Towards Improved Decision-Making in Degraded Drylands of Southern Africa: an Indicator Based Assessment for Integrated Evaluation of Restoration and Management Actions in the Kalahari Rangelands1

DREBER, Nielsa,b, KONG, Taryn M.c, KELLNER, Klausa, HARMSE, Christiaan J.a, VAN EEDEN, Alberta, and OCAMPO-MELGAR, Anahid

a School of Biological Sciences, North-West University, Potchefstroom, South Africa, e-mail:;; (CH); (AvE) b Biodiversity, Evolution and Ecology of Plants, Biocentre Klein Flottbek and Botanical Garden, University of Hamburg, Hamburg, Germany, e-mail: c School of Natural Resources and the Environment, Office of Arid Lands Studies, University of Arizona, Tucson, United States of America, e-mail: d Arid Lands Resource Sciences Program, University of Arizona, Tucson, United States of America, e-mail:

Abstract — The loss of ecosystem resilience and land productivity is a major problem in drylands of southern Africa. This needs to be addressed in an integrated way linking science to society. Identifying best practices for land restoration and sustainable land management in a process combining local and scientific knowledge is therefore very important as regional perspectives are created and knowledge shared among affected land-users. A corresponding bottom-up framework was suggested by the multinational EU-project PRACTICE, which has been tested in arid Kalahari rangelands of South Africa. Following the identification of a multi-stakeholder platform (MSP) related to the livestock farming community of the study area, the participants’ baseline assessment and site-specific indicators for the evaluation of locally applied restoration and management actions to combat desertification were obtained in a participatory process. The MSP ranked the relative importance of the indicators on an individual basis using the pack-of-cards method, and re-ranked these indicators following group discussions. The individual ranking results were combined and integrated with biophysical and socio-economic measurements for the indicators through a multi-criteria decision analysis (MCDA), which ranked the alternative actions according to their relevancy and performance. The MCDA outcome was shared back with the MSP to stimulate group discussion and re-evaluation of the restoration and management actions aiming at improving management decision-making in Kalahari rangelands. The steps of this participatory approach are documented and results and the overall potential for implementation in local and national policy frameworks is critically discussed.

Keywords — best practice, local knowledge, multi-stakeholder platform, participatory process, social learning, South Africa.

1  Introduction

Subsistence in the rural dryland areas worldwide depends on the effective and sustainable utilization of the natural resources, which are increasingly threatened by land degradation, mainly due to extensive droughts and mismanagement, such as overgrazing. The African continent is spatially the most impacted with more than 70% of its agricultural drylands being already desertified (Hoffman and Ashwell, 2001). In particular the socio-ecological systems of sub-Saharan countries underwent dramatic changes in recent decades due to the combined effects of land use change (away from traditional practices like pastoralism towards sedentism) and climate change (Archer and Tadross, 2009; Oba et al., 2000). This development is also symptomatic for South Africa, with serious environmental, social and economic implications for the country (Hoffman and Ashwell, 2001; Milton and Dean, 1995; Palmer and Bennett, 2013).

Figure 1: Location of the study area (Mier municipality) in the Northern Cape Province (NC) of South Africa. Shown vegetation types follow Mucina and Rutherford (2006).

Land degradation is a complex and dynamic process driven by a multitude of factors including biophysical and socio-economic factors. Recognizing their interrelated impacts, any monitoring and assessment framework of land degradation should be based on an integrated and multidisciplinary approach (Vogt et al., 2011; Von Maltitz, 2009). In this respect, Sommer et al. (2011) point out that local participation is both ethically desirable and practically essential. Although South Africa has a relatively long history of combating land degradation emerging in the second half of the 20th century, it took a while until the needs of environment and development (e.g. community participation) started to be considered together (Hoffman and Ashwell, 2001). However, many land-use practices in rural areas are still driven by inappropriate policy frameworks, which emphasize the urgent need for local-level governance institutions assisting land users in sustainable land management (Palmer and Bennett, 2013; Von Maltitz, 2009).

Attempts in South Africa’s drylands to mitigate degradation at local scale should make use of land-user knowledge and actively involve them at all levels of project planning, decision-making and execution (Botha et al., 2008; Reed et al., 2006; van Rooyen, 1998). This principle of linking science to society by combining traditional/local knowledge with ecological (academic) expertise for improved decision-making and sustainable environmental management is nowadays perceived a necessity in order to combat the loss of ecosystem resilience and land productivity, for creating local to regional perspectives and to share knowledge for the benefit of the socio-ecological environment (Fraser et al., 2006; Reed, 2008). A corresponding bottom-up framework is therefore suggested by the international, European Commission-funded project PRACTICE (Prevention and Restoration Actions to Combat Desertification. An Integrated Assessment) aiming at filling the gap of systematic evaluation of practices for land restoration and sustainable land management in a participatory and integrated manner ( In the course of the project, an evaluation protocol was developed and tested for its general applicability in different socio-economic and biophysical contexts and syndromes of land degradation in selected dryland sites worldwide (Rojo et al., 2012).

The present paper reports the application of the evaluation protocol exemplified by a case study from degradation-prone Kalahari rangelands of South Africa. First, the single steps of the participatory PRACTICE approach are documented and results presented. Second, lessons learned and shortcomings observed are highlighted and concluded with a way forward. Finally, the overall potential for implementation in local and national policy frameworks is briefly discussed.

Figure 2: (A) Composition of the multi-stakeholder platform and (B) relative importance of indicators averaged over individual stakeholder perceptions (group composite weights) before (first iteration) and after (second iteration) group discussions. Bars represent means ± SD; asterisks indicate significant differences at *p < 0.05 and **p < 0.01 (Mann-Whitney-Wilcoxon test).

2  The socio-ecological environment

The PRACTICE approach is presently applied in two different cultural and biophysical settings of the Kalahari farming area in South Africa comprising different land tenure systems in the Mier municipality of the Northern Cape Province (test completed) and the wider Molopo area of the North West Province (evaluation running). The focus of the present case study lies on the Mier rural area just south of the Kgalagadi Transfrontier Park, between the Namibian and Botswana borders (Fig. 1). It forms part of the arid southwestern Kalahari basin receiving between 150-200 mm of rainfall per year mainly from January to April. Study sites encompass two broad rangeland types:

  1. the duneveld characterized by nutrient-poor aeolian sand forming linear dunes intersected by dune streets and covered by open, savanna type vegetation (Auob and Gordonia Duneveld sensu Mucina and Rutherford (2006); Fig. 1), and
  2. the rather level hardveld on loamy, stony soils with a prominent dwarf shrub vegetation (Kalahari Karroid Shrubland sensu Mucina and Rutherford (2006); Fig. 1).

Livestock farming with mostly mixed herds of sheep, goats and cattle, as well as game ranching are the two main land use activities in especially the duneveld area (Van Rooyen, 2000). However, most properties (leased or owned) are relatively small (< 2500 ha) and allow for smallholding to semi-commercial farming only. They have little infrastructure, such as fencing to make paddocks and watering points for better distribution of livestock and rotational grazing options. Many of the local farmers are not able to expand or develop their farms because of financial constraints. This causes inflexibility in employing ecologically sustainable management practices and forces farmers to utilize continuous grazing systems with inadequate water availability. In combination with a highly variable climate, rangelands in the study area easily deteriorate (Van Rooyen, 1998), which may result in profound regime shifts characterized by a loss of ecosystem resilience and rangeland productivity. Related degradation processes include either the loss of vegetation cover and an associated dune remobilization (Bhattachan et al., 2012) or an increase in woody vegetation (bush encroachment) (Van Rooyen, 2000), as well as changes in plant diversity and abundance patterns of plant functional groups (Rutherford and Powrie, 2010).

3  The integrated assessment

Table 1: Stakeholder rating of actions using a 1-to-5 Likert scale (pre- / post- integrative assessment (% of responses)).
Categories CriteriaIndicator/proxyUnit Good managementShrub control
Supporting services Soil fertility Soil condition % organic carbon 0.1±0.02 0.2±0.08
Regulating services Resource regenerationGrass abundance plants m-2 14.0±3.9 25.7±11.9
 Ecosystem integrity Woody abundance rank**1±0.42±0.7
Provisioning services Grass phytomass Biomass production kg ha-1 2755.7±374.1 2238.4±343.5
  Forage production Grazing capacity**** ha LSU-1 4.7±1.4 4.4±1.2
  Meat production Animal condition rank** 1.2±0.4 2.4±1.0
Economy Water availability Water availability rank* 1.1±0.25 2.1±0.8
  Labor & material costs Economic costs rank** 1.1±0.3 2.3±0.9
Sustainability Personal well-being rank** 1.1±0.3 2.5±1.0<
Subsidies Governmental help rank** 1.9±1.0 1.9±1.4
Biodiversity Species diversity Grass diversity H’ 1.04±0.2 0.9±0.6

* effect of action on the indicator as assessed by stakeholders (1 = affecting the most to 6 = no effect) ** overestimated for Kalahari duneveld; applied method needs to be adapted (relations remain the same)

The PRACTICE integrated assessment protocol (IAPro) (Bautista and Orr, 2011) was applied to identify suitable and widely accepted practices for desertification mitigation in Kalahari rangelands of the Mier area. The protocol integrates both the human and biophysical dimensions of desertification in seven steps, four of them being fully participatory. The participatory part includes the identification of local stakeholders (SHs) and their engagement, a baseline evaluation of actions and indicators, a weighting of indicators, as well as a collective integrated evaluation of actions (Rojo et al., 2012).

3.1  Identification and evaluation of actions and indicators

In Mier, IAPro step 1 and 2 were conducted simultaneously in a semi-structured interview due to logistic constraints. A diverse set of SHs was identified through a local consultation process and chain referrals, which resulted in a selection of 28 SHs forming the multi-stakeholder platform (MSP) to be involved throughout the study (Fig. 2A). Information was obtained on the participants’ baseline evaluation of locally applied land management and restoration actions, and site-specific indicators identified from the participants to be selected for action evaluation from the individual semi-structured interviews.

The interviews revealed that the most commonly applied actions to mitigate desertification in the Mier area include: (1) chemical shrub control (duneveld only), (2) good management (i.e. rotational grazing, resting periods, herding), (3) dune stabilization by brush packing (duneveld only), and (4) revegetation by transplanting grasses (predominantly in hardveld). A total of 30 ecological and socio-economic indicators were mentioned and/or proposed by the interviewees. A final set of meaningful indicators was short-listed by taking into account aspects of redundancy, popularity, data availability and collectability. Combined with expert-selected common indicators based on the ecosystem services approach, this resulted in a condensed list of 11 refined indicators for evaluating the identified actions (Fig. 2B and Table 1).

Table 2: Stakeholder rating of actions using a 1-to-5 Likert scale (pre- / post- integrative assessment (% of responses)).
 Excellent choiceVery good choice Moderate choiceBad choice Very bad choice
Good management 62.1 / 100 24.1 / 0 13.8 / 0 0 / 0 0 / 0
Shrub control55.2 / 7520.7 / 25 10.3 / 0 3.4 / 0 13.8 / 0
Dune stabilization 44 / 1040 / 70 12 / 10 0 / 10 4 / 0

In IAPro step 3, the MSP established the relative weight of the final set of indicators. Following Figueira and Roy (2002), indicator prioritization was processed via the revised Simos’ procedure or pack-of-cards method. Participants were asked independently to rank cards symbolizing the indicators according to their perceived importance and to insert blank cards to reinforce ranking differences. After computation of both the individual and integrated SH perspectives (sensu Figueira and Roy, 2002), the resulting normalized weighting (Fig. 2B, first iteration) was presented to the MSP during a workshop. Participants were encouraged to discuss each other’s results in small sub-groups, followed by an open group discussion and a second round of weighting for potential re-ordering the indicators should their perspectives have changed. The integrated collective weighting of indicators was indeed different from the first iteration (Fig. 2B). This might indicate a potential social learning effect through the exchange of views and experiences between land users, as well as individual and group reflections on the issues discussed. The SHs clearly rated the indicators ‘grazing capacity’ and ‘biodiversity’ (referring to fauna and flora) higher and ‘abundance of grasses’ lower in the second run, with the latter two being assigned a significantly different mean weight after the group discussions (Fig. 2B). In both iterations, the availability and supply of water was ranked highest as this indicator promotes the application of better management strategies, such as rotational grazing. Second highest was ‘animal condition’, which is a function of the ‘grazing capacity’ of the rangeland (third rank). Aspects related to personal factors such as finances and future well-being of the family played obviously an important role for Kalahari farmers when evaluating the actions (Fig. 2B).

3.2  Combining stakeholder perspectives with quantitative data

Ecological indicators were also quantitatively evaluated in the field (IAPro step 4) based on the FIXMOVE methodology for standardized rangeland assessments (Morgenthal and Kellner, 2008). Field measurements followed a preferential sampling approach by asking the SH to determine the sampling sites according to their perception of what their most degraded area, rangeland in best condition or successful restoration site is. No quantitative data could be gathered for social or economic indicators for reasons of being not directly measureable (e.g. personal well-being) or confidentiality (e.g. economic costs). For these, SHs were asked to rank each action in order of decreasing impact on the indicator (compare Table 1).

In IAPro step 5, the results from the indicator ranking in step 3 were combined and integrated with respective biophysical measurements for the indicators through a multi-criteria decision analysis (MCDA) using ELECTRE IS (Aït Younes et al., 2000). This is an outranking procedure based on pairwise comparisons of all the alternative restoration and management actions according to their relevancy (indicator relative importance representing SH perspectives) and performance (indicator quantification). The method identifies the indicators that are either indifferent (meaning have a similar performance in both actions being compared) or supportive for a certain action over another. Based on this, outranking relationships are constructed, where one action outranks another when it is at least as good as the other, as underlined by enough supportive arguments (integrated data).

As shown for the duneveld, a good management action (GM) in the Kalahari rangelands outranked dune stabilization but was equally good as shrub control (SC) (Fig. 3A), although there were a couple of indicators where the preference was weak in favor of GM over SC (Fig. 3B). Accordingly, GM provided best results in terms of biomass production and grazing capacity, whereas SC effectively decreased the competition for moisture and soil nutrients between shrubs and grasses, thus favoring the latter. SC also contributed to an increase in soil organic carbon due to litter accumulation. It should be noted that the MCDA does not identify the best option possible on an absolute scale but rather provides a relative ranking and information on which action outranks each other as a function of the indicators considered (Bautista and Orr, 2011).

3.3  Integrated evaluation of actions

A final step of IAPro (step 6) is the collective integrated assessment of the locally implemented actions based on the results of the weighting exercise of indicators, the subsequent MCDA, and related group discussions during the workshop. This step integrates scientific and local knowledge, biophysical data and SH perspectives for the purpose of a more informed re-evaluation. An adapted language, easily accessible figures and other visualizations were used to summarize all participatory outputs. As was also done in IAPro step 2, SHs were asked to rate each action on a scale from 1 to 5 (very bad to excellent choice). The rating was performed in an open, interactive setting allowing for ongoing discussions and exchange of opinions.

All respondents perceived the application of ‘good management’ as an excellent choice. The majority of the SH (75%) also rated ‘shrub control’ as excellent, whereas the rating of ‘dune stabilization’ was less consistent and overall lowest (Table 2). Compared to the first assessment done in IAPro step 2 (pre-integrative assessment, Table 2), the final action rating remained basically the same, but the underlying pattern was more distinct showing clearer preferences (Table 2).

4  Discussion and conclusion

4.1  Challenges and lessons learned

The case study showed that the tested IAPro is a promising tool for a locally-contextualized assessment and evaluation process of actions applied to mitigate land degradation. As concluded by Rojo et al. (2012), the general strength of PRACTICE is its bottom-up approach for assessing both environmental sustainability and social acceptance and integration. The iterative nature of weighting options, group discussions, and collective integrated assessments is central to establish a more diversified understanding and hybrid knowledge based on local and scientific expertise (Fraser et al., 2006). The participatory steps of IAPro furthered the information exchange among SHs. It can be assumed that being more informed would allow land users to adapt more effectively to changing environmental conditions and observed signs of rangeland deterioration, respectively. The exchange between SHs also has the potential to create opportunities for co-operative actions for the benefit of the local farming community. Indeed, the overall response of the Mier MSP was positive, appreciating in particular their direct involvement, learning-effect and immediate transformation of input into outcome of practical value. This feedback shows that locally important factors were captured by identifying indicators relevant to the SHs.

The positive perception by the SHs was certainly furthered by the fact that the conducted combination of indicators specific to local-scale needs with scientific-based indicators allowed to address locally-contextualized key ecosystem services. Such an integrated indicator selection is regarded a necessity to achieve sustainable land management goals (Sommer et al., 2011). The original list of 30 indicators suggested by the interviewees documented the wealth of information available among land users as also found elsewhere in the Kalahari region (Reed et al., 2008). This can be seen as a significant benefit, as the complex variety of land user perspectives is addressed (Reed et al., 2006) and the set of indicators is less likely too simple and incomplete (Fraser et al., 2006). However, as was done in the present study, a short-listing of the indicator set is necessary to avoid too much detail at the expense of the overview (Sommer et al., 2011). Short-listing criteria used included popularity and collectability, which bore conflicts. A couple of identified indicators, particularly those addressing popular socio-economic issues, such as economic costs, personal well-being or water availability, could not be directly quantified as they either did not specifically constituted the rangeland condition or were confidential in nature. Unfortunately, the qualitative ranking procedure chosen for these indicators showed to be too abstract for some of the participating SHs, and thus needs to be adapted for the sake of confirmability. Challenging for the integrative process of identifying a suitable indicator set were the different types of indicators suggested. Defining an indicator as being reflective of the degradation status, personal well-being must be regarded an indirect indicator reflecting secondary or consequential interactions which "carry a heightened risk of incorporating the expression of conditions associated with but not intrinsically part of the condition" that is assessed (Mabbutt, 1986). Further, water availability, which was suggested by most SHs as the most critical indicator, is rather a predictor indicating the risk or vulnerability to degradation than an indicator (Mabbutt, 1986).

Another critical point of overall importance for the successful application of the PRACTICE approach was related to the remote location of the study sites in combination with time- and budget constraints, which exacerbated timing and frequency of the quantitative evaluation of indicators in the field. Due to the intrinsic spatio-temporal environmental heterogeneity of drylands, biophysical once-off assessments are not representative and do not account for the need of multi-temporal information (Sommer et al., 2011). A shortcoming with respect to representativeness is also given by the biophysical assessment based on a preferential sampling approach, which has some disadvantages related to the subjective course of action. On the other hand, this type of approach gives the SHs some kind of control and increases their acceptance in site selection (van Rooyen, 1998). It showed to be difficult to identify enough suitable sites for replications in order to reduce the variability in data caused by the heterogeneity of both the biophysical environment and with respect to differences in management strategies and restoration design. In the course of the running case study in the Molopo area, the number of replicates was significantly increased. However, this kind of effort is very time consuming and requires a good knowledge of an area.

It turned out that the Mier community was already “over-workshoped” by many projects in recent years taking place in the area. This can result in a consultation fatigue, especially if their involvement in participatory processes gains them little reward (Reed, 2008). In addition, the huge distances between localities in the Kalahari farmers have to take into consideration when invited to workshops may result in low participation rates at critical participatory steps of IAPro, which is especially true if no financial expenses at least for travel costs can be rewarded. Varying participation rates can be problematic for a sound interpretation of results. For example, the weighting of indicators in at least two iterations including group discussions (IAPro step 3) should ideally be implemented in a single meeting with all the SHs being present (Bautista and Orr, 2011). Unfortunately, due to logistic and financial constraints the gap between first and second round of the ranking exercise was longer than one year. In addition, not all SHs involved during the first iteration could participate also during the second. Thus, although group means of normalized weighting results were taken into account, the observed shifts in indicator ranks might partly be attributed the time gap, due to which in the meantime other factors may have caused the change in mind, as well as an incomplete set of SHs. Outcomes then have to be interpreted with care.

Land-user perceptions of what is meant by certain aspects, such as “land degradation” or “biodiversity”, can differ substantially (Reed et al., 2008). As in the Mier case, this can be particularly evident when comparing different land use systems (e.g. commercial vs. communal), and even in similar biophysical and socio-economic contexts, pointing out that implications are only of relevance at a certain local scale to particular SHs and must be adapted from case to case (Reed et al., 2006; Von Maltitz, 2009). In this respect, the question arises how conducive it is when the MSP contains categories of farmers having a completely different socio-economic background, as is the case comparing freehold ((semi-)commercial) and communal farmers. Contrary to commercial farmers, communal farmers largely don’t have the ability to respond decisively to land degradation due to the absence of governance institutions and property rights, competition for resources, absentee livestock owners or economic inflexibility (Palmer and Bennett, 2013). Environmental awareness is also a question of ownership related responsibilities. Shared ownerships in a communal land management setup reduce individual liability and motivation to conserve natural resources (Van Rooyen, 1998). These factors mitigate against the technical implementation of cost-intensive restoration actions or appropriate management actions without external support and legal policy frameworks. For example, although the present study revealed that shrub control is perceived an excellent action and is preferred above others (compare figure 3 and Table 2), high associated costs for chemicals are a core constraint often hampering its application, particularly on communal land. Thus, the dependency on funding and resources supplied by national and provincial government is a critical factor for sustainable land management in the Mier area. Therefore, it might be advantageous to realign the MSP by stratifying involved farmers according to specific socio-economic settings and farming objectives. This would certainly optimize the identification of suitable and affordable practices based on an improved problem-specific contextualization.

Figure 3: Outranking relationships of the MCDA performed for the duneveld. (A) Graphical expression (GM = good management; SC = shrub control; DS = dune stabilization; direction of arrows indicate an outranking relation, two way arrows mean an equal performance), (B) example of a comparison of pairs of alternatives

Overall, the preliminary indications suggest that the PRACTICE approach is promising for evaluating restoration and management actions in Kalahari rangelands. However, replication and comparisons are essential for proper evaluation of participatory approaches (Reed, 2008). Thus, it remains to be seen how well it performs under the different socio-economic and biophysical setting of the Molopo area, which will then allow for a more detailed and critical evaluation of the framework. Accordingly, and as requested in step 7 of IAPro, dissemination of results will take place after completing this second Kalahari case study in order to support knowledge sharing between land users, farming communities, extensions officers in comparable social, economic and environmental contexts, including researchers and the wider desertification community.

4.2  The way forward

Combating desertification needs to take place particularly at the local level, but concerted action must be enabled by appropriate institutional mechanisms and structures adapted to the particular land tenure forms in South Africa (Von Maltitz, 2009). Thus, the challenge is now to find ways to promote the use of PRACTICE results and the approach itself at local scales and to couple it with local-level institutions. Facing the need for the formulation of environmentally sound and socio-economically accepted prevention and restoration projects to combat desertification in South Africa, an adjusted PRACTICE approach might also represent a promising tool for the exploration of potential alternatives for management and/or restoration options, i.e. other than identified by the local MSP. There is quite a high degree of overlap between the framework of the PRACTICE approach and the participatory approach for land degradation adaptation in other parts of the Kalahari rangelands developed by Reed et al. (2007, 2008), showing many conceptual and methodological parallels. The PRACTICE approach differs mainly in representing in first instance an evaluation tool for management and/or restoration actions (i.e. action assessment), whereas the work of Reed et al. (2007, 2008) ultimately adds up to providing a monitoring tool (i.e. degradation assessment). Combining best practices from both frameworks might be promising to create a more holistic framework for improved decision-making and sustainable land management embedded in a long-term monitoring program.

5  Policy oriented recommendations

Any national and international project that assesses land degradation and restoration actions which will help land users to improve land management decision making must comply to the National Action Programs (NAPs) of the three International Conventions (i.e. UNCCD, CBD and UNFCCC). The NAP of especially the UNCCD emphasizes popular participation and the creation of an enabling environment designed to enable local people to reverse land degradation. The NAP also acknowledges the medium-term strategic framework of government, which is an outcome based approach, similar to that followed by the PRACTICE project. There is already a wealth of information related to combating desertification, land degradation and mitigating the effects of drought (DLDD), which has been collected by universities, government, non-governmental and research organizations through national and international research projects (e.g. PRACTICE, LandCare, Desert Margins Program, Kalahari-Namib Project or LADA (Land Degradation Assessment in Drylands)). The challenge now is that South Africa’s government, in collaboration with all these organizations, creates a supportive policy environment including a comprehensive information system, which makes the information available and more accessible in a format understandable by any SH. This should be facilitated by the established multi-stakeholder-based National Coordinating Body (NCB), whose function is to advise government on the appropriate policies, strategies and action plans for the implementation of South Africa’s NAP for the UNCCD. The Committee on Science and Technology (CST) is a subsidiary body of the NCB, and as such has the function, among others, to help with the improvement of coordination, integration and prioritization of DLDD issues and collating related scientific information that is of use and needed by all SH. The NCB as a whole should therefore play a leading role in the design, formulation and establishment of a South African information system on DLDD, sustainable land management and dryland restoration.


This paper is based on the correspondent extended abstract in the conference proceedings of the UNCCD 2nd Scientific Conference, Bonn, April 2013. We would like to thank Bennie Snyders and the farming community of the Mier area, staff of the extension bureaus and all participants of the PRACTICE workshop in Askham 2012. We are grateful Hendrie Cotzee, Marie du Toit and Yolande van der Watt for input and support. The study was conducted in the frame of the PRACTICE project funded by the European Commission (grant GA22226818).


Aït Younes et al. (2000): ELECTRE IS: Manuel d’utilisation, Document du LAMSADE 118 and 118bis: Université Paris-Dauphine.

Archer, E.R.M. and Tadross, M.A. (2009): Climate change and desertification in South Africa – science and response, African Journal of Range and Forage Science, 26(3): 127-131.

Bhattachan, A. et al. (2012): The Southern Kalahari: a potential new dust source in the Southern Hemisphere?, Environmental Research Letters, 7(2): 1-7.

Bautista, S. and Orr, B.J. (2011): IAPro: PRACTICE Integrated Assessment Protocol, PRACTICE project deliverable 2.3, Valencia: European Commission Research Area, Centro de Estudios Ambientales del Mediterráneo (CEAM).

Botha, M.S. et al. (2008): Capturing lessons from land-users to aid the development of ecological restoration guidelines for lowland Namaqualand, Biological Conservation, 141(4): 885-895.

Figueira, J. and Roy, B. (2002): Determining the weights of criteria in the ELECTRE type methods with a revised Simos’ procedure, European Journal of Operational Research, 139(2): 317–326.

Fraser, E.D.G. et al. (2006): Bottom up and top down: Analysis of participatory processes for sustainability indicator identification as a pathway to community empowerment and sustainable environmental management, Journal of Environmental Management, 78(2): 114-127.

Hoffman, T. and Ashwell, A. (2001): Nature divided. Land degradation in South Africa, Lansdowne/South Africa: University of Cape Town Press.

Mabbutt, J.A. (1986): Desertification indicators, Climate Change, 9(1-2): 113-122.

Milton, S.J. and Dean, R.J. (1995): South Africa’s arid and semiarid rangelands: why are they changing and can they be restored? Environmental Monitoring and Assessment, 37(1-3): 245-264.

Morgenthal, T.L. and Kellner, K. (2008): Operational Manual for FIXMOVE. Proposed methodology to sample vegetation for the national fixed site monitoring programme, Potchefstroom, Report No. GW/A/2008/: Department of Agriculture.

Mucina, L. and Rutherford M.C. (2006): The Vegetation of the South Africa, Lesotho and Swaziland, Pretoria/South Africa: South African National Biodiversity Institute.

Oba, G. et al. (2000): New perspectives on sustainable grazing management in arid zones of sub-Saharan Africa, BioScience, 50(1): 35-51.

Palmer, A.R. and Bennett, J.E. (2013): Degradation of communal rangelands in South Africa: towards an improved understanding to inform policy, African Journal of Range and Forage Science, DOI:10.2989/10220119.2013.779596.

Reed, M.S. (2008): Stakeholder participation for environmental management: A literature review, Biological Conservation, 141(10): 2417-2431.

Reed, M.S. et al. (2006): An adaptive learning process for developing and applying sustainability indicators with local communities, Ecological Economics, 59(4): 406-418.

Reed, M.S. et al. (2007): Integrating local and scientific knowledge for adaptation to land degradation: Kalahari rangeland management options, Land Degradation and Development, 18(3): 249-268.

Reed, M.S. et al. (2008): Participatory indicator development: what can ecologists and local communities learn from each other?, Ecological Applications, 18(5): 1253-1269.

Rojo, L. et al. (2012): Prevention and restoration actions to combat desertification. An integrated assessment: the PRACTICE Project, Sécheresse, 23(3): 219-226.

Rutherford, M.C. and Powrie, L.W. (2010): Severely degraded dunes of the southern Kalahari: local extinction, persistence and natural re-establishment of plants, African Journal of Ecology, 48(4): 930-938.

Sommer, S. (2011): Application of indicator systems for monitoring and assessment of desertification from national to global scales, Land Degradation and Development, 22(2): 184-197.

Van Rooyen, A.F. (1998): Combating desertification in the southern Kalahari: connecting science with community action in South Africa, Journal of Arid Environments, 39(2): 285-297.

Van Rooyen, A.F. (2000): Rangeland degradation in the southern Kalahari, Ph.D. thesis, Pietermarizburg/South Africa: University of Natal.

Vogt, J.V. et al. (2011): Monitoring and assessment of land degradation and desertification: towards new conceptual and integrated approaches, Land Degradation and Development, 22(2): 150-165.

Von Maltitz, G. (2009): Institutions for sustainable land management: reflections on institutional aspects of implementing the UNCCD in South Africa, African Journal of Range and Forage Science, 26(3): 159-168.


Dreber, N.; Kong, T. M.; Kellner, K.; Harmse, C.; van EEDEN; A.; Ocampo-Melgar, A. (2014): Towards improved decision-making in degraded drylands of southern Africa: an indicator based assessment for integrated evaluation of restoration and management actions in the Kalahari rangelands. In: Planet@Risk, 2(1), Special Issue on Desertification: 21-28, Davos: Global Risk Forum GRF Davos.

This article is based on a presentation given during the UNCCD 2nd Scientific Conference on "Economic assessment of desertification, sustainable land management and resilience of arid, semi-arid and dry sub-humid areas", held 9-12 April 2013 in Bonn, Germany (