Abstract
This article uses experimental archaeology to provide empirical data on the impact of soil salinization on ancient maize agriculture. Hopi blue maize was experimentally grown in a drought-prone soil containing an extremely high level (50%) of evaporite mineral salts (CaCO3, CaSO4, and NaCl) in order to evaluate their effect on agricultural productivity. A control group of plants was grown in the absence of mineral salts. All of plants, which were companion planted with beans and squash, were adversely impacted by evaporite mineral salts. Maize grown alone in CaCO3−, CaSO4−, and NaCl-enriched soil exceeded the height of the control plants because the mineral salts acted as a fertilizer. Plants grown in NaCl-enriched soil produced cobbs that were reduced in size and kernel number. These findings suggest that CaCO3 and CaSO4, the most common evaporite mineral salts found in soil, would not have adversely impacted Ancestral Puebloan maize agriculture.
Introduction
Maize (Zea mays) was the single most important cultigen in North American prehistory. It is believed to have been domesticated in the Rio Balsas region in South-Central Mexico ∼9000 BP (Doebley, 1990; Matsuoka et al., 2002). By ∼4500 BP, maize had spread into Southwestern North America and by ∼1000 BP, it had become well established in Eastern North America (Pagán-Jiménez et al., 2015). The rapid prehistoric spread of maize is related to the fact it can be stored in surplus, traded, and it can be transformed and consumed in numerous foods. However, maize also has a number of weaknesses including intolerance to cold temperatures, drought, strong wind, and nutrient-poor soils (Mann, 2011).
Maize is particularly vulnerable in high elevation, dry land environments because of extreme temperature fluctuations, prolonged periods without rain, and strong storms during the growing season. Soil salinization is also an inherent problem, which potentially arises from long-term maize agriculture and irrigation systems. In addition to poisoning tillable land, soil salinization can lead to an overall decline in irrigable acreage. Nutrient-deficient soils result when sodium and chlorine ions displace potassium and phosphorus ions, which are vital nutrients for the growth of maize (Farooq et al., 2015). Maize roots also absorb salt ions such as chlorine and transport them to the leaves where they accumulate in toxic levels, become scorched, and thus reduce the effectiveness of the photosynthetic process (Farooq et al., 2015). Consequently, some archaeologists assume that soil salinization had a serious impact on maize-based agricultural economies in the dry land environments of North America (McCool et al., 2018).
The issue of soil salinization and ancient tillable land has become the focus of a recent and contentious debate on the interpretation of the Ancestral Puebloan economy of Chaco Canyon. The mineralogical and geochemical analyses of soil and water samples from tillable land in Chaco Canyon have produced conflicting and diametrically opposed interpretations. This confusion results from misinterpretations of the chemical and mineralogical composition of the salts, which occur in ancient agricultural soils (Tankersley, 2017; Tankersley et al., 2016).
Salts originate in ground water (e.g., surficial and aquifer) from the processes of infiltration and evapotranspiration. In dry land environments, saline-rich water evaporates and salts precipitate in soil in the form of evaporite minerals (Tucker, 2001). Although most archaeologists assume that the salts in ancient maize agricultural fields are deleterious evaporite chloride minerals such as halite (NaCl), sylvite (KCl), carnallite (KMgCl3·6H2O), and kainite (KMg(SO4)Cl·3H2O), they are predominately carbonate minerals such as calcite (CaCO3), dolomite (CaMg(CO3)2), and magnesite (MgCO3) and sulfate minerals such as gypsum (CaSO4·2H2O), anhydrite (CaSO4), kieserite (MgSO4·H2O), langbeinite (K2Mg2(SO4)3), and polyhalite (K2Ca2Mg(SO4)6·H2O) (Tankersley et al., 2016; Tucker, 2001). This article addresses the impact of evaporite mineral salts (i.e., chlorides, carbonates, and sulfates) on maize agriculture. We take an experimental archaeological approach in order to address this issue.
Experimental archaeology
Experimental archaeology, also known as replication study, provides a mid-range theoretical foundation for postulating inferences concerning a hypothesis or add support to a conclusion concerning an archaeological or geoarchaeological pattern or body of data. Experimental archaeology offers an opportunity to provide empirical testing and theoretical insights into our interpretations of past agricultural-based economies (Foulds, 2013). Experimental archaeology affords a realistic means to collect data, which can be used to test the underlying assumptions and hypotheses from which our theoretical interpretations about maize agricultural-based economies have been formulated. We use an experimental archaeological approach to provide a practical methodology to test the impact of soil salinization on maize agriculture. This investigation provides a living analytical process and recreation of what has been perceived as toxic levels evaporite minerals in order to test whether or not they would have prevented successful maize agriculture in a drought-prone landscape.
Field methods
An experimental maize field (∼150 m2) was tilled to a depth of ∼30 cm in a fine silty, mixed, mesic soil at the University of Cincinnati Center for Field Studies located in the Whitewater River valley of Hamilton County, Ohio. We selected the location of the experimental agricultural field because of its drought-prone geological setting. The experimental field is located on a well-drained late Pleistocene glacial outwash terrace. The substratum consists of a ∼30 m thick layer of coarse calcareous sand and gravel (Lerch et al., 1982).
We cultivated the Ap soil horizon, which is a friable brown (7.5YR 4/2) loam with a subangular blocky structure (Lerch et al., 1982). Today, this soil horizon supports a restored tall grass prairie, which includes Cirsium vulgare (thistle), Echinacea purpurea (purple cone flower), Indigofera tinctoria (indigo), Opuntia humifusa (Eastern prickly pear cactus), Schizachyrium scoparium (little blue stem), Silphium laciniatum (compass plant), and Tradescantia virginiana (spiderwort). The annual precipitation averages ∼965 mm and the annual temperature averages ∼12°C.
Kernels of traditional heirloom Hopi blue maize, also known as Rio Grande Blue, Tarahumara Maiz Azul, and Yoeme Blue, were selected for the soil salinization experiment. Traditional Hopi blue maize is high in protein, has deep roots, and tolerates excessive heat and drought conditions (Nankar et al., 2016). In order to replicate extreme natural and anthropogenic soil salinization conditions, three clusters of seven Hopi blue maize kernels were planted using traditional Puebloan agricultural methods at a depth of ∼3.8 to 5.1 cm, and each cluster was planted at least ∼30 cm apart in a 1:1 mixture of reagent grade NaCl (0.5 kg), CaSO4 (0.5 kg), and CaCO3 (0.5 kg) and topsoil, respectively. It is important to note that this high concentration (50%) of mineral evaporite salts exceeds even the most extreme conditions of ancient North American soil salinization.
In order to test the impact of soil salinization on maize, which is companion grown with C3 cultivars, three additional clusters of seven kernels of Hopi blue maize were planted in tandem using traditional Iroquoian-mounded agricultural methods with seven Cucurbita pepo seeds (squash) and seven Phaseolus vulgaris seeds (bean) at a depth of ∼3.8 to 5.1 cm and each cluster was planted at least ∼30 cm apart in a 1:1 mixture of NaCl (0.5 kg), CaSO4 (0.5 kg), and CaCO3 (0.5 kg) and soil, respectively. A control group of eight clusters of seven maize kernels were planted at a depth of 3.8 to 5.1 cm and planted at least ∼30 cm apart and at a distance of approximately 7 m from the experimentally salinized soil.
An automated Hobolink Model weather monitoring station (Hobolink RX3000) was used to record daily humidity and air and soil temperature. Weekly measurements of the growth of the maize plants were made using a steel metric tape from the time the seedlings sprouted (within six days) to maturity. A 3-m high, 1-inch nylon mesh exclosure was constructed around the experimental garden to prevent herbivores and omnivores from damaging the experimental plants. No fertilizers, herbicides, or pesticides were used in the experiment.
Results
All of the experimental Hopi blue maize plants companion grown with beans and squash using an Iroquoian agricultural method were adversely impacted by the evaporite mineral salts. Indeed, none of the Hopi blue maize stalks achieved the height of the control plants (171 cm). Plants grown in CaSO4 attained the greatest height (165 cm), those grown in CaCO3 reached a height of 140 cm, and those grown in NaCl reached a height of 130 cm (Figure 1). On the other hand, all of the Hopi blue maize plants, which were grown in evaporite mineral salts in the absence of beans and squash exceeded the height of the control plants (>200 cm).
Growth record of Hopi blue maize plants companion grown with squash and beans in evaporite mineral salt-enriched drought-prone soil.
Hopi blue maize plants grown alone in CaCO3 using traditional Puebloan agricultural techniques attained the greatest height (250 cm). Those grown in CaSO4 reached a height of 241 cm and plants grown in NaCl reached a height of 222 cm (Figure 2). By comparison, the control group of Hopi blue maize plants reached a height of 171 cm. Apparently, the air and soil temperature and humidity did not affect the growth of any of the plants (see Figures 3 and 4).
Growth record of Hopi blue maize plants grown alone in evaporite mineral salt-enriched drought-prone soil, regression lines, and R2 values.
Ground and soil temperature record relative to the record of Hopi blue maize plants grown in evaporite mineral salt-enriched drought-prone soil.
Humidity record relative to the record of Hopi blue maize plants grown in evaporite mineral salt-enriched drought-prone soil.
An R2 value was calculated for each Hopi blue maize plant group to determine how close the experimental growth trend data fitted the regression line. More than 90% of the experimental maize growth data variability were explained by the regression model: blue maize control, R2 = 0.932; blue maize grown in NaCl, R2 = 0.963; blue maize grown in CaSO4, R2 = 0.949; and, blue maize grown in CaCO3, R2 = 0.946 (Figure 2).
We found similar results in the productivity of the experimental Hopi blue maize plants in terms of kernel viability and cobb length (Figure 5). In every case, plants grown in evaporite salts exceeded the productivity of the Hopi Blue maize control. The control plants produced between two and six mature stalks (mean = 4 stalks) from the seven kernels that were planted. All of the Hopi blue maize kernels planted in CaSO4 and NaCl produced seven mature stalks from seven kernels. The Hopi blue maize kernels planted in CaCO3 produced only four mature stalks out of the seven kernels planted. Likewise, the length of the cobbs of plants grown in CaSO4 and CaCO3 exceeded the length of the Hopi Blue maize control group cobbs. Although NaCl-enriched soil did not adversely impact the viability of the Hopi Blue maize kernels, it did adversely impact the length of the cobbs and ultimately the number of kernels (Figure 5).
Viability (number of viable kernels) and productivity (cobb length) of Hopi blue maize plants grown in evaporite mineral salt-enriched drought-prone soil.
Discussion
One caveat of this study lies in the variation introduced by the possible variation introduced by regional water chemistry. Although the levels of ammonia, phosphorus, and nitrogen are comparable, the dissolved oxygen, sodium, and sulfate levels are higher in the Southwest than they are in the Ohio River valley (Table 1). In order to compensate for this disparity, we experimentally enriched the soil with calcium sulfate and sodium chloride that greatly exceeded that found in the southwestern agricultural drainage basin (Table 1).
Cross regional comparison of water chemistry between the San Juan Basin and Great Miami Basin watersheds.
aAfter Levings et al. (1996).
bAfter MCD Staff (2005).
Despite the geographically separated regions, the sediments of the two areas are comparable in terms of pH and substrates (Table 2). Thus, the two areas have comparable sediment characteristics such permeability and porosity that could have affected the experimental results. Although surface run-off that may have varied between watered fields in the two regions, ancestral Puebloans managed run-off with water management strategies such as check dams, waffle gardening with earthen walls, and terracing.
Cross-regional comparison of agricultural substrates and pH values for San Juan Basin and Great Miami Basin watersheds.
aAfter Tankersley et al. (2016).
bAfter MCD Staff (2005).
It is widely known from ethnohistoric sources that maize, beans, and squash, commonly referred to as the three sisters, were a significant aspect of sustainable Native American economies in North America at historic contact. In Eastern North America, companion planting of these crops was practiced by Iroquoian-speaking cultures (Mt. Pleasant, 2006). Theoretically, the maize stalk provides vertical support for the bean vines to climb, and the roots of the bean plants enrich the soil with the much-needed nitrogen. Broad-leaved squash vines spread around the plants, which help to retain soil moisture and decomposing squash leaves provide mulch. In addition, the thorny vines of the squash provide protection from herbivores and omnivores. However, the Iroquoian method of gardening is a maladaptive horticultural strategy in salt-rich environments.
Our experiments demonstrate that soil naturally or anthropogenically enriched by a high concentration of evaporite mineral salts (i.e., carbonates, chlorides, and sulfates) may have had deleterious effects on prehistoric companion planting agriculture (see Figure 1). However, our experiments also show that a high concentration of evaporite mineral salts would not have adversely impacted ancient maize if the plants were grown separately from C3 cultivars such as beans and squash as is the case in Puebloan agricultural practices. Although NaCl may have reduced maize productivity, that is, cobb size and kernel number, CaCO3 and CaSO4 more likely increased maize productivity. Evaporite mineral salts in soils, such as CaCO3 and CaSO4, act as fertilizers (see Figures 2 and 5).
Calcium carbonate and sodium sulfate are used to aid seed emergence by breaking up compacted clay soils. They decrease the bulk density of soil and help plants absorb plant nutrients. Furthermore, these mineral salts improve soil development by stabilizing organics and reduce acidity, erosion, soil-runoff, and water logging (see Tankersley et al., 2016). Indeed, a large number of modern fertilizers use other salts such as ammonium nitrate and potassium sulfate for contemporary horticulture.
Unlike the companion planting of maize and C3 cultigens, which has been ethnohistorically documented in Eastern North America among Iroquoian-speaking cultures, Puebloans incorporate a very different agricultural system. Although Puebloans grow multiple varieties of maize, they are not grown together with beans, gourds, or squash (Eggan, 1950; Titiev, 1944). The symbolism of growing maize alone is reflected in the ceremonial, economic, and ritual activities of Puebloan culture (Ortiz, 1979).
Given that Pueboans have retained a remarkable degree of cultural continuity over thousands of years despite the impacts of colonialism, European diseases, and the generational pressures to acculturate, it is likely that Ancestral Puebloans had similar agricultural practices as their descendants (Bradfield, 1971; Hack, 1942). In this scenario, evaporite mineral salts CaCO3 and CaSO4 would have been beneficial to Ancestral Puebloan agriculture. Although NaCl would have lessened the productivity of the size of maize cobbs and number of kernels, it would have been far less abundant in the soil as other non-deleterious evaporite mineral salts such as carbonates and sulfates. Indeed, it is unlikely that Ancestral Puebloan agricultural fields ever reached the concentration of NaCl used in this study. One possible caveat, which is currently impossible to evaluate, is that contemporary Hopi blue maize is a cultigen that is able to succeed in high salt agricultural settings that are ubiquitous across the ancestral Puebloan area.
Conclusion
We used experimental archaeology to recreate maize growing in a highly salinized drought-prone agricultural environment. We directly tested maize’s tolerance to the three most common evaporite mineral salts—CaCO3, CaSO4, and NaCl. Experimental archaeology is by far the most time and cost-effective way to test the impact of soil salinization on maize agriculture because it uses a controlled methodology and produces scientifically valid results. By reproducing extreme soil salinization conditions using a variety of evaporite mineral salts, we can better estimate the feasibility and productivity of ancient maize agriculture.
High concentrations of carbonate, chloride, and sulfate evaporite mineral salts in agricultural soil are unquestionably deleterious to maize when it is grown using the Iroquoian method of gardening as a companion crop with beans and squash. Similarly, high concentrations of NaCl can reduce the productivity of maize in terms of cobb length and kernel numbers. However, CaCO3 and/or CaSO4, the two most common and abundant of evaporite mineral salts in soils, would have been beneficial to ancient agriculturalists. In addition to acting as soil fertilizers, these salts would have increased the productivity of maize in terms of cobb length and kernel numbers. Understanding the impact of salinized soil on maize productivity is crucial to our interpretations of ancient agricultural-based economies.
Footnotes
Acknowledgments
The authors would like to thank the invaluable help of Roger Ruff, David Lentz, Denis Conover, Marcy Taylor, and Marick Hart. The authors are particularly grateful for the comments and suggestions of reviewer, Phillip H. Shelley, and North American Archaeologist editor, Anthony T Boldurian, which were most insightful and helpful.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Charles Phelps Taft Foundation, the John C. Court Family Foundation, and the University of Cincinnati, University Honors Program.