Restoration of degraded agricultural terraces

Journal of Environmental Management 138 (2014) 32e42

Contents lists available at ScienceDirect

Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman

Restoration of degraded agricultural terraces: Rebuilding landscape structure and process M.C. LaFevor Department of Geography and the Environment, University of Texas at Austin, Austin, TX 78712, United States

a r t i c l e i n f o

a b s t r a c t

Article history: Received 30 July 2013 Received in revised form 31 October 2013 Accepted 14 November 2013 Available online 16 December 2013

The restoration of severely degraded cropland to productive agricultural capacity increases food supply, improves soil and water conservation, and enhances environmental and ecological services. This article examines the key roles that long-term maintenance plays in the processes of repairing degraded agricultural land. Field measurements from Tlaxcala, Mexico stress that restoring agricultural structures (the arrangements of landforms and vegetation) is alone insufficient. Instead, an effective monitoring and maintenance regime of agricultural structures is also crucial if the efforts are to be successful. Consequently, methods of wildland restoration and agricultural restoration may differ in the degree to which the latter must plan for and facilitate a sustained human involvement. An improved understanding of these distinctions is critical for environmental management as restoration programs that employ the technologies of intensive agriculture continue to grow in number and scope. Ó 2013 Elsevier Ltd. All rights reserved.

Keywords: Agriculture Restoration Environment Terrace Mexico

1. Introduction Agricultural landscapes are the primary medium through which many societies receive critical environmental and ecosystem services (Swift et al., 2004; Swinton et al., 2007; Bohlen et al., 2009). Cultivation processes produce food and create agroecosystems that affect soil, water, climate, biodiversity, and a wide range of market and non-market driven services (Gliessman, 1998; Wood et al., 2000; Altieri, 2004). The degradation of agroecosystems represents both a decrease in the potential food supply and a degradation of the natural resources upon which society depends. The restoration of severely degraded agricultural lands to productive capacity, in the absence of a return to more ‘natural’ conditions, offers the potential for improved soil, water, and biodiversity conservation (Wade et al., 2008; Pywell et al., 2011). But as artificially structured, human-created environments, agrosystems require some degree of continued human involvement to develop (Doolittle, 2006). In prioritizing the short-term environmental remediation of degraded agricultural environments, restoration programs often neglect to plan for and facilitate processes of longterm maintenance. Maintenance is a critical, though often overlooked part of building and sustaining agroecological structures and processes.

E-mail address: [email protected]. 0301-4797/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jenvman.2013.11.019

Consequently, greater emphasis on the structures and processes of intensive agricultural environments, in many respects distinct from wildland or natural environments, is critical for environmental management. If the ultimate goal of restoration, broadly defined, is to build self-supporting ecosystems that are resilient to perturbation without further assistance (Ruiz-Jaen and Aide, 2005; SER, 2004), the restoration of agroecosystems indeed requires modified methods and criteria. Institutional efforts to restore severely degraded cropland may effectively apply the principles of restoration ecology, but they are likely to require a more sustained human involvement. This study examines the role of maintenance in restoring the structures and processes of agricultural terraces along severely degraded hillslopes in the Mexican state of Tlaxcala (Fig. 1). Field measurements of surface erosion and sediment accumulation on newly restored terraces provide structural evidence of flawed restoration processes. Newly built, incipient terraces have been constructed upon degraded agricultural land. Erosion mitigation structures, retention ditches (zanjas) and vegetated berms (bordos), however, are degrading at unsustainable rates, with no plans for maintenance or upkeep. This study stresses that the processes of rebuilding agricultural landscapes, especially as a high-input form of environmental remediation, must also plan for and facilitate sustained site maintenance to be effective over the long term (1012 yrs). This fundamental characteristic of intensive agrosystems is often not considered in the search for quick and inexpensive approaches to repair environmental damage.

M.C. LaFevor / Journal of Environmental Management 138 (2014) 32e42

33

Fig. 1. Map of the study area.

1.1. Agroecological restoration with traditional technologies The repair of ecosystem processes is a critical component of successful environmental restoration (Bradshaw, 1997; Whisenant, 1999; Herrick et al., 2006). But with about 33% of Earth’s surface under some form of agricultural production (FAO, 2013), an improved understanding of agroecological processes has become increasingly relevant in environmental management. Restoration ecologists have incorporated many of the land-shaping technologies of intensive agriculture into wildland repair efforts (Jordan et al., 1987; Harper, 1982), although with mixed results (Whisenant, 1999 p. 16). Far from a panacea for environmental problems (Altieri, 1995), some traditional or indigenously developed technologies nonetheless offer advantages over conventional approaches in that many represent cost-effective, low external input (LEI), and ecologically sustainable forms of environmental management (Gliessman, 1998; Reijntjes et al., 1992). But transferring traditional strategies into modern restoration contexts can be problematic (Wilken, 1989; Kaimowitz, 1990), as distinctions between agroecological structures and processes, especially those involving incremental changes (Doolittle, 1984), are often overlooked or inadequately understood. Topedown efforts that attempt to mimic traditional structures can prove ineffective or counterproductive over the long term (Chapin, 1988; Doolittle, 1989). Traditional farming approaches represent useful baseline analogs for contemporary development, restoration, or adaptive management efforts (Altieri et al., 2012; Berkes et al., 2000; Reij et al., 1996; Erickson, 1988). But comparative studies of their effectiveness are key, as accepting and implementing them on purely emotional or ideological grounds is irresponsible (Butzer,

1996). Moreover, once a strategy is chosen, planning for, funding, and evaluating a program can be problematic due to insufficient technological understanding or lack of an adequate system of program evaluation (Wilken, 1989). Ultimately, agricultural composition, structures, and other forms of landesque capital may require time to develop (Blaike and Brookfield, 1987 p. 9), as centuries of incremental modifications and fine-tuning of agrosystems are difficult, if not impossible to replicate with a few hours of bulldozer or backhoe work. Where the basic technologies are appropriate, successful implementation or modification requires collaborative, adaptive, long-term thinking, especially at the institutional level (Critchley, 1999). Agricultural terracing is an ancient and widespread approach to intensive cultivation. Terrace forms range from step-like horizontal planting surfaces (treds) with supporting vertical walls (risers), to parallel rows of plants that only slightly modify the degree of hillslope (Treacy and Denevan, 1994). By leveling hillslopes, all terraces seek to create better planting surfaces that mitigating surface erosion and deepen soils (Spencer and Hale, 1961), in effect, conserving soil and water. But as carefully built environments, terraced landscapes are also prone to degradation. The artificial leveling of hillslopes creates greater potential for erosion as the terraced surface competes with geomorphic processes that seek to return the hillslope to its natural gradient (Borejsza, 2006). Terraced landscapes can be among the most fragile of built environments (Treacy, 1989), although their vulnerability to degradation depends on a wide range of factors often stemming from mismanagement (Barbier, 1990) or abandonment (Inbar and Llerena, 2000; Hunter, 2013). Restoring terrace agriculture requires a fine-grained understanding of the human agency,

34


technological development, and biogeomorphology associated with farming in different environments. The soils of China’s loess plateau, for example, are globally among the most erodible (Zhang et al., 1998), and many agricultural practices, including some forms of terracing, have contributed to widespread land degradation (Shi and Shao, 2000; Chen et al., 2007). Field research and modeling studies suggest that modifications to existing terrace systems, along with building new terraces, may effectively mitigate precipitation deficits and improve future soil conservation practices (Lü et al., 2009; Huang and Zhang, 2004; Fu et al., 2003; Li and Lindstrom, 2001; Quine et al., 1999; Veek et al., 1995). In mountainous Mediterranean settings, studies of terraced agrosystems examine the effects of land abandonment on runoff, soil erosion, and soil regeneration (Gallart et al., 1994; Lasanta et al., 2001; Ruecker et al., 1998) and the mechanics of conserving or restoring ancient terraces to enhance soil and water conservation (Zurayk, 1994; Martínez-Casasnovas and SánchezBosch, 2000; Hammad et al., 2004). Studies of terraced agrosystems in Sub-Saharan Africa explore terrace building as a form of participatory development (Pretty and Shah, 1997) and terraced landscapes as a means of cultural preservation and enrichment (Watson, 2009). Other research examines farmers’ willingness to adopt and implement new terracing techniques on their lands (Critchley and Brommer, 2003; Gebremedhim and Swinton, 2003; Gebreegziabher et al., 2009), analyzes the household level risks associated with building new structures (Teklewold and Köhlin, 2011), and stresses the general importance of terrace maintenance in conservation and agricultural development (Evans and Winterhalder, 2000; Critchley et al., 2008). In Latin America, studies of terrace rehabilitation focus on the potential for agricultural development in marginal areas (Treacy, 1987; Kendall, 2004), soil conservation and watershed management (Masson, 1984; Cotler et al., 2013), the economic costs of rehabilitation and management (de la Torre and Burga, 1986), and the role of government organizations and NGOs in carrying out terrace building programs (Altieri, 1999). The Central Mexican Highlands provides an illustrative example of many of these trends, where government agencies actively promote the restoration of terraced landscapes as an effective means of soil and water conservation. The region is home to about 60% of Mexico’s population and about 51% of the nation’s cropland, but it contains only about 12% of the nation’s water resources (Tiscareño-López et al., 1999). Governmental programs of terrace restoration promise improved soil and water conservation, increased agricultural production, enhanced environmental services, and stronger defense against climate change and desertification. These are lofty goals, and are associated with billions of pesos (US $ hundreds of millions) of investment during the last decade (FAO-SAGARPA, 2008). But there remain critical distinctions between institutional notions of land restoration and results on the ground e in the current case, between restoring the physical structures of terraces and restoring the processes of hillslope agriculture in the region. 1.2. Autogenic processes of environmental repair One key driver of process-oriented environmental repair is the idea of autogenic succession, which describes the use of low-input, self-sustaining processes that initiate and intensify, through positive feedback, a natural recovery of primary ecological functions (Whisenant, 1999 p. 22; Harris et al., 1996 pp. 168e187; Bradshaw, 1996). The use of vegetation is critical, as its role as ecosystem engineer can serve to further modify, maintain, and transform (repair) biophysical environments (Jones et al., 1994; Tanner, 2001; Wright and Jones, 2006). Autogenic succession contrasts with

expensive, subsidy driven programs that focus on restoring the structures, and not necessarily the processes of ecosystem function (Whisenant, 1999). Since the cost of restoring degraded lands often exceeds the immediate return on investment (see Werner, 1992), repair efforts often promote the use of autogenic, vegetative succession as a cost-effective, natural, and low-input alternative to expensive restoration programs (Whisenant, 1999). In the current case study, agricultural terrace restoration incorporates two basic forms of natural or autogenic repair processes. First, conservation groups use native maguey (Agave spp.) to mitigate the effects of accelerated erosion. A resilient, xeric plant, agave thrive in semiarid environments and gain biomass quickly (Nobel, 2003). Their broad, shallow root systems serve as excellent soil binders as they interlock to form networks of soil reinforcement (Patrick, 1977 pp. 88e9). In traditional terrace agriculture of the region, farmers utilize strategically planted rows of agave to serve as barriers to hillslope runoff, improving soil and water retention in the process (Wilken, 1987; Skopyk, 2010).1 Today, state-run conservation programs in Tlaxcala, Mexico often imitate the practices of farmers by utilizing native agave species as a means of low cost, assisted autogenic repair of hillslopes (Sánchez-Morales et al., 2008; Vázquez-Alvarado et al., 2011). A second form of natural processes farmers use to build terraces is the creative manipulation of hillslope erosion. Terrace building, without heavy machinery, usually requires the accumulation of soils behind topographical impediments, behind a vertical barrier (stone riser, earthen border, or plant), or within an erosionretaining ditch (zanja) from where the accumulated sediment is redistributed. Farmers can creatively use plows to shape hillslope contours and encourage the downward movement of soil (Williams, 1990), or on degraded lands, the buildup may happen more quickly. While erosion (accelerated or natural) is often viewed negatively, in the context of terrace building it can be an effective means of moving and reconfiguring soil (Frederick and Krahtopoulou, 2000; Denevan, 2001 pp. 177, 180). In sum, biophysical processes of vegetative (agave) and hillslope (erosion) succession can serve as low-cost and low-input means of terrace construction. But equally important is how farmers assist or direct these processes, under different biophysical environments, towards the development of an agroecosystem. 2. Reclaiming hardpan surfaces (tepetates)2 The reclamation and restoration of indurated, volcanic soils is a primary goal of environmental restoration programs in mountainous regions of Latin America where soil degradation is severe (Nimlos, 1992; Quantin et al., 1999). In the Mexican state of Tlaxcala alone, infertile hardpan already covers about 15e20% of the surface area, with another 54% of the surface in danger of exposing hardpan through improvident land use (Werner, 1992; Haulon et al., 2007). Once exposed, hardpan soils prevent cultivation and allow only limited growth of vegetation, rendering the land unproductive. Hardpan reclamation is a focus of local, regional, and national soil and water conservation efforts. For smallholder farmers, the

1 In Tlaxcala, Mexico these terraces are known as metepantle or metepantli, especially when lined with rows of agave. The term derives from the Nahuatl language (metl ¼ agave, pantli ¼ line or berm, or the space between), and commonly refers to any terraced agricultural land composed of a planting surface (tred) and a row of agave (border or riser) that follows natural hillslope contours (West, 1970 p. 366; Patrick, 1977 pp. 39e40; Evans, 1990; Whitmore and Turner, 2001 p. 141; Borejsza, 2006 pp. 45e6). 2 The regional name for hardpan (tepetate) derives from the indigenous Nahuatl language, from tetl and petatl, which loosely translates as “rock mat” (Williams, 1972).


reclamation of tepetate promises more productive agricultural lands (Werner, 1992; Zebrowski, 1992). But at the regional scale, in northern Tlaxcala, reclamation efforts focus on protecting the critical watersheds that feed the headwaters of the Zahuapan River, the main tributary of the Atlangatepec reservoir to the south, which is the state’s largest source of freshwater. Accelerated erosion of hillslopes surrounding the Atlangatepec basin result in sediment deposition in the reservoir, where storage capacity decreases annually. Linking the freshwater security of Tlaxcala to soil erosion in the Atlangatepec-Tlaxco Basin, municipal and state authorities actively promote soil conservation and hillslope restoration programs in nearby communities (Observador Tlaxcalteca, 2013). Field studies show that successful reclamation of hardpan soils requires long-term and intensive land management, which often focuses on finding technical solutions to problems of low tepetate fertility and permeability in order to improve plant growth and water infiltration (Werner, 1992). Haulon et al. (2007) found that careful additions of soil organic carbon (SOC) over time improved soil nutrients, fertility, and vegetative growth, but also aggregate cohesive strength, which decreased erosion. But with insufficient groundcover, often the outcome of initial rehabilitation efforts, Quantin et al. (1999) found erosion rates of disturbed hardpan soils were often elevated. In addition, socioeconomic and political factors often impede thorough reclamation efforts, as in the short to medium term, reclamation is likely to be unprofitable (Werner, 1992). As an example, beginning in the 1970s, Tlaxcala’s Program for the Rehabilitation of Tepetate broke up 12,850 ha of soils with heavy machinery in response to farmers’ demands for more productive lands. Officials carried out the program without any formal investigations into the economic validity of the plan. Decades later, experimental results showed the returns from crops would have paid for the reclamation efforts only after fifty years of farming, making the program difficult to justify economically (Werner, 1992). While studies show it is possible to reconvert tepetates into productive soils using a wide variety of management solutions (Rainey, 2002; Zebrowski, 1992), the practice requires years of intensive, high input management that is often at odds with institutional time frames. The historical and archaeological records also give clues as to how indigenous groups may have practiced hardpan soil reclamation. Ethnographic accounts, based on historical documents and fieldwork, show hardpan management included specialized mulching, fertilizer additions, and other forms of long-term and intensive management (Williams, 1972). In contrast, archaeological excavations reveal that Prehispanic farmers also reclaimed hardpan surfaces by building terraces directly on top of them, instead of attempting to transform the hardened soil itself. Over what may have been generations of land management, farmers incrementally built stone walls that impeded erosion from upslope. The carefully built walls led to soil accumulation and eventually allowed the formation of agricultural terraces (Borejsza, 2006 p. 450). In sum, hardpan soil reclamation in highland Mexico has included: 1) the physical and chemical manipulation of the soil itself, and 2) the creation of agricultural land (terraces) directly on top of the hardpan surface. But critical to all the above practices e modern, historical, and archaeological e was long-term soil management and maintenance as part of an extended process. In the past, there appear to have been no easy, short-term fixes to the problems of hardpan soil exposure. The following field study demonstrates that this characteristic of hardpan soil management largely persists today, despite some advantages of modern technologies, and it represents the first available examination of tepetate reclamation using the traditional ditch-and-border technique of semi-terrace (metepantle) construction.

35

3. Case study: agricultural terrace restoration in Xalostoc, Tlaxcala In July of 2007, municipal authorities directed a metepantle, semiterrace restoration project on the hardpan-covered hillslopes of the ex-hacienda of Xalostoc, in the north-central Tlaxco Valley where erosion had revealed a patch of tepetate spanning 0.66872 km2. Interviews with two former project directors show the ultimate goal was to restore Xalostoc to the condition of surrounding lands, productive semi-terraces of mostly corn, barley, and maguey manso (Agave salmiana). But instead of breaking up and tilling the entire hardpan surface, as had been attempted in previous reclamation efforts (e.g., Haulon et al., 2007), officials contracted local farmers and a backhoe operator to excavate and shape rows of contour ditches and borders, leaving the majority of the soils intact. On the borders of excavated hardpan fill, farmers planted agave shoots, carefully spaced 1 m apart, in the traditional way. The team also dug four large retention ponds (jagüeyes), strategically placed in the paths of hillslope gullies. The agave-lined ditches and borders were to first mitigate soil erosion, increase soil retention, and improve soil moisture and vegetative growth at the site. Over time, a natural process of environmental recovery was envisioned to cover the hillslope, with the succession of agave and grasses leading the way towards site stabilization. Once stabilized, the full restoration of terrace agriculture would be possible using the ditch and border structure as a base. The following study sought to assess the progress of the Xalostoc restoration project from its beginning in July of 2007 to January 2012. 3.1. Methods and experimental design: estimating surface erosion and sediment accumulation The Xalostoc terrace restoration project was mapped with a total station in February 2011, 46 months after the initial reclamation attempt. Three vertical transects (A, B, and C) were laid at the upslope field margins of the project, where spacing allowed for the broadest sample of continuous ditch-and-border structures (Fig. 2). Transects A and B were 68.03 m apart and transects B and C were 41.56 m apart. Transects A, B, and C measured 96.93, 107.84, and 128.26 m, respectively. Each transect crossed 8 terrace complexes. Each complex included 1 agave-lined border, 1 retaining ditch, and 1 catchment area (Fig. 3). The surface areas of each catchment upslope from the retaining ditches were measured with the total station, as were the volumes of each ditch. The volume of sediment accumulated in each ditch until that time was estimated based on soil depth sampling with a 7/1600 30 steel tile probe (AMS# 403.06) every 5 m of the length of each ditch. The sediment volume estimates for each ditch were then compared with the surface areas of each catchment and erosion estimates for the first 46 months (4 rainy seasons) were calculated. The total volume of sediment accumulated in each ditch was then compared with the total original volume (storage capacity) of each ditch, and rates of ditch infilling were calculated. 3 sediment samples were extracted (one from each transect) using a pipe corer (5.500 600 ), and the average bulk density (dry) was estimated to be 1.5 g cm3. After the initial measurements, rows of 25.4 cm (10 in) pins were inserted into each of the 24 ditches (Transects A, B, and C; Rows 1e8). A total of 94 pins were installed, each spaced 30 cm (width) by 4 m (length). Monthly sediment accumulation was estimated based on the average change around each pin multiplied by the surface area of the ditch. Silt fences were not employed because of cost considerations and because the drainage areas at Xalostoc exceeded those recommended for the technique (see Robichaud and Brown, 2002). Instead, sediment pins (Takekawa et al., 2010) provided the most practical and effective means of estimating sediment aggradation in the ditches. Because sediment

36


Fig. 2. Terrace restoration project in Xalostoc, Tlaxcala.

traps had already been excavated into the impervious, hardpan substrate during the restoration project, this method represented the most economical and effective means of monitoring rates of erosion and ditch infilling. Monthly precipitation totals were calculated from daily readings from a 280 mm capacity, Tenite Rain Gauge (#110672) with a cylindrical case and inner receiver, mounted on an open house roof near the town of Atlangatepec (6.2 km distant), which receives rainfall totals very similar to Xalostoc (Gay-García et al., 2004). These observed readings were compared with 30-year monthly normals from the same town

(CLICOM, 2013) in order to assess whether 2011 was representative of longer-term conditions (Fig. 4). In addition to measuring surface erosion, rates of ditch infilling, and precipitation, the terrace borders were monitored monthly for rilling and other signs of wasting. This process continued from February 2011 through January 2012. 3.2. Results: erosion rates, ditch capacities, and land use The cumulative monthly erosion of each catchment as a function of the volume of sediment captured in each ditch was normalized


Fig. 3. Profile view of transect C. (not to scale).

by the catchment area (cm3 cm2) (Fig. 5). For all 3 transects (A, B, and C) and 8 rows (1e8), the rates of erosion for the “Row 8” catchments (the furthest upslope) were less than the catchments for all other rows (1e7). The annual erosion rates for ditches A8, B8, and C8 were 0.17, 0.16, and 0.14 cm3 cm2; and until 2012, less than 1.0 cm3 cm2 of soil had been lost from these three catchments. The upslope catchments generated 52.81% less surface erosion on average than did the downslope catchments (Fig. 3). By February 2011, 16 of the 24 ditches had already filled at least 50% of their capacity with sediment (Fig. 6). By 2012, the ditches furthest upslope (A8, B8, and C8) had filled 98% of their capacity on average, with 20% filling during the 2011 rainy season alone. Ditch C8 had completely filled by January 2012. Although the downslope ditches (A1e7, B1e7, and C1e7) filled at a slower rate (8e12%), and averaged only 64.9% filled to capacity, all ditches were projected to completely fill by 2018, assuming a similar erosion rate. Having completely filled with sediment, these ditches would lose the capacity to impede overland flow, capture runoff, hold water, and store sediment.

37

The erosion data was further aggregated according to land use (Fig. 7). The property immediately upslope from the Xalostoc project was intermittently cultivated, with healthy agricultural soils, while the downslope catchments were exclusively reworked hardpan soils, disturbed during the initial reclamation effort in 2007. Erosion varied significantly among the reworked hardpan e standard deviation from the average ranged from 0.46 cm month1 during the driest months, to 0.67 cm month1 during the wettest months (Figs. 4 and 7). The variability of erosion rates of the disturbed hardpan soils were likely due to the physical differences in border construction, compaction, rilling, and the irregularity of the hardpan surfaces themselves, some of which were pitted and served as uneven conduits for overland flow. The different rates of ditch infilling for the agricultural lands (A8, B8, C8) and the hardpan lands (A1e7, B1e7, C1e7) were largely due to the different catchment sizes. Although the erosion rates of upslope farmland were an average 52.8% less than the disturbed hardpan soils, the catchment sizes (A8, B8, and C8) were about an order of magnitude larger (Chart 1). With regard to the original ditch dimensions, the depth and width remained consistent. Only the ditch length and the vertical spacing between rows differed substantially, and this spacing ultimately determined the catchment area for each retaining ditch. In building the terraces, project coordinators did not align the ditches and borders according to erosion calculations; rather, construction was carried out less formally, on site. The inconsistency in this spacing largely drove the variability of ditch infilling rates among the reworked hardpan soils. But because the retaining ditches furthest upslope collected sediment from relatively large catchment areas (A8 ¼ 1494; B8 ¼ 2770; C8 ¼ 2291 m2), these ditches filled an average 33.1% faster. By January 2012, borders exhibited rilling and some flattening, especially along patches were agave were absent or in poor condition. No correlations among different transect rows and catchments were apparent. Rills were up to 10 cm deep, on both the downslope and upslope faces of the earthen borders. Reworked

Fig. 4. Observed precipitation in Atlangatepec, Tlaxcala (February 2011 to January 2012). 30-year monthly and cumulative normals provided by CLICOM (2013).

38


Fig. 5. Cumulative volume of sediment captured in the retaining ditches (zanjas) normalized by catchment area over transects A, B, and C and rows 1e8. Row 8 is the furthest upslope and row 1 is the furthest downslope. February 2011 through January 2012.

hardpan soils are often nutrient poor without fertilizer additions (Báez-Pérez et al., 2007; Haulon et al., 2007), and sections of the borders were unstable and prone to erosion. Grasses did not accumulate on the nutrient-poor, sandy loam borders as project directors had hoped, although grasses did grow in the organically rich sediments (lama) that collected in the ditches (Fig. 8). There were at least two reasons why the erosion and ditch infilling rates observed in 2011 may underestimate normal conditions. First, total precipitation for 2011 was about 20.6% less than the 30-year normal (Fig. 4). During a typical year erosion and ditch infilling rates of similar efforts are likely greater. Second, as sediment storage devices fill (in this case, the ditches) they lose the capacity to store

Fig. 6. Cumulative proportion of sediment infilling to retaining ditch (zanja) capacity over transects A, B, and C and rows 1e8. Row 8 is the furthest upslope and row 1 is the furthest downslope. February 2011 to January 2012.

runoff (Robichaud and Brown, 2002). During heavy precipitation events overflow of the structures carries with it finer grained sediments, which are carried away by runoff downslope (RamosScharrón and MacDonald, 2007). 3.3. Discussion: implications for land management This study provides estimates of erosion and infilling rates for a typical ditch-and-border hardpan reclamation project, which are now growing in number and scope throughout Mexico. The current study has 3 general implications for these efforts. First, the high rate of ditch infilling at the upslope margins of the Xalostoc project (A8, B8, and C8) illustrates how land use on neighboring lands can impact an adjacent site. Although erosion rates of the agricultural


39

Fig. 7. Average monthly erosion rates for agricultural/bush fallow lands and reclaimed hardpan (tepetate) lands. Error bars represent standard deviations from the averages. February 2011 to January 2012.

fields upslope were 52.8% slower than for the reworked hardpan soils, these upslope catchments generated enough sediment to fill the Row 8 retaining ditches to 98% capacity after 5 years. Future maintenance of Xalostoc should focus first on the upslope field margins as this area received greater impact from the erosion of adjacent lands. Future programs of environmental repair should carefully consider the effects of landscape connectivity, especially along site margins, in planning remediation programs.

Second, most erosion at Xalostoc appears to have resulted from the initial disturbance caused by the restoration attempt. Previous studies of undisturbed hardpan give erosion rates of between 9.07 and 13.61 Mg ha1 yr1 (Figueroa Sandoval, 1975; cited in Cordova and Parsons, 1997 p. 182) and 14.52 Mg ha1 yr1 (Baumann et al., 1992; cited in Werner, 1992 p. 326). In contrast, erosion rates of the disturbed hardpans at Xalostoc averaged 58.9 Mg ha1 yr1 during the first 4 years after reclamation and 54 Mg ha1 yr1 during 2011

Chart 1. Erosion of individual terrace catchments over transects A, B, and C and rows 1e8. Row 8 is the furthest upslope and row 1 is the furthest downslope. February 2011 to January 2012.

40


Fig. 9. Conceptual model of the terrace restoration process. Over time, maintenance shifts primary function of terraces from erosion mitigation to agriculture. Periodic maintenance then holds balance between accretion and erosion.

soil accretion and soil erosion as dominant and secondary processes, and it allows for the creation of agricultural land. 4. Conclusions: rebuilding agricultural structures and processes

Fig. 8. Newly constructed ditch and border (zanja y bordo) metepantle terrace (#C7) on partially reclaimed hardpan (tepetate) in Xalostoc, Tlaxcala. Note vegetation, soil, and moisture conditions of the retaining ditch (zanja)(top). Fully developed metepantle terrace on adjacent agricultural lands (bottom). (Photographs by the author, July 2011).

alone. Rates had slowed an average of 4.9 Mg ha1 yr1 from the first 4 years to the fifth year, echoing a similar slowing of erosion rates after disturbance found by Haulon et al. (2007). But by 2012, the initial disturbance at Xalostoc had generated 217.0e 244.2 Mg ha1 more erosion than likely would have resulted if the land had been left undisturbed. In the absence of a more effective and prolonged management system with respect to soil erosion, leaving the Xalostoc hardpan soils as they were probably would have been the better option. Third, while the cleaning and redistribution of ditch sediments is a critical part of terrace maintenance (Wilken, 1987 p. 83) e ditches must be cleaned or they lose the capacity to store soil and water from runoff e at the same time, sediment redistribution can be an effective, although incremental means of border reinforcement and terrace development. A conceptual model illustrates how in terrace restoration, soil initially lost to erosion upslope can be gained as terrace mass through the use of hillslope impediments (Fig. 9). In the case of ditch and border semi-terraces, redistribution of ditch sediments (lama) can reinforce earthen borders, providing structural support and nutrient-rich fertilizer for vegetation. Once stabilized, ditches may permanently fill and vegetated borders may alone serve to impede erosion, but borders must be maintained or sheetwash and gullying can erode the entire structure (West, 1970 p. 367). This is of primary concern on the impervious, hardpan-covered slopes of Xalostoc where overland runoff forms quickly. Whether in the early stages of terrace restoration or years into agricultural production, continued maintenance is essential. Active maintenance balances

The Xalostoc restoration project illustrates why a low-input, self-sustaining recovery of severely damaged lands can be difficult to achieve in an agroecological context, despite its relevance in wildland restoration efforts. Restoring degraded agricultural land requires different methods and criteria. Agroecosystems are largely humanized environments, created, maintained, and managed by and for human use. As such, they require the continued attention of cultivators if they are to produce the domesticated plants and landforms that characterize them, regardless of whether the initial intent of restoration is to produce food or to mitigate environmental degradation. Autogenic processes of succession may provide an effective, natural means of ecological restoration. But the restoration of agroecosystems (especially those requiring high levels of inputs) more often requires a long-term and sustained effort at system maintenance. Terrace farmers direct and shape autogenic or natural processes of vegetative (e.g., agave and grasses) and hillslope (erosion) succession. But periodic maintenance is critical to terrace development and ultimately, agricultural sustainability. In this sense, agricultural land cannot be produced through autogenic processes alone, despite the efficiency of natural processes in shaping biophysical environments. As anthropogenic components of the biosphere, agricultural land must be created (and maintained) through human actions (Doolittle, 2006; Turner II et al., 1990). Depending on the level of agricultural intensity, this trait, in part, distinguishes the restoration of natural ecosystems from the restoration of agroecosystems. Future programs of agroecological restoration must plan for and facilitate some degree of sustained site maintenance. Restoring agricultural structures (the arrangement of vegetation or landforms) and relying on autogenic processes of environmental repair alone is insufficient. In the current case study, long-term planning beyond this initial stage was absent. Project directors took care to replicate many traditional terrace structures, but failed to adequately maintain the site e a critical, though often overlooked part of building and improving agricultural land, and an important step towards sustaining the agroecological services upon which society depends.


Acknowledgments This material is based upon work supported by the National Science Foundation under Grant No. 1031676, a Fulbright-Hays Doctoral Dissertation Research Abroad Fellowship, a Boren Fellowship for International Study, and various support from the University of Texas at Austin. I thank Carlos E. Ramos-Scharrón of the Department of Geography and the Environment at the University at Texas at Austin and Paul F. Hudson of Leiden University for helpful comments on drafts of this paper. I also thank two anonymous reviewers for providing thoughtful insights and suggestions. References Altieri, M., 1995. Agroecology: the Science of Sustainable Agriculture. Westview, Boulder. Altieri, M., 1999. Applying agroecology to enhance the productivity of peasant farming systems in Latin America. Environ. Develop. Sustain. 1, 197e217. Altieri, M., 2004. Linking ecologists and traditional farmers in the search for sustainable agriculture. Front. Ecol. Environ. 2 (1), 35e42. Altieri, M., Funes-Monzote, F., Petersen, P., 2012. Agroecologically efficient agricultural systems for smallholder farmers: contributions to food sovereignty. Agron. Sustain. Develop. 32 (1), 1e13. Báez-Pérez, A., Etchevers-Barra, J.D., Prat, C., Hidalgo-Moreno, C., 2007. Formation of aggregates and carbon sequestration in ameliorated tepetates in the Río Texcoco Basin, Mexico. Rev. Mexi. Cienc. Geológ. 24 (3), 487e497. Barbier, E., 1990. The farm-level economics of soil conservation: the uplands of Java. Land Econ. 66 (2), 199e211. Baumann, J., Werner, G., Muñoz-Nava, H., Vera Reyes, A., Flores, G.F., 1992. Resultados preliminares sobre la erosión en tepetates en el Bloque de Tlaxcala. Terra 10, 347e354. Berkes, F., Colding, J., Folke, C., 2000. Rediscovery of traditional ecological knowledge as adaptive management. Ecolog. Appl. 10 (5), 1251e1262. Blaike, P., Brookfield, H., 1987. Land Degradation and Society. Methuen, New York. Bohlen, P.J., Lynch, S., Shabman, L., Clark, M., Shukla, S., Swain, H., 2009. Paying for environmental services from agricultural lands: an example from the northern Everglades. Front. Ecol. Environ. 7 (1), 46e55. Borejsza, A.J., 2006. Agricultural Slope Management and Soil Erosion in Tlaxcala, Mexico. Doctoral Dissertation. UCLA, Los Angeles. Bradshaw, A.D., 1996. Underlying principles of restoration. Can. J. Fish. Aqua. Sci. 53 (S1), 3e9. Bradshaw, A.D., 1997. What do we mean by restoration? In: Urbanska, K.M., Webb, N.R., Edwards, P.J. (Eds.), Restoration Ecology and Sustainable Development. Cambridge University Press, Cambridge, pp. 8e14. Butzer, K.W., 1996. Ecology in the long view: settlement histories, agrosystemic strategies, and ecological performance. J. Field Archaeol. 23 (2), 141e150. Chapin, M., 1988. The seduction of models: Chinampa agriculture in Mexico. Grassroots Develop. 12 (1). Chen, L., Wei, W., Fu, B., Lü, Y., 2007. Soil and water conservation on the Loess Plateau in China: review and perspective. Prog. Phys. Geograph. 31 (4), 389e403. CLICOM, 2013. Base de datos climatológica nacional (sistema CLICOM). www. clicom-mex.cicese.mx (accessed 29.10.13.). Cordova, C.E., Parsons, J.R., 1997. Geoarchaeology of an Aztec dispersed village on the Texcoco piedmont of central Mexico. Geoarchaeology 12 (3), 177e210. Cotler, H., Cram, S., Martinez-Trinidad, S., Quintanar, E., 2013. Forest soil conservation in central Mexico: an interdisciplinary assessment. Catena 104, 280e287. Critchley, W., 1999. Harnessing traditional knowledge for better land husbandry in Kabale District, Uganda. Mount. Res. Develop. 19 (3), 261e272. Critchley, W., Negi, G., Brommer, M., 2008. Local innovation in ‘green water’ management. In: Bossio, D., Geheb, K. (Eds.), Conserving Land, Protecting Water. CAB International, Cambridge, pp. 107e119. Critchley, W., Brommer, M., 2003. Understanding traditional terracing. LEISA Mag. 19, 14e15. de la Torre, C., Burga, M. (Eds.), 1986. Andenes y Camellones en el Peru Andino: Historia Presente y Futuro. CONCYTEC, Lima. Denevan, W.M., 2001. Cultivated Landscapes of Native Amazonia and the Andes. Oxford University Press, Oxford. Doolittle, W.E., 1984. Agricultural change as an incremental process. Ann. Asso. Am. Geograph. 74 (1), 124e137. Doolittle, W.E., 1989. Comments: Chinampas revisited: subversion of logic and abandonment of Chinampa agriculture. DESFIL Newslet. 3 (2), 3. Doolittle, W.E., 2006. Agricultural manipulation of floodplains in the southern Basin and Range Province. Catena 65 (2), 179e199. Erickson, C.L., 1988. Raised field agriculture in the Lake Titicaca Basin: putting ancient agriculture back to work. Expedition 30 (3), 8e16. Evans, S.T., 1990. The productivity of maguey terrace agriculture in central Mexico during the Aztec period. Lat. Am. Antiq. 1 (2), 117e132. Evans, T.P., Winterhalder, B., 2000. Modified solar insolation as an agronomic factor in terraced environments. Land Degrad. Develop. 11 (3), 273e287.

41

FAO [Food and Agriculture Organization of the United Nations], 2013. FAOSTAT Online Database (Total World Agricultural Area). FAO. http://faostat.fao.org/site/ 377/DesktopDefault.aspx?PageID¼377#ancor (accessed 15.05.13.). FAO-SAGARPA (Secretaría de Agricultura, Ganadería, Desarrollo Rural, Pesca y Alimentación), 2008. Informe de evalucación de consistencia y resultados 2007. In: Programa integral de agricultura sostenible y reconversion en zonas de siniestralidad recurrente. www.fao-evaluacion.org.mx/pagina/informeevaluacion-programas (accessed 29.07.13.). Figueroa Sandoval, B., 1975. Pérdidas de suelo y nutrimentos y su relación con el uso del suelo en la cuenca del Rio Tezcoco. Thesis. UACH, Chihuahua. Frederick, C.D., Krahtopoulou, A., 2000. Deconstructing agricultural terraces: examining the influence of construction method on stratigraphy, dating and archaeological visibility. In: Halstead, P., Frederick, C.D. (Eds.), Landscape and Land Use in Postglacial Greece. Sheffield Academic Press Ltd, Sheffield, pp. 79e94. Fu, B., Wang, J., Chen, L., Qiu, Y., 2003. The effects of land use on soil moisture variation in the Danangou catchment of the Loess Plateau, China. Catena 54 (1), 197e213. Gallart, F., Llorens, P., Latron, J., 1994. Studying the role of old agricultural terraces on runoff generation in a small Mediterranean mountainous basin. J. Hydrol. 159 (1), 291e303. Gay-García, C., Hernández-Vázquez, M., Jiménez-López, J., Lezama-Gutiérrez, J., Magaña-Rueda, V., Morales-Acoltzi, T., Orozco-Flores, S., 2004. Evaluation of climatic forecasts of rainfall for the Tlaxcala State (México): 1998e2002. Atmósfera 17 (3), 127e150. Gebreegziabher, T., Nyssen, J., Govaerts, B., Getnet, F., Behailu, M., Haile, M., Deckers, J., 2009. Contour furrows for in situ soil and water conservation, Tigray, northern Ethiopia. Soil Tillage Res. 103 (2), 257e264. Gebremedhim, B., Swinton, S.W., 2003. Investment in soil conservation in northern Ethiopia: the role of land tenure security and public programs. Agri. Econ. 29 (1), 69e84. Gliessman, S.R., 1998. Agroecology: Ecological Processes in Sustainable Agriculture. Ann Arbor Press, Chelsea. Hammad, A.A., Haugen, L.E., Børresen, T., 2004. Effects of stonewalled terracing techniques on soil-water conservation and wheat production under Mediterranean conditions. Environ. Manage. 34 (5), 701e710. Harper, J.L., 1982. After description. In: Newman, E.I. (Ed.), The Plant Community as a Working Mechanism, Special Publication no. 1 of the British Ecological Society. Blackwell, Oxford, pp. 11e25. Harris, J.A., Birch, P., Palmer, J., 1996. Land Reclamation and Restoration: Principles and Practice. Longman, Singapore. Haulon, M., Werner, G., Flores-García, G., Vera-Reyes, A., Felix-Henningsen, P., 2007. Assessment of erosion rates during rehabilitation of hardened volcanic soils (tepetates) in Tlaxcala. Rev. Mexi. Cienc. Geológ. 24 (3), 498e509. Herrick, J.E., Schuman, G.E., Rango, A., 2006. Monitoring ecological processes for restoration projects. J. Nat. Conserv. 14 (3), 161e171. Huang, M., Zhang, L., 2004. Hydrological responses to conservation practices in a catchment of the Loess Plateau, China. Hydrolog. Process. 18 (10), 1885e1898. Hunter, R., 2013. Land use change in New Spain: a three-dimensional historical GIS analysis. Profess. Geogr.. Published online: http://www.tandfonline.com/doi/ citedby/10.1080/00330124.2013.784951#.UjHiOrzFbrM. Inbar, M., Llerena, C.A., 2000. Erosion processes in high mountain agricultural terraces in Peru. Mount. Res. Develop. 10 (1), 72e79. Jones, C.G., Lawton, J.H., Shachak, M., 1994. Organisms as ecosystem engineers. Oikos 69, 373e386. Jordan, W.R., Gilpin, M.E., Aber, J.D., 1987. Restoration ecology: ecological restoration as a technique for basic research. In: Restoration Ecology: a Synthetic Approach to Ecological Research. Cambridge University Press, Cambridge, pp. 3e21. Kaimowitz, D. (Ed.), 1990. Making the Link: Agricultural Research and Technology Transfer in Developing Countries. Westview Press, Boulder. Kendall, A., 2004. Restauración de canales y andenes agícolas prehispánicos en los Andes usando tecnología tradicional y apropiada. In: Llerena, C.A., Inbar, M., Benavides, M.A. (Eds.), Conservacón y abandono de andenes. Universidad Nacional Agraria La Molina, Universidad de Haifa, . Lima, pp. 98e106. Lasanta, T., Arnáez, J., Oserín, M., Ortigosa, L.M., 2001. Marginal lands and erosion in terraced fields in the Mediterranean mountains: a case study in the Camero Viejo (Northwestern Iberian System, Spain). Mount. Res. Develop. 21 (1), 69e76. Li, Y., Lindstrom, M.J., 2001. Evaluating soil quality-soil redistribution relationship on terraces and steep hillslope. Soil Sci. Soc. Am. J. 65 (5), 1500e1508. Lü, H., Zhu, Y., Skaggs, T.H., Yu, Z., 2009. Comparison of measured and simulated water storage in dryland terraces of the Loess Plateau, China. Agri. Water Manage. 96 (2), 299e306. Martínez-Casasnovas, J.A., Sánchez-Bosch, I., 2000. Impact assessment of changes in land use/conservation practices on soil erosion in the Penedés-Anoia vineyard region (NE Spain). Soil Tillage Res. 57 (1), 101e106. Masson, L., 1984. La recuperacion de los Andenes para la ampliacion de la frontera agricola en la sierra. COFIDE, Lima. Nimlos, T.J., 1992. Reclamation of indurated, volcanic-ash materials in Latin America. In: Lal, R., Stewart., B.A. (Eds.), Soil Restoration. Springer, New York, pp. 153e170. Nobel, P.S., 2003. Environmental Biology of Agaves and Cacti. Cambridge University Press, Cambridge. Observador Tlaxcalteca, 2013. El ejido de San Juan Bautista en Atlangatepec se pierde entre la erosión severa y falta de apoyos oficiales del gobierno y federación. www.observadortlaxcalteca.com/portal/municipios-tlaxcala/el-ejido-desan-juan-bautista-en-atlangatepec-se-pierde-entre-la-erosion-severa-y-faltade-apoyos-oficiales-del-gobierno-y-federacion (accessed 14.06.13.).

42


Patrick, L.L., 1977. A Cultural Geography of the Use of Seasonally Dry, Sloping Terrain: the Metepantli Crop Terraces of Central Mexico. Doctoral Dissertation. University of Pittsburgh. Pretty, J., Shah, P., 1997. Making soil and water conservation sustainable: from coercion and control to partnerships and participation. Land Degrad. Develop. 8 (1), 39e58. Pywell, R.F., Meek, W.R., Loxton, R.G., Nowakowski, M., Carvell, C., Woodcock, B.A., 2011. Ecological restoration on farmland can drive beneficial functional responses in plant and invertebrate communities. Agri. Ecosys. Environ. 140 (1), 62e67. Quantin, P., Prat, C., Zebrowski, C., 1999. Soil restoration and conservation: the “tepetates” e indurated volcanic soils e in Mexico. In: Harper, D., Brown, T. (Eds.), The Sustainable Management of Tropical Catchments. Wiley, New York, pp. 109e119. Quine, T.A., Walling, D.E., Zhang, X., 1999. Tillage erosion, water erosion and soil quality on cultivated terraces near Xifeng in the Loess Plateau, China. Land Degrad. Develop. 10 (3), 251e274. Rainey, S.J., 2002. Local perspectives on the erosion and conservation of soil resources in two highland Guatemalan Municipios. In: Steinberg, Michael K., Hudson, Paul F. (Eds.), Cultural and Physical Expositions: Geographic Studies in the Southern United States and Latin America. Geoscience Publications, Baton Rouge, pp. 97e117. Ramos-Scharrón, C., MacDonald, L., 2007. Measurement and prediction of natural and anthropogenic sediment sources, St. John, U.S. Virgin Islands. Catena 71 (2), 250e266. Reij, C., Scoons, I., Toulmin, C. (Eds.), 1996. Sustaining the Soil: Indigenous Soil and Water Conservation in Africa. Earthscan Publications, London. Reijntjes, C., Haverkort, B., Waters-Bayer, A., 1992. Farming for the Future: an Introduction to Low-external-input and Sustainable Agriculture. Macmillan Press, London. Robichaud, P., Brown, R., 2002. Silt Fences: an Economical Technique for Measuring Hillslope Soil Erosion. General Technical Report RMRS-GTR-94. US Forest Service, Fort Collins, CO. Ruecker, G., Schad, P., Alcubilla, M.M., Ferrer, C., 1998. Natural regeneration of degraded soils and site changes on abandoned agricultural terraces in Mediterranean Spain. Land Degrad. Develop. 9 (2), 179e188. Ruiz-Jaen, M.C., Aide, T.M., 2005. Restoration success: how is it being measured? Restor. Ecol. 13 (3), 569e577. Sánchez-Morales, P., Ocampo-Fletes, I., Sánchez-Hernandez, M., MartínezSaldaña, T., 2008. Proceso autogestivo para la conservación de suelos y agua en sistemas campesinos sustentables: los casos de Vicente Guerrero y la Reforma, Tlaxcala. Ra Ximhai 4 (2), 165e181. SER (Society for Ecological Restoration International Science & Policy Working Group), 2004. The SER International Primer on Ecological Restoration. Available from: http://www.ser.org (accessed July 2013). Shi, H., Shao, M., 2000. Soil and water loss from the Loess Plateau in China. J. Arid Environ. 45 (1), 9e20. Skopyk, B., 2010. Undercurrents of Conquest: the Shifting Terrain of Indigenous Agriculture in Colonial Tlaxcala, Mexico. Doctoral Dissertation. York University, Toronto. Spencer, J.E., Hale, G.A., 1961. The origin, nature, and distribution of agricultural terracing. Pacif. Viewpt. 2 (1), 1e40. Swift, M.J., Izac, A., van Noordwijk, M., 2004. Biodiversity and ecosystem services in agricultural landscapes: are we asking the right questions? Agri. Ecosys. Environ. 104 (1), 113e134. Swinton, S.M., Lupi, F., Robertson, G.P., Hamilton, S.K., 2007. Ecosystem services and agriculture: cultivating agricultural ecosystems for diverse benefits. Ecolog. Econ. 64 (2), 245e252. Takekawa, J., Woo, I., Athearn, N., Demers, S., Gardiner, R., Perry, W., Ganju, N., Shellenbarger, G., Schoellhamer, D., 2010. Measuring sediment accretion in early tidal marsh restoration. Wetlan. Ecol. Manage. 18 (3), 297e305.

Tanner, C.C., 2001. Plants as ecosystem engineers in subsurface-flow treatment wetlands. Water Sci. Technol. 44 (11e12), 9e17. Teklewold, H., Köhlin, G., 2011. Risk preferences as determinants of soil conservation decisions in Ethiopia. J. Soil Water Conserv. 66 (2), 87e96. Tiscareño-López, M., Báez-González, A.D., Valle, M.V., Potter, K.N., Stone, J.J., Vargas, M.T., Alonso, R.C., 1999. Agricultural research for watershed restoration in central Mexico. J. Soil Water Conserv. 54 (4), 686e692. Treacy, J.M., 1987. Building and rebuilding terraces in the Colca Valley of Peru. Yearbook: Conf. Latin American. Geograph. 13, 51e57. Treacy, J.M., 1989. Agricultural terrace in Peru’s Colca Valley: promises and problems of an ancient technology. In: Browder, J.O. (Ed.), Fragile Lands of Latin America: Strategies for Sustainable Development. Westview Press, Boulder, pp. 209e229. Treacy, J.M., Denevan, W.M., 1994. The creation of cultivable land through terracing. In: Miller, N.F., Gleason, K. (Eds.), The Archaeology of Garden and Field. University of Pennsylvania Press, Philadelphia, pp. 91e110. Turner II, B.L., Clark, W.C., Kates, R.W., Richards, J.F., Mathews, J.T., Meyer, W.B. (Eds.), 1990. The Earth as Transformed by Human Action. Cambridge University Press, Cambridge. Vázquez-Alvarado, R., Blanco-Macías, F., Ojeda-Zacarías, M., Martínez-López, J., Valdez-Cepeda, R., Santos-Haliscak, A., Háuad-Marroquín, L., 2011. Reforestación a base de nopal y maguey para la conservación de suelo y agua. Rev. Salud Púb. Nutric. 5 (edición especial), 185e203. Veek, G., Zhou, L., Ling, G., 1995. Terrace construction and productivity on loessal soils in Zhongyang County, Shanxi Province, PRC. Ann. Asso. Am. Geograph. 85 (3), 450e467. Wade, M.R., Gurr, G.M., Wratten, S.D., 2008. Ecological restoration of farmland: progress and prospects. Phil. Trans. R. Soc. B Biolog. Sci. 363 (1492), 831e847. Watson, E.E., 2009. Living Terraces in Ethiopia: Konso landscape, culture and development. Boydell and Brewer, New York. Werner, G., 1992. Suelos volcánicos endurecidos (tepetates) en el estado de Tlaxcala: Distribución, rehabilitación, manejo y conservación. Terra 10 (edición especial), 318e331. West, R.C., 1970. Population densities and agricultural practices in pre-Columbian Mexico, with emphasis on semi-terracing. In: Verhandlungen Des XXXVIII Internationalen Amerikanistenkongresses, vol. 2, pp. 361e369. Whisenant, S., 1999. Repairing Damaged Wildlands: a Process-orientated, Landscape-scale Approach. Cambridge University Press, Cambridge. Whitmore, T.M., Turner, B.L., 2001. Cultivated Landscapes of Middle America on the Eve of Conquest. Oxford University Press, Oxford. Wilken, G.C., 1987. Good Farmers: Traditional Agricultural Resource Management in Mexico and Central America. University of California Press, Berkeley. Wilken, G.C., 1989. Transferring traditional technology: a bottom-up approach for fragile lands. In: Browder, J.O. (Ed.), Fragile Lands in Latin America: Strategies for Sustainable Development. Westview Press, Boulder, pp. 44e57. Williams, B., 1972. Tepetate in the valley of Mexico. Ann. Asso. Am. Geograph. 62 (4), 618e626. Williams, L.S., 1990. Agricultural terrace evolution in Latin America. Yearbook: Conf. Latin American. Geograph. 16, 82e93. Wood, S., Sebastian, K., Scherr, S.J., Batjes, N.H., 2000. Agroecosystems. World Resources Institute, Washington D.C. Wright, J.P., Jones, C.G., 2006. The concept of organisms as ecosystem engineers ten years on: progress, limitations, and challenges. BioScience 56 (3), 203e209. Zebrowski, C., 1992. Los suelos volcáncos endurecidos en América Latina. Terra 10, 15e23. Zhang, X., Quine, T.A., Walling, D.E., 1998. Soil erosion rates on sloping cultivated land on the Loess Plateau near Ansai, Shaanxi Province, China: an investigation using 137Cs and rill measurements. Hydrolog. Process. 12 (1), 171e189. Zurayk, R.A., 1994. Rehabilitating the ancient terraced lands of Lebanon. J. Soil Water Conserv. 49 (2), 106e112.