Elsevier Editorial System(tm) for Ecosystem...

Elsevier Editorial System(tm) for Ecosystem Services Manuscript Draft Manuscript Number: ECOSER-D-14-00195R1 Title: Multi-Criteria Decision Analysis to identify dryland ecosystem service trade-offs under different rangeland land uses Article Type: Full Length Article Keywords: Sustainable Land Management, land degradation, Kalahari, benefit transfer, southern Africa, Multi-Criteria Decision Analysis Corresponding Author: Dr. Nicola Favretto, Corresponding Author's Institution: United Nations University (UNU-INWEH) First Author: Nicola Favretto Order of Authors: Nicola Favretto; Lindsay Stringer; Andrew Dougill; Martin Dallimer; Jeremy Perkins; Mark Reed; Julius Atlhopheng; Kutlwano Mulale Abstract: Land degradation undermines ecosystem service provision, limiting economic returns from semi-arid rangelands. We apply a Multi-Criteria Decision Analysis (MCDA) to assess the value of ecosystem services in semi-arid rangelands using both monetary and non-monetary techniques. We do this in the Kalahari rangelands of Kgalagadi District, southern Botswana, where concerns over poverty are growing for pastoral communities due to increasing levels of land degradation. We provide an empirical contribution to understanding the linkages between policy, land use and the provision of ecosystem services in drylands based on the perspectives of local stakeholders identified through interviews and a workshop consultation. Findings suggest communal grazing provides the widest range of monetary and non-monetary values linked to ecosystem service delivery. Current economic incentives and policy initiatives supporting the livestock sector, linked to fencing and borehole drilling, create perverse incentives that over-emphasise commercial food production at the expense of other services. We identify a need for policy reforms to support livelihood diversification through the provision of a wider range of ecosystem services, and for further research to explore potential market opportunities for veld products and carbon trading. MCDA offers a useful holistic assessment framework that could be applied more widely to semi-arid rangelands globally. Suggested Reviewers: Jasper Kenter Lecturer in Ecological Economics, University of the Highlands and Islands [email protected] Dr Kenter has a relevant expertise on the use of deliberative and participatory approaches to valuing ecosystem services, using both monetary and non-monetary tools Chasca Twyman Senior Lecturer, University of Sheffield [email protected] Dr Twyman is an expert of natural resource management (water, ecosystems and wildlife) and policy in Southern Africa. She has a PhD from the University of Sheffield focused on natural resource management in Kalahari, Botswana.

Stacey Noel Centre Director for SEI Africa and Research Fellow, Stockholm Environment Institute [email protected] The work of SEI in Africa led by Stacey Noel is highly relevant to the research presented in this paper. Stacey Noel's research focuses on ecosystem services, water, and livelihoods. Michael Christie Professor of Environmental and Ecological Economics, Aberystwyth University [email protected] Prof Christie is Executive Member of the UK Network of Environmental Economics and the Ecosystem Service Partnership, and Associated Editor of the Ecosystem Services journal. He is specialised in the economic and social valuation of biodiversity and ecosystem services. Patrik Klintenberg Senior Lecturer, Mälardalen University [email protected] Response to Reviewers: Please find the Detailed Response to Reviewers attached to the bottom of this submission.

175 Longwood Road S. Telephone +1 905 667 5511 Suite 204 Fax +1 905 667 5495 (direct) Hamilton, Ontario L8P 0A1 E-mail [email protected] Canada Web www.inweh.unu.edu

Cover letter

To: Ecosystem Services

19th June 2015 Dear Editor, Resubmission of ECOSER-D-14-00195 I would like to submit a revised version of this manuscript for consideration in Ecosystem Services. We have considered the referee comments and addressed each fully, as described in our Response to Reviewers, submitted with this revised version of our manuscript. A version of the revised manuscript with track changes is available on request if you would like to see all the changes we made in more detail. I hope that these revisions will satisfy you that this is a highly original piece of work, which will have a significant impact. If you have any questions about this manuscript, please don't hesitate to get in touch. With kind regards,

Yours sincerely,

Dr Nicola Favretto

Cover Letter

Highlights (Resubmission of ECOSER-D-14-00195)

Key rangeland ecosystem services in southern Botswana's Kalahari are identified

Socio-economic, ecological & policy dimensions of land use are integrated using MCDA

Communal grazing delivers the widest range of monetary and non-monetary values

Policy priorities to promote sustainable land management are identified

*Highlights

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

1

Abstract

Land degradation undermines ecosystem service provision, limiting economic returns from

semi-arid rangelands. We apply a Multi-Criteria Decision Analysis (MCDA) to assess the

value of ecosystem services in semi-arid rangelands using both monetary and non-

monetary techniques. We do this in the Kalahari rangelands of Kgalagadi District, southern

Botswana, where concerns over poverty are growing for pastoral communities due to

increasing levels of land degradation. We provide an empirical contribution to

understanding the linkages between policy, land use and the provision of ecosystem

services in drylands based on the perspectives of local stakeholders identified through

interviews and a workshop consultation. Findings suggest communal grazing provides the

widest range of monetary and non-monetary values linked to ecosystem service delivery.

Current economic incentives and policy initiatives supporting the livestock sector, linked to

fencing and borehole drilling, create perverse incentives that over-emphasise commercial

food production at the expense of other services. We identify a need for policy reforms to

support livelihood diversification through the provision of a wider range of ecosystem

services, and for further research to explore potential market opportunities for veld

products and carbon trading. MCDA offers a useful holistic assessment framework that

could be applied more widely to semi-arid rangelands globally.

Keywords

Sustainable Land Management, land degradation, Kalahari, benefit transfer, southern

Africa, Multi-Criteria Decision Analysis

*ManuscriptClick here to view linked References

http://ees.elsevier.com/ecoser/viewRCResults.aspx?pdf=1&docID=646&rev=1&fileID=20557&msid=E2DEF023-0287-4B4B-AAE6-934D48B8A630

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

2

1. Introduction

Drylands cover approximately 40% of the world’s terrestrial surface, with 15% of this area

threatened by land degradation (MA, 2005). Land degradation undermines the provision of

a range of ecosystem services (ES) for billions of people who depend on these

environments for their livelihoods and subsistence, and is driven by a variety of socio-

economic, political and environmental factors (Foley et al., 2005). The United Nations

Convention to Combat Desertification (UNCCD, 1994) provides an international policy

framework for countries to tackle problems of dryland degradation and desertification. At

the national level, country parties to the UNCCD are obliged to develop National Action

Programmes (NAPs) to outline the national status of land degradation, and provide a

Sustainable Land Management (SLM) strategy to address the problem (Stringer et al.,

2007). SLM is a situation in which land users are able to meet their needs and derive

socio-economic benefits through the use of land resources, while simultaneously ensuring

the long-term productive potential of these resources and the maintenance of their

environmental functions (WOCAT, 2010). In addition to NAPs, it is important to develop

sector-specific policies that treat land degradation as a cross-cutting issue that can be

mainstreamed across sectoral public policies (Akhtar-Schuster et al., 2011). This

recognition of the cross-cutting nature of land degradation has led to demands for the

integrated assessment of different kinds of dryland uses and management, and for

information that can help policy makers to prioritise actions to enhance dryland ES delivery

(and/or avoid ES losses), and promote SLM within decision making. Efforts to bridge key

information gaps in economic decision-making are being led by the World Bank through

the Wealth Accounting and the Valuation of Ecosystem Services (WAVES) programme,

which is establishing an environmental accounting system across Africa’s drylands,

including in Botswana and Rwanda (WAVES, 2014). However, more integrated, holistic

methodologies that bring together socio-economic, environmental and policy dimensions

are required, incorporating those aspects that have traditionally been quantified with those

that have not (Bateman et al., 2013; Costanza et al., 2014).

Various analytical frameworks have been developed for quantifying and valuing ES (e.g.,

Bagstad et al., 2013), which have provided useful information for the public and policy

makers (e.g., TEEB, 2010). For example, the monetary value of ES can be conceptualised

as the way in which they contribute to different elements of the ‘Total Economic Value’. ES

may increase an individual’s welfare through direct provision of goods (e.g., food or

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

3

recreational use), or indirectly through their provision of benefits such as the regulation of

water and carbon cycles (TEEB, 2010). People may also value ES for their non-use (also

termed “passive-use”) benefits. However, these conventional monetary approaches to

valuation have been criticised for an inability to capture shared values, which people hold

for others and the communities and society in which they live (Kenter et al., in press). This

presents the need to integrate these various dimensions through mixed-method

approaches that combine deliberative-based techniques, where varied stakeholders’

perspectives are developed. Fish et al. (2011) identify Multi-Criteria Decision Analysis

(MCDA) as an effective decision making tool to evaluate the non-monetary and monetary

costs and benefits of different management options. Similarly, MCDA has been applied as

a central mixed-valuation method in the UK National Ecosystem Assessment. It therefore

offers a useful integrative approach that allows cultural and shared values related to ES to

be assessed in a systematic way based on key socio-economic, policy and environmental

priorities (Kenter et al., 2014).

Following the de Groot et al. (2010) ES classification, we present an analytical framework

that utilises MCDA to identify the multiple monetary and non-monetary dimensions of land

use and management in southern Kgalagadi District, Botswana, where rangelands deliver

a variety of ES that underpin livelihoods (Chanda et al., 2003). We identify, value and

score ES benefits provided by four types of land management: i) private (fenced) cattle

ranching, ii) (unfenced) communal livestock grazing, iii) (private) game farming and iv)

Wildlife Management Areas (WMAs). We then discuss the costs and trade-offs associated

with ES delivery under each of these options. Our approach provides decision makers with

a valuable analytical example that can be used to better understand ES provision across

semi-arid rangelands, while findings can be used to inform measures that could reduce the

degradation of particular ES and advance SLM.

2. Materials and Methods

2.1. Study area

We draw upon data collected during 2013-2014 along an east-west transect of southern

part of Kgalagadi District, Botswana (Figure 1), which incorporates a total area of c. 66,000

km2 (Government of Botswana (GoB), 2003) and an estimated human population of

30,000 (GoB, 2012a). Concerns over poverty and land degradation are growing for dryland

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

4

communities in the District and rangeland degradation has led to extensive bush

encroachment (Reed and Dougill, 2010; Thomas and Twyman, 2004); reducing good

quality grazing and increasing rural poverty levels (Chanda et al., 2003), resulting in

degradation being a matter of national concern. Land uses include communal grazing

areas (unfenced cattle posts) (c. 14,800 km2), privately owned (fenced) cattle ranches (c.

8,900 km2), private game ranches (c. 800 km2) and Wildlife Management Areas (WMAs)

(c. 14,800 km2) designated as protected conservation areas around the National Parks

(Kgalagadi Transfrontier Park, c. 26,700 km2).

Figure 1: Land use of Kgalagadi District, southern Botswana and study sites Source: adapted from KGLB (2013)

2.2. Methods

We use MCDA as a framework that allows monetary-based techniques to be integrated

with non-monetary ecological and shared values (de Groot et al., 2010; Kenter et al.,

2014). It allows us to rank alternative options by quantifying, scoring and weighting a range

of quantitative and qualitative criteria, and has been previously successfully applied in

landscape planning research to assess varied land use alternatives in relation to ES

provision (Fontana et al., 2013). Scoring was undertaken by the project team (composed

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

5

of national and international researchers with expertise in land policy, livelihoods, ES

valuation, land degradation assessment, range ecology, geomorphology and

environmental economics), while weighting was undertaken in consultation with

stakeholders from a range of institutions, including government ministries and

departments, and NGOs.

The steps followed and data used in the MCDA development are presented below.

2.2.1. Problem definition

The research problem to be tackled was defined as: "Which land uses and land

management strategies are best placed to generate the widest range of economic and

non-economic values linked to specific ES delivered in Kalahari rangelands in southern

Kgalagadi District, Botswana?"

2.2.2. Identification of options, criteria definition and assessment

Four land uses were defined as MCDA options: i) communal livestock grazing; ii) private

cattle ranches; iii) private game ranches; and iv) WMAs. These options include all the

different land uses in the study area, which are also typical of semi-arid rangelands

globally. The performance of the land use options was measured by their capacity to

deliver the identified ES in 2013. Nine criteria were identified, supported by 14 indicators

(Table 1).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

6

Table 1. Criteria and indicators used to assess capacity to deliver ecosystem

services in Kgalagadi District, southern Botwana, as assessed through MCDA

Criterion (ecosystem

service) Indicator Type of measurement

Food (commercial)

Net profit of meat production (US$/ha/yr)

Stocking level (Ha/Livestock Unit)

Quantitative-monetary and Quantitative-non-monetary

Food (wild) Gathering of veld products

Subsistence hunting

Qualitative

Fuel Firewood collection Qualitative

Construction material

Collection of thatching grass and poles for fencing

Qualitative

Groundwater Value of water extracted (US$/ha/yr) Quantitative-monetary

Plant and livestock diversity

Species and genetic diversity between forage species

Genetic diversity between livestock breeds

Qualitative

Climate regulation

Value of carbon sequestration (US$/ha/yr)

Quantitative-monetary

Recreation

Revenues from Community Based Natural Resource Management, trophy hunting & photographic safari (US$/ha/yr)

Ecotourism potential

Wild animal diversity

Quantitative-monetary and qualitative

Cultural/Spiritual benefits

Presence of landscape features or species with cultural/spiritual value

Qualitative

Using the ecosystem service categorisations as criteria underplays the values associated

with the interactions between ES across both the District and wider Kalahari context,

where mobility, links and flows, of both wildlife and water, shape the delivery of other ES.

However, this approach facilitates the application of both quantitative and qualitative

measurements. Similarly, we recognise that use of the farm scale as the unit of analysis

does not allow assessment of the aggregate interactions of land management options

across the landscape as a whole. It does however provide a deep understanding of ES

delivery at the scale at which land use decisions are made.

Primary and secondary data were gathered using mixed methods. A comprehensive

literature review and secondary data (including rainfall data, land tenure information,

national economic statistics and population data) allowed us to gain a picture of the socio-

economic and ecological characteristics of the study area. Where data were lacking, the

benefit transfer method was used. This allowed valuation estimates for the ‘study site’ to

be adapted to those obtained by other studies (‘policy sites’) for the same ES in similar

environments (Richardson et al., 2014).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

7

A total of 37 semi-structured interviews (see S1) were carried out across 8 study sites.

Criteria for respondent selection were based on relevant expertise (in the case of

government officers and village development committees) and ownership of cattle and/or

game. The sample comprised: communal livestock farmers (n=20), private cattle ranchers

(n=10), private game ranchers (n=3), government officers (n=3) and village development

committee leaders (n=1). Quantitative data were collected to investigate the monetary

costs and gains from land use activities noted in interviews and included examination of

financial statements where available. These data informed the MCDA criteria assessment

and the benefit transfer study site economic characteristics. Qualitative information

gathered in interviews also covered the different land management strategies adopted and

their main implications for land degradation and ES provision. Policy analysis allowed

identification of priorities in different sectors and the economic mechanisms currently used

to advance their implementation. In addition, 12 ecological assessments were carried out

across the study region to verify assumptions made on land use and patterns of ecological

change / degradation. The following points detail the methods and data used for the

assessment of each criterion.

Criterion 1: Food (commercial). The mean net profit of the meat production indicator was

derived by subtracting the annual operating expenses under each land use option from the

total operating revenues. Values provided by private cattle and game farms through

financial statements1 were measured for different fiscal years within the period 2010-2013

and compared after being deflated to 2013 prices using the Botswana Consumer Price

Index (CPI). Private operating expenses included feeds and medicines, motor vehicles and

transport, fuel and oil, electricity, repairs and maintenance, and included salaries and

wages used for bush removal. In the case of communal grazing areas, revenues and

expenses were calculated based on interview data integrated with literature values (GoB,

2007). Revenues included the average livestock herd size owned (cattle and smallstock;

assessed through interview data) multiplied by the mean off-take rate (identified in GoB,

2007) and valued according to the 2013 market price. Expenditure identified from interview

data included the mean cost of borehole drilling (with a 10-year depreciation period),

borehole equipment and maintenance, kraals (enclosures for livestock), feed and

medicines, fuel and labour. Minimum and maximum expected profit values were estimated

by calculating a 95% confidence interval of the standard deviation of the profit mean with

1 2013 Exchange rate applied: Pula/US$ = 0.11

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

8

uncertainty expressed by the standard error. The stocking level non-monetary indicator

provided the mean stocking values, assessed through interviews and analysis of

secondary data contained in GoB (2007).

Criterion 2 (Food, wild), Criterion 3 (Fuel) and Criterion 4 (Construction material). The

indicators ‘gathering of veld products’, ‘subsistence hunting’, ‘firewood collection’ and

‘collection of thatching grass and poles for fencing’ were qualitatively assessed using a 5-

point scale (1 very low, 5 very high) according to information gained through interviews

and from previous studies in this area that focused explicitly on community use of wild

(veld) products, fuel and construction material (e.g., Thusano Lefatsheng Trust, 2005;

Twyman, 2000).

Criterion 5: Groundwater. The economic value of the groundwater extracted was estimated

based on interview data. The water is often highly saline and is used mostly for watering

cattle. The average number of boreholes used per ha under each land use was calculated

and multiplied by the borehole's extraction capacity (L/hr). Total L/ha/yr of water extracted

was derived by multiplying the latter value by 365 days and assuming a daily pumping time

of 16 hours (from interview data). The result was valued according to the 2006 market

price of non-potable (raw) water (GoB, 2006) deflated to the real 2013 price. A 95%

confidence interval was calculated to provide minimum and maximum expected values,

together with its standard error.

Criterion 6: Plant and livestock diversity. This criterion classifies biodiversity (the variability

among living organisms including genetic diversity) as a final ES (de Groot et al., 2010;

Mace et al., 2012). In line with this view, the two indicators of biodiversity (forage species

and livestock breeds) indicate the capacity to ensure resilient food production against

future climate change and or diseases. These indicators were assessed using a 5-point

scale (1 very low, 5 very high) according to the findings of the ecological assessments at

each study site (see Dougill et al., 2014) and from interviews which quantified the livestock

breeds present at each site.

Criterion 7: Climate regulation. The monetary value of net carbon (C) sequestration was

assessed through the benefit transfer method (Richardson et al., 2014). Above ground

biomass (vegetation) C storage estimates were based on the project’s field ecological

studies (Dougill et al., 2014) rather than by assuming bush encroachment scenarios used

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

9

in economic assessments nationally (e.g., Arntzen et al., 2010; Turpie et al., 2006).

Previous scenarios assume that soil C gains through photosynthesis in the lightly grazed

WMA scenario are close to being balanced by C losses (through respiration and in fire

events) with only low levels of soil C accumulation, typical of Kalahari sand soils (Dougill

and Thomas, 2004). Net amounts of total soil C sequestered per annum were estimated

using recent studies in southern Botswana (Thomas, 2012). Three scenarios were used:

intense grazing (communal livestock grazing and private cattle ranches) where net gains

were identified as 0.25 t C/ha/yr, light grazing (game ranches) where net gains are 0.20 t

C/ha/yr and very light grazing (WMAs) where gains are 0.05 t C/ha/yr, with the higher C

storage at intensively grazed sites resulting from bush encroachment leading to higher

above ground biomass C storage. These figures were multiplied by the total land surface

under each use and valued according to the 2013 C price set in the European Union

Emissions Trading Scheme (EU ETS) (US $6.7/t).

Criterion 8: Recreation. Revenues from Community-Based Natural Resource Management

(CBNRM) gained through trophy hunting and/or photographic safaris for the controlled

hunting area in the WMA were assessed by examining the 2006 hunting agreement

provided by the Khawa Kopanelo Development Trust. These values validate the CBNRM

estimates from another study based in southern Botswana (Madzwamuse et al., 2007).

Real 2013 prices were calculated using the Botswana CPI. The ecotourism potential

indicator was assessed using a 5-point scale in interviews, as well as the WMA land use

management plan and a local development plan designed by a private safari operator. The

third indicator, wild animal diversity, considers biodiversity as a good in itself (Mace et al.,

2012). It was assessed using a 5-point scale developed by using the 2012 Aerial Census

of Animals in Botswana (GoB, 2012b) that maps the spatial distribution of the main 26 wild

animal species.

Criterion 9. Cultural/Spiritual benefits. The same qualitative methods for the assessment of

the indicator ‘presence of landscape features or species with spiritual value’ were used as

for criteria 2, 3 and 4 using previous studies from Arntzen et al. (2010) and Madzwamuse

et al. (2007).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

10

2.2.3. Criteria weighting

Each of the criteria were weighted to reflect their relative importance to society for the final

ranking. Weights were assigned based on the rating technique (Hajkowicz et al., 2000),

with scores obtained in a policy workshop held in Botswana in 2014. Twenty-eight

invitations were sent to major local experts in the field of land degradation, as identified by

the research team. Individual scores were provided through a questionnaire administered

to the 15 stakeholders who attended the workshop. This is a similar number of responses

used in other studies (e.g., Fontana et al., 2013; Hajkowicz et al., 2000). Stakeholders

were distributed among relevant sectors affected by land degradation, including agriculture

and water (n=6), wildlife and forestry (n=6), as well as the cross-cutting sectors of statistics

(n=2) and energy (n=1). Ratings were obtained on a 9-point scale ranging from most

important (9) to least important (1) criteria. The geometric mean was used to aggregate

the individual priorities of each stakeholder into a single representative weight for the

entire group. The weights for the criteria were then normalised so that their total was equal

to 1.

2.2.4. Derivation of overall preference scores for each land use option

The quantitative and qualitative criteria were scored on a homogeneous 100-point scale. A

score of 0 represented the worst level of performance encountered, and 100 represented

the best (as per UK Government (GoUK), 2009). End points were established for all criteria

based on the criteria assessment. A linear value function was used to translate the

measure of achievement of each criterion (j) into a MCDA value score (0-100). A

homogenous score range (i.e., using the same end points across different criteria) was

used for all the monetary criteria, in order to reflect equal variations of MCDA score as a

result of a variation of US $/ha/yr 1 in one given monetary criterion. For each land use

option (i), each criterion score (sij) was multiplied by the criterion's weight (wj). The land

use options' overall preference score (Si) was derived by summing these products for all

the criteria (n) under each land use option (Eq. 1).

𝑆𝑖 = ∑ 𝑤𝑗 𝑠𝑖𝑗

𝑛

𝑗=1

(1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

11

3. Results

The performance of each criterion under the different land use alternatives is detailed in

Table 3, alongside the type of valuation and data collection method(s) used to inform the

MCDA.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

12

Table 3. Criteria performance for four land use types in Kgalagadi District, southern Botswana, as assessed through the MCDA


service)

Indicator / ES category (de Groot et al., 2010)

Communal livestock grazing

Private cattle ranches

Private game ranches

WMAs Valuation/collection

methods used to inform the MCDA

Provisioning services

Food (commercial) Net profit of meat production (US$/ha/yr)

Mean: 0.64 (-0.56 ; 1.95)*

(0.64)**

Mean: 1.21 (0.66 ; 1.75)*

(0.16)**

Mean: -2.07 (-7.89 ; 3.75)*

(1.33)**

0

(0.0)**

Interviews & market prices

Stocking level (Ha/LSU) Mean: 11 (9 ; 13)

Mean: 14 (8 ; 20)

Mean: 9.5 (7 ; 12)

Mean: 160 (120 ; 200)

Interviews & literature

Food (wild) Gathering of veld products High Low Low Medium


Subsistence hunting High Very low Very low Very high


Fuel Firewood collection Very high Medium Medium High




Very high Medium Low High Interviews &

literature

Groundwater Value of water extracted (US$/ha/yr) Mean: 0.84 (0.63 ; 1.05)*

(0.21)**

Mean: 0.97 (0.22 ; 1.71)*

(0.37)**

0.15

(0)**

0

(0)**

Interviews & market prices



Low Medium High Very high Ecological

assessments


Low High Very low Low Interviews

Regulating services

Climate regulation Value of carbon sequestration (US$/ha/yr)

1.7 1.7 1.3 0.3 Benefit transfer &

market prices

Cultural services

Recreation Revenues from CBNRM trophy hunting and photographic safari (US$/ha/yr)

0 0 0 0.04 Interviews

Ecotourism potential Low Very low High Very high Interviews

Wild animals diversity Medium Very low Very high Very high Literature

Cultural/ Spiritual benefits

Presence of landscape features or species with cultural/spiritual benefits

Very high Very low Medium Very high Interviews

* 95% confidence interval of the standard deviation of the mean ** Standard error

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

13

Table 4 details the weights attributed to each criterion. The normalised values indicate

groundwater ranks highest, followed by commercial food production, plant and livestock

diversity, wild food, fuel and construction material. Climate regulation, recreation and

spiritual inspiration were ranked lowest.

Table 4. Criteria weighting based on the rating technique with inputs from policy stakeholders

Criteria Ground water

Food1

Plant & livestock diversity

Food (wild) Fuel

Constr.

material2

Climate reg.

3

Recreation Spiritual4 Total

Mean weight (normalised) 0.18 0.17 0.15 0.12 0.11 0.10 0.08 0.06 0.03 1.0

Mean weight (non-

normalised) 7.6 7.0 6.2 4.9 4.4 4.1 3.3 2.6 1.4 -

Standard deviation

1.2 1.8 1.3 1.5 1.6 1.4 1.9 1.8 1.8 -

Abbreviations: 1commercial,

2construction material,

3climate regulation,

4cultural/spiritual benefits

The final scores (quantitative and qualitative criteria ranked on a 100-point scale) are

outlined in Table 5. The weighted values (single land use options’ scores multiplied by

normalised criterion weight) are indicated in brackets.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

14

Table 5. Final scoring of the MCDA (scale 0-100) (weighted values in brackets)


service)

Indicator / ES category

Communal livestock grazing

Private cattle

ranches

Private game

ranches WMAs

Provisioning

Food (commercial)

Net profit of meat production 73 (12) 78 (13) 50 (9) 68 (12)

Stocking level 98 (17) 96 (16) 99 (17) 20 (3)

Mean 86 (15) 87 (15) 75 (13) 44 (7)

Food (wild) Gathering of veld products 75 (9) 25 (3) 25 (3) 50 (6)

Subsistence hunting 75 (9) 0 (0) 0 (0) 75 (9)

Mean 75 (9) 13 (2) 13 (2) 63 (8)

Fuel Firewood collection 100 (11) 50 (6) 50 (6) 75 (8)



100 (10) 50 (5) 25 (3) 75 (8)

Groundwater Value of water extracted 75 (14) 76 (14) 69 (12) 68 (12)

Plant & livestock diversity


25 (4) 50 (8) 75 (11) 100 (15)


25 (4) 75 (11) 0 (0) 25 (4)

Mean 25 (4) 63 (9) 38 (6) 63 (9)

Regulating

Climate regulation

Value of carbon sequestration 82 (7) 82 (7) 79 (6) 70 (6)

Cultural

Recreation Revenues from CBNRM trophy hunting and photographic safari

67 (4) 67 (4) 67 (4) 68 (4)

Ecotourism potential 25 (2) 0 (0) 75 (5) 100 (6)

Wild animals diversity 50 (3) 0 (0) 100 (6) 100 (6)

Mean 25 (2) 0 (0) 58 (4) 81 (5)

Cultural/ Spiritual inspiration

Presence of landscape features or species with spiritual value

100 (3) 0 (0) 50 (2) 100 (3)

TOTAL (weighted)

(67) (52) (39) (49)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

15

The performance of each land use option with regard to its capacity to deliver the range of

ES under consideration is summarised in Figure 2. Communal livestock grazing emerged

as the land use that provides the greatest benefit to society as it achieved the highest

MCDA score, followed by WMAs, private cattle ranches and private game ranches. High

scores achieved by communal grazing areas are mainly linked to their use for commercial

food production, with management practices also allowing wild food production, fuelwood

collection, construction material provision, climate regulation and cultural/spiritual non-use

values to be retained.

Figure 2. Weighted performance of ecosystem service delivery under four alternative land uses in Kgalagadi District, southern Botswana, as assessed

through the MCDA

We recognise that the society’s perspectives in this study are extrapolated from a relatively

small panel of people. This data were combined with extensive secondary data and

judgement of experts, to reduce the risk that different results might have been generated if

the MCDA analysis drew on a different panel of individuals. In order to demonstrate

confidence in our results, sensitivity analysis was carried out as an important step to check

the robustness of the analysis. This involved several iterations, where different weightings

and scores were applied in order to assess possible deviations in the final MCDA ranking

against the results in Figure 2. This included doubling the weighting of one or multiple

criteria at one time under different scenarios, and also doubling or halving the MCDA score

0

10

20

30

40

50

60

70

80

Communallivestockgrazing

Private cattleranches

Private gameranches

WMAs

Tota

l (w

eig

hte

d)

valu

e o

f ES

Type of land use

Cultural/Spiritual benefits

Recreation

Climate regulation



Fuel

Groundwater

Food (wild)

Food (commercial)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

16

of multiple options (i.e., commercial food, wild food, construction material, climate

regulation, recreation and cultural/spiritual benefits) (Figure 3).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

17

a. Weighting of plant and livestock diversity doubled b. Weighting of recreation doubled

c. Weighting of ground water halved d. Weighting of cultural/spiritual benefits doubled

e. Weighting of wild food doubled f. Scoring of multiple options changed: halved under commercial food and climate regulation, and doubled under wild food, construction material, recreation and cultural/spiritual benefits

Figure 3. Sensitivity analysis of the MCDA ranking of four land uses in Kgalagadi

District, southern Botswana, performed by applying varied MCDA scoring and weighting (X Axis ‘Type of land use’, Y Axis ‘Total (weighted) value of ES’)

0

10

20

30

40

50

60

70

80

90


Privatecattle

ranches

Privategame

ranches

WMAs

Cultural/SpiritualbenefitsRecreation

Climate regulation

Plant and livestockdiversityConstruction material

Fuel

Groundwater

Food (wild)

Food (commercial)

0

10

20

30

40

50

60

70

80

90


Privatecattle

ranches

Privategame

ranches

WMAs


Climate regulation


Fuel

Groundwater

Food (wild)

Food (commercial)

0

10

20

30

40

50

60

70

80


Privatecattle

ranches

Privategame

ranches

WMAs

Cultural/Spiritualbenefits

Recreation

Climate regulation

Plant and livestockdiversity

Constructionmaterial

Fuel

Groundwater

Food (wild)

Food (commercial)

0

10

20

30

40

50

60

70

80

90


Privatecattle

ranches

Privategame

ranches

WMAs


Climate regulation


Fuel

Groundwater

Food (wild)

Food (commercial)

0

10

20

30

40

50

60

70

80

90


Privatecattle

ranches

Privategame

ranches

WMAs

Cultural/Spiritualbenefits

Recreation

Climate regulation

Plant and livestockdiversity

Constructionmaterial

Fuel

Groundwater

Food (wild)

Food (commercial)

0

10

20

30

40

50

60

70

80

90


Privatecattle

ranches

Privategame

ranches

WMAs


Climate regulation


Fuel

Groundwater

Food (wild)

Food (commercial)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

18

Results from these analyses indicated insensitivity of the results in the ranking (i.e., the

ranking remains unchanged under all the different iterations). This suggests that if a

broader panel of individuals is interviewed the results of this MCDA will not change.

4. Discussion

The MCDA shows that Kalahari communal rangelands are important for a broader range

of ES than simply commercial food production, despite that the value of these other ES is

currently underestimated from an economic perspective. When non-marketed ES are

weighted according to their relative importance to other ES, communal grazing areas rank

as the land use that generates the highest value (both in provisioning and cultural terms),

followed by WMAs. This indicates that the value of a land use is not only linked to the

availability of an ES, but also to its relative importance to society. It is recognised that the

relative importance of ES assessed might vary for different social groups, however the

sensitivity analysis indicates that the overall ranking of MCDA land use options does not

change when varied iterations (i.e., setting different levels of relative importance of ES) are

tested. This suggests that communal grazing areas and WMAs are able to deliver a higher

value to those in society (e.g., communal farmers) whose livelihoods rely substantially on

ES that are not linked to commercial food production.

4.1. Land management strategies and policy drivers: impacts on ES trade-offs

Livestock production is the major land use, partly due to the strong policy support (Tribal

Grazing Land Policy (TGLP), 1975 and National Policy on Agricultural Development

(NPAD), 1991) and subsidy schemes granted to the livestock sector (e.g., Services to

Livestock Owners in Communal Areas (SLOCA), 2002, Livestock Water Development

Programme (LWDP), 2002 and the Livestock Management and Infrastructure

Development Programme (LIMID), 2007). National policies have widely promoted fencing

and land privatisation for livestock production, assuming that private ranches would adopt

more sustainable and profitable land management practices through rotational grazing and

reduced stocking levels. The literature suggests that the per animal productivity gains on

fenced ranches are only slightly better than those on communal grazing areas (Hubbard,

1982); our MCDA findings support this assertion (Table 3).

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

19

Commercial food production is a complementary ES to plant and livestock diversity (Mace

et al., 2012). Diversity of forage species (Scherf et al., 2008) and livestock breeds (Notter,

1999) is vital for the nutritional status of cattle and to ensure their resilience, particularly

during drought. The highest MCDA scores under the plant and livestock diversity criterion

were achieved by private cattle ranches and WMAs as opposed to private game ranches

and communal grazing areas. In the WMAs, the land is not usually grazed by livestock, so

limited degradation is found and nutritious forage species are preserved (Dougill et al.,

2014). The private cattle ranches support the use of SLM by implementing management

strategies that aim to conserve grazing surfaces: "Since we increased the resting period

across paddocks to 90 days, the land condition has drastically improved and we have

plenty of grass...new perennial grasses are growing such as Brachiaria nigropedata,

Digitaria eriantha and Schmidtia pappophoroides", (interview data, private cattle ranch,

Werda, 2013). Despite these good practices, ecological assessments show that privately-

owned cattle ranches are just as heavily bush-encroached as communal grazing areas

(Dougill et al., 2014).

SLM practices are supported on private ranches by sizeable investments to remove

encroaching bushes to increase the grazing surface and enhance grass quality. The cost

of this accounts for US $22.9/ha through selective spraying done by hand or US $36.6/ha

through spraying from the air. Only a minority of farmers can afford such investments,

despite encouraging results: "Some people do not de-bush as they have no money. I am

borrowing [the money used for treatment]... removing bushes pays back shortly" (interview

data, private cattle ranch, Werda, 2013). Land management practices such as bush

removal can enhance provisioning ES through improved cattle production resulting from

better forage, but this reduces the amount of carbon stored in above ground biomass.

Sizeable investments also allow the management of mixed cattle breeds in private

ranches. Mixed breeding is less common in communal areas due to a lack of fences: "We

are not able to keep specific breeds other than Tswana because our cattle are mixed with

the ones owned by other farmers" (interview data, communal farmer, Werda, 2013).

Fresh water is a scarce resource and groundwater extracted through boreholes is often

highly saline, so while Table 3 suggests groundwater extraction has the second highest

economic value, the figures need to be interpreted with caution. Extraction through

borehole technology is expensive, and drilling investments translate into an economic loss

when the water found is not usable, either due to excessive salinity or because the

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

20

quantity of water found is very small: "In 2011 we drilled 5 boreholes, but 2 have too salty

water and 3 are empty" (interview data, communal farmer, Khawa, 2013). Interview data

indicate that on average a single private cattle rancher uses 5 boreholes with a mean

extraction capacity of 4,400 L/hr while a communal farmer uses 2 boreholes (of which 1

commonly belongs to a syndicate) with a mean extraction capacity of 2,500 L/hr. Pumping

time ranges from 9hr/day (winter) to 24 hr/day (summer) for both communal and private

users, despite government recommendations that for adequate groundwater recharge,

pumping should not exceed 8hr/day. Water extraction costs for livestock in communal

areas have been subsidised through various programmes (e.g., LIMID, LWDP and

SLOCA). Subsidies cover up to 60% of the borehole drilling costs (GoB, 2013), allowing a

positive MCDA score for this land use. The lower score of groundwater extraction for

private game ranches is because game have a lower water demand than cattle, so less

water is extracted.

Important trade-offs between commercial food production (including groundwater

extraction) and the conservation of forage species diversity are observed in the MCDA: a

high score for the former is usually associated with a lower score for the latter. Policy-led

expansion of boreholes across semi-arid communal grazing areas has led to a high

concentration of cattle around water points and movement away from multi-species

production systems and viable CBNRM projects. Fencing for commercial production and

the effective private status of individually owned (or syndicate owned) cattleposts, has

greatly reduced the amount of communal grazing land available to livestock (Perkins,

1996). Ecological assessments carried out to inform the MCDA indicate that this produces

a noticeable retreat of grass cover as well as bush encroachment up to at least 5km away

from water points in semi-arid communal grazing areas (Dougill et al., 2014).

Apart from livestock production, communal and WMA land uses are important sources of

veld products. They provide a supplementary source of (wild) food (and water), particularly

in the dry season or during times of drought and can also be used as medicines (e.g.,

Hoodia gordonii is used as an appetite suppressant and Devil’s Claw (Harpagophytum

procumbens) is an analgesic, sedative and anti-inflammatory) with significant commercial

value. Trade-offs between cattle farming and veld products are observed in the weighted

MCDA scores. The economic value and availability of veld products have declined with the

degradation induced by the expansion of borehole drilling. As a consequence, these

products are found increasingly further away from communal settlements (Perkins et al,

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

21

2002). Villagers’ willingness to invest time in veld product collection is limited by the lack of

a formal market, which hampers the generation of financial returns: "We do not harvest for

sale because it is very difficult to sell these products...we only consume them in the

household" (interview data, communal farmer, Khawa, 2013). Similar trade-offs are

observed between policy-driven cattle production and the utilisation of construction

material (i.e., thatching grass and poles for fencing) and fuel (firewood). These two ES

rank higher under communal grazing and in WMAs, showing their importance in the

management strategies adopted under these land uses. The supply of these ES is

threatened, as observed in the case of veld products, by degradation and privatisation.

Subsistence hunting plays a significant role in WMAs and the neighbouring communal

grazing areas and, together with the gathering of veld products, contributes to the highest

values achieved by these land uses under the criterion ‘wild food production’. This activity

is regulated through a licensing system through the Department of Wildlife and National

Parks (DWNP). Interviews indicate that strict government control discourages the villagers

to pursue this management strategy: "There are too many rules that constrain hunting"

(interview data, communal farmer, Khawa, 2013). The hunting ban, applied by the

government since January 2014, with the aim to stop an observed decline in wildlife

numbers (GoB, 2012b), has further reduced the hunting activities across the country.

CBNRM approaches have been developed through the Wildlife Conservation Policy of

1986 since the 1990s and a CBNRM Policy was adopted in 2007. CBNRM puts the local

communities adjacent to WMAs in charge of managing wildlife so that resource

conservation and financial returns can be simultaneously promoted. Exclusive rights were

granted over a wildlife quota. The community could decide whether to hunt the quota for

subsistence use or to sell the quota rights to a private operator (e.g., a safari company).

CBNRM trophy hunting and photographic safaris have generated tangible benefits to the

WMA community of Khawa. Our analysis reveals that these activities can generate up to

US $31,000 annually: "Hunting is our only source of revenue. It allows us to sustain our

livelihoods" (interview data, communal farmer, Khawa, 2013). CBNRM approaches offer

additional opportunities to generate economic benefits from the ecotourism sector, where

tourists are attracted by wildlife diversity and cultural experiences (e.g., guided visits and

4x4 driving as identified in Thusano Lefatsheng Trust, 2005). The WMAs also aimed to

conserve wildlife diversity so that these recreational and cultural values can be preserved

and enhanced.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

22

CBNRM activities and natural resource use diversification remain constrained by the policy

and market incentives in the livestock sector. Declines in wildlife numbers and diversity are

caused by the expansion of livestock production and exacerbated by the promotion of

fencing which, together with the establishment of veterinary cordon fences, have blocked

wildlife migration routes (Perkins, 1996). The policy incentives within the livestock sector,

alongside recent decisions on hunting, contrast with the WMAs' wildlife conservation

objective, and limit the economic viability of CBNRM as a management strategy under this

land use: "Wildlife numbers are declining. The government took key animals - lions and

leopards - out of the hunting quota for conservation purposes. Since then, the safari

hunting business has not anymore been profitable, neither for the private operators or the

community" (interview data, private safari operator, Werda, 2013).

The hunting ban and climatic constraints (especially during severe droughts) will constrain

CBNRM development. Low levels of rainfall and high inter-annual variability (Mogotsi et

al., 2013) exacerbate the degradation observed as a result of suboptimal land

management strategies: "The area is vulnerable to drought. When it rains the veld is fine

even if we are overgrazed, but during droughts many of our cattle die" (interview data,

communal farmer, Kokotsha, 2013). While climate change requires long-term approaches

so that rangeland users can adapt, the MCDA findings show that current policy fails to

provide strong support that guides livelihoods away from livestock production into

alternative activities.

In addition to the provisioning and cultural services identified, the Kalahari provides other

values through climate regulation. Whether carbon sequestration can be profitable is

questionable, due to low global prices, uncertainty over markets and standards, and poorly

developed methodologies, particularly for monitoring, reporting and verification (Stringer et

al., 2012). Care needs to be taken that increased woody biomass resulting from land

degradation is not encouraged through carbon credit schemes. Other monetary

approaches could be used to extract economic benefits from carbon sequestration and

biodiversity enhancement, for example through Payments for Ecosystem Services (PES)

initiatives. However, little evidence of how these schemes could work is available for

rangeland systems (Dougill et al., 2012). It must also be considered that potential markets

might not exist for all ES, where features such as the spiritual value of landscape are

difficult to be valued economically. Future policy interventions will need to be placed within

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

23

the cultural context of ES values, so that they create incentives for land users to invest in

rangeland resources, rather than misusing them (Ostrom et al., 1999).

4.2. Policy change for SLM

Policies based upon effective privatisation and commercialisation have been largely

unsuccessful in Botswana (Malope and Batisani, 2008; Perkins, 1996). This study has

shown that southern Botswana's rangelands pose complex land management challenges

when viewed through an ES lens, particularly when some ES are providing current

economic benefits whereas the potential of others is yet to be tapped. Adequate policy and

economic mechanisms are needed so that the values assessed through the MCDA can be

translated into concrete economic benefits. These might include the establishment of a

functioning market for ES that can generate immediate benefits through local

commercialisation (i.e., wild food, construction material and fuel). However, the MCDA

indicates that policy measures should not only be focused on the establishment of a

market for these ES, but also on the promotion of management practices that do not

reduce access to the broader range of ES.

Specifically, declining opportunities for hunting and gathering are linked to the reduction of

wildlife numbers and the loss of plant diversity resulting from intensive cattle rearing in

remote areas. This situation, together with a limited capability to develop tourism and

CBNRM activities, translates into increased poverty for the rangeland users and higher

levels of dependency on government support and drought relief programmes. Well-

intentioned poverty alleviation schemes such as the provision of livestock to the poorest in

society have served to increase stocking rates (and degradation) around communal

(village) water points, but reduced the biodiversity value of the system.

The key to maintaining wild animal biomass in the south-western Kalahari is to maintain its

mobility over large areas, with its conservation linked to the development of wildlife-based

economic activities such as tourism. Existing policies that support borehole drilling and

fencing directly conflict with those needs and current drought relief interventions tend to

address only short-term sector-specific needs. They fail to develop multi-species

production systems and to create a longer term drought-resistant socio-economic structure

(Jacques, 1995). There is an urgent need to promote alternative land management

strategies that improve SLM and the welfare of the communities through economic and

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

24

livelihood diversification. To achieve this requires an assessment of the ways in which

different policies and incentives interact and conflict, such that a more enabling policy

context can be developed through the promotion of synergies. Successful outcomes (both

socio-economic and environmental) depend not only on the type of land use promoted by

policy, but also on the range of management strategies that are adopted under each land

use, and the interactions between them.

5. Conclusions and ways forward for SLM

This paper has applied MCDA to Kalahari rangelands of Botswana's southern Kgalagadi

District, to assess the linkages between rangeland land uses and the provision of ES. The

study provides an empirical contribution to our understanding of which land uses and land

management strategies are best placed to generate the widest range of monetary and

non-monetary values from specific ES. While private land use enables the generation of

higher incomes that can be invested towards SLM practices, these findings contrast with

the common perception that communal grazing leads to land degradation. Overall,

communal grazing land was identified as the most appropriate land use option as it

achieved the highest MCDA score, followed by WMAs, private cattle ranches and private

game ranches.

The significant government support provided to the cattle sector in the form of borehole

technology for groundwater extraction and through the promotion of fencing (i.e., land

privatisation) has increased cattle populations and degradation around water points. Veld

products, construction material and fuelwood remain undervalued from an economic

perspective due to the lack of a market. Creation of a market with commercial potential is

needed so that the provisioning values of these ES are translated into wider economic

benefits to society. However, generation of perverse incentives that encourage

overharvesting must be avoided when the commercial market is created.

Livestock production and wildlife areas could be more clearly separated by limiting

borehole development within communal grazing land use in areas in proximity to WMAs.

This would increase livelihood diversification opportunities to the poorest households by

increasing the land management opportunities derived from provisioning ES, as well as

potentially rewarding cultural ES. Whether such diversification could include revenues from

the trading of carbon credits remains unclear and requires further investigation.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

25

By integrating monetary with non-monetary valuation techniques, MCDA provides a

mechanism for identifying the values of stakeholders from a range of policy sectors, which

in turn support wider dryland populations who depend on ES. Dryland policies that fail to

take a holistic approach to valuing such services run the risk of inadvertently exacerbating

land degradation, resulting in unforeseen social and economic costs. In promoting SLM

practices, the focus of future policies should go beyond the profitability of commercial food

production and consider the broader livelihood impacts (the social distribution of wealth

and diversification), and the ecological implications of all the ES analysed, including non-

marketed provisioning services and cultural values. Such an approach would better

recognise that livelihoods are not just about income generation but also are supported by a

range of other social, cultural, environmental and physical assets.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

26

Acknowledgments

This research was funded by the Economics of Land Degradation Initiative, contract

number 81163498, and carried out under research permit number EWT 8/36/4xxv(60) of

the Government of Botswana. We thank the staff of the Deutsche Gesellschaft für

Internationale Zusammenarbeit (GIZ), United Nations Convention to Combat

Desertification (UNCCD) and United Nations University (UNU) for their support, and

particularly Emmanuelle Quillerou for her thoughtful comments on this paper.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

27

References

Akhtar-Schuster, M., Thomas, R.J., Stringer, L.C., Chasek, P. and Seely, M. 2011.

Improving the enabling environment to combat land degradation: Institutional,

financial, legal and science-policy challenges and solutions. Land Degradation &

Development. 22(2), pp. 299-312. Doi: 10.1002/ldr.1058

Apthorpe, R. 1996. Reading development policy and policy analysis: On framing, naming,

numbering and coding. The European Journal of Development Research. 8(1), pp.

16-35. Doi: 10.1080/09578819608426651

Arntzen, J., Barnes, J., Turpie, J., Ruthenberg, P. and Setlhogile, P. 2010. The economic

value of the MFMP area. In: Centre for Applied Research and Department of

Environmental Affairs. Makgadikgadi Framework Management Plan Volume 2

[online]. Gaborone. [Accessed 01 April 2014]. Available from:

http://www.mewt.gov.bw/uploads/files/Final%20Chapter%202%20Value.pdf

Bagstad, K.J., Semmens, D.J., Waage, S., Winthrop, R. 2013. A comparative assessment

of decision-support tools for ecosystem services quantification and valuation.

Ecosystem Services 5, pp. 27–39. DOI: 10.1016/j.ecoser.2013.07.004

Bateman, I.J., Harwood, A.R., Mace, G.M. Watson, R.T., Abson, D.J., Andrews, B.,

Binner, A., Crowe, A. Day, B.H., Dugdale, S., Fezzi, C., Foden, J., Hadley, D.,

Haines-Young, R., Hulme, M., Kontoleon, A., Lovett, A., Munday, P., Pascual, U,.

Paterson, J., Perino, G., Sen, A., Siriwardena, G., van Soest, D., Termansen, M.

2013. Bringing Ecosystem Services into Economic Decision-Making: Land Use in the

United Kingdom. Science. 341(6141), pp. 45-50. Doi: 10.1126/science.1234379

Chanda, R., Totolo, O., Moleele, N., Setshogo, M. and Mosweu, S. 2003. Prospects of

subsistence livelihood and environmental sustainability along the Kalahari transect:

The case of Matsheng in Botswana’s Kalahari rangelands. Journal of Arid

Environments. 54(2), pp. 425-445. Doi: http://dx.doi.org/10.1006/jare.2002.1100

Costanza, R., de Groot, R., Sutton, P., van der Ploeg, S., Anderson, S.J., Kubiszewski, I.,

Farber, S. and Turner, R.K. 2014. Changes in the global value of ecosystem

services. Global Environmental Change. 26, pp. 152-158. Doi:

http://dx.doi.org/10.1016/j.gloenvcha.2014.04.002

de Groot, R.S., Fisher, B., Christie, M., Aronson, J., Braat, L., Haines-Young, R., Gowdy,

J., Maltby, E., Neuville, A., Polasky, S., Portela, R. and Ring, I. 2010. Integrating the

ecological and economic dimensions in biodiversity and ecosystem service

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

28

valuation. In: TEEB. ed. The Economics of Ecosystems and Biodiversity Ecological

and Economic Foundations. London and Washington: Earthscan.

Dougill, A.J. and Thomas, A.D. 2004. Kalahari Sand Soils: Spatial Heterogeneity,

Biological Soil Crusts and Land Degradation. Land Degradation and Development.

15 (3), pp. 233-242. Doi: 10.1002/ldr.611

Dougill, A.J., Fraser, E.D.G. and Reed, M.S. 2010. Anticipating vulnerability to climate

change in dryland pastoral systems: using dynamic systems models for the Kalahari.

Ecology and Society. 15(2). URL: http://www.ecologyandsociety.org/vol15/iss2/art17/

Dougill, A.J., Stringer, L.C., Leventon, J., Riddell, M., Rueff, H., Spracklen, D.V., Butt, E.

(2012). Lessons from Community-based Payment for Ecosystem Service Schemes:

From Forests to Rangelands. Philosophical Transactions of the Royal Society B.

367, pp. 3178-3190.

Dougill, A.J., Perkins, J.S., Akanyang, L., Eckardt, F., Stringer, L.C., Favretto, N.,

Atlhopheng, J. and Mulale, K. 2014. Land Use, Rangeland Degradation and

Ecosystem Service Provision: New Analyses from southern Kalahari, Botswana.

Leeds, UK: Report for the Economics of Land Degradation Initiative [online].

[Accessed 4 April 2014]. Available from: http://www.see.leeds.ac.uk/research/sri/eld/

Fish, R., Burgess, J., Chilvers, J., Footitt, A., Haines-Young, R., Russel, D., Turner, K.,

Winter, D.M. 2011. Participatory and deliberative techniques to embed an

ecosystems approach into decision-making: Full technical report. London, Defra.

Foley, J.A., DeFries, R., Asner, G.P., Barford, C., Bonan, G., Carpenter, S.R., Chapin,

F.S., Coe, M.T., Daily, G.C., Gibbs, H.K., Helkowski, J.H., Holloway, T., Howard,

E.A., Kucharik, C.J., Monfreda, C., Patz, J.A., Prentice, C., Ramankutty, .N. and

Snyder, P.K. 2005. Global Consequences of Land Use. Science. 309(5734), pp.

570-579. Doi: 10.1126/science.1111772

Fontana, V., Radtke, A., Bossi Fedrigotti, V., Tappeinera, U., Tasserc, E., Zerbeb, S. and

Buchholzd, T. 2013. Comparing land-use alternatives: Using the ecosystem services

concept to define a multi-criteria decision analysis. Ecological Economics. 93, pp.

128-136. Doi: http://dx.doi.org/10.1016/j.ecolecon.2013.05.007

GoB. 2003. Kgalagadi District Development Plan 6: 2003-2005 [online]. Gaborone:

Ministry of Local Government. [Accessed 6 June 2014]. Available from:

http://www.gov.bw/Global/MLG/Kgalagadi_DDP.pdf

GoB. 2006. Policy Brief on Botswana’s Water Management [online]. Gaborone:

Department of Environmental Affairs. [Accessed 6 June 2014]. Available from:

http://www.mewt.gov.bw/uploads/files/DEA/water%20policy%20brief%20final.pdf

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

29

GoB. 2007. Natural Resource Accounting of Botswana’s Livestock Sector [online].

Gaborone: Department of Environmental Affairs and Centre for Applied Research.

[Accessed 6 June 2014]. Available from:

http://www.mewt.gov.bw/uploads/files/natural_resource_accounting_of_bws_livestoc

k_sector.pdf

GoB. 2012a. 2011 Botswana population and housing census [online]. Gaborone: Central

Statistics Office. [Accessed 6 June 2014]. Available from: http://www.cso.gov.bw/

GoB. 2012b. Aerial Census of Animals in Botswana: 2012 Dry Season. Gaborone:

Department of Wildlife and National Parks.

GoB. 2013. Botswana Integrated Water Resources Management & Water Efficiency Plan

[online]. Gaborone, Botswana: Department of Water Affairs. [Accessed 6 June

2014]. Available from:

http://www.water.gov.bw/images/Reports/Final%20Botswana%20IWRM%20WE%20

Plan%20Vol%202%20Version%204%20(Web-Optimised).pdf

GoUK. 2009. Multi-criteria analysis: a manual [online]. London: Department for

Communities and Local Government. [Accessed 6 June 2014]. Available from:

http://eprints.lse.ac.uk/12761/1/Multi-criteria_Analysis.pdf

Hajer, M.A. 2006. Doing discourse analysis: Coalitions, practices, meaning. In: M. Van den

BrinK and T. Metze. eds. Words matter in policy and planning: Discourse theory and

method in the social sciences. Utrecht: Netherlands Geographical Studies 344, pp.

65-76.

Hajkowicz, S.A., McDonald, G.T., Smith, P.N. 2000. An evaluation of multiple objective

decision support weighting techniques in natural resource management. Journal of

Environmental Planning and Management 43(4), pp. 505-518.

Hubbard, M. 1982. Stock limitation — any economic alternatives for Botswana? Pastoralist

Development Network Paper, 14c. London: Agricultural Administration Unit, ODI.

Jacques, G. 1995. Drought in Botswana: Intervention as Fact and Paradigm. Pula:

Botswana Journal of African Studies. 9(1), pp. 33-60. URL:

http://archive.lib.msu.edu/DMC/African%20Journals/pdfs/PULA/pula009001/pula009

001004.pdf

Kenter, J.O, Brady, E., Bryce, R., Christie, M., Church, A., Irvine, K.N., Cooper, N., Davies,

A., Evely, A., Everard, M., Fazey, I., Hockley, N., Jobstvogt, N., Molloy, C., O’Brien,

L., Orchard-Webb, J., Ravenscroft, N., Ranger, S., Reed, M.S., Ryan, M. and

Watson, V. (in press). What are shared and social values of ecosystems? Ecological

Economics.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

30

Kenter, J.O., Brady, E., Bryce, R., Christie, M., Church, A., Irvine, K.N., Cooper, N.,

Davies, A., Evely, A., Everard, M., Fazey, I., Hockley, N., Jobstvogt, N., Molloy, C.,

O’Brien, L., Orchard-Webb, J., Ravenscroft, N., Ranger, S., Reed, M.S., Ryan, M.,

Watson, V. (in press). What are shared and social values of ecosystems? Ecological

Economics.

Kenter, J.O., Reed, M. S., Irvine, K.N., O'Brien, E., Brady, E., Bryce, R., Christie, M.,

Church, A., Cooper, N., Davies, A., Hockley, N., Fazey, I., Jobstvogt, N., Molloy, C.,

Orchard-Webb, J., Ravenscroft, N., Ryan, M., & Watson, V. 2014. UK National

Ecosystem Assessment Follow-on. Work Package Report 6: Shared, Plural and

Cultural Values of Ecosystems. UNEP-WCMC, LWEC, UK.

MA. 2005. Ecosystems and human well-being: A framework for assessment [online].

Washington, DC: World Resources Institute. [Accessed 6 June 2014]. Available

from: http://pdf.wri.org/ecosystems_human_wellbeing.pdf

Mace, G.M., Norris, K. and Fitter, A.H. 2012. Biodiversity and ecosystem services: a

multilayered relationship. Trends in Ecology and Evolution. 27(1), pp. 19-26. Doi:

10.1016/j.tree.2011.08.006.

Madzwamuse, M., Schuster, B. and Nherera, B. 2007. The Real Jewels of the Kalahari.

Drylands Ecosystem Goods and Services in Kgalagadi South District, Botswana

[online]. France: IUCN The World Conservation Union. [Accessed 6 June 2014].

Available from: http://www.cbd.int/financial/values/botswana-valueservices.pdf

Malope, P. and Batisani, N. (2008) Land reforms that exclude the poor: the case of

Botswana. Development Southern Africa. 25(4), pp. 383-397. Doi:

10.1080/03768350802316179

Mogotsi, K., Nyangito, M.M. and Nyariki, D.M. 2011. The perfect drought? Constraints

limiting Kalahari agro-pastoral communities from coping and adapting. African

Journal of Environmental Science and Technology. 5(3), pp. 168-177. Doi:

10.5897/AJEST10.216

Notter, D.R. 1999. The importance of genetic diversity in livestock populations of the

future. Journal of Animal Science. 77(1), pp. 61-69. URL:

http://www.journalofanimalscience.org/content/77/1/61

Ostrom, E., Burger, J., Field C.B., Norgaard, R.B., Policansky, D. 1999. Revisiting the

Commons: Local lessons, global challenges. Science 284(5412), pp. 278-282.

Perkins, J.S. 1996. Botswana: fencing out the equity issue. Cattleposts and cattle ranching

in the Kalahari Desert. Journal of Arid Environments. 33(4), pp. 503-517. Doi:

http://dx.doi.org/10.1006/jare.1996.0086

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

31

Perkins, J.S., Stuart-Hill, G. and Kgabung, B. 2002. The impact of borehole based cattle

keeping on the environment and society of the Kalahari. In: D. Sporton. and D.S.G.

Thomas. Sustainable livelihoods in Kalahari Environments: contributions to global

debates. OUP, Oxford.

Richardson, L., Loomis, J., Kroeger, T. and Casey, F. 2014.The role of benefit transfer in

ecosystem service valuation. Ecological Economics. 115, pp.51-58. In Press,

Corrected Proof. Doi: http://dx.doi.org/10.1016/j.ecolecon.2014.02.018

Scherf, B., Rischkowsky, B., Hoffmann, I., Wieczorek, M., Montironi, A. and Cardellino, R.

2008. Livestock Genetic Diversity in Dry Rangelands. In: C. Lee and T. Schaaf. eds.

The Future of Drylands. Netherlands: Springer. pp. 89-100.

Stringer, L.C., Dougill, A., Thomas, A., Stracklen, D., Chesterman, S., Ifejika-Speranza, C.,

Rueff, R., Riddell, M., Williams, M., Beedy, T., Abson, D., Klintenberg, P.,

Syampungani, S., Powell, P., Palmer, A., Seely, M., Mkwambisi, D., Falcao, M.,

Sitoe, A., Ross, S. and Kopolo, G. 2012. Challenges and opportunities in linking

carbon sequestration, livelihoods and ecosystem service provision in drylands.

Environmental Science and Policy. 19-20, pp.121-135. Doi:

http://dx.doi.org/10.1016/j.envsci.2012.02.004

Stringer, L.C., Thomas, D.S.G. and Twyman, C. 2007. From global politics to local land

users: applying the United Nations Convention to combat desertification in

Swaziland. The Geographical Journal. 173(2), pp.129-142. Doi: 10.1111/j.1475-

4959.2007.00226.x

TEEB. 2010. The Economics of Ecosystems and Biodiversity Ecological and Economic

Foundations. Edited by Pushpam Kumar. London and Washington: Earthscan.

Thomas, A. D. 2012. Impact of grazing intensity on seasonal variations in soil organic

carbon and soil CO2 efflux in two semiarid grasslands in southern Botswana.

Philosophical Transactions of the Royal Society B: Biological Sciences. 367(1606),

pp. 3076-3086. URL: http://hdl.handle.net/2160/11980

Thomas, D.S.G. and Twyman, C. 2004. Good or bad rangeland? Hybrid knowledge,

science, and local understandings of vegetation dynamics in the Kalahari. Land

Degradation and Development. 15(3), pp. 215-231. Doi: 10.1002/ldr.610

Thusano Lefatsheng Trust. 2005. Land use and management plan: Controlled Hunting

Area KD 15. Prepared for Khawa Kopanelo Development Trust. Gaborone: Thusano

Lefatsheng Trust.

Turpie, J.K., Barnes, J., Arntzen, J., Nherera, B., Lange, G. and Buzwani, B. 2006.

Economic value of the Okavango Delta, Botswana and its implications for

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

32

management [Online]. Accessed 6 June 2014]. Available from:

http://www.anchorenvironmental.co.za/

Twyman, C. 2000. Livelihood Opportunity and Diversity in Kalahari Wildlife Management

Areas, Botswana: Rethinking Community Resource Management. Journal of

Southern African Studies. 26(4), pp. 783-806. URL:

http://www.jstor.org/stable/2637571

UNCCD. 1994. Elaboration of an International Convention to combat desertification in

countries experiencing serious drought and/or desertification, particularly in Africa

[online]. Nairobi: UNEP. [Accessed 6 June 2014]. Available from:

http://www.unccd.int/Lists/SiteDocumentLibrary/conventionText/conv-eng.pdf

Walker, W.E. 2000. Policy Analysis: A Systematic Approach to Supporting Policymaking in

the Public Sector. Journal of Multi-Criteria Decision Analysis. 9(1-3), pp. 11-27. Doi:

10.1002/1099-1360(200001/05)9:1/3<11::AID-MCDA264>3.0.CO;2-3

WAVES. 2014. Botswana - Natural Capital as a Diversification [Online]. [Accessed 7

October 2014]. Available from: http://www.wavespartnership.org

WOCAT. 2010. Sustainable land management [Online]. [Accesed 6 June 2014]. Available

from: https://www.wocat.net/en/about-wocat/vision-mission/sustainable-land-

management.html

https://www.wocat.net/en/about-wocat/vision-mission/sustainable-land-management.html

https://www.wocat.net/en/about-wocat/vision-mission/sustainable-land-management.html

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

33

Supplementary Material

S1. Checklist of MCDA questions assessed through semi-structured interviews in

Kgalagadi District, southern Botswana

Supplementary Material S1. Checklist of MCDA questions assessed through semi-structured interviews in Kgalagadi District, southern Botswana

Interview ID: ______________________________ Interpreter (if any): _______________________________________________ GPS Waypoint: ______________________________________________________ Village: _______________________________________________________________ Date: ___________________________________________________________ 1. Land use: ⧠ Communal grazing area ⧠ Private cattlepost ⧠ WMA ⧠ Fenced cattle ranch ⧠ Fenced game ranch ⧠ Other: _________________ 2. Ownership: ⧠ Freehold ⧠ Leasehold ⧠ Other:______________________________

Participant demographics 3. Name: 4. Age: _________ 5. Gender: ⧠ Male ⧠ Female 6. Ethnic group: ⧠ Basarwa ⧠ Bakgalagadi ⧠ Batswana ⧠ Other: ______________ 7. Education (degree obtained): ________________________________________________

Livestock 8. Number of livestock owned and sold in Year 2012/2013

Livestock species

No. Owned No. Looked after

No. sold Price per unit

No. died

Cattle Bulls Oxen Tollies Cows Heifers

Various Goats Poultry Pigs Horses Sheep Donkeys Others: 9. What mix of cattle breeds (why did you choose these particular breeds)? 10. Cause of death for each livestock species: 11. Have you lost your livestock due to wildlife attacks in 2012/2013 (number and type of livestock)? 12. What wildlife species have caused this problem? How do you cope with this problem? 13. Has this situation changed over the past 15 years (how)? 14. Calving rates and calf survival? 15. Milk yields (quantity produced, consumed and sold, price per litre)? 16. What are the main market channels for selling your cattle, goats and sheep (e.g. BMC, butchers, individuals)? 17. What are the price differences when sold to different users? 18. Sales permits are released by veterinary officers. Is the release of these permits easy or difficult? Explain. 19. What is the role played by the BMC? What are the constraints? 20. Do you export? Who/where to? Constraints? 21. Do you produce pelts and wool? Quantity sold in 2012/2013 and price? To what market?

21b. Have you established any other market for other livestock products (e.g. horns, offal, hooves, etc.)?

Livestock management 22. What grazing management system do you adopt? ⧠ Non-rotational grazing ⧠ Rotational grazing 23. What stocking strategy (describe, indicate stocking rate - e.g. ha per livestock unit (LSU))? Outline the strengths and key challenges. ⧠ Conservative ⧠ Tracking (increase stoking rate when rain increases and sell when rain decreases) ⧠ Opportunistic ⧠ "Savory" method ⧠ Other: __________ 23b. Has your management strategy changed over the past 15 years (how)? 24. When do you water your cattle (how many times per day, what time). How much time does the cattle spend in between drinking? 24b. When do you kraal your animals (why)? 25. What is the minimum herd size to survive drought?

Supplementary Material

25b. What is the minimum herd size to generate positive cash balance?

Costs What are the major costs in your business?

• 26. Animal health: How much do you spend on veterinary drugs per year (for what type of livestock)? • 27. Supplementary feeds: Type, quantity and price? Do you import it or buy locally? Do you use any

veld product for fodder? • 28. Water

No of boreholes and dwells, when were they built, cost of drilling Total capacity (liters) and how many livestock can they feed How many liters of water do you extract per day/year Depth of water and reliability (has it ever gone dry) Water quality No of watering points Investment to build pipes (total pipes length and price per meter or total values) Maintenance costs

• 29. Transport (Transport expenses for the sale and purchase of livestock; No of cattle trucks (size, capacity) and 4x4, market price)

• 30. Fuel (Cars + Water pumps on boreholes: How many hours per day does the pump work? How many litres of diesel are consumed per day?):

• 31. Labour How many employees/waged workers (gender and age) Full time/part-time, employed on livestock or non-livestock, skilled/unskilled Average salary Turnover of staff Payments in kind such as food, housing, animals

• 32. Fencing (cattle) Electrified? ⧠Y ⧠N Height? Total fenced area (total length of perimeter fence) How many paddocks? Fencing cost per km Maintenance costs over the past 15 years?

• 33. Fencing (game) Electrified? ⧠Y ⧠N Height? Total fenced area (total length of perimeter fence) How many paddocks? Fencing cost per km Maintenance costs over the past 15 years?

Hunting and game farming and tourism 34. Wildlife type and population size in game ranch in Year 2012/2013 (+ historical data for the past 15 years)

Wildlife Population size

Market price per animal

No. hunted in 2012/2013

(subsistence)

No. hunted in 2012/2013 (commerce)

Hunting fee No. purchased

Ostrich Kudu Giraffe Zebra Gemsbok Cheetah Springbok Impala Eland Blue wildebeest

Rhino Lion Leopard Others:

34b. Do you slaughter the animals and sale the meat? Where? Quantity sold in 2012/2013 and price? Slaughtering costs?

36. Do you offer tourist packages for hunting? Price? Provide a detailed list of services included and expenses. 37. What additional services are provided to the hunters and at what price (e.g. transport, professional hunter/guide, trackers, pellets making, trophies)? 38. Do you have any hunting/photography concession in a WMA? Where? How long for? How much did you pay for it? What are the benefits and the key problems? 39. Accommodation expenses: price of room per night + food + travel packages 40. No of tourists in 2012/2013 and over the past 15 years (photographic and hunting) 35. Do you offer tourist packages for photographic safari? Price? Please provide a detailed list of services included and expenses.

35b. How do you manage both photographic and hunting? (timeframe, areas). 35c. Any additional information / further comment you would like to add?

Others

41. Does your household collects or utilises any of the following woody or veld products?

Subsistence Activity √ if consumed (quantify)

√ if sold locally for cash (quantify)

Cost per unit

Firewood Wild fruits Thatching grass, timber and woody plants for construction

Medicinal Plants Edible Plants Honey Craft making (specify) Other (specify) 42. Changes over the past 15 years in the use of veld products? Are these changes linked to land degradation (explain)? 43. Is the access to market easy or difficult? 44. What are the key factors that affect productivity and financial performance in both cattle and game farming? 45. What major events have affected your productivity and financial performance in the past 15 years and how did you respond? 47. What is the current level of veld health in your farm? Have the type and amount of plant and grass species in your farm/area changed over the past 15 years? What are the costs of environmental degradation? 48. Data on average rainfall and droughts 49. Invasive species, how do you fight bush encroachment? Do you poison the invasive plants? Mechanical clearing? How much does it costs? 50. Fire frequency and severity. 51. Compared to 15 years ago, is your household more or less prosperous today? ⧠ More prosperous ⧠ Less prosperous ⧠ Equal ⧠ Don’t know

51b. What are the reasons behind this change?

Detailed Response to Reviewers

Dear Editor and Reviewers, We deeply appreciate the time and effort you have spent on reviewing our manuscript. Your comments have helped us to improve and strengthen the paper. The substantive points you raised (in quotes below) were addressed as follows: EDITOR ‘Strong reservations have been made on how the MCDA method was applied, especially regarding the weighting of criteria. Reviewer #1 has clearly explained and highlighted the specific aspects of the weighting process. These will require re-analysis’

Thanks for raising this concern, which has been fully addressed by re-analysing the data and updating accordingly Table 5, Figure 2, as well as Sections 2, 3 and 4. Details on these changes are outlined below (see Reviewer #1, first comment).

‘I also recommend that results of sensitivity analysis are brought to fore in the manuscript, and not presented as supplementary information, considering that a whole paragraph is dedicated to its findings (pages 14 &15). On the other hand, details of the semi-structured interviews (the questionnaires) used should either be appended or presented in supplementary information’

Results of sensitivity analysis have been brought to fore in the manuscript (from ‘Supplementary Material – Figure S1’ to ‘Section 3 – Figure 3, p.17’. Details of the semi-structured interviews (i.e., questions checklist) have been presented in the supplementary material (S1: Checklist of MCDA questions assessed through semi-structured interviews in Kgalagadi District, southern Botswana).

‘GoB: page 3, line 58: State in full at the first mention. GoUK, page 10, line 30: State in full at the first mention’

Government of Bostwana (GoB), p.3, and UK Government (GoUK), p.10, have been stated in full at the first mention.

REVIEWER #1 ‘In MCDA, the weight of a criterion should describe how important the value change in each criterion from its worst value level to its best value level is compared to the corresponding changes in the other criteria. Thus, the weights should not describe some general level importance of the criteria, but they should be explicitly connected to the ranges applied in the criteria and how they are mapped to the criteria scores (i.e. scoring). However, in this manuscript, this essential element of the weighting process seems to be missing. For example, in criterion "Net profit of meat production" the scoring is carried out so that score 0 corresponds with -7.89 US$/ha/yr, and score 100 corresponds with 3.75 US$/ha/yr. Thus, the score range 0-100 corresponds with a change of 11.64 US$/ha/yr. However, in criterion "Revenues from CBN/RM Trophy…" the score 0 corresponds with value 0 US$/ha/yr, and score 100 corresponds with value 0.09 US$/ha/yr. Thus, in this criterion, the score range 0-100 corresponds with a change of only 0.09 US$/ha/yr. This is under 1/100 of the range in "Net profit of meat production". Thus, if a decision maker considers an increase of one US$/ha/yr in these criteria to be equally important, he/she should give over 100 times more weight to criterion "Net profit of meat production" than to "Revenues from CBN/RM Trophy…"…The above was the most extreme example of this phenomenon, but this can be seen in the other criteria too. For example, alternative "Private cattle ranches" has approximately 1 US$/ha/yr higher value than alternative "WMAs" both in "Net profit of meat production" and

*Detailed Response to Reviewers

in "Value of water extracted". However, in criterion "Net profit of meat production", the effect to the overall weighted value is a difference of 1 score point (12 vs. 13), whereas in criterion "Value of water extracted" the effect is 10 score points (0 vs. 10)’

We agree that the inconsistency in the score ranges across different criteria (the ones measured in quantitative monetary terms) would result in an unbalanced overall weighted value of these criteria, as policymakers would not consider these ranges explicitly when giving weights to the criteria. This issue has been addressed by setting a homogeneous score range (with same end points) for all the following monetary criteria: food (commercial), groundwater, climate regulation, and recreation. The new score range is the same as the one used for commercial food production, which accounts for the widest confidence interval – or end points (see Table 3): (score 0 = US$/ha/yr -7.89 ; score 100 = US$/ha/yr 3.75). Under the current scoring system, a variation of US$/ha/yr 1 in one criterion (e.g. recreation) accounts for the same variation of score under the other criteria (e.g. groundwater, etc.). Clarification has been added in Section 2.2.4 (p.10): ‘A homogenous score range (i.e. using the same end points across different criteria) was used for all the monetary criteria, in order to reflect equal variations of MCDA score as a result of a variation of US $/ha/yr 1 in one given monetary criterion’. The scores, both non-weighted and weighted, have been updated in Table 5. An updated version of Figure 2 has been created.

As a result, the ranking of the intermediate land uses has changed (i.e. WMAs and private cattle ranches are inverted, while communal livestock grazing and private game ranches remain unchanged) as follows: 1. communal livestock grazing, 2. WMAs, 3. private cattle ranches, 4. private game ranches. This has been reflected in Section 3 (p.15) and Section 5 (p.24). As regards the sensitivity analysis, in our first submission minor ranking variability in the intermediate options was observed under certain iterations. The updated sensitivity analysis (see Figure 3, p.17 - former S1) now proves stronger, as: ‘the ranking remains unchanged under all the different iterations’ (Section 3, p.18). This is also stressed in Section 4 (p.18), which has been updated as follows: ‘It is recognised that the relative importance of ES assessed might vary to different social groups, however the sensitivity analysis indicates that the overall ranking of MCDA land use options does not change when varied iterations (i.e., setting different levels of relative importance of ES) are tested. This suggests that communal grazing areas and WMAs are able to deliver a higher value to those in society (e.g., communal farmers) whose livelihoods rely substantially on ES that are not linked to commercial food production’.

REVIEWER #3 ‘Page 2, line 47: the comma after the word "quantified" is unnecessary. Page 2, line 52 (and elsewhere throughout the manuscript): a comma should follow "i.e." or "e.g." Page 3, line 30: a comma should be placed after "Botswana" Page 4, line 4: the comma after "grazing" is unnecessary Page 6, line 58: The Richardson reference should be Richardson et al., here, elsewhere, and in the references cited. Page 7, line 52: define what a "kraal" is Page 9, line 20 (and elsewhere): This should be "US $6.7/t" - not "US$ 6.7/t" Page 9, line 57: criteria is plural, so it should read "criteria were"

, line 50: "data" are plural - this should read "These data were" Page 15, line 37: this should read "might vary for different social groups" Page 16, line 15: the Hubbard 1982 reference should be followed by a semicolon, not a comma. Page 17, line 44: the comma after "forage species diversity" is unnecessary Page 24, line 48: Some of the authors in the Costanza et al. 2014 citation have an extra letter after their name. For instance, "Sutton," "Kubiszewski," "Farber," and "Turner" are the correct spellings. Pages 26-27: The Kenter et al. in press citation is listed twice’

All these minor changes have been addressed in full. ‘Page 4, Figure 1: This map should be reworked. The resolution is poor and should be sharpened to improve readability. It would be helpful to outline the district's boundary and also to have an inset map that shows the district's location in relation to the national boundary’

The map has been replaced with a new one (higher resolution – also available in .tiff format). The new figure clearly outlines the District’s boundary. An inset map showing the district’s location has been included.

‘Page 6, Table 1 (and elsewhere on page 6): Shouldn't the "criterion" be labeled as "ecosystem services?" It's less clear to me with them labeled as they are now than if they were labeled as ecosystem services’

This should be labelled ‘Criterion’ as the column presents the various MCDA ‘criteria’, coherently with the wording used throughout the manuscript. Double-labelling including ‘ecosystem service’ has been added to the headings of Table 1, Table 3 and Table 5 in order to improve clarity.

‘Page 8, line 54: You're not using the benefit transfer method here, you're using the monetary value of carbon, in this case, market prices’

The benefit transfer method was used to derive the ‘net amounts of total soil C sequestered per annum’ (see p.9) from similar environments in southern Botswana (Thomas, 2012) – that is what we define after Richardson et al. (2014) as the ‘policy sites’ (see p.6). C values from Thomas (2012) were then valued in monetary terms by using the monetary value of carbon (as stressed, through market prices). This is outlined in Section 2.2.2, Criterion 7, p.8-9.

‘Page 12, Table 4: Can you assign a standard deviation to the criteria weights as well as a mean value? This would be useful to show how much agreement or disagreement there was about these weightings’

Mean value (non-normalised) and standard deviation have been assigned in Table 4 (p.13).

‘Page 21, line 9: You state that "The key to maintaining wild animal biomass in the south-western Kalahari is to maintain its mobility over large areas." Existing policies that support borehole drilling and fencing directly conflict with those needs, right? It seems like it would clarify the argument to make this point’

This has been clarified – by adding the suggested sentence - in Section 4, p.23. ‘Page 22, lines 3-4: You state that "Creation of a market with commercial potential is needed so that the provisioning values of these ES are translated into wider economic benefits to society." I think it's absolutely critical to point out that creation of a commercial market would need to be done very carefully, to avoid creating perverse incentives to encourage overharvesting’

The following sentence has been added to Section 5, p.24: ‘However, generation of perverse incentives that encourage overharvesting must be avoided when the commercial market is created’.

OTHERS

Contract number has been added to the Acknowledgements Section. I hope that you feel that we have adequately considered and addressed your comments. Thanks very much for your reviews. Best regards.

Elsevier Editorial System(tm) for Ecosystem...

Documents

Transcript of Elsevier Editorial System(tm) for Ecosystem...