In West Africa, the most extreme predicted climate change effects are expected to occur in desert and grassland areas. According to Heubes et al. (2011), grassland is projected to expand into the desert over an area of 2 million km² by 2050. However, uncertainty about future rainfall patterns is a major challenge because there is a wide range of estimates projected from different interclimate models (Adger et al. 2003). What do these projections mean for West Africans? For agricultural and water sector actors in many parts of Africa, interclimate model differences in rainfall variability often remain a barrier to the effective use of climate change information by farm managers and other stakeholders. For local populations, this uncertainty is greater, especially when traditional methods of predicting rainfall, e.g., phenology of local trees, have also become unreliable. For African researchers and extension services, the understanding of climate processes, driving forces, and meaningful coping and adaptive strategies remains insufficient (Twomlow et al. 2008, Tschakert and Dietrich 2010).
In dealing with this climate-related uncertainty, implementing well-designed experiments with local populations using an adaptive management approach (Holling 1978) is one of the suggested strategies (Pahl-Wostl 2007). Increasing the adaptive capacity of the actors includes learning based on a more inductive approach, e.g., learning by doing, and anticipation of, or a forward-looking stance toward, the possible implications of extreme effects of climate change, e.g., collective learning improves when actors become aware of how to prepare for unexpected events. However, this also requires an in-depth exploration of the perceptions and reactions of the affected stakeholder groups and the public (Berkes and Folke 1998). Consequently, there are very few published efforts to address such uncertainties in which social constructions and related subjective perceptions of the affected people are explicitly included (Pahl-Wostl 2006). If they are included, this type of knowledge is often only considered in a shallow way because most research efforts are narrowly confined within specific disciplines. At the same time, learning tools exploring anticipatory adaptation are still limited, especially in places and populations that have low levels of adaptive capacity such as in West African countries (Tschakert and Dietrich 2010, d’Aquino and Bah 2013, 2014).
In many ecosystems, unsustainable resource management may result from farmers’ management decisions, their choice of practices, and their ability to respond appropriately to current and future threats (Bernard et al. 2014). Typically, there are multiple actors operating within social-ecological systems who often have distinct specific interests that may be contrasting or competing (Villamor et al. 2014). Hence, participatory adaptive approaches to land-use conflict prevention are needed that facilitate learning among stakeholders (Folke et al. 2002, Cundill et al. 2012). According to Reed et al. (2010), social learning is a change in understanding that extends beyond individuals to become established within broader social units or communities of practice through social interactions among actors by means of their social networks. Because our interest is to explore the possible adaptive strategies and behavior of farm households to future unpredictable rainfall patterns together with demographic and policy trends, we also applied the concept of anticipatory or future-looking learning, which serves as a framework and develops skills to understand future possibilities and the ability to collaborate in creating a preferred future (McGray et al. 2007, Shostak 2009). Accordingly, the latter type of learning involves cycles of discovery, integration, and renewal that keep the actors thinking forward in an ever-changing environment (Shostak 2009). Anticipatory learning assumes that if learning outcomes look to the past (memory) too much, what becomes important is mastering a body of knowledge, i.e., local ecological knowledge, whereas if learning focuses forward, knowledge moves into new possibilities, i.e., emergent. We relied on these definitions of two types of learning throughout our research. Although these types of learning are gaining increasing attention for their application in coping with complexity under climate change, determining how social learning might be better facilitated for this purpose and with what tools is a challenge (Garmendia and Stagl 2010, Reed et al. 2010, Tschakert and Dietrich 2010, Cundill et al. 2012).
Games, particularly role-playing games (RPGs), have become well-recognized as natural resource management tools for better understanding the behavior of human actors (Barreteau et al. 2007). RPGs, especially with game boards, have been used to simulate and help human actors visualize and react to potential future uncertainties based on their existing knowledge and experiences (Vieira Pak and Castillo Brieva 2010, Villamor and van Noordwijk 2011). With respect to land-use decisions in the face of climate change, RPGs can help stakeholders identify options for resilience-building responses to extreme climate change impacts. According to Schelling (1961), games rapidly generate complex interactions and dynamics among decision centers that must reflect the reality of the system being represented. RPGs, especially board games, are like mirrors of the social-ecological systems they represent, in which roles simulate key actors in the real world. Players can interact in a collective and iterative way that improves their understanding of the processes that link the social and ecological systems (Bousquet et al. 2001, 2003, Barreteau et al. 2007). They may reflect on their experience with and their understanding of these systems, while acquiring new knowledge as a result of the interactions that may emerge, as well as imposed scenarios, and subsequently modify their perspective of these systems. Moreover, integrating uncertainties into such games may facilitate change in the way players respond to potential negative impacts, which may in turn stimulate social learning. Self-design RPGs are examples that facilitate social learning both for scientists and local people by incorporating new rules, e.g., collective rules or land-use zoning, in the games (d’Aquino and Bah 2014).
We explored, through the development and application of an RPG board game, the adaptive strategies of local farmers collectively in response to anticipated unfavorable conditions that might result from climate uncertainties. Individuals often tend to be negative or avoid responding to questions about pessimistic future scenarios, e.g., drought, deluge, or extreme sea-level rise. If, collectively, people engage in a game setting with their neighbors and/or colleagues, these issues can be dealt with in an entertaining way that encourages individual participation. Through a game, we emphasized climatic uncertainty, i.e., erratic rainfall patterns, that allowed us to explore how local farmers perceive threats, identify possible coping strategies, and respond in a collective way. The RPG board game is one of the tools identified under the work package (WP) 6.2 of the West African Science Service Center on Climate Change and Adapted Land Use (WASCAL) program (http://www.wascal.org). The goal of WP 6.2 is to identify resilient landscapes under climate change impacts. Among the other tools that are simultaneously tested and implemented are agent-based models and scenario-building exercises (Badmos et al. 2014, 2015, Villamor et al. 2015). Because RPGs are useful tools for envisaging the behavior and responses of farm households to the negative effects of climate variability, we explored whether RPGs can facilitate social and anticipated learning both with the researchers implementing the game and the target stakeholders as players, by addressing the different modes of learning. Our hypothesis is that games can contribute to a better understanding of the resilience of human and environmental systems to climate change and increased variability in a dynamic and collective way. We explored the following questions: (1) How do games facilitate social and anticipated learning? (2) What knowledge, e.g., coping strategies, experiences, and/or changes of understanding of the involved actors are elicited through the games?
We conducted participatory field trials of a grazing management game in the Vea catchment of the districts of Bongo and Bolgatanga, Upper East Region (UER) of Ghana (Fig. 1). The study area is directly bordered by Burkina Faso to the north and Togo to the east. Most of the region belongs to the semiarid West African Guinea savanna belt, with the exception of a small swath of land in the very northeastern part of study area that belongs to the Sudan savanna (Adu 1972). With a total land area of 8842 km², this region represents 3.7% of Ghana’s territory (GSS 2012). Bolgatanga Municipal District has a population of 131,550 with an average household size of 5, approximately 50% of which is rural. Bongo District has a population of 84,545 with an average household size of 6, approximately 94% of which is rural (GSS 2012).
Rainfall in the study area is unimodal with a rainy season peak (60% of annual total) between July and September (Fig. 2). Over the past 40 years, mean annual rainfall has been 1044 mm, which is suitable for a single wet-season crop (IFAD 2007). The rainy season in the UER is relatively short and marked by variations in its onset, duration, and the intensity of rainfall. This creates interannual variation in agricultural production potential (IFAD 2007). Mean annual temperatures are approximately 28°C to 29°C, and the absolute minimum temperatures are approximately 15°C to 18°C (Mdemu 2008). The hottest period of the year is around March and April, and the coolest period occurs around August.
The communities in this region practice agropastoralism (Eguavoen 2013, Yembilah and Grant 2014). Agricultural activities are the main source of income in the area and are carried out both during the rainy season (rain fed) and the dry season (irrigated), but most agricultural work is concentrated during the rainy season. Within the immediate farmer’s environment, traditional cereals are the basis for the agricultural system. The traditional cereals are cultivated around the compound, whereas groundnut (Arachis hypogaea), mixed or monoculture, may be cultivated within the compound or at some distance from the house. Other crops that are traditionally cultivated include guinea corn, millet, and rice. The cropping system practiced by the local farmers is mostly for subsistence. In terms of livestock production, the most common include cattle; sheep; goats; poultry, primarily guinea fowl (Numida meleagris) and ducks; swine; and donkeys. In the region, cattle ownership is a measure of wealth and social status (Yilma 2005). Together with smaller livestock, cattle are often used to pay bridal dowries. During periods of extreme stress, the smaller livestock are more easily liquidated to address immediate household needs.
The grazing game we employed is a modified version of the “overgrazing game” developed by Van Noordwijk (1984) as a teaching method for university students in the context of the Sudan savanna. The objective of the game is to reveal the processes that lead to overgrazing and desertification. To apply the game for exploring the coping strategies of local farmers, we modified the game to match our study area and research objectives. The modified settings of this game simulate dryland conditions where agricultural production is highly dependent on rainfall with additional factors that introduce flexibility and complexity into the game. We based this assumption on several studies conducted in the region regarding the lack of reliability of rainfall, including the timing, associated to interannual variability of both distribution and total amounts of rainfall (Dietz et al. 2004, Van der Geest and Dietz 2004, Amikuzino and Donkoh 2012).
Figure 3 depicts the conceptual model of the modified game. It involves the actors, i.e., farmers and market; resources, i.e., cows, grass, patches of land, and rainfall; processes, i.e., reproduction and regrow of grass; and strategies, i.e., keep the cows, sell the cows, or locate the cows in the valley or on hills. The basic setting is that the rainfall patterns that farmers rely on for their livelihoods become more unpredictable, and the land degradation process is influenced by rainfall and the number of grazers. The farmers’ livelihoods are dependent on cattle raising, and crop production is limited to household subsistence use in a restricted area. The basic processes in the game involve grass development cycles based on rainfall: less rainfall limits grass growth, and more rainfall increases bush development and cow reproduction. The resources conceptualized in the model are indicators to be monitored, such as total produced, total sold, and reproductive increases in herd size. The arrows depicted in Figure 3 describe the interactions among causes and effects, including, but not limited to, simple decision strategies such as selling cows, maintaining cows, and grazing management. To make the game handy, we used the term “cows” in the game and assumed that a bull was available for reproduction; the term “cattle” refers to both cows and their calves.
The game board representing all available land is organized by a grid of 8 × 8 cells, each of which measures 5 × 5 cm, for a total of 64 cells or “patches” of land on the game board. A total of 16 patches at the center of the game board represents a “valley,” where water is assumed to be available throughout the year.
A six-sided die is used to determine the amount of rainfall and grass production before each round of the game (Fig. 4). Herd indicators, e.g., pebbles, are used to represent herds of cows. Each herd is composed of five cows as a starting herd size.
Land patches are colored coded according to land cover type: red patches represent desert, and green patches represent bush. Colored pins are used to indicate the quantity of grasses and crops. A score sheet (see Appendix 1) is provided to monitor the status of individual players, such as the number of cows produced (yield) and the number of cows sold, and an observer cross-checked the conversations of the players.
Each time step or round of the game represents 1 annual cycle. Each year is divided into 2 seasons: a rainy season and a dry season. The rainy season begins in April and lasts for a period of 7 months. Afterward, the dry season begins in November and ends in March. In northern Ghana, the rainy season is also referred to as the growing season.
In dryland areas, rainfall is low and erratic. In the game, the die is used to determine the amount of rainfall once a year for each land patch on the board. The amount of grass growth varies along a range from 1 to 6 markers. For example, if the die indicates the number 1, then each field of the board will have a unit of grass during that round of the game.
Every month, each cow in a herd requires 1 unit of grass. The herd can move through 2 neighboring patches per month. If the full requirements of the cows are not met, they can be fed at half a ration, but this will affect both reproduction and sale value. If individual cows are not fed at all, they perish. The crop residues remaining after the harvest of corn, millet, peanut or groundnut (A. hypogaea), and rice can be used to feed cows, but only during the month that crops are harvested.
At the end of each dry season, the cows that have been fed full rations for the past 6 months give birth to a calf. At the end of each rainy season, cows can be sold at the discretion of the player. If cows have not been fully fed over the previous 6 months, their value (count) is reduced by one-half. If a herd consists of 6 cows or more, it may be split into 2 subherds that graze separately. Subherds must be reunited if they are reduced to fewer than 3 cows.
After the first year, there are the following additional rules for determining the vegetation on the basis of rainfall:
In terms of players, each game included 5 to 15 players. Each game had a game master, an observer to document the conversation each round, and a recorder to maintain scores and facilitate the process. The games were also facilitated by 2 additional research assistants who were locals and native speakers in each of the study sites. The game trial locations were selected based on the results of an initial household survey conducted within the study area (Badmos et al. 2013). A total of 23 game trials involving 243 individual farmers, excluding children and other bystanders, were conducted from August to October 2013. These individual farmers were also part of the initial household survey, and we asked if they were willing to participate in the game. Table 1 presents the key characteristics of the surveyed households. Each of the games represents the subvillages in the catchment study area. Typically, 1 or 2 female players participated in each game.
Before the beginning of each game, the players were asked to locate 4 patches with their choice of crops, e.g., 1 unit of millet, 1 unit of corn, 1 unit of rice, and 1 unit of groundnut. Each player began with a herd of 5 cows that would graze in one of the suitable patches. The objectives were to manage the herd, maximize the production of cows, and avoid desertification. The game master explained the objectives and the rules of the game, including the score sheets (see Appendix 1). The score sheets were used to track the indicators, e.g., primary production of grass, amount of grass used, number of calves produced, number of desert patches, number of bush patches, number of cows sold, and amount of fertilizer bought, during each round.
A pretest was conducted before each full game trial to make certain that the rules of the game were clearly understood by the players. Each full game consisted of 5 rounds. Each round was composed of 12 months. Depending on the number of players per game, players rotated turns per month or per year to graze (move) on the game board. Each herd began grazing inside the valley, i.e., central 16 patches on the board, for the first month of year/round 1 before the herd could be moved to graze outside the valley. The game master would score the result after each round and announce the status of achieving the players’ goals, i.e., the number of calves produced and the number of desert or bush patches created. During the course of each annual round, the game master would also announce the beginning and end of the rainy and dry seasons and ask whether players wanted to sell cows. At the end of each game, a reflection exercise was conducted to clarify and verify the strategies/decisions made by the players and for them to assess the overall game. Typically, we asked multiple-choice and open-ended questions regarding the quality of the game, i.e., playability, perceived value as a learning tool, and so forth; reflection of reality; cooperation; role of the government; local ecological knowledge; and ways to improve the game (see Appendix 2).
We used the game scenarios that were identified during the participatory scenario exploration exercise before the grazing game implementation (Badmos et al. 2014). During the game, the game master announced scenarios for the following years:
We compared results of the games by plotting key indicators, i.e., the number of desert patches against accumulated yields or the total number of cows produced, throughout the game. From these indicators, we determined the best performers on the following (1) subvillages that had above-average yields under limited rainfall conditions and (2) subvillages that had the smallest number of desert patches under limited rainfall conditions. For simplification, we categorized the average rainfall generated by the die, i.e., the average die value of the farmers per year, as follows: 1.0-2.5 as very dry, 2.6-3.0 as dry, 3.1-4.5 as wet, and 4.6-6.0 as very wet.
Yields and average annual rainfall values generated from 23 game trials are presented in Figure 5. Approximately 14 (61%) games had mean annual rainfall values between 3 and 4, which fall under the dry and wet categories. Eight games had mean annual rainfall values in the wet category, and only 1 game had values in the very wet category. Among the games that had mean annual rainfall values in the very dry and dry categories, 5 subvillages had above-average yields, i.e., in terms of the number of cows produced. Among games with mean annual rainfall values in the dry category, the best cow producer was the subvillage Yorogo, with a total of 110 cows.
The mean percentage of patches that were classified as desert, i.e., degraded land, by the end of the game was approximately 11%. Figure 6 presents a summary of the final percentages of desert patches. Only 7 of the games (30%) with mean annual rainfall in the dry category had below-average percentages of desert patches. The top performer among these games was the subvillage of Bongo II, which finished the game with only 2 desert patches. There were multiple cases of games ending with higher percentages of desert patches despite higher rainfall, underlining the fact that the causes of desertification can be unrelated to climatic factors (see case 3 of Fig. 7). Only 1 game was completed without any desert patches.
From the 23 games that were conducted, we selected 3 games that are representative of the resulting rainfall trends. Figure 7 presents a summary of land-use trajectories (Fig. 7B) and yields from cow production (Fig. 7C) according to rainfall categories. In case 1, the pattern of average rainfall (Fig. 7A) shows a concave shape with decreasing trend, case 2 has a moving average rainfall in a convex shape with increasing trend, and case 3 has a moving average with a stable but increasing trend. In cases 1 and 2, the land-use trajectories and yields from cow production mirrored the rainfall patterns, suggesting that as rainfall decreased the process of desertification intensified and yield from cow production decreased. However, in case 3, the percentage of desert patches increased even though rainfall was relatively abundant. Because the yields from cow production were high in case 3, this appears to be an example of anthropogenic land degradation or overgrazing.
During the course of the games, we observed the players’ strategies under decreasing rainfall conditions. Some of the farmers were adept at strategizing their cattle movements with respect to timing and ration levels (Table 2). The most common decision-making process was that farmers would wait until after the rainfall was determined, by the die, before deciding on a coping strategy, i.e., selling or keeping cattle. This suggests that farmers would prefer to make decisions based on relevant information.
The most common coping strategy for rainfall variability/uncertainty was selling livestock (Table 3). In fact, the participants commented on the importance of paying attention to the condition of their livestock. One comment that summarized the reaction of many participants was that “though cows are quite expensive to acquire, we let them graze around the place and we don’t notice them until we realized that they were already dead.” Some of the women players opted to sell their livestock and save the money for coping with the dry seasons, or drought events, whereas other women proposed seeking alternative livelihoods, i.e., basket weaving. Among males players, many opted to seek migrant labor jobs.
Another coping strategy identified was the cultivation of new drought-resistant crop varieties. Maize is a relatively new crop in the study area. Farmers expressed a preference for maize because it only requires 4 months rainfall and matures before the end of the rainy season; crops that require longer periods to mature can present problems.
Approximately 98% of the players perceived the game as a reasonably good reflection of reality. The time required to introduce the game and conduct the pretest typically ranged from 30 to 45 minutes. As the game boards were designed to simply represent the landscape in the study sites, where the inner patches (4 × 4 central areas) represent the valley and the outer patches (8 × 8) represent the hills or uplands, we allowed players to locate their crops according to their customary practices. For example, the corn and rice were cultivated in separate patches, whereas millet and peanut were always cultivated together. Rice was always cultivated in the valley because water is available there throughout the year. The overlapping arrangement of millet and peanut on the game board matched our observations of the farmers’ plots.
When we introduced scenarios in years 3 and 4 of the game sequence, we observed that at first, inconvenience was felt by all the players, especially the older ones. In year 3, we introduced the population increase by adding a new household as a newcomer. As observed, the newcomer was always played by a younger farmer in the group and always followed the suggestions of the older households. During the course of the games, households would compete for available resources while we observed how they cooperated with each other. The following are some of the direct statements from the farmers:
If he has one bull and another farmer also has [one], we can combine them to plough for each other.
Giving my herd to another farmer helped me keep [some] animals for me.
Allowing the household to settle in my farm plot to let his land restore new grasses
To live in peace and consult for information with my neighbors especially when the rain is delayed
Plan together and respect each other’s views.
Asking a neighbor for food support and money to buy food
The erratic rainfall patterns generated through the games were one of the features that players considered realistic. For instance, some participants explained that the rainfall in 2013 was delayed by more than a month compared to the typical annual pattern. For this reason, some farmers were reluctant to plant early millet, a traditional crop. Every time a desert was “created,” it was immediately recognized as a serious problem. Some participants shouted, “Disaster!” when this occurred, or else they paused and laughed at each other and expressed that the person who threw the die had bad luck. During the game, some players expressed their desire to shift their traditional cultivation practices to maize production to secure their harvest. Some participants, particularly women, also suggested restoring irrigation canals to allow them to continue cultivating crops during the dry season. To secure the production of at least 1 calf every year, the players kept at least 1 cow at full rations for the required 6-month period. There were many instances of players using crop residues to feed their cows to maintain grass availability in other patches. Other behaviors observed during the games are presented in Table 4.
Based on a very abstract representation of the dryland social-ecological system, one of the main findings was how the local people relied on their ecological knowledge, which affirmed some prior ecological knowledge of the facilitators, i.e., researchers. During the course of the games, players often shared specific ecological insights that indicated means for coping with uncertainty.
In the context of the drylands and climatic uncertainty, how do games contribute to better understanding of the resilience of human and environmental systems to climate change and increased variability, as well as facilitate social and anticipatory learning? Furthermore, what knowledge, experiences, and/or changes of understanding of the actors involved are elicited through the games?
Basically, the grazing game is instrumental for eliciting the players’ subjective perceptions, goals, and expectations within the specific context described. It is a simple and straightforward tool for better understanding the perceptions and behaviors of the local people in a relatively realistic context. Moreover, this tool facilitates social learning particularly for the scientists involved. The key aspects for assessing the degree to which the grazing game is an effective tool for facilitating social learning are summarized in Table 5. This information was generated during the reflection sessions of each game, in which the players exchanged information with the facilitators/game master, i.e., researchers and modelers, and it was considered the most important part of the entire process. The facilitators and researchers investigated the reasons behind participants’ reactions more deeply and, at the same time, verified and validated the behaviors and perceptions of the players observed during the game. Although most of the local ecological knowledge listed previously was not that new to scientific ecological knowledge, the integration of the two knowledge systems harnessed an understanding of the system (Stringer and Reed 2007) that may be useful for the development of management plans, e.g., provision of maps of various water bodies to communities. For example, a study conducted in the savannas of northern Nigeria on traditional knowledge for predicting rainfall variability also found that local people identified guinea fowl as a climatic condition indicator (Sanni et al. 2012), which concurs with the local farmers in the region. In addition, our results support the outcomes of the participatory scenario exercises (Badmos et al. 2014), such as consideration of water availability over the lifetime of a selected crop before deciding whether to cultivate it, as well as consideration of other varieties of crops and land suitability.
However, fully addressing the definition of social learning of Reed et al. (2010) is not possible in the current implementation design of the RPGs, and we considered this a limitation. One reason is that transformative forms of learning (see Table 5, one of the three social learning modes) are hard to achieve because they involve changes in the underlying values, belief systems, norms, higher order thinking processes, and worldviews of the individuals. Although RPGs, if well designed, can facilitate double-loop learning (Schelling 1961), according to Pahl-Wostl (2009) this kind of societal learning involves transitions of entire regimes, e.g., from optimal land-use planning to integrated landscape planning. In fact, belief systems or core normative beliefs may be so deep that changing them may require time and specific perturbations (Sabatier 1988, Villamor 2006).
On the other hand, in terms of anticipatory learning, the farm households’ behavior and patterns observed in the game provide a different insight. As mentioned previously, some of the local ecological knowledge from the game overlaps with the existing scientific ecological knowledge (Hesse and MacGregor 2006, Flintan et al. 2013). This suggests that either the local respondents have mastered strongly their local knowledge (Shostak 2009) or that scientific knowledge is also reliant on the local existing knowledge. Because the average farmer in the study area depends largely on farming (Table 1), by anticipating more erratic rainfall patterns in the future the players saw the importance of their small animals, i.e., guinea fowl, during extreme drought events for immediate cash (as reflected in Table 2).
In addition, as expressed by the players, the dynamic consequences of human decision making in response to unpredictable rainfall patterns made the game more realistic and engaging. The players who generated unpredicted rainfall patterns through the use of a die made the game more exciting. Thus, they were more reactive to the rainfall while being proactive in terms of strategizing cow production efforts. In comparison, another study used RPGs to integrate seasonal climate forecasts into the context of the livelihoods of small-scale producers in the environment of southern Africa (Ziervogel 2004), with a design based on suggestions about how to respond to different rainfall patterns. In that case, each game was composed of three rounds: the first round reflected a weather forecast of normal rainfall, the second round included above-normal rainfall, and the third round included below-normal rainfall. In each round, the players, i.e., small-scale farmers, were asked to reassess their decisions. The game provoked suggestions about how to adapt to those scenarios; however, the author commented that there were not many suggestions because it was hard for the players to relate to the impacts of climate change, especially in terms of their livestock.
We aimed, through the use of RPGs, to explore the adaptive strategies of the local farmers and their responses to highly unpredictable rainfall patterns and not to make decisions for them. Unlike self-design RPGs, participants were not involved in the formulation of the model; rather, we used this simple game for them to refine and improve the tool for further replications in other WASCAL pilot sites. It was so simple that the players easily related to the game concept, and it was favorable especially if the end target was the integration of different knowledge systems (Villamor et al. 2014). With respect to representing actual complex systems using soft system tools, the debate, i.e., simplification versus realism, remains and is understudied (Voinov and Bousquet 2010). Nevertheless, Barreteau et al. (2013) suggested the following indicators to assess the quality of the tools used, which we considered in the RPG:
We designed the grazing game and conducted field trials to explore the perceptions and coping strategies of the local people in the drylands of the UER of Ghana under climate variability. The game enabled us to examine farmers’ behavior under unpredictable rainfall patterns with clear targets for their livelihoods with consequences such as land degradation. Through the game trials, we identified coping strategies and local ecological knowledge for increased climatic variability. Furthermore, the game provided the farmers with an opportunity for observing the implications of their land-use decisions on their livelihoods. The replication of the game around the study area enabled us to assess whether the game can facilitate social learning. We found that the game was instrumental for eliciting the players’ subjective perceptions, goals, and expectations within the specific context described but was limited in facilitating social learning, particularly the transformative form of learning.
The authors are grateful for the financial support provided by the West African Science Service Center on Climate Change and Adapted Land Use (WASCAL). We also thank Meine van Noordwijk and Utkur Djanibekov for their comments and suggestions, and John Maalog and Christopher Abotisum for translating the game rules in their local dialect and facilitating the game sessions. We highly appreciate the valuable comments and suggestions of three anonymous reviewers.
Adger, W. N., S. Huq, K. Brown, D. Conway, and M. Hulme. 2003. Adaptation to climate change in the developing world. Progress in Development Studies 3:179-195. http://dx.doi.org/10.1191/1464993403ps060oa
Adu, S. V. 1972. Eroded savanna soils of the Navrongo-Bawku area, northern Ghana. Ghana Journal of Agricultural Sciences 5:3-12.
Amikuzino, J., and S. A. Donkoh. 2012. Climate variability and yields of major staple food crops in Northern Ghana. African Crop Science Journal 20:349-360.
Angelucci, F. 2012. Analysis of incentives and disincentives for maize in Ghana. Technical Note Series. Monitoring African Food and Agricultural Policies, Food and Agriculture Organization of the United Nations, Rome, Italy.
Badmos, B. K., S. K. Agodzo, G. B. Villamor, and S. Odai. 2015. An approach for simulating soil loss from an agro-ecosystem using multi-agent simulation: a case study for semi-arid Ghana. Land 4:607-626. http://dx.doi.org/10.3390/land4030607
Badmos, B. K., G. B. Villamor, S. K. Agodzo, and S. N. Odai. 2013. Preliminary study on the impact of climate change and socio-economy on some farming communities in northern Ghana. In 2nd Climate Change and Population Conference on Africa (Accra, Ghana, 3-7 June). Climate and Development Knowledge Network, London, UK.
Badmos, B. K., G. B. Villamor, S. K. Agodzo, S. N. Odai, and S. S. Guug. 2014. Examining agricultural land-use/cover change options in rural northern Ghana: a participatory scenario exploration exercise approach. Journal of Interdisciplinary Environmental Studies 8:15-35.
Barreteau, O., P. Bots, K. Daniell, M. Etienne, P. Perez, C. Barnaud, D. Bazile, N. Becu, J.-C. Castella, W. Daré, and G. Trebuil. 2013. Participatory approaches. Pages 197-234 in B. Edmonds and R. Meyer, editors. Simulating social complexity: a handbook. Springer, Berlin, Germany. http://dx.doi.org/10.1007/978-3-540-93813-2_10
Barreteau, O., C. Le Page, and P. Perez. 2007. Contribution of simulation and gaming to natural resource management issues: an introduction. Simulation & Gaming 38:185-194. http://dx.doi.org/10.1177/1046878107300660
Berkes, F., and C. Folke. 1998. Linking social and ecological systems: management practices and social mechanisms for building resilience. Cambridge University Press, Cambridge, UK.
Bernard, F., M. van Noordwijk, E. Luedeling, G. B. Villamor, G. W. Sileshi, and S. Namirembe. 2014. Social actors and unsustainability of agriculture. Current Opinion in Environmental Sustainability 6:155-161. http://dx.doi.org/10.1016/j.cosust.2014.01.002
Bousquet, F., O. Barreteau, P. d’Aquino, M. Etienne, S. Boissau, S. Aubert, C. Le Page, D. Babin, and J.-C. Castella. 2003. Multi-agent systems and role games: collective learning processes for ecosystem management. Pages 248-285 in M. Janssen, editor. Complexity and ecosystem management: the theory and practice of multi-agent approaches. Edward Elgar, Cheltenham, UK.
Bousquet, F., R. Lifran, M. Tidball, S. Thoyer, and M. Antona. 2001. Agent-based modelling, game theory and natural resource management issues. Journal of Artificial Societies and Social Simulation 4(2). [online] URL: http://jasss.soc.surrey.ac.uk/4/2/0.html
Cundill, G., G. S. Cumming, D. Biggs, and C. Fabricius. 2012. Soft systems thinking and social learning for adaptive management. Conservation Biology 26:13-20. http://dx.doi.org/10.1111/j.1523-1739.2011.01755.x
d’Aquino, P., and A. Bah. 2013. A participatory modeling process to capture indigenous ways of adaptability to uncertainty: outputs from an experiment in West African drylands. Ecology and Society 18(4):16. http://dx.doi.org/10.5751/ES-05876-180416
d’Aquino, P., and A. Bah. 2014. Multi-level participatory design of land use policies in African drylands: a method to embed adaptability skills of drylands societies in a policy framework. Journal of Environmental Management 132:207-219. http://dx.doi.org/10.1016/j.jenvman.2013.11.011
Dietz, T., D. Millard, S. Dittoh, F. Obeng, and E. Ofori-Sarpong. 2004. Climate and livelihood change in North East Ghana. Pages 149-172 in A. J. Dietz, R. Ruben, and A. Verhagen, editors. The impact of climate change on drylands: with a focus on West Africa. Kluwer Academic, Dordrecht, The Netherlands. http://dx.doi.org/10.1007/1-4020-2158-5_12
Eguavoen, I. 2013. Climate change and trajectories of blame in Northern Ghana. Anthropological Notebooks 19:5-24.
Flintan, F., R. Behnke, and C. Neely. 2013. Natural resource management in the drylands in the Horn of Africa. Brief prepared by a Technical Consortium hosted by CGIAR in partnership with the FAO Investment Centre. Technical Consortium Brief 1. International Livestock Research Institute, Nairobi, Kenya.
Folke, C., S. Carpenter, T. Elmqvist, L. Gunderson, C. S. Holling, and B. Walker. 2002. Resilience and sustainable development: building adaptive capacity in a world of transformations. AMBIO: A Journal of the Human Environment 31:437-440. http://dx.doi.org/10.1579/0044-7447-31.5.437
Garmendia, E., and S. Stagl. 2010. Public participation for sustainability and social learning: concepts and lessons from three case studies in Europe. Ecological Economics 69:1712-1722. http://dx.doi.org/10.1016/j.ecolecon.2010.03.027
Ghana Statistical Service (GSS). 2012. 2010 Population & housing census: summary report of final results. GSS, Accra, Ghana.
Hesse, C., and J. MacGregor. 2006. Pastoralism: drylands’ invisible asset? Developing a framework for assessing the value of pastoralism in East Africa. Issue Paper No. 12. International Institute for Environment and Development, London, UK.
Heubes, J., I. Kühn, K. König, R. Wittig, G. Zizka, and K. Hahn. 2011. Modelling biome shifts and tree cover change for 2050 in West Africa. Journal of Biogeography 38:2248-2258. http://dx.doi.org/10.1111/j.1365-2699.2011.02560.x
Holling, C. S. 1978. Adaptive environmental assessment and management. Wiley-Interscience, Chichester, UK.
International Fund for Agricultural Development (IFAD). 2007. Ghana: Upper East Region Land Conservation and Smallholder Rehabilitation Project (LACOSREP). IFAD, Rome, Italy. [online] URL: http://www.ifad.org/evaluation/public_html/eksyst/doc/prj/region/pa/ghana/s026ghbe.htm
McGray, H., A. Hammill, R. Bradley, E. L. Schipper, and J.-E. Parry. 2007. Weathering the storm: options for framing adaptation and development. World Resources Institute, Washington, D.C., USA.
Mdemu, M. V. 2008. Water productivity in medium and small reservoirs in the Upper East Region (UER) of Ghana. Dissertation. Universitäts-und Landesbibliothek Bonn, Bonn, Germany.
Pahl-Wostl, C. 2006. The importance of social learning in restoring the multifunctionality of rivers and floodplains. Ecology and Society 11(1):10. [online] URL: http://www.ecologyandsociety.org/vol11/iss1/art10/
Pahl-Wostl, C. 2007. Transition towards adaptive management of water facing climate and global change. Water Resources Management 21:49-62. http://dx.doi.org/10.1007/s11269-006-9040-4
Pahl-Wostl, C. 2009. A conceptual framework for analysing adaptive capacity and multi-level learning processes in resource governance regimes. Global Environmental Change 19:354-365. http://dx.doi.org/10.1016/j.gloenvcha.2009.06.001
Reed, M. S., A. C. Evely, G. Cundill, I. Fazey, J. Glass, A. Laing, J. Newig, B. Parrish, C. Prell, C. Raymond, and L. C. Stringer. 2010. What is social learning? Ecology and Society 15(4):r1. [online] URL: http://www.ecologyandsociety.org/vol15/iss4/resp1/
Sabatier, P. A. 1988. An advocacy coalition framework of policy change and the role of policy-oriented learning therein. Policy Sciences 21:129-168. http://dx.doi.org/10.1007/BF00136406
Sanni, S. A., K. O. Oluwasemire, and N. O. Nnoli. 2012. Traditional capacity for weather prediction, variability and coping strategies in the front line states of Nigeria. Agricultural Sciences 3:625-630. http://dx.doi.org/10.4236/as.2012.34075
Schelling, T. C. 1961. Experimental games and bargaining theory. World Politics 14:47-68. http://dx.doi.org/10.2307/2009555
Shostak, A. 2009. Four defining dimensions of anticipatory learning, by Marsha Lynn Rhea. Educational Futuristics. 19 July. [online] URL: http://www.educationalfuturistics.com/2009/07/19/four-defining-dimensions-of-anticipatory-learning-by-marsha-lynn-rhea/
Stringer, L. C., and M. S. Reed. 2007. Land degradation assessment in Southern Africa: integrating local and scientific knowledge bases. Land Degradation & Development 18:99-116. http://dx.doi.org/10.1002/ldr.760
Tschakert, P., and K. A. Dietrich. 2010. Anticipatory learning for climate change adaptation and resilience. Ecology and Society 15(2):11. [online] URL: http://www.ecologyandsociety.org/vol15/iss2/art11/
Twomlow, S., F. T. Mugabe, M. Mwale, R. Delve, D. Nanja, P. Carberry, and M. Howden. 2008. Building adaptive capacity to cope with increasing vulnerability due to climatic change in Africa – a new approach. Physics and Chemistry of the Earth, Parts A/B/C 33:780-787. http://dx.doi.org/10.1016/j.pce.2008.06.048
Van der Geest, K., and T. Dietz. 2004. A literature survey about risk and vulnerability in drylands, with a focus on the Sahel. Pages 117-146 in A. J. Dietz, R. Ruben, and A. Verhagen, editors. The impact of climate change on drylands: with a focus on West Africa. Kluwer, Dordrecht, The Netherlands. http://dx.doi.org/10.1007/1-4020-2158-5_11
Van Noordwijk, M. 1984. Ecology textbook for the Sudan. Grafische Kring Groningen, Amsterdam, The Netherlands.
Vieira Pak, M., and D. Castillo Brieva. 2010. Designing and implementing a role-playing game: a tool to explain factors, decision making and landscape transformation. Environmental Modelling & Software 25:1322-1333. http://dx.doi.org/10.1016/j.envsoft.2010.03.015
Villamor, G. B. 2006. The rise of protected area policy in the Philippine forest policy: an analysis from the perspective of Advocacy Coalition Framework (ACF). Forest Policy and Economics 9:162-178. http://dx.doi.org/10.1016/j.forpol.2005.04.003
Villamor, G. B., P. A. Dah-gbeto, A. Bell, U. Pradhan, and M. van Noordwijk. 2015. Gender-specific spatial perspectives and scenario building approaches for understanding gender equity and sustainability in climate-smart landscapes. Pages 211-224 in P. A. Minang, M. van Noordwijk, O. E. Freeman, C. Mbow, J. de Leeuw, and D. Catacutan, editors. Climate-smart landscapes: multifunctionality in practice. World Agroforestry Centre (ICRAF), Nairobi, Kenya.
Villamor, G. B., I. Palomo, C. A. López Santiago, E. Oteros-Rozas, and J. Hill. 2014. Assessing stakeholders’ perceptions and values towards social-ecological systems using participatory methods. Ecological Processes 3:22. http://dx.doi.org/10.1186/s13717-014-0022-9
Villamor, G. B., and M. van Noordwijk. 2011. Social role-play games vs individual perceptions of conservation and PES agreements for maintaining rubber agroforests in Jambi (Sumatra), Indonesia. Ecology and Society 16(2):27. http://dx.doi.org/10.5751/ES-04339-160327
Voinov, A., and F. Bousquet. 2010. Modelling with stakeholders. Environmental Modelling & Software 25:1268-1281. http://dx.doi.org/10.1016/j.envsoft.2010.03.007
Yembilah, R., and M. Grant. 2014. The impact of herder sedentarization on natural resource access in northeastern Ghana. Society & Natural Resources: An International Journal 27:621-635. http://dx.doi.org/10.1080/08941920.2014.888793
Yilma, T. 2005. Modeling farm irrigation decisions under rainfall risk in the White-Volta Basin of Ghana: a tool for policy analysis at the farm-household level. Dissertation. University of Bonn, Bonn, Germany.
Ziervogel, G. 2004. Targeting seasonal climate forecasts for integration into household level decisions: the case of smallholder farmers in Lesotho. Geographical Journal 170:6-21. http://dx.doi.org/10.1111/j.0016-7398.2004.05002.x