increases the temperature near the surface (ibid.).
Considering these effects, it is clear that the current state of degradation of the world’s landscapes is detrimental for the functioning and resilience of ecosystems on which we depend. This has led to the recognition that the protection of the remaining natural landscapes is insufficient sustain the flow of ecosystem services and goods essential for human wellbeing (Gann et al., 2019). Instead, to meet our growing population in its needs it is essential to secure a net gain in the functioning of native ecosystems by investing in both the protection and the repair of the environment. Fortunately, we know that more than 2 billion hectares of land worldwide are suitable for restoration (UNCCD, 2012). Landscape restoration is defined as “a planned process that seeks to recover landscape-level ecological integrity and the capacity of a landscape to
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
10
provide long-term, landscape-specific ecosystem services essential for improving human wellbeing” (Gann et al., 2019). Correspondingly, landscape restoration involves both ecological and social targets and goals.
In the past 15 years, ecosystem restoration has been recognized as a global priority in sustaining the future health of our planet (Aronson & Alexander, 2013). This is enhanced by many ambitious conventions and targets set up by international organisations and authorities, including the 2030 Agenda for Sustainable Development (United Nations, 2015b), the Strategic Plan for Biodiversity 2020 and Aichi Biodiversity Targets (Convention on Biological Diversity, 2011), the UN Framework Convention on Climate Change (United Nations, 1992), the Paris Agreement (United Nations, 2015a), the UN Convention to Combat Desertification (UNCCD, 1994), Land Degradation Neutrality (UNCCD, 2016), and the Global Partnership on Forest and Landscape Restoration (GPFLR) (Besseau, Graham, & Christopherson, 2018). The largest international initiatives are the Bonn Challenge, which aims to restore 350 million hectares of degraded or deforested lands by 2030 (“The Bonn Challenge,” 2011), and the African Forest Landscape Restoration Initiative, which has the goal to restore 100 million hectares of land by 2030 (AFR100, n.d.). The largest effort to date bringing all these efforts together has been announced recently: the UN Decade on Ecosystem Restoration will accelerate existing commitments and initiatives to achieve transformational ecosystem restoration in 350 million hectares of land between 2021-2030 (“UN Decade on Restoration,” n.d.).
Ecosystem restoration has proven to be a challenging task due to often poorly defined goals and a lack of quality monitoring (Wortley, Hero, & Howes, 2013). In 2004, the Society for Ecological Restoration (SER) defined the first international guidelines with key concepts and fundamental principles upon which ecological restoration is based, including nine attributes of a restored ecosystem to determine whether restoration has been successful (SER, 2004). These guidelines have increased both the quantity and quality of scientific assessments of restoration (Wortley et al., 2013). The effectivity of landscape restoration has been assessed in several meta-analyses, in which it is found that restoration increases biodiversity, vegetation structure, and ecosystem services (Barral, Rey Benayas, Meli, & Maceira, 2015; Crouzeilles et al., 2016). One meta-analysis also found, however, that the success of active restoration efforts is highly variable between different local conditions (H. P. Jones et al., 2018).
Therefore, it is necessary to thoroughly analyse a landscape’s local conditions when engaging in landscape restoration. Accordingly, the purpose of this research is to investigate the drivers of land use and land cover change in the Langkloof Valley, South Africa. This valley, located between the Kouga mountains and the Tsitsikamma mountains in the Eastern Cape Province contains lot of valuable natural resources, including the production of fruit and livestock (de Kock, 2015; Schafer, 2014), water storage and use (van Vuuren, 2011) and harvesting of wild and cultivated honeybush tea (Joubert, Joubert, Bester, de Beer, & De Lange, 2011). Unfortunately, these resources are under pressure, due to a changing climate and the degradation of land and ecosystem services. The organisation Living Lands has been working in the Langkloof since 2016, with the aim of implementing socio-ecological restoration projects. It is essential for a landscape restoration project to understand the local drivers of land use change in the past, present, and future. Yet, no thorough review of these drivers has been done to date in the Langkloof. This study therefore aims to identify the most important drivers of land use change in the Langkloof. To simulate future land use changes based on these drivers, a spatial land use model is implemented. This is done on the basis of four scenario’s, which represent different possible futures depending on the course of the climate and social and political decisions made.
1.2. Research aim and questions
The aim of this project is to provide scientific insight into the driving factors land use and cover change and their impacts in the Langkloof valley, South Africa. This knowledge can serve decision making processes on the restoration strategy in this area. The results can also provide an example for comparison to LUCC studies in other regions, in order to identify synergies and differences in driving factors and their impacts. To
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
11
be able to provide such an integrated perspective on the socio-economic, natural, and political drivers of land use change in the Langkloof, the following two main research questions will be answered. These questions will be answered by a combination of methods, as outlined below.
1. “What are the main drivers of land use change and their interactions in the Langkloof?” 2. “What are the predicted land use changes in the Langkloof 20 years into the future?”
1.3. Research design
This study consists of two parts with their own methods. In Chapter 2 a conceptual model was developed in order to identify and structure the main drivers of LUCC in the Langkloof. This was done through an iterative process of literature review and interviews with stakeholders and experts for scoping of knowledge and gathering of feedback on the model. In Chapter 3 land use change in the Langkloof region is simulated with the spatial model Dyna-CLUE. Two types of simulations were run: a predictive model, which had the purpose of getting a realistic idea of future land use changes, and explorative models with scenarios. The scenario storylines were developed based on the conceptual model. Then, these storylines were translated into parameter settings for the model. Finally, the results of Chapter 2 and 3 are reflected upon and put into context in the discussion in Chapter 4, which also provides the final conclusions and policy recommendations.
Figure 7 – Workflow of the research process. The structure is based on the conceptual framework of modelling with stakeholders by Hamilton, ElSawah, Guillaume, Jakeman, & Pierce (2015). This framework is particularly useful for this study, because it provides a framework for combining different modelling methods.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
12
Chapter 2. A conceptual model of the drivers of land
use & land cover change in the Langkloof
Daar is 'n groot verskeidenheid van mense, bedrywe en verskillende idees. Jy het soveel diversiteit in die natuurlike omgewing ook. As jy net 500 meter daarheen kan gaan, is dit heeltemal anders. Ons het die groot kloof in die middel, en van daar het ons kleiner afdraaipaadjies na verskeie kleiner klowe.
In elke kleiner klofie is daar amper soos 'n nuwe gemeenskap en 'n nuwe sosiale groep, wat ook met iets anders boer1.
- Ancia Cornelius, landscape mobilizer at Living Lands, and Thelani Grant, landscape innovator at Living Lands
2.1. Introduction
The identification of the relevant driving factors of land use change of an area is a fundamental step in the set-up of a spatial LUCC model (P. H. Verburg, Schot, Dijst, & Veldkamp, 2004). These drivers can be direct, by influencing biodiversity and ecosystem processes, and indirect, by altering direct drivers and other indirect drivers (IPBES, 2016). Drivers should not be seen as isolated influences, but rather considered as dynamic factors interacting with and within each other on various spatial and temporal scales (IPBES, 2016). To understand the dynamic interactions between driving factors, it is useful to identify a landscape’s trade-offs and synergies between different ecosystem services. Trade-trade-offs happen when one service increases at the expense of others, and synergies can be found when multiple services decrease or increase together (Smith et al., 2013). Similar land cover change trade-offs and synergies exist between different stakeholder groups and different geographic regions.
For this purpose, extensive scoping of the situation and developments of the Langkloof were done with the goal of developing a conceptual model which captures the relevant drivers of land use change in the Langkloof. The author spent 3,5 months in the Langkloof as a visiting scholar at the Living Lands office in Tweeriviere, a small town in the middle of the Langkloof. Information was gathered from various sources to identify the key processes related to land use change in the Langkloof.
The outcomes for this conceptual chapter are a conceptual model which represents the top-down driving factors of land use change in the Langkloof and the scenario storylines based on these drivers. The aim of the conceptual model of drivers is to explore and illustrate all relevant drivers of land use change, which can serve as the basis for the scenario storylines in Chapter 3. The main ‘drivers diagram’ can be found in figure A1, and in this chapter for each driver’s category a separate diagram is presented which shows the drivers of this category and the ones with which they directly interact.
2.1.1. The Langkloof: area overview
The Langkloof (Afrikaans for ‘long valley’) is a valley in the Eastern and Western Cape province of South Africa, extending 160 km between the Tsitsikamma and Kouga mountain ranges (Fig. 1). The name ‘Langkloof’ refers to a social-geographic boundary and does not align perfectly with either the catchment boundaries or municipalities (Cockburn, 2018). However, the Langkloof is often referred to in a broader sense as the Kouga and Kromme catchment areas (Cockburn, 2018; Living Lands, 2017a; WRC Project # K5/2527, n.d.). This is the definition used in this study, because water is a central driver of LUCC and therefore relevant for LUCC modelling.
1 There is a large diversity of people, businesses, and different ideas. There is also great diversity in the natural
environment. If you go 500 meters in any direction, the environment is completely different. There is the large valley in the middle, and then there are many smaller roads which turn into smaller valleys. Each small valley forms an entire new community and a new social group, which also farms with different crops or livestock.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
13
Figure 1 - Map of South Africa indicating the location of the Langkloof in the Eastern and Western Cape provinces. From Cockburn (2018), with permission.
Figure 2 - View of the Langkloof, looking North-West, with indications of the most important ecosystem services. From Cockburn (2018), with permission.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
14
Most of the Langkloof falls within the Kou-Kamma municipality (KLM). It is therefore suitable to use the socio-demographic data of this municipality to describe the Langkloof. There are 46,000 people living in the KLM, of which 61% are coloured, 31% black, and 8% white (Kou-Kamma local municipality, 2017). The main language is Afrikaans (spoken by 74% of the population as their home language) (Kou-Kamma local municipality, 2017). The Eastern Cape is highly agriculture-oriented: 27.1% of households are agricultural households (Statistics South Africa).
The Kouga and Kromme catchments, together with the Baviaanskloof catchments, provide 70% of the water supply to the city Port Elizabeth (Mander et al., 2010). This city houses 1 million inhabitants and is the largest city in the Eastern Cape Province. On top of water, the Langkloof provides many valuable ecosystem services to its surrounding region. The most important ecosystem services used in the landscape by private land owners are the production of fruit and livestock (de Kock, 2015; Schafer, 2014), water storage and use (van Vuuren, 2011) and harvesting of wild and cultivated honeybush tea (Joubert, Joubert, Bester, de Beer, & De Lange, 2011; fig. 2).
The Langkloof has been farmed since 1760 (Cockburn, 2018). While the dominant practices were initially livestock and mixed crop farming, it developed into an important deciduous fruit-growing region, mostly for export markets (Cockburn, 2018; Schafer, 2014). In the North-Eastern section of the Langkloof (‘the ‘Suurveld’, fig. 2), however, sheep farming continued on extensive rangelands (Cockburn, 2018). In 2012, 35.5% of the country’s total apples and pears were produced in the Langkloof (de Kock, 2015). Other fruit types grown in the area are apricots, peaches, nectarines, plums, and prunes (ibid). The Langkloof is consists of many subregions with separate communities. Cockburn mapped the socio-ecological subregions in the Langkloof on the basis of interviews (Cockburn, 2018). This map shows the diversity in agricultural commodities in the different subregions, including crop farming, honeybush tea production from wild and cultivated honeybush, lifestyle farming, and wildlife ranching (Fig. 4).
Figure 4 - Map of the greater Langkloof area indicating different agricultural sub-communities, primary land use activities, and towns or villages. Each coloured patch on the map indicates a different subcommunity. From Cockburn (2018), with permission.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
15
Unfortunately, farmers in the Langkloof area are increasingly exposed to environmental and social stressors (de Kock, 2015). The most severe environmental stressors are currently regular severe drought periods, leading to water scarcity, and increasingly severe climate events including floods, hail, wildfires, and heatwaves (ibid.). This greatly impacts farmers in the Langkloof through the destruction of harvest and reduced or poor-quality harvest (ibid.).
The Langkloof has high biodiversity value, characterised by a mosaic of fynbos, grassland and thicket vegetation, and high rates of indigenous species (Mander et al., 2010). The Valley Thicket, which also occurs in the Kouga catchment, is specifically of great importance since it hosts the highest incidence of rare and endemic plants of all thicket habitats in the world (East of the Cape: Conservering Eden., 2009).
Fynbos is by far the most abundant vegetation, because of the valley’s nutrient poor sandstone soils. The Fynbos Biome in the Kouga catchment is represented by 18 different fynbos vegetation classes (Veerkamp, 2013). Typical for fynbos is its dependence on fires for regeneration (Veerkamp, 2013). Finally, another important vegetation type in the Langkloof is wetlands, mostly palmiet wetlands. Wetlands have a strong ability to capture the currently scarcest resource in the Langkloof: water (Rebelo, Le Maitre, Esler, & Cowling, 2015). Sadly, however, a lot of wetlands were transformed to farms because of better the soil quality, higher nutrient concentrations, and a better water holding capacity (Veerkamp, 2013). Additionally, a lot of wetland is lost due to alien plant infestation (ibid.). Some dense, untouched smaller wetland vegetation still occurs along the Kouga river and its tributaries where there are not many roads (ibid.). Most wetland vegetation is currently situated around artificial water bodies such as water storage dams, irrigation furrows and drainages (ibid.).
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
16
Unfortunately, a lot of land in the Langkloof is degraded: in 2010, a report estimated that the Kouga and Kromme catchments together include 17471 hectares of degraded vegetation (Mander et al., 2010). A large part, approximately 4339 hectares, is thought to be degraded due to the infestation of alien invasive vegetation. More than 50 different exotic plant species are found in the Langkloof, but the most widespread species acacia mearnsii, commonly known as black wattle. (van de Witte, 2015; Veerkamp, 2013). This plant is known for its high rate of water consumption (Rebelo et al., 2015).
Over the past 50 years. there has been considerable land use change in the Kromme catchment (Rebelo et al., 2015). The most significant
changes were an increase in black wattle and farming areas (orchards, dryland, and irrigated farming). This has driven the loss of palmiet wetlands, which decreased by 84% and a decrease in riparian vegetation by 92% (fig. 6). Next to the effects of land use change on hydrology, reverse effects of hydrology on LUCC can be expected as well. For example, a decrease of the groundwater level may encourage farmers to grow crops that are less dependent on water. This causal relation from hydrology to land use in the Langkloof has not been studied yet. In return, land use changes affect these factors as well, the decline in wetlands due to an increase in irrigated land as found by Rebelo et al. (2015).
Figure 5 – Indigenous fynbos vegetation on the hills and pastures in the valley floor.
Figure 6 - Land-use/land-cover change (LUCC) in the Kromme river catchment over the past century. LUCC was categorized using aerial photography. From: Rebelo et al. (2015).
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
17
2.1.2. Living Lands: collaborations working on living landscapes
The NGO Living Lands has been working in the Langkloof since 2016. Their main goal is to establish ‘Living Landscapes’ (Living Lands, 2017a). The term ‘Living Landscape’ refers to a landscape which includes a variety of healthy ecosystems and land uses. Of importance here is the sustainable functioning of ecological, agricultural, and social systems in the area (Living Lands, 2017a). They are involved with multiple projects which contribute in restoring landscapes, bring water back to the farmers, communities living on the landscape, and the citizens of Port Elizabeth downstream (Living Lands, 2017a). Their mission is to bring synergy and collective action to the landscape through knowledge creation for Living Landscapes, mobilizing civil society, implementing landscape innovation and rehabilitation, enabling and facilitating social learning processes, and fostering mutually beneficial partnerships and participatory networks.
Living Lands implements landscape restoration through the 4 Returns approach (Living Lands, 2018b) which is developed by Commonland Foundation (Commonland, 2017). 4 Returns is a science-based framework which aims to restore degraded ecosystems by focusing on 4 key returns: a return of inspiration, a return of social capital, a return of natural capital, and a return of financial capital (ibid.). Furthermore, a 20-year commitment is given to a landscape, as this time is minimally needed for a landscape to recover (ibid.). The restoration is done in 3 zones: a natural zone, in which forest and nature restoration, a combined zone with mixed agriculture and nature, and an economic zone with sustainable production (ibid.).
In summary, it is clear that the Langkloof is a diverse landscape in terms of its nature and its residents. It accordingly faces a wide range of complex interconnected socio-economic, natural, and political challenges, of which many are direct or indirect drivers of LUCC change. In order to understand these challenges and their potential effects on the future land use and cover, an integrated perspective on the complex socio-ecological landscape is essential.
2.2. Methods
The conceptual model was created in a dynamic process of field observations, expert interviews, and literature study, and optimized during many rounds of feedback conversations with landscape experts. The field observations included the following activities:
• living & interacting with the landscape and its residents;
• attending a meeting on water restrictions by the Department of Water and Sanitation in the Eastern Cape;
• fieldwork with the WRC project team;
• collaboration on the creation of a movie on social engagement for the WRC project, which included 4 farmer interviews.
Expert knowledge was gathered in various manners: official interviews, group discussions, or e-mail contact in which feedback was given on the conceptual models. A list of the 12 experts consulted about the model can be found in table A1. ‘Landscape experts’ are defined here as an individual that has professional knowledge about the Langkloof, in some way relating to land use change.
Next to interviews on the conceptual model, three interviews with farmers were held to explore the personal drivers for land use decisions, three farmers were interviewed. These interviews followed the following method, designed to identify personal drivers around land use decisions. The question asked was: 'what would you do if you would get a field of land added to your farm?'. After they answered, a range of ecosystem services were shown on small paper. These words were shown:
- Voedsel (food)
- Grondstowwe (raw materials) - Varswater (fresh water)
- Medisinale hulpbronne (medicinal resources)
- Plaaslike klimaat en luggehalte (local climate and air quality)
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
18
- Ekosisteem (ecosystem) - Plant diversiteit (plant diversity) - Toerisme (tourism)
- Vrugbaarheid van die grond (fertility of the soil)
- Hout (wood) - Water (water)
- Ontspanning (recreation)
- Erosie beheer (erosion management) - Regulering (regulating)
- Geld (money) - Godsdiens (religion)
- Aanpasbaarheid teen die vloede (resilience against floods and droughts). Interviewees were asked to pick out 5 of these words which represented the motivation for their answer the most. They were also asked to rank them according to importance. Afterwards we asked them to explain why they picked out these words. These exact steps were repeated for the same question but instead of a piece of fallow land, we asked the farmers to imagine getting 'a piece of land full of black wattle'.
2.3. Results
The Langkloof is a highly complex landscape, with many natural, social, political, and economic factors driving land use change. It is therefore a challenge to capture all these complexities into a conceptual model. The iterative method with numerous rounds of feedback were used to capture and filter the most influential factors. A balance between completeness and comprehensibility was strived for. The resulting model of the drivers of land use change in the Langkloof and the connections between them can be found in Appendix 1. The four categories of the model (natural, socio-economic, political, and knowledge & innovation) are displayed and described in detail below. While they are described separately, it is clear from the full model that they are highly intertwined. It must be noted that these drivers are specific for land use change in the Langkloofregion, emphasizing change and the local scale. Factors that are more stagnantly important for land use in the Langkloof or driving factors of land use change from a larger scale are explained in the text but not included in the diagram.
2.3.1. Natural drivers
As an agricultural area, the welfare of the Langkloof is highly dependent on favourable natural conditions. Unfortunately, the Langkloof is increasingly subject to climate variability. Also alien invasive plants are a problem, especially because they consume a lot of water. Directly and indirectly, these factors are changing the land cover (Fig. 8).
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
19
2.3.1.1. Climatic drivers: rainfall, temperature, and extreme weather events
In the period 2006-2012, the Langkloof was affected by the worst drought in 134 years (de Kock, 2015). The region also suffered from a sequence of environmental stressors in that period, including floods, hail, wildfires, and heatwaves (ibid.). The environmental consequences of these stressors include the destruction of harvest and reduced or poor-quality harvest (ibid.). These events have more severe implications than the ecological damage alone; it has led to a loss of agricultural success followed by a loss of employment and entrepreneurial opportunities (ibid.). The combination of these shocks caused an estimated loss of R (Rand) 600 million. These losses are affecting the livelihoods of small-scale farmers more strongly than large scale farmers, as they have lower access to insurance and other financial resources to recover and are therefore less resilient. This is exacerbated with each extreme event (ibid.).
It is expected that the Langkloof will be increasingly challenged by environmental stressors due to the consequences of global climate change (Midgley, 2016). Specifically, an increase in temperature is expected, more frequent and longer drought periods, more intense rainfall and heavy floods, possible increases in hail and strong winds, and increasingly favourable conditions for wildfires (ibid.). Projections indicate an expected average increase in temperature of about 1-2 ○C for the period 2040-2060 (Fig. 9). This
analysis is based on the assumption of Representative Concentration Pathway 8.5, the most pessimistic of the Figure 8 - Natural drivers of land use change in the Langkloof, as identified through expert feedback process. * IAPs = invasive alien species, defined as pines and black wattle, **Natural area = wetlands, fynbos, riparian vegetation & forest, thicket., *** Due to a lack in experience - not in all cases.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
20
four greenhouse gas concentration trajectories adopted by the IPCC in its Fifth Assessment Report (IPCC, 2014).
Rainfall is predicted to become more variable, as predicted by this same model, with increases expected in some months of the year but decreases in others (Fig. 10). This was also shown on a South Africa-wide model from the Department of Environmental Affairs (DEA), which predicts anomalies in rainfall towards the end of the century (DEA, 2013). In the Langkloof specifically it can be seen that less rainfall is expected during autumn (March, April, May), and less during summer (June July, August) and spring (September, October, December). These anomalies are generally increasing over time (Fig. 11). This is of concern, because it decreases the predictability of rainfall which is crucial for successful farming, but also because it could mean there will be more destructive events, such as thunderstorms, floods, or severe droughts (DEA, 2013). Rainfall has shown to be directly related to agricultural success in South Africa, and horticulture especially is vulnerable to variability in water resources (Blignaut, Ueckermann, & Aronson, 2009). Therefore, it is of high importance for the Langkloof to build resiliency against water shortages.
Figure 9 - Temperatures in Joubertina (central Langkloof) as observed between 1981 and 2001, with projections of temperature changes towards the period 2040-2060, under the assumption of RCP 8.5 (Midgley, 2016)
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land use and land cover change in the Langkloof valley, South Africa
21
While the climate variability in the Langkloof generally has disadvantageous effects on farming practices, in some cases it can open new opportunities. For example, it was found that in many wine-making regions the generally rising temperatures worsen the conditions for grape ripening, but for other regions the predicted climatic changes could open up the opportunity for productive vineyards (G. V. Jones, White, Cooper, & Storchmann, 2005). In the Langkloof, a large horticultural company has started growing blueberries for the first time, as a strategy to adapt to the higher temperatures in the Langkloof (Eurofruit, n.d.).
Figure 11 – Projected median change in the average seasonal rainfall (mm) over South Africa for DJF, MAM, JJA and SON, for the period 2040-2060 (A) and 2075–2095 relative (B) relative to 1971–2005. Red shows negative change and blue shows positive change. Modelled according to RCP 8.5. The green circle indicates the Langkloof.
B
A
Figure 10 - Rainfall in Joubertina (central Langkloof) as observed between 1981 and 2001, with projections of rainfall changes towards the period 2040-2060, under the assumption of RCP 8.5 (Midgley, 2016).
22
2.3.1.2. Environmental drivers: alien invasive species and wetland degradation
A major natural driver of land use change in the Langkloof is the widespread and growing invasion of alien invasive plants (IAPs) (van de Witte, 2015). Most important IAPs in the Langkloof catchment are woody trees such as black wattle (acacia mearnsii, Fig. 12), and to a lesser degree pine species. It is estimated that 110.8 km2 of the Kouga catchment and 66.9 km2 of the Kromme catchment is
fully covered with black wattle, and 1.7 km2 and 3.3 km2 respectively is covered with invasive pine species
(Living Lands, 2018a). Black wattle grows most densely in river valleys and on mountain hills (van de Witte, 2015). It forms a major problem for
farmers in the Langkloof for many reasons. Black wattle competes for space and nutrients on farmland, leading to yield loss as well as a loss in biodiversity. Also, the trees consume much higher amounts of water than indigenous vegetation of the region, leading to a water scarcity, which is specifically tedious in the increasing number of droughts. It is estimated that the black wattle uses 200 mm more water than the indigenous wetland vegetation, which in turn uses approximately 260 mm more than the dryland fynbos (Rebelo et al., 2015). Acacia mearnsii also caused reduction in water flow of the Kromme catchment and increased vulnerability to extreme rainfall.
A related major environmental challenge in the Kromme catchment is the loss of wetlands. Black wattle, due to its high uptake of water, is mainly found in wet areas. Therefore, it has taken over large amounts of wetland areas. In combination with an increase of irrigated farmland, this has led to a 84% loss of wetlands in the past 50 years (Rebelo et al., 2015). A loss of wetlands has major effects on the Kromme catchment’s ability to retain water and vulnerability to extreme rainfall.
Both the clearing of black wattle and the restoration of wetlands, therefore, is necessary to retain the catchment’s water-related ecosystem services. According to Rebelo’s hydrological model implemented on the Kromme river catchment, rehabilitating floodplain wetlands could decrease flood intensity and increase average yield by up to 30%, roughly 1 million m3/yr (Rebelo et al., 2015). Moreover,
research in the Baviaanskloof, an area bordering the north of the Langkloof, found that the restoration of floodplain ecological infrastructure could reduce flood peaks by 14–20% and increase the average groundwater table level over the entire floodplain by 1 meter (Eurofruit, n.d.). Mander et al. (2017) found that the implementation of interventions to rehabilitate, maintain, and protect key ecological infrastructure could generate significant gains in water quantity in this catchment system; annual streamflow could be enhanced by over 11 million m3 per year, and base flow gains of over 42 million m3 per year. The study also showed that it makes economic sense to invest in ecological infrastructure as a complementary approach to supporting built infrastructure, as the cost/benefit ratio is lower than those of many built infrastructure measures, such as borehole installation and water transfer schemes.
Clearing of wattle is a tedious job, however, as the species has a very large seed bank of which a large proportion of may become dormant in the soil and seed may remain viable for more than 50 years (van de Witte, 2015). Therefore, many farmers cannot afford the time and effort needed to successfully clear the plants. An great contribution to clearing wattle is the Working for Water Programme which was founded in 1995 (“Working for Water (WfW) programme | Department of Environmental Affairs,” n.d.). This programme, led by the DEA, aims to control the spread of invading alien plants and reduce the impact of existing invasive alien plants. It does so through labour intensive, mechanical clearing and
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land
use and land cover change in the Langkloof valley, South Africa
23
chemical control. This leads to the creation of jobs, especially in local communities. WfW also partners with various governmental departments on the national and provincial level, research foundations, and private companies to enhance the capacity and commitment to solve IAP problems.
The South African government has launched more of these so-called ‘payment for ecological services’ programs. In the Langkloof, another active program is Working for Wetlands (“Working for Wetlands | Department of Environmental Affairs,” n.d.). The main driving force for this was the concern for continuing loss and degradation of wetlands to agriculture, as wetlands were considered valueless until recently. The growing recognition that wetlands provide invaluable ecosystem services, and that it is possible to rehabilitate them, provided the foundation for the programme.
The implementing agent of the Working for Water and Working for Wetlands programs in the Langkloof is the Gamtoos Irrigation Board (GIB) (“Gamtoos Water,” n.d.). Their project in the Kouga river is one of the oldest and largest WFW projects in the Eastern Cape. The project focuses on alien wattle, pine and eucalyptus treatment in the Kouga River and its tributaries. The GIB is also implementing alien clearing and wetland restoration in the Kromme river. Both the environmental restoration by the WFW programs and the job creation are important to the Langkloof region.
2.3.2. Socio-economic drivers
A landscape is formed by the people living on it, either directly through landowner’s decisions about the use of their land, or indirectly through the social dynamics and culture in the landscape which determine the development of the landscape on a higher level. Social and economic factors are highly intertwined, because the people on the landscape are driving the economy, and economic welfare drives the livelihoods of the people. These interdependencies are made clear in the socio-economic summary from the conceptual model (Fig. 12). This section will first explore the personal drivers of landowners in detail, then provide an overview of the broader social dynamics of all landscape residents, and finally describe the economic dynamics.
Figure 13 - Socio-economic drivers of land use change in the Langkloof, from an expert feedback process.* IAPs = invasive alien species, defined as pines and black wattle, **Natural areas = wetlands, fynbos, riparian vegetation & forest, thicket, *** Agricultural areas = orchards (apples, pears, stone fruit), honeybush, Lucerne, pasture.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land
use and land cover change in the Langkloof valley, South Africa
24
2.3.2.1. Landowner’s personal considerations
Interviewee 1 was a livestock farmer from the Suurveld region, interviewee 2 was a citrus and stone fruit farmer from the Kouga, and interviewee 3 was a livestock farmer from the Middel-Kromrivier. To the first question, the interviewee 1 responded that he would use it for natural grazing for his livestock. He argued that he would have to burn the land frequently, to prevent the renosterbos and fynbos from coming up and facilitate the seeding of the grass. He did mention that the location of the field is of high importance, because the land up in the mountains is less arable than the land in the valley. Interviewee 2 responded that he would do nothing with the land, because cultivating it would give him more to stress about, and he values his free time highly. Interviewee 3 would plant grazing on the land for cattle. The main limiting factor to do this, however, would be whether he would get water with it as well. Just like interviewee 1, he mentioned that if there is fynbos on the land, that is not so good for grazing. To the second question, the first interviewee responded that he would clear the wattle. He would not try to produce something out of the cut wattle, because that is not profitable according to him. The only opportunity he would see for the wattle would be furniture making.
The second interviewee would get a team of workers to clear the wattle if it would be growing upstream to his crops, because it would block the river. Since the water must go around it, it would cause erosion onto the land around it. If the wattle would be growing downstream, however, clearing would not be a priority and he would only take it out himself if he would have time and not hire a team to do it. The third interviewee agreed about the necessity to clear the wattle as well. He emphasized the importance of aftercare of the cleared area, because this prevents the wattle from coming back and allows other species to come up.
Table 1 - Answers of landowner interviews. Farmers were asked what they would do with a piece of either empty of fallow land, and to pick out and rank 5 ecosystem services which most strongly motivated their answers.
The ecosystem services that interviewees picked out to motivate these answers are listed in table 1. The most frequently mentioned word is water (5 times). This is not a surprise given the current drought in the region, and overall scarcity of water even in non-drought years. Interviewee 3 explained that you cannot do anything without water, and interviewee 1 explained that important to his
Interviewee 1 2 3
Question Empty land Invaded
land
Empty land Invaded land
Empty land Invaded land Ranking
1 Ecosystem Tourism Religion Water Recreation Resilience
against floods and droughts
2 Food Wood Resilience
against floods & droughts Fertility of the soil Water Water 3 Plant diversity Fertility of the soil
Water Money Erosion
control
Erosion control
4 Tourism Ecosystem Fertility of
the soil
N.A. Regulation Regulation
5 Fertility of
the soil
Water Plant
diversity
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land
use and land cover change in the Langkloof valley, South Africa
25
consideration not to use the fallow land is that he would need to use water for cultivation on that land, which means he has less water left for the rest of his land. Second most mentioned is fertility of the soil (4 times). Interviewee 1 explained that it is important to keep the natural fertility in the soil to maintain sustainability. Therefore, it is important to not use too many chemicals, because this is taking away the fertility. Interviewee 3 agreed with this and mentioned that it is also a waste to sow expensive seeds in a poor soil. Fertility of the soil is also an important motivation for removing wattle, as wattle is drinking all the water (interviewee 3), and it grows on the most arable areas anyhow, so that is also the most favourable areas to clear (interviewee 1). Notable is also that money is only mentioned twice, and both times on the fifth place. The first interviewee explained that farmers are not just trying to earn more money; there is a growing population so a growing demand for food, and farmers feel the responsibility to meet this demand. Furthermore, it is also clear that the variety of motivations is big; amongst the highest ranked words, not one word is mentioned twice.
Since recently, social engagement is organised by Living Lands through a research project supported by the Water Research Commission, which has as the aim of modelling the hydrology of three large catchments, the Baviaanskloof, Kouga, and Kromme (BKK), through a participatory process (Living Land; referred to in this study as the WRC project). For the participatory part of the project, they engage actively in understanding stakeholder’s priorities, visions, and interests, as well as sharing knowledge and social learning in regular workshops, learning journeys, and individual dialogue interviews. Bosch (2018) evaluated the social engagements in the Langkloof through dialogue interviews with participants during and around these workshops. Relating to visions on land use, she found that the water shortage was unanimously regarded as the biggest challenge in the Langkloof. However, no consensus was found on the cause of this problem. Answers included climate change, the spread of IAPs, anthropogenic water demand, social challenges such as poverty and unemployment, and the role of the government. All these suggestions are included in the driver diagram (Fig. 13).
2.3.2.2. Social dynamics of non-landowners
In the previous section, the perspectives of the landowners were mapped out. This was relevant because their visions and priorities are directly related to the land use pattern in the Langkloof. The social dynamics amongst non-landowners are equally important, because they highly intertwine with important driving factors of land use change. They also represent a much larger part of the population; 85.5 % of the population in the Kouga and 71.3% of the population in Kou-Kamma does not live on a farm (Statistics South Africa, n.d.).
Poverty and unemployment are still highly prominent in the Langkloof. In 2011 15,6% of the KLM citizens earned less than R9,600, which is close to the national Upper-Bound Poverty Line (UPBL)2 of that
year (R9348). This same survey disclosed an unemployment rate of 15% and youth unemployment of 17.5% (Statistics South Africa, n.d.). This is slightly higher than the unemployment rate in the whole Eastern Cape in 2011 (24,8%, Statistics South Africa, 2018), because the agricultural activity brings along work opportunities (Cockburn, 2018). However, this also means that job opportunities are highly dependent on agriculture. It was found that the most common loss of jobs for seasonal workers on fruit farms in the Langkloof are cost-cutting measures by farmers who suffered from reduced production
2 The UBPL refers to the food poverty line plus the average amount derived from non-food items
of households whose food expenditure is equal to the Food Poverty Line (FPL). The FPL refers to the amount of money that an individual will need to afford the minimum required daily energy intake (Statistics South Africa, 2018a).
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land
use and land cover change in the Langkloof valley, South Africa
26
(Disaster Mitigation for Sustainable Livelihoods Programme, 2012). One problem related to this is the seasonality of agricultural work (Cockburn, 2018). Another is the effects of the droughts. In 2009-2010, many people migrated from the Langkloof to the George area due to the loss of jobs in drought-affected orchards (Disaster Mitigation for Sustainable Livelihoods Programme, 2012).
A powerful solution to the unemployment is provided by ‘payment of ecological services programs’, such as Working for Water and Working for Wetlands as described in section 2.3.1.2. One of the primary objectives of the Working for Water programs is poverty alleviation, for which it receives substantial political support (Turpie, Marais, & Blignaut, 2008). Each year, the programs create thousands of jobs to previously unemployed people in South Africa (Turpie et al., 2008).
2.3.2.3. Market dynamics
The choice of land use or agricultural commodity is strongly dependent on the market opportunities for the product. For this reason, it is important to examine the agricultural outlook of the most prominent agricultural commodities in the Langkloof, namely fruit orchards, livestock, and honeybush cultivation. An outlook for South Africa was made by the Bureau for Food and Agricultural Policy (BFAP) (BFAP, 2018).
The Langkloof fruit industry is highly export oriented (HORTGRO & BFAP, 2018). There has been an increase in competition globally, especially from Southern hemisphere role-players Argentina, Chile, and Peru, as well as the expansion of the pome fruit industry in former Eastern European countries and Russia (HORTGRO & BFAP, 2018). According to HORTGRO, a Knowledge group for South Africa's deciduous fruit industry, and the Bureau for Food and Agricultural Policy (BPAP), there will be increased competition for scarce resources (land, water, capital) in the future, and an increased focus on water management practices, as the efficiency and productivity of water needs to be improved (HORTGRO & BFAP, 2018).
While domestic consumption of livestock products is expected to increase, the livestock industry of South Africa is very dependent on export prices as well (BFAP, 2018). Export prices of meat are expected to drop a little after 2018, due to increased global meat supply. After 2020, however, prices will rise again due to a growing population in Asia and Africa as well as a growing per capita meat consumption globally (BFAP, 2018).
Finally, the market dynamics of honeybush cultivation in the Langkloof is very different from orchards and livestock, because there is no competition from other countries (DAFF, 2016). It is similar in most of the product is being exported: 75 % according to a recent report (McGregor, 2017). The export market for honeybush is also expanding. This has led to an increase in the export tonnage of honeybush tea, in particular from 2010 onwards, as can be seen in figure 14.
Roos van der Deijl | MSc. Earth Science thesis: A scenario-based spatial model of land
use and land cover change in the Langkloof valley, South Africa
27
Figure 14 - Export tonnage and prices of Honeybush tea in South Africa from 1999-2017. From: (McGregor, 2017).
2.3.2.4. Global and national drivers
There are some socio-economic factors which are highly relevant for agricultural success and farmer’s decisions about land use, but which are not included in the drivers diagram, because they relate to the national or global processes, while the diagram relates to drivers on the Langkloof-level, but. Therefore, a summary of these factors and their recent developments are described below.
