Agronomy of Halophytes as Constructive Use of Saline Systems

Item Type text; Electronic Dissertation

Authors Bresdin, Cylphine

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 07/10/2021 03:23:03

Link to Item http://hdl.handle.net/10150/577318 Agronomy of Halophytes as Constructive Use of Saline Systems

by Cylphine Bresdin

A Dissertation Submitted to the Faculty of the department of Soil, Water and Environmental Sciences

In Partial Fulfillment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY with a major in ENVIRONMENTAL SCIENCE

In the Graduate College The University of Arizona 2015

1 THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Cylphine Bresdin, titled Agronomy of Halophytes as Constructive Use of Saline Systems and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy.

______date: 7/29/2015 Edward Glenn

______date: 7/29/2015 Janick Artiola

______date: 7/29/2015 Kevin Fitzsimmons

______date: 7/29/2015 Margaret Livingston

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

______date: 7/29/2015 Dissertation Director: Edward Glenn

2 STATEMENT BY AUTHOR This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library. Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all instances, however, permission must be obtained from the author.

SIGNED: Cylphine Bresdin

3 Table of Contents

Abstract...... 7

introduction...... 9

I. Problem and Context ...... 9

II. Review o f Literature ...... 10

III. Explanation of Dissertation ...... 12

Present study...... 15

References...... 18

Appendix A: Salicornia bigelovii...... 22

Abstract ...... 22

I. Introduction ...... 23

II. Materials and Methods ...... 24

II.A. Relevant Biology of Salicornia bigelovii ...... 24

II.B. Source of W ild Accessions ...... 25

II.C. Experimental Design ...... 25

II.D. Greenhouse Procedures and Crop Observations ...... 26

II.E. Harvest and Processing ...... 27

II.F. Seed Purity and Proximate Analysis ...... 27

II.G. Environmental Measurements ...... 28

II.H. Statistical Methods ...... 29

III. Results ...... 29

III.A. Survival and Growth ...... 29

4 Table of Contents - Continued

III.B. Temperature Effects ...... 31

III.C. Oil Content ...... 31

IV. Discussion ...... 32

V. Conclusion ...... 33

V. Acknowledgments ...... 34

VI. References ...... 35

VII. Supplemental Material ...... 37

Appendix B: Distichlis palmeri...... 45

Abstract ...... 45

I. Introduction ...... 46

II. Materials and Methods ...... 50

III. Results ...... 51

IV. Discussion ...... 53

V. Acknowledgments ...... 55

V. References...... 56

VI. Supplemental Material ...... 58

Appendix C: Design Concept ...... 64

Abstract ...... 64

I. Introduction ...... 65

II. Methods / Approach ...... 66

III. Results / Design ...... 69

5 Table of Contents - Continued

III.A. Concept ...... 70

III.B. Calculations ...... 71

III.C. Design ...... 73

IV. Discussion ...... 76

V. Conclusion ...... 78

IV. Acknowledgements ...... 79

VI. References ...... 80

VII. Supplemental Material ...... 82

6 Abstract

Extensive coastal sabkhas in the northern Gulf of California in North

America are colonized by Distichlis palmeri, an endemic perennial grass that produces a grain that was harvested as a staple food by native Cocopah people. Previous short-term trials have shown good vegetative growth but low grain yields. During outdoor trials under anaerobic saline soil conditions of paddy-style irrigation, D. palmeri exhibited high salt tolerance, grain and biomass production. Reproductive maturity was reached four years after initial establishment of plants from seed and a 1:3 mixture of male and female plants produced 231-310 g m-2 of grain, with nutritional content similar to domesticated grains, confirming the feasibility of developing D. palmeri as a perennial grain and biomass crop for salinized soils and water supplies. Salicornia bigelovii Torr., a cosmopolitan annual coastal marsh succulent, produces seed with high oil content and has been suggested as a potential cash crop for fuel production from saline irrigation but its domestication and development into a cost effective commodity has been slow. A breeding and selection program for agronomic traits that will provide multiple landscape and ecosystem services that could enhance cost benefits of the agronomy ofS. bigelovii was initiated during a two year period while producing seed for a pilot system at the Masdar Institute in

Abu Dhabi, U.A.E. A concept for a saline landscape designed to consume and concentrate saline waste streams was developed and demonstrates the feasibility and potential to support agronomy of halophytes within a built landscape

7 ecology akin to coastal marsh systems. Exploration and development of potential services halophytes could provide and field testing of selected halophytes for their potential to produce food, fuel, fiber and habitat under designed and managed domestication in our salinized soils with saline waste irrigation needs our continued investigation.

8 introduction

I. Problem and Context

Increased demand for fresh water by expanding urban centers competes with agricultural irrigation demands already strained by soil salinization.

Infiltration of waters from agricultural runoff, industry and natural weathering has increased groundwater salinity, thus intensifying need for production of potable water by industrial desalination which creates a concentrated waste stream. According to Lefebvre and Moletta (2006) the leather, textile, petroleum, agro-food, and chemical industries discharge large volumes of saline wastewater.

Management of the very high volume of saline effluent has been approached from an engineering point of view, as a disposal problem, but if considered a resource for landscape use, we can devise applications for these waters in pre- existing saline environments and garner from the opportunity. Understanding what is required to turn salinized land into productive land while reducing volume of saline waste water through halophyte agronomy is one need. How to integrate halophyte agronomy into a functional multi-component landscape system that does not exacerbate environmental degradation or expose wildlife to toxins is the problem. Exploration of potential services halophytes could provide and the field testing of selected halophytes for their potential to produce food, fuel, fiber and habitat under managed domestication in our salinized soils with saline irrigation needs our continued attention.

9 II. Review o f Literature

Halophytes are plants that grow in a variety of saline environments from submerged sea-beds to coastal marshes to deserts and their potential as food and fuel crops under annual and perennial saline agriculture has been the focus of numerous studies over the last 20 years in tandem with studies investigating genetic modification to increase salt tolerance in high income crops. Genetic modification to impart more salt tolerance in glycophytic crops has had little success. Use of industrial waste brine has been successfully used to irrigate landscapes and agricultural crops in arid regions of the Southwest

(Gerhart, et al., 2006; Glenn, et al., 2009; Riley, et al., 1997). Letters in Science

(Federoff, et al. 2010); Rozema and Flowers, 2008) and American Scientist (Van

Tassel and DeHaan, 2013) have advocated for development of halophyte farming and management techniques for saline agriculture. There is a need to develop environmentally sound agronomic and crop management techniques for saline water agriculture in conjunction with selective breeding of halophytes (Jaradat &

Shahid, 2012; Shahid et al., 2013) to produce high yielding varieties of halophytes for food and fuel appropriate to field conditions where brackish or other low quality water can be used for irrigation (Brown et al., 2014; Glenn et al., 1992,

1998, 1999, 2009, 2013; Grattan et al., 2008; Jordan et al., 2009; Masters et al., 2007;

Rozema and Flowers, 2008; Weber et al., 2007; Zerai et al., 2010). This indicates we may need to alter our diets toward what crops we can grow in our salinized soils with saline irrigation.

10 Salicornia bigelovii Torr. is a cosmopolitan coastal marsh succulent halophyte that has potential as a seawater irrigated crop that can yield: fiber, feed, oilseed for fuel and food for human consumption (Brown et al., 2014; Glenn et al., 1991, 1999, 2013; Jaradat & Shahid, 2012; Shahid et at., 2013; Weber et al.,

2007; Zerai, 2010). Distichlis palmeri, a dioecious perennial grass with Kranz anatomy that predicts C4 photosynthesis, has been considered as a potential crop for salt water agriculture (Leake, 2004; Yensen, 1986, 1987). Attempts have been made to introduce D. palmeri into cultivation as a grain crop. However, in two years of large-scale field trials in Australia, Leake (2004) reported vegetative growth but very low seed production and Pearlstein et al. (2012) reported similar results from plants grown for two years under greenhouse conditions in Tucson, AZ. Yensen and Weber (1986, 1987) showed that D. palmeri grain had nutritional qualities similar to wheat and other grains. Yensen (2006) patented several selected lines of D. palmeri. Because D. palmeri produces large caryopses from terminal panicles like wheat, the grains are easily harvested and processed and was a wild harvest staple of native Cocopah people of the northern Gulf of California. D. palmeri is endemic to extensive coastal esteros adjacent to the desert environment of the Sonoran coastline of North America (Brusca et al.,

2006 ; Felger, 2000, 2007; Pearlstein et al., 2012; Nagler et al., 2006, 2009). This estero of the Colorado River is flooded twice a day by a tidal range of up to 7-9 meters with hypersaline seawater (36-42 g L-1) through a network of tidal creeks bringing seawater as far as 10 km inland at high tide and exposing vast mudflats

11 and saltflats at low tide. Vegetation, dominated by D. palmeri, is confined to the margins of the tidal creeks (Zamora-Arroyo et al., 2005).

In his manuscript, van der Gaag (2010) proposes that we can responsibly deal with waste brine by creation of remedial saline wetlands, but system salinity increases due to evaporation and consumptive water use by plants leading to increased soil salinity problems especially at inland locations and movement of water from source to application via an open canal system also subjects water to evaporation and a consequent increase in salinity. An ecological solution to the problem of high volume saline waste streams might be creation of coastal depressions akin to evaporation ponds in the intertidal zone, essentially creating synthetic tide pools or tidal marshes to dispose of brine waste in a manner confluent with natural tidal regimes (Franklin, 1993; Mexicano et al., 2013;

Yechieli & Wood, 2002) that also provide ecosystem services (Yapp et al., 2012).

III. Explanation of Dissertation

Investigations stem from the success of Cienega de Santa Clara (Baeza, et al., 2013; Mexicano, et al., 2013) which receives irrigation from farm runoff, and the salinity gradient that underlies sequential cell structure of traditional salterns (Rodrigues et al., 2011; Rodriguez-Valera et al., 1981,1985; Zafrilla et al., 2010), agro-forestry principles and integrated farm drainage management

(IFDM) which is an on-farm technique adapted from agro-forestry that uses field runoff in a sequential manner to irrigate progressively more salt tolerant crops resulting in water use efficiency and minimizing runoff due to consumptive

12 water use along the saline gradient (Benes et al., 1999; Jacobsen et al., 2004).

Principles of saline agro-forestry have been tested in seawater aquaponic systems in Eritrea, Africa and a new pilot system has been built at the Masdar Institute in

Abu Dhabi, U.A.E. Oxnard, California recently built a pilot saline effluent urban landscape for treatment purposes. A linear brine wetland and public amenity to convey desalination effluent from the desalination facility in Goodyear, Arizona to the Salt River is under design. If landscape based systems that combine saline effluent volume reduction with halophyte agronomy and habitat creation in arid salinized drylands of the Southwestern US exist, they are obscure.

The author began a breeding and selection program from wild harvested

Salicornia bigelovii accessions while producing seed and biomass for the

Masdar Institute, Abu Dhabi. The author performed technical duties: seeded, transplanted, tended, harvested, made observations, recorded data and analyzed or oversaw analysis of data produced from the two consecutive crop years it was grown in the greenhouse at the Environmental Research Lab at the University of Arizona, Tucson, AZ, U.S.A. Produced seed was prepared and shipped to the

Masdar Institute by the author. The author took lead in preparation of progress reports and manuscript preparation (Appendix A).

The author took over the care of Distichlis palmeri approximately six months after it had been set-up paddy-style outdoors at the Environmental

Research Lab at the University of Arizona, Tucson, AZ, U.S.A. Evaporated and consumed saline water was replenished on an as needed basis through

13 application of city utility water at the base of stalks. The author harvested stalks/ stems, seed-heads and grain, made observations, recorded and analyzed data, and was responsible for the methods and materials, results and discussion sections of the manuscript and is contact author (Appendix B).

The author devised the idea, developed the concept, wrote, submitted and is the contact for the manuscript in Appendix C. The design concept describes a serial ecotope system based on salinity range in order to sustainably dispose of brine waste through volume reduction. Habitat and ecosystem services are created while increasing system salinity from source value to sink value, and thus, turns a waste problem into an ecological amenity.

14 Present study

The methods, results and conclusions of this study are presented in the manuscripts appended to this dissertation. The following is a summary of the most salient findings of this body of work and conclude that the potential of saline landscape systems designed to dispose of saline waste streams in an ecologically sound manner have the potential to support agronomy of cash crop halophytes for food, fuel and fiber as well as habitat and research grounds.

As part of an arid land biofuels program under development in the

U.A.E., 20 wild accessions of Salicornia bigelovii were evaluated as a potential oilseed, biomass and saline irrigated crop over two annual crop cycles in a common garden plot experiment under greenhouse conditions in Tucson,

Arizona, U.S.A.; primary focus was on seed yield. As a secondary focus, selection for other desirable agronomic characteristics including heat tolerance, large flower spikes with synchronized flowering and seed with high oil content with the aim of selecting founder lines for a selection and breeding program.

Differences in seed yield and biomass between accessions were more significant crop year 1 than year 2 when plants less than 100 g were excluded from analysis of randomized blocks using one way analysis of variance (ANOVA).

When ANOVA was based on temperature blocks, there was a clearly defined temperature effect, ambient greenhouse temperature above 40°C reduced biomass and seed production in all accessions. All accessions were sensitive to high humidity and this might have been due to the consistently moist soil

15 conditions maintained by drip irrigation into one gallon pots in the greenhouse.

Yellowing of meristems in juvenile plants was accompanied by slow growth and was abated with the use of synthetic sea-salt containing magnesium, boron and several other micro nutrients. Agronomic traits were more consistent within accessions than between accessions, with differences between accessions less pronounced during Crop 2 than Crop 1. Accessions originating from Louisiana were lanky with little flowering when grown under cooler temperatures and displayed more pronounced heat sensitivity than other accessions. The most northern derived accession displayed little vegetative growth through the hot summer but when the weather cooled at summer’s end, it produced large flowers yielding large seeds with low oil content. It was found that small seed had the greatest percentage of oil and accessions with small seed produced a greater number of seed. Seed shatter was not a problem, but rain caused dried flowers to degrade and drop seed.

In a continuing effort to domestic Distichlis palmeri for grain production for human consumption it was found that when the plant is cultivated outdoors under saline (26-34 g L-1 Total Dissolved Solids (TDS)) paddy-style irrigation, it requires four years to mature and produce a crop (greater than 97% of females were fertile) with yields and nutritional values similar to rice and wheat. Harvest must be high on the stalk because blanket harvest at stalk base slowed recovery and depressed the next season’s yield. When mature plants are maintained under greenhouse conditions, stigmas present prior to pollen maturity resulting

16 in unfertilized females and no harvest. Stalks of potted plants tend to whiten and become waxy when deprived of fertilizer, giving indication that high nutrient effluent or runoff might be an acceptable source of irrigation under domestication. Mixtures of male and female plants (1:3) produced 231-310 g m-2 of grain and demonstrate the potential for further developing D. palmeri as a global crop for salinized soils and water supplies.

Landscape pattern was incorporated as the vehicle for an evapotranspiration induced sequence of ecotopes along a directional saline gradient as a constructive method to reduce reverse osmosis concentrate waste volume while increasing salinity during transport from source (RO facility) salinity to salinity of the sink (sabkha). In the process, biota is relinquished to self-organized marsh habitat and the system of pattern creates potential for wildlife, plant and microbial crops within a defined surface area. The concept and model are logical on paper and have the ability to be adapted to inland environments, supporting the notion that the landscape can be used to constructively dispose of waste saline streams in a responsible manner. However, site specific conditions, including water chemistry, will require pilot systems to finesse details.

17 References

Benes, S., Peters, D., & Grattan, S., (1999). Integrated On-Farm Drainage Management: Using Plant Transpiration to Reduce Drainage Volumes. College Of Agricultural Sciences and Technology California State University, Fresno, Center for Irrigation Technology. Fresno: California Agricultural Technology Institute. Brown, J. J., Glenn, E. P., Smith S. E., (2014). Feasibility of halophyte domestication for high-salinity agriculture. In M.A. Khan et al. (Eds.), Sabkha Ecosystems: Volume IV: Cash Crop Halophyte and Biodiversity 73 Conservation, Tasks for Vegetation Science 47, DOI 10.1007/978-94-007-7411-7_5. Brusca, R. C., Cudney-Bueno, R., Moreno-Baez, M., (2006). Gulf of California Esteros and Estuaries: Analysis, State of Knowledge and Conservation Priority Recommendations. Final Report to the David and Lucille Packard Foundation, Arizona-Sonora Desert Museum, Tucson, AZ. Available on-line at: http://www. rickbrusca.com/http___www.rickbrusca.com_index.html/Blank_files/Packard%20 Final%20Report%20SMALL.pdf Federoff, N. V., Battisti, D. S., Beachy, R. N., Cooper, P. J. M., Fischhoff, D. A., Hodges, C. N., Knauf, V. C., Lobell, D., Mazur, B. J., Molden, D., Reynolds, M. P., Ronald, P. C., Rosegrant, M. W ., Sanchez, P. A., Vonshak, A., Zhu, J.-K., (2010). Radically Rethinking Agriculture for the 21st Century, Science 327: 883. DOI: 10.1126/science.1186834. Felger, R. S., (2000). Flora of the Gran Desierto and Río Colorado of Northwestern Mexico. University of Arizona Press, Tucson. Felger, R. S., (2007). Living resources at the center of the Sonoran Desert: native American plant and animal utilization. In: R. S. Felger, B. Broyles (Eds.), Dry Borders: Great Natural Reserves of the Sonoran Desert. University of Utah Press, Salt Lake City, pp. 147-192 Franklin, J. F., (1993). Preserving Biodiversity: Species, Ecosystems, or Landscapes? Ecological Applications, 3:2 202-205. Gerhart, V. J., Kane, R., & Glenn, E. P., (2006). Recycling Industrial Wastewater for Landscape Irrigation in a Desert Urban Area. Journal of Arid Environments, 67: 473-486. Glenn, E. P., O’Leary, J. W ., Watson, M. C., Thompson, T. L., Kuehl, R. O., (1991). S. bigelovii Torr.: an oilseed halophyte for seawater irrigation. Science 251, 1065– 1067.

18 Glenn, E., Hodges, C., Lieth, H., Pielke, R., Pitelka, L., (1992). Growing halophytes to remove carbon from the atmosphere. Environment 34: 40–43. Glenn, E. P., Brown, J., O’Leary, J. W ., (1998). Irrigating crops with seawater. Scientific American 279: 56–61. Glenn, E., Brown, J., Blumwald, E., (1999). Salt tolerance and crop potential of halophytes. Critical Review in Plant Sciences 18: 227–255. Glenn, E. P., McKeon, C., Gerhart, V., Nagler, P. L., (2009). Deficit irrigation of a landscape halophyte for reuse of saline waste water in a desert city. Landscape Urban Planning 89: 57–64. Glenn, E. P., Anday, T., Chaturvedi, R., Martinez-Garcia, R., Pearlstein, S., Soliz, D., Nelson, S. G., Felger, R. S., (2013). Three halophytes for saline-water agriculture: an oilseed, a forage, a grain. Environmental and Experimental Botany 92: 110-121. Grattan, S. R., Benes, S. R., Peters, D. W ., Diaz, F., (2008). Feasibility of irrigating pickelweed (S. bigelovii Torr.) with hyper-saline drainage water. Journal of Environmental Quality 37: 149–156. Jacobsen, T., Basinal, L., Drake, N. R., Cervinka, V., Buchnoff, K., & Martin, M. A. (2004). Salt Management Using IFDM. In 2004 Landowner Manual (pp. 2.1-2.16). Fresno, California: Center for Irrigation Technology. Jaradat, A. A., Shahid, M., (2012). The dwarf saltwort (Salicornia bigelovii Torr.): evaluation of breeding populations. ISRN Agronomy, article ID 151537. Jordan, F. L., Yoklic, M., Morino, K., Seaman, R., Brown, P., Glenn, E. P., (2009). Consumptive water use and stomatal conductance of Atriplex lentiformis irrigated with industrial brine in a desert irrigation district. Agriculture and Forest Meteorology 149: 899–912. Leake, J., (2004). NyPa “W ild Wheat” Proving Trials, Final Report, NyPa Australia Limited, Adelaide, South Australia. Lefebvre, O., & Moletta, R., (2006). Treatment of Organic Pollution in Industrial Saline Wastewater: a literature review. Water Research, 40: 3671-3682. Masters, D., Benes, S. R., Norman, H., (2007). Biosaline agriculture for forage and livestock production. Agriculture Ecosystems Environment 119: 234–248. Mexicano, Lourdes, Glenn, Edward P., Hinojosa-Huerta, Osvel, Garcia- Hernandez, Jaqueline, Flessa, karl, Hinojosa-Corona, Alejandro, (2013). Long- term sustainability of the hydrology and vegetation of Cienega de Santa Clara, an anthropogenic wetland created by disposal of agricultural drain water in the delta of the Colorado River, Mexico. Ecological Engineering, 59: 111-120.

19 Nagler, P. L., Glenn, E. P., Brusca, R. C., Hinojosa-Huerta, O., (2006). Coastal wetlands of the northern Gulf of California: inventory and conservation status. Aquatic Conservation: Marine and Freshwater Ecosystems 16: 5-28. Pearlstein, S. L., Felger, R. S., Glenn, E. P., Harrington, J., Al-Ghanem, K. A., Nelson, S. G., (2012). Nipa (Distichlis palmeri): A perennial grain crop for saltwater irrigation. Journal of Arid Environments 82: 60-70. Riley, J. J., Fitzsimmons, K. M., & Glenn, E. P., (1997). Halophyte Irrigation: an Overlooked Strategy for Management of Membrane Filtration Concentrate. Desalination, 110: 197-211. Rodrigues, C. M., Bio, A., Amat, F., & Vieira, N., (2011). Artisinal Salt Production in Aveiro/Portugal - an Ecofriendly Process. DOI:10.1186/1746-1448-7-3, Retrieved from Saline Systems: http://www.salinesystems.org. Rodriguez-Valera, F., Ruiz-Berraquero, F., & Ramos-Cormenana, A., (1981). Characteristics of the Heterotrophic Bacterial Populations in Hypersaline Environments of Different Salt Concentrations. Microbial Ecology, 7: 235-243. Rodriguez-Valera, F., Ventosa, A., Juez, G., & Imhoff, J. F., (1985). Variation of Environmental Factors and Microbial Populations with Salt Concentrations in a Multi-Pond Saltern. Microbial Ecology, 11: 107-115. Rozema, J., Flowers, T., (2008). Crops for a salinized world. Science 322: 1478– 1480. Shahid, M., Jaradat, A. A., Rao, N. K., (2013). Use of marginal water for Salicornia bigelovii Torr. planting in the United Arab Emirates. In: Shahid, S. A., Abdelfattah, M. A., Taha, F. K. (Eds.), Developments in Soil Assessment and Reclamation: Innovative Thinking and Use of Marginal Soil and Water Resources in Irrigated Agriculture. Springer, Dordrecht, pp. 451-462. van der Gagg, J. J., Paulissen, M., & Slim, P. A., (2010). Halophyte Filters as Saline Treatment Wetlands applications and constraints. Wageningen, The Netherlands: Alterra. Van Tassel, David, DeHaan, Lee, (2013). W ild Plants to the Rescue. American Scientist, 101:3 218-223. Weber, D., Ansari, R., Gul, B., Khan, M., (2007). Potential of halophytes as sources of edible oil. Journal Arid Environments 68: 315–321. Yapp, G., Walker, J., & Thackway, R., (2012). Linking Vegetation Type and Condition to Ecosystem Goods and Services. Ecological Complexity, 7: 292-301. Yechieli, Y. & Wood, W . W ., (2002). Hydrologic Processes in Saline Systems: playas, sabkhas, and saline lakes. Earth-Science Reviews, 58: 343-365.

20 Yensen, S. B., Weber, C. W ., (1986). Composition of Distichlis palmeri grain, a saltgrass. Journal of Food Science 51: 1089-1090. Yensen, S. B., Weber, C. W ., (1987). Protein quality of Distichlis palmeri, a salt grass. Nutrition Reports International 35: 863-872. Zafrilla, B., Martinez-Espinosa, R. M., Alonso, M. A., & Bonete, M. J., (2010) Biodiversity of Archaea and Floral of Two Inland Saltern Ecosystems in the Alto Vinalopo Valley, Spain. DOI:10.1186/1746-1448-6-10, Retrieved from Saline Systems: http://www.salinesystems.org. Zamora-Arroyo, Francisco, Jennifer Pitt, Steve Cornelius, Edward Glenn, Osvel Hinojosa-Huerta, Marcia Moreno, Jaqueline García, Pamela Nagler,Meredith de la Garza, and Iván Parra, (2005). Conservation Priorities in the Colorado River Delta, Mexico and the United States. Prepared by the Sonoran Institute, Environmental Defense, University of Arizona, Pronatura Noroeste Dirección de Conservación Sonora, Centro de Investigación en Alimentación y Desarrollo, and World W ildlife Fund—Gulf of California Program. Zerai, D. B., Glenn E. P., Chatervedi R., Lu Z., Mamood, A. N., Nelson, S. G., Ray D. T., (2010). Potential for the improvement of S. bigelovii through selective breeding. Ecological Engineering 36: 730-739.

21 Appendix A: Salicornia bigelovii

Comparison of Seed Production and Agronomic Traits

of 20 Wild Accessions of Salicornia bigelovii Torr.

Grown under Greenhouse Conditions

in

Halophytes for Food Security in Dry Lands M.A. Khan, M. Ozturk, B. Gul, & M. Z. Ahmed (Eds.) DOI: http://dx.doi.org/10.1016/B978-0-12-801854-5.00005-4 © 2016 Elsevier Inc. All rights reserved.

Cylphine Bresdin, Edward P. Glenn, J. Jed Brown

Abstract

Salicornia bigelovii is a succulent annual salt marsh halophyte that produces oilseed on brackish and seawater irrigation. We compared oilseed and biomass production of 20 wild accessions of S. bigelovii grown under greenhouse conditions and irrigated with brackish (10 g L-1 Total Dissolved Solids (TDS)) water for two crop cycles while initiating a selection program to develop desirable agronomic traits for field cultivation in the United Arab Emirates

(U.A.E.). Best-performing accessions came from the Gulf of Mexico, U.S.A. and gave biomass yields of 1200-1800 g/m2 and seed yields of 188-244 g m-2, similar to open-field yields under seawater irrigation. Agronomic traits were more

22 consistent within accessions than between accessions, with differences between accessions less pronounced during Crop 2 than Crop 1. We found that ambient greenhouse temperature above 40°C reduced biomass and seed production in all accessions.

Keywords: Salicornia bigelovii; Saline irrigation; Halophyte; Oilseed yield;

Biomass yield

I. Introduction

Salicornia bigelovii has potential as a seawater irrigated crop that can yield: fiber, feed, oilseed and food for human consumption (Glenn et al., 1991,

1999, 2013; Jaradat & Shahid, 2012; Weber et al., 2007; Shahid et at., 2013; Zerai,

2010). There is a need to develop environmentally sound agronomic and crop management techniques for seawater agriculture in conjunction with selective breeding of S. bigelovii (Brown et al., 2014; Jaradat & Shahid, 2012; Shahid et al.,

2013) to produce high yielding varieties appropriate to field conditions where seawater, brackish or other low quality water can be used for irrigation (Glenn et al., 1992, 1994, 1998, 1999, 2009, 2013; Grattan et al., 2008; Jordan et al., 2009;

Masters et al., 2007; Rozema and Flowers, 2008;Zerai et al., 2010). We investigated

S. bigelovii as a potential oilseed and biomass crop as part of an arid land biofuels program under development in the U.A.E. We evaluated 20 wild accessions of

S. bigelovii over two annual crop cycles under greenhouse conditions in Tucson,

Arizona, U.S.A. Our primary focus was on seed yield. As a secondary focus, we

23 selected for other desirable agronomic characteristics including heat tolerance, large flower spikes with synchronized flowering and seed with high oil content with the aim of selecting founder lines for a selection and breeding program.

II. Materials and Methods

II.A. Relevant Biology of Salicornia bigelovii

Salicornia bigelovii Torr. (Amaranthaceae), a small annual succulent halophyte native to coastal salt marshes of North America and the Caribbean, is usually less than 60 cm tall under natural conditions and has leafless succulent segmented stems that may display lateral branching. Both stems and branches terminate in long upright spikes that are comprised of three-flowered cymes bearing perfect flowers in response to photoperiod (Munz, 1974; W iggins, 1980;

York et al., 2000). In its native range, S. bigelovii begins flowering in June and the spike continues to elongate from the apex producing new flowers over 30–60 days, causing a population of S. bigelovii to contain flowers and seeds at all stages of maturation during the majority of the flowering season (Zerai et al., 2010).

Stigmas appear about 10 days before anthers emerge, allowing out-crossing to occur, and at anthesis, self-fertilization will occur in the absence of out-crossing

(Zerai et al., 2010). Each flower on the spike produces a single seed held by a persistent fleshy calyx until the flower spike dries in the autumn then seeds detach from the spike (Sheperd et al., 2005).

24 II.B. Source of W ild Accessions

Seventeen S. bigelovii accessions were collected in 2011 from wild populations in Puerto Rico, the Atlantic coast of the US and the US portion of the Gulf of Mexico (Table 1). Two additional accessions were collected from the Florida Keys in 2012. Seed previously produced from an accession that had been collected from Galveston, Texas, U.S.A. and held in cold storage, was also included in these trials.

II.C. Experimental Design

Plants of each annual crop were grown in 4 L pots containing a 1:1 mix of soil and medium sand in a greenhouse in Tucson, Arizona, US, an arid location with generally clear skies. A block design was used to account for a positive

6°C gradient from the cooling pad end to the fan end of the evaporatively- cooled greenhouse. For each crop cycle, accessions were randomized using rand function (Microsoft Excel) into blocks that consisted of five columns and ten rows, 50 pots per block. Pots were spaced 0.30 meters apart within columns and rows on four parallel steel mesh tables orientated along the temperature gradient, giving each plant an area of 0.093 m2. This spacing allowed plants to form closed canopies, allowing yields to be expressed on an area basis as well as on a per plant basis, for comparison with open-field yields. Crop 1 (2012) consisted of 1,200 plants. Crop 2 (2013) consisted of 1,100 plants.

25 II.D. Greenhouse Procedures and Crop Observations

For Crop 1, the first 18 accessions listed in Table 1 were sown on March

26, 2012 in seeding trays and irrigated with municipal water by overhead spray three times a day. Poor germinating accessions: West Barnstable, South Padre 2,

South Padre North, Web Landing, Well Fleet, Quimby, Brewster and Elmer Island were reseeded on April 10-12, 2012. South Padre 2 and South Padre 3 were combined and labeled as South Padre and Well2 and Well Fleet were combined under the label of Well Fleet. Individual plant height was measured every other week during 2012 until flowering became prominent.

Crop 2 was sown on March 19, 2013 following the procedure used for

Crop 1, and consisted of the top nine performing accessions from Crop 1 plus two new accessions from Florida. Seedlings were transplanted May 16 through

June 16 for 2012 and May 12-21 for 2013. During 2012, overhead spray was maintained for the transplants with an additional manual spray of saline water,

-1 -1 10 g L NaCl plus MgSO4·7H2O, soluble fertilizer to supply 25 mg L N, P, K, and fungicide (Ridomil, BASF, Inc., Florham Park, New Jersey). Pots were put into place on tables and drip irrigation began June 4, 2012. Crop 2 seedlings were transplanted into pots with the same soil mixture as the prior year and directly put on drip with synthetic sea salt (Crystal Sea Marinemix, Marine Enterprises

International, Baltimore, MD) at 10 g L-1 with the same fertilizer and fungicide additive used for Crop 1. Irrigation regime was three times per day at a rate to keep soil moist and allow some leaching without depleting salts. Salinity of

26 leachate was 9-11 g L-1.

II.E. Harvest and Processing

Crop 1 plants were individually harvested in early November 2012 when most of the spikes had formed seeds but before seeds began to drop from spikes.

They were cut at the base, placed in paper bags, weighed and allowed to dry for two months. Plants, except the Atlantic accessions were processed individually in batches by accession. Individual dried plants were weighed, passed through a hammer mill and the milled content was then hand sifted and winnowed in front of a greenhouse exhaust fan to separate seed from chaff. Seeds erew put into a paper coin envelope and stored in a cold-room. Small accessions from the

Atlantic region (Quimby, Chincoteaque, Well Fleet, Web Landing, West Barnstable,

Brewster, and Puerto Rico) were pooled per accession for processing.

Crop 2 plants were harvested early December 2013 and processed the same as the Crop 1 plants with the exception that small plants less than 65 g dry weight were pooled and processed as one plant.

II.F. Seed Purity and Proximate Analysis

Ten randomly chosen envelopes per accession were individually analyzed for seed purity for Crop 1. A sample (seed plus chaff) was weighed, seed and chaff were separated by hand then seeds were weighed and counted. Seed weight was divided by sample weight to obtain purity. Weight per seed was obtained by dividing seed weight by seed count. Average number of seeds per

27 plant per accession was calculated by dividing seed number by sample weight and then multiplying by weight of envelope contents. The mean purity of the ten assays was taken to represent the accession and applied to all seed envelopes

(individual plants) of that accession (Table 2).

For Crop 2 purity assays, a sample of about 0.6 ml of seed from ten randomly chosen envelopes per accession was mixed and weighed. Chaff and seed from a sample of the mixed seed were separated and each component was weighed. Seed weight divided by the sum of seed weight and chaff was taken as accession purity. Eight groups of 25 seeds were weighed and weight was divided by 25 to estimate individual seed weight for that accession (Table 2). Harvest index (HI) was calculated on a per accession level as, total weight of seed yield divided by total dry above-ground biomass before separating seeds from plant.

Mean seed yield was multiplied by 10.6 to obtain seed yield per m2 in order to approximate accession yields on an area basis. Average number of seed per m2 was obtained by dividing seed yield per m2 by average seed weight.

Cleaned seed used for purity calculations from Crop 1 and Crop 2 were combined per accession and sent to Litchfield Analytical Service (Litchfield,

Michigan, U.S.A.) for total fat content by ethanol extraction and a proximate feed analysis was done on the biomass waste from Crop 2.

II.G. Environmental Measurements

we used a plant growth photometer (International Light INC.,

Newburyport, MA, U.S.A.) to measure light transmission into the greenhouse at

28 irregular intervals over each crop cycle. Air temperatures were measured with maximum-minimum thermometers installed at the pad end and the fan end of the array of plants. Maximum daily temperature recorded at approximately weekly intervals from July 25, 2012 to August 24, 2012 were used to establish the temperature gradient during the period of flowering and seed filling for Crop 1.

II.H. Statistical Methods

For analysis, data from individual blocks were consolidated into nine complete blocks running parallel to the temperature data in the greenhouse.

Crop 1 and Crop 2 were analyzed by two-way analysis of variance (ANOVA) with accession (seed source) and block along the temperature gradient as categorical variables and seed yield, biomass yield and HI as dependent variables. When ANOVAs were significant, (P < 0.05) means were separated using the Holm-Sidak Test for comparing multiple means. Statistical procedures were carried out using SigmaPlot software (Systat, Inc., Santa Cruz, CA).

III. Results

III.A. Survival and Growth

An early display of pale yellow shoot tips during Crop 1 suggested that there was either a reduction of light in the greenhouse or that there was a mineral deficiency. Blue and red light were reduced in the greenhouse compared to outside by about 38% and far red was reduced by about 54%. Yellow tips receded when we substituted a synthetic sea salt compound (Instant Ocean, Union Pet

29 Group, Inc., Cincinnati, OH, U.S.A.) for the in-house saline mix made from

Epsom salts. No yellow shoot tips were evident during the course of Crop 2. Of

1,200 established transplants in Crop 1, 996 (83%) survived the 2012 growing season and produced seed; 14 plants had no seed. Of 1,100 (84%) established transplants in Crop 2, 922 survived the 2013 growing season and produced seed, whereas 11 plants had no seed. Crop 2 produced seed with a higher mean weight per seed than Crop 1. Non-seed bearing plants tended to be either small plants from the Atlantic coast or lanky plants grown at the cool end of the greenhouse.

Crop 1 produced a total of about 20 kg seed with an estimate of 4.59 x 107 total seeds and a mean weight of 0.442 milligrams per seed (Table 2). ANOVA of accessions for Crop 1 was significant for seed yield, biomass and HI (Table

3). Grand Isle had a higher HI than South Padre, Corpus Christie Bridge and Boca

Chica West in 2012. Crop 2 produced a total of about 16 kg seed with an estimate of 3.40 x 107 total seeds and a mean weight of 0.482 milligrams per seed (Table

2). ANOVA of accessions for Crop 2 was also significant for biomass, seed yield and HI (Table 3). Although the ANOVA for HI by accessions was significant (P

= 0.003) for Crop 2, the means separation test did not find differences among any of the accessions at P <0.05 for Crop 2. There were no significant differences between mean seed yield per plant between crop years (P>0.05).

The two Florida accessions were not included in Crop 2 analysis because they performed poorly. The Florida accessions produced a total of 403.4 g of seed from 42 plants or 9.6 grams per plant, which is 50% less than the mean yield

30 for Crop 1 accessions and the other accessions in Crop 2. Seed yield, biomass and HI of individual accessions for Crop 1 and Crop 2 are in Figures 1 and 2, respectively.

Differences among accessions tended to be more pronounced for Crop 1 than for Crop 2. Quimby, from Massachusetts on the Atlantic Coast, had only half the seed and biomass yield as accessions from the Gulf of Mexico, and was not included in Crop 2. All of the Texas accessions performed well, yielding about

19-23 grams of seed per plant (Table 3).

III.B. Temperature Effects

During the period of flowering and seed filling, maximum daily temperatures ranged from 33.9-38.9o C at the pad end to 40.6-43.8o C at the fan end of the greenhouse. For both Crop 1 and Crop 2, the effect of blocking for temperature was significant for seed yield, biomass, and HI (Figure 3). Seed yields peaked between 39°C and 42°C (Figure 3A) for all accessions except

Quimby, for which yield decreased across the gradient (not shown). Biomass decreased over the whole temperature range (Figure 3B) and HI was maximal at

42.5o C (Figure 3C). Scatter plots of height versus temperature for Crop 1 showed that height increased with decreasing temperature (Figure 4).

III.C. Oil Content

Seed oil content as percent dry weight of combined Crops 1 and 2 for the nine best performing accessions ranged from 22.3% to 25.9% while oil content

31 for Quimby was 15.9%. Proximate analysis shows the biomass waste to be high in digestible nutrients and judged by the ash and sodium content, also high in salts.

IV. Discussion

Mean seed weight from greenhouse grown accessions was lower than observed for accessions subjected to repeated harvest and mechanical threshing in previous field trials (Glenn et al., 1998). We attribute this to mass selection under repeated harvesting and threshing, in which the smaller seeds tend to be culled out at each harvest (Zerai et al., 2010).

over two crop cycles, more than 1,918 plants survived from 2,300 transplanted, showing that greenhouse conditions were favorable for evaluating the yield potential of these accessions. The greenhouse had a two-year average seed yield of 216.3 g m-2 from individually processed plants. The best-performing accessions came from the Gulf of Mexico and yielded the equivalent of 188-

244 g m-2 of seeds and 1200-1800 g m-2 of biomass, similar to open-field yields under seawater irrigation (Glenn et at., 1991). Field trials of many of these same accessions were grown in outdoor plots in the United Arab Emirates with saline water irrigation, and similarly, Gulf of Mexico accessions performed best

(Brown, unpublished data). Field yields of seed from other seed oil crops: hemp, canola, mustard, sesame and poppy are in the range of 115-180 g m-2 (FAOSTAT,

2012). The oil content of these greenhouse grown plants was slightly lower than reported for field-grownS. bigelovii, but still within the range of other seed oil crops (Glenn et al., 1991). Hence, this study confirms the high yield potential of

32 this species under brackish water irrigation.

It was noted that plants that produced small seeds produced more seeds than plants that produced larger seeds. This is relevant to the goals of a breeding program because these seeds are small and difficult to process compared to some other oilseed crops. Corpus Christi Bridge, the tallest accession, and the South

Padres produced the largest seeds both years. Elmer Island showed the greatest improvement in HI and average seed weight from Crop 1 to Crop 2. Quimby from

Virginia was not included in Crop 1 analysis or grown in Crop 2 because it was a poor performing accession, but it had distinctive characteristics. It displayed little vegetative growth, but when the photoperiod and ambient temperatures began to change in August it grayed and with no further vegetative growth it produced large flower spikes that produced a low number of large black seeds;

0.933 milligrams per seed. Hence, this germplasm should be retained in a future breeding program.

V. Conclusion

Seed from S. bigelovii accessions collected from the Atlantic coast of the

US; the US portion of the Gulf of Mexico coast and Puerto Rico were grown and evaluated in a common garden experiment under greenhouse conditions in

Tucson, Arizona, U.S.A. Seed harvested from highest performing accessions from

Crop 1 were used to produce Crop 2. Seed weights were larger for Crop 2 than

Crop 1, showing the potential of improving the crop through selective breeding and mass selection. Accessions from the Gulf of Mexico tolerated the high

33 greenhouse temperatures encountered in this study, but temperatures above 40o

C reduced seed yield and should be avoided if possible. Seeding earlier in the season could allow plants to establish before extreme ambient temperatures,

40-47°C, are reached and maintained for extended periods in the United Arab

Emirates and similar climates. A successful strategy for hot climates could be to sow seed early (e.g., October) and harvest in April or May to avoid heat stress in summer, as was practiced in Eretria (Zerai, 2010).

Continued cultivation of crops from individual top seed producers of each accession should be carried out in the greenhouse and concurrently in the field to establish whether sufficient genetic variation within accessions exists and to create founder lines of top producers, and to develop environmentally sound management techniques. In addition to improving seed yield and seed size, breeding goals should address improving heat sensitivity, selecting for early-flowering (day neutral) plants and selection for shorter but more numerous flower spikes, to reduce the long period from flowering to maturation as new flowers are produced along elongating spikes.

V. Acknow ledgments

This research was funded by a grant from Masdar Institute’s Sustainable

Bioenergy Research Consortium, Abu Dhabi, U.A.E. and partially fulfills requirements for the Degree of Doctor of Philosophy, University of Arizona.

34 VI. References

Brown, J. J., Glenn, E. P., Smith S. E., (2014). Feasibility of halophyte domestication for high-salinity agriculture. In M. A. Khan et al. (Eds.), Sabkha Ecosystems: Volume IV: Cash Crop Halophyte and Biodiversity 73 Conservation, Tasks for Vegetation Science 47, DOI 10.1007/978-94-007-7411-7_5. FAOSTAT, (2012). http://fao.org/site/291/default.aspx. Last visited August 2014. Glenn, E. P., O’Leary, J. W ., Watson, M. C., Thompson, T. L., Kuehl, R. O., (1991). S. bigelovii Torr.: an oilseed halophyte for seawater irrigation. Science 251: 1065– 1067. Glenn, E., Hodges, C., Lieth, H., Pielke, R., Pitelka, L., (1992). Growing halophytes to remove carbon from the atmosphere. Environment 34: 40–43. Glenn, E. P., Olsen, M., Frye, R., Moore, D., (1994). Use of halophytes to remove carbon from the atmosphere: results of a demonstration experiment. Electric Power Research Institute, TR-103310, Research Report 8011-03, Palo Alto, CA, 6 chapters, various pages. Glenn, E. P., Brown, J., O’Leary, J. W ., (1998). Irrigating crops with seawater. Scientific American 279: 56–61. Glenn, E., Brown, J., Blumwald, E., (1999). Salt tolerance and crop potential of halophytes. Critical Review in Plant Sciences 18: 227–255. Glenn, E. P., McKeon, C., Gerhart, V., Nagler, P. L., (2009). Deficit irrigation of a landscape halophyte for reuse of saline waste water in a desert city. Landscape Urban Planning 89: 57–64. Glenn, E. P., Anday, T., Chaturvedi, R., Martinez-Garcia, R., Pearlstein, S., Soliz, D., Nelson, S. G., Felger, R. S., (2013). Three halophytes for saline-water agriculture: an oilseed, a forage, a grain. Environmental and Experimental Botany 92: 110-121. Grattan, S. R., Benes, S. R., Peters, D. W ., Diaz, F., (2008). Feasibility of irrigating pickelweed (S. bigelovii Torr.) with hyper-saline drainage water. Journal of Environmental Quality 37: 149–156. Jaradat, A. A., Shahid, M., (2012). The dwarf saltwort (Salicornia bigelovii Torr.): evaluation of breeding populations. ISRN Agronomy, article ID 151537. Jordan, F. L., Yoklic, M., Morino, K., Seaman, R., Brown, P., Glenn, E. P., (2009).

35 Consumptive water use and stomatal conductance of Atriplex lentiformis irrigated with industrial brine in a desert irrigation district. Agriculture and Forest Meteorology 149: 899–912. Masters, D., Benes, S. R., Norman, H., (2007). Biosaline agriculture for forage and livestock production. Agriculture Ecosystems Environment 119: 234–248. Munz, P., (1974). A flora of Southern California. University of California Press, Berkeley, CA. Rozema, J., Flowers, T., (2008). Crops for a salinized world. Science 322: 1478– 1480. Shahid, M., Jaradat, A. A., Rao, N. K., (2013). Use of marginal water for Salicornia bigelovii Torr. planting in the United Arab Emirates. In: Shahid, S. A., Abdelfattah, M. A., Taha, F. K. (Eds.), Developments in Soil Assessment and Reclamation: Innovative Thinking and Use of Marginal Soil and Water Resources in Irrigated Agriculture. Springer, Dordrecht, pp. 451-462. Sheperd K. A., Macfarlane T. D., Colmer T. D., (2005). Morphology, anatomy and histochemistry of Salicornioideae (Chenopodiaceae) fruits and seeds. Annals of Botany 95: 917-933. Weber, D., Ansari, R., Gul, B., Khan, M., (2007). Potential of halophytes as sources of edible oil. Journal Arid Environments 68: 315–321. York, J., Lu, Z., Glenn, E. P., John, M. E., (2000). Daylength affects floral initiation in S. bigelovii Torr. Plant Biology, 41–42 (abstract). Zerai, D. B., Glenn ,E. P., Chatervedi, R., Lu, Z., Mamood, A. N., Nelson, S. G., Ray, D. T., (2010). Potential for the improvement of S. bigelovii through selective breeding. Ecological Engineering 36: 730-739.

36 VII. Supplemental Material

Table 1. Locality data for accessions and number of plants planted in 2012 and 2013.

No. Planted 2012 2013 ID Name State Location 129 114 BC Boca Chica TX N25 59.873 W97 09.607 123 118 BCW Boca Chica West TX N25 59.221 W97 11.339 37 B Brewster MA N41 45.601 W70 06.930 1 C Chincoteaque VA N37 56 13.95 W75 24 35.05 128 104 CCB Corpus Christi Bridge TX N27 37 54.01 W97 14 13.09 122 197 EI Elmer Island LA N29 10.742 W90 04.197 128 104 G Galveston TX N 29.269 W95.0065 122 94 GI Grand Isle LA N29 14.709 W89 59.240 122 118 MI Mustang Island TX N27 37 16.85W97 12 38.82 16 PR Puerto Rico PR N17 57.125 W67 11.745 48 Q Quimby VA N37 33 53.12W75 44 32.13 123 131 Sp SP 2 TX N26 10.071 W97 10.560 Sp SP 3 TX N26 09.390 W97 10.435 53 130 SpN SP north TX N26 11.889 W97 10.863 29 WL Web Landing VA N37 24 7.28W755 2 0.98 WL well2 MA N41 54.372 W69 59.940 16 WF Well Fleet MA N41 53.780 W70 00.547 3 WB West Barnstable MA N41 42.959 W70 22.046 45 FlB Big Coppitt Key FL N24 35.836W081 39.311 45 FlL Little Torch Key FL N24 39.985W081 23.615

37 Table 2. Data for 2012 and 2013 accessions for mean biomass, seed yield, purity, seed weight and total seed produced.

2 2012 Biomass (gDW) Seed (g) Purity % Seed g/m wt/seed # seed Total seed (g) Total # seed BC 157.9 22.1 77.4 234.1 4.8E-04 4.6E+04 2429 5.1E+06 BCW 154.5 20.2 78.5 214.5 4.6E-04 4.4E+04 2287 4.9E+06 CCB 163.4 21.6 79.4 229.2 5.3E-04 4.1E+04 2379 4.5E+06 EI 168.0 17.5 68.6 185.6 3.8E-04 4.6E+04 1698 4.5E+06 G 123.6 23.0 75.9 243.8 3.3E-04 6.9E+04 2783 8.3E+06 GI 146.5 21.9 80.3 232.2 4.3E-04 5.1E+04 2278 5.3E+06 MI 136.6 18.6 65.1 197.1 4.4E-04 4.2E+04 2324 5.2E+06 Sp 173.7 23.7 75.7 250.8 4.4E-04 5.3E+04 2579 5.8E+06 SpN 156.2 21.2 67.6 224.3 4.8E-04 4.4E+04 1079 2.2E+06 mean 153.4 21.1 74.3 223.5 4.4E-04 4.8E+04 19837 4.6E+07

2 2013 Biomass (gDW) Seed (g) Purity % Seed g/m wt/seed # seed Total seed (g) Total # seed BC 145.7 20.3 95.7 214.8 3.6E-04 5.6E+04 2067 5.7E+06 BCW 134.7 18.6 91.2 197.6 4.6E-04 4.1E+04 1995 4.3E+06 CCB 120.8 19.0 94.3 201.7 5.2E-04 3.7E+04 1770 3.4E+06 EI 131.0 17.7 89.3 188.1 5.4E-04 3.3E+04 1083 2.0E+06 G 107.5 20.0 94.6 211.6 4.4E-04 4.5E+04 1358 3.1E+06 GI 136.0 20.6 88.9 218.8 4.8E-04 4.3E+04 1404 2.9E+06 MI 150.1 19.6 93.9 207.5 4.8E-04 4.1E+04 1918 4.0E+06 Sp 125.6 20.2 94.6 214.5 5.4E-04 3.7E+04 2226 4.1E+06 SpN 146.9 20.4 88.8 215.8 5.2E-04 3.9E+04 2280 4.4E+06 mean 133.1 19.6 92.4 207.8 4.8E-04 4.1E+04 16100 3.4E+07

38 Table 3. Results of two-way ANOVAs for seed yield, biomass and harvest index (HI) for Crops 1 and 2 for Salicornia bigelovii grown in Tucson, Arizona, U.S.A.

DF: F P Crop/Dependent Accession, Block, Accession, Block Accession, Block Variable Residual Crop 1 Seed Yield 8,9,879 7.61, 10.0 < 0.001, < 0.001 Biomass 9, 8, 882 17.3, 14.1 < 0.001, < 0.001 HI 9, 8, 865 19.9, 32.3 < 0.001, < 0.001 Crop 2 Seed Yield 8, 7, 795 2.50, 3.09 0.003, 0.011 Biomass 8, 7, 795 3.76, 24.5 < 0.001, < 0.001 HI 8, 7, 779 2.96, 9.59 0.003, < 0.001

39 Table 4. Proximate analysis showing nutritive value of Salicornia bigelovii straw.

Feed Analysis Dry Basis Metabolizable Energy 1731 Kcal kg-1 Digestible Energy 2.1 Mcal kg-1 Net Energy for Lactation 1.06 Mcal kg-1 Net Energy for Maintenance 0.90 Mcal kg-1 Net Energy for Gain 0.20 Mcal kg-1 Effective Net Energy 37.43 % Crude protein 4.86 % Crude fiber 20.47 % Crude carbohydrates 30.25 % Digestible carbohydrates 39.55 % Fat 1.9 % Total Digestible Nutrients 47.96 % Ash 33.23 % Phosphorus (P) 0.22 % Calcium (Ca) 0.25 % Potassium (K) 2.46 % Magnesium (Mg) 0.48 % Sodium (Na) 10.72 %

40 Figure 1. Mean seed yield (A), biomass yield (B) and harvest index (C) of Crop 1 accessions of Salicornia bigelovii grown in a greenhouse in Tucson, Arizona, U.S.A. in 2012. Error bars are standard errors of mean. Different letters erov bars indicate means were significantly different at P<0.05.

41 Figure 2. Mean seed yield (A), biomass yield (B) and harvest index (C) of Crop 2 accessions of Salicornia bigelovii grown in a greenhouse in Tucson, Arizona, U.S.A. in 2013. Error bars are standard errors of mean. Different letters erov bars indicate means were significantly different at P<0.05. The ANOVA of Harvest Index was not significant (P > 0.05) so means were not separated.

42 Figure 3. Means across accessions of seed yield (A), biomass yield (B) and harvest index (C) versus maximum day temperatures during the flowering and seed filling stages of growth for Salicornia bigelovii grown in a greenhouse in Tucson, Arizona, U.S.A. Error bars are standard errors of means.

43 Figure 4. 2012 Height (cm) of individual plant/row/table at time of last measure showing that the correlation between height and temperature favors a cooler temperature.

44 Appendix B: Distichlis palmeri

Distichlis palmeri: an endemic grass in the coastal

sabkhas of the northern Gulf of California and a potential new grain crop for saltwater agriculture in

Sabkha Ecosystems V: The Americas

© 2016 Elsevier Inc. All rights reserved.

Cylphine Bresdin, Edward P. Glenn

Abstract

Extensive coastal sabkhas in the northern Gulf of California in North America are colonized by Distichlis palmeri, an endemic grass that produces a grain that was harvested as a staple food by native Cocopah people. It has been considered as a potential perennial grain crop for salt water agriculture. Previous short- term trials have shown good vegetative growth but low percentage of flowering stems resulted in low grain yields. In these trials, we grew D. palmeri outdoors in paddy-style (flooded) conditions in 26-34 g -1L sea salt solutions. Reproductive maturity was reached four years after initial establishment of plants from seed, with nearly all stems producing male or female flowers. Mixtures of male and female plants (1:3) produced 231-310 g m-2 of grain, with nutritional content

45 similar to domesticated grains. These yields are within the range of other grain crops and demonstrate the potential for further developing D. palmeri as a global crop for salinized soils and water supplies.

I. Introduction

while sabkhas are usually associated with North Africa and the Middle

East, they occur on all continents except Antarctica (Yechieli and Wood, 2002).

An extensive set of coastal salt flats occurs in the Northern Gulf of California, where they are called esteros or negative estuaries (Brusca et al., 2006; Glenn et al., 2006). In this extreme desert environment (less than 100 mm yr-1 of rainfall) and with a tidal range of up to 7-9 m, esteros typically form at the mouths of rivers that are no longer connected to the sea. They are flooded and drained twice a day with hypersaline seawater (36-42 g L-1) through a network of tidal creeks bringing seawater as much as 10 km inland at high tide and exposing vast mudflats and saltflats at low tide (Figure 1). Occurring mainly along the

Sonoran (eastern) coastline above 28o N, esteros extend from the city of Guaymas to the mouth of the Colorado River and occupy over 114,000 ha. Unlike normal estuaries with active river systems, these esteros often contain extensive salt flats and are saltier at their heads than at their mouths (Figure 1A), with vegetation confined to the margins of the tidal creeks.

The southernmost esteros, below 29o N, contain mangroves (Avicennia germinans, Rhizophora mangle, Laguncularia racemosa and Conocarpus erectus) as well as 17 low-growing, halophytic shrubs, succulent forbs, and grasses (Glenn et

46 al., 2006). Above 29o N, 14 halophytes are commonly found in the esteros, four of which are endemic to the Sonoran Desert. The dominant species in the low zone of these esteros is an endemic grass, Distichlis palmeri (Palmer’s salt grass) known only from the northern Gulf of California (Felger, 2000, 2007). It was named nipa

(nee-pah) by the indigenous Cocopah people, who harvested its grain as a major food staple in summer. Distichlis is a genus of dioecious perennials with Kranz anatomy that predicts C4 photosynthesis. The leaves have bicellular microhairs that excrete salts, which commonly are seen on the leaf surfaces (e.g., Bell and

Columbus, 2008). The stems (culms) of D. palmeri produce terminal panicles.

Female panicles are usually 5-13.5 cm long; the spikelets break apart above the glumes and between the florets, and are 6-9-flowered, with the terminal one often a sterile rudiment. As with other grasses, the fruit of D. palmeri is called a caryopsis and at maturity it is called a grain. In naturally occurring populations most or often nearly all stems become synchronously reproductive, with flowering occurring in March and April and grain ripening in May (Pearlstein et al., 2012). In the esteros, male and female plants grow in mixed stands with approximately equal numbers of males and females.

D. palmeri is distinguished from common saltgrass (D. spicata), from which it has been derived, by the large size of its grain. While D. spicata caryopses are minute, the caryopsis of D. palmeri weighs about 10 mg and is the size of a rice grain (Pearlstein et al., 2012). Because they are borne on terminal panicles like wheat, the grains are easily harvested and processed, which is why they

47 were a major food source for the Cocopah. D. palmeri might have evolved from

D. spicata in response to the annual summer floods, from snow melt in the

Rocky Mountains that entered the northern Gulf of California in the Colorado

River. D. palmeri grain germinates readily in brackish water and it was able to colonize large areas of shifting mudflats in the Colorado River delta, formed from sediments carried in the river. These floods were curtailed by construction of upstream dams and diversions of water for agriculture, and the river only occasionally reaches the sea today. Nevertheless, D. palmeri has persisted as the dominant halophyte in the Colorado River delta, forming clonal stands covering several thousand hectares along the banks of the river in its intertidal reach and on Montague Island at the mouth of the river (Glenn et al., 2006). It is also the dominant halophyte in the low intertidal zone of the esteros south of the

Colorado River delta, but above the mangrove line.

interest in D. palmeri as a modern grain crop for saltwater agriculture began in the 1970s when the idea emerged of developing new crops for saline soils and waters from wild halophytes (Felger, 1979, 2000, 2007). All our major grain crops are annual grasses, but a case has been made for developing energy- efficient non-tillage perennial grain crops (Glover et al., 2010; Van Tassel &

DeHaan, 2013). Not only is D. palmeri perennial but it also can be grown paddy- style like rice, as it has aerenchyma tissues that allow it to grow in permanently flooded conditions (Pearlstein et al., 2012). Furthermore, it is extremely salt tolerant, growing on seawater with salinities of 38-42 g L-1 under natural

48 conditions. Dense stands of D. palmeri on Montague Island yield an estimated

1250 kg ha-1 of grain (Pearlstein et al., 2012), overlapping the low end of cultivated grain crops. Several attempts have been made to introduce D. palmeri into cultivation as a grain crop. Yensen and Weber (1986, 1987) showed that D. palmeri grain had nutritional qualities similar to wheat and other grains. Yensen

(2006) patented several selected lines of D. palmeri and D. spicata for forage and grain production. However, in two years of large-scale field trials in Australia,

Leake (2004) reported vegetative growth but very low seed production.

Pearlstein et al. (2013) revived interest in D. palmeri by collecting new germplasm from the Colorado River delta and conducting greenhouse trials. The high salt tolerance, high biomass production under anaerobic soil conditions, and high nutritional value of the grains were confirmed. However, in two years of trials, grain production was very low. After two years, only 2% of stems of female plants became reproductive, compared to nearly 100% each year in the natural stands from which the grain was collected. It was speculated that D. palmeri might require several years of vegetative growth before reaching its full reproductive potential. The longest growth trials had only continued for two years. The present study reports grain yields and nutritional value of D. palmeri grown for six years in tubs flooded with saline water, simulating a paddy-style agronomic system. During outdoor trials, plants reached full reproductive potential and produced high yields of grain with high nutritional content, confirming the feasibility of developing D. palmeri as a perennial grain crop for

49 salinized soils and water supplies.

II. Materials and methods

Plant material originally sown by Pearlstein (2012) in November 2009 was allowed to mature in a greenhouse under flooded conditions at 10 g -1L TDS until

March 2012 when the plants were relocated outdoors (Environmental Research

Laboratory, Tucson, AZ). Plant stock was divided and planted into seven, 60 cm diameter, 60 cm deep pots (with drainage holes) containing a 2:1 ratio of sand to potting soil. Pots were set into a larger diameter, 60 cm deep tub filled with flood irrigation water of 10 g L-1 synthetic sea salt (Crystal Sea® Marinemix, Marine

Enterprises International, Baltimore, MD) consisting of 83% NaCl, 10% MgSO4,

3.5% CaCl2, 3.1% KCl and trace amounts of other salts (Figure 4F). Consumed and evaporated water was replenished with 10 g L-1 irrigation water for the first nine months. This allowed time for soil moisture and salinity to stabilize.

Irrigation was then changed to city water, approximately 0.7 g L-1, to make up for water lost to evaporation. After one year outdoors a small sample of plants were transplanted and moved back into the greenhouse and continued to receive the same irrigation regime. Concentration (g L-1) was measured in 2014 with a

Traceable probe (VWR® Traceable® Portable Conductivity Meter, model: 23226-

505) during the flowering season from mid-February to harvest in May. Probe calibration was with synthetic sea salt mix and NaCl solutions. Stem density was counted in two, 10 cm2 areas per pot, averaged and translated to total stems.

D. palmeri grain was harvested over two crop cycles in 2013 and 2014.

50 Crop 1 was harvested in May, 2013 by cutting all stems in each pot at the base.

Total female and male stems per pot were counted. The number of caryopses per spikelet and number of spikelets per stalk from five random stems from an all female pot were used to estimate grain yield. Crop 2, May 2014, was harvested by cutting only mature, grain containing panicles high on the stem. The number of caryopses per pot was calculated from the weight of spikelets per pot where grain was 66% of spikelet weight for hand-cleaned samples. Proximate analyses of grain and stems were conducted by Litchfield Analytical Services (Litchfield,

MI).

III. Results

Unlike previous trials, nearly all stems in each pot were reproductive in

2013 and 2014. One pot was 100% male plants while others were dominated by female plants (Table 1) to give an overall male to female ratio of approximately

1:3. Male flowers (Figure 3B) presented about seven days before female flowers

(Figure 3C). At anthesis, versatile stamens are a greenish cream and stigmas are purple giving the outdoor stand a purple cast. Flower presentation in the greenhouse precedes outdoor presentation by about ten days and stigmas appear before anthers form pollen. Females are less resilient than males when harvested by cutting at the base of the stem. Re-growth of females after Crop 1 harvest was slow and had not fully recovered by Crop 2 and is likely to be the reason for the lower harvest year 2. Stems with mature panicles should be cut high on the stem or at the base of the stalk as was done for Crop 2 harvest.

51 For Crop cycle 1, five random stems averaging 1 m tall from an all female pot were used to determine number of caryopsis per flower spike (Figure 3A,

3D, 3E). Number of spikelets per panicle ranged from 5 -7 (mean = 6, Std. = 1).

Number florets per spikelet ranged from 2-6 (mean = 4, Std. = 0.9). Number of caryopsis per floret was 1. Number of grain per panicle ranged from 18-30 (mean

= 23, Std. = 5). We had 2131 female stems in Crop 1which produced an estimated

49,000 mature grains or an average of 31,000 grains per square meter. Grain weight ranged from 6-14 mg (sample size = 42, mean = 10 mg, Std. = 2.3) A mean weight of 10 mg gives an estimated grain yield of 310 g m-2. Grain size ranged from 6-9 mm (sample size = 42, mean size = 8 mm, Std. = 0.9). Stem density in

Crop 2 was 4791 stems per m2. The mean number of caryopses in pots containing females was estimated at 6200, equivalent to 23,100 grains per m2 or 231 g m-2 including male plants.

Proximate analysis of D. palmeri grain (Table 2) showed 7-12.5% crude protein with an effective net energy between 60-75%, with low ash and sodium content, and high iron. Digestible carbohydrates were the main constituents of both the grain and the biomass (stems). Crude fiber was 38% of the stem material which was higher in calcium, sodium and magnesium than the grain.

Salinity of paddy water in the tub increased from 28 to 34 g L-1 over the flowering and grain filling season in 2014, due to consumptive water use by the plants and direct evaporation of water from the tub during high ambient spring temperature in Tucson, AZ (Figure 4).

52 IV. Discussion

D. palmeri is a perennial rhizomatous halophytic tidal marsh grass endemic to the mud flats, esteros of the Colorado River Delta. At low tide monoculture stands are left standing in pools of evaporating saltwater causing the plant available water to exceed the salinity of seawater. Pearlstein

(2012) reported that female stands in its natural environment were nearly all reproductive and had projected grain yields of 125 g m-2. Yensen (2006) reported field yields of 200-400 g -2m but did not give details, whereas Leake (2004) could not replicate those field yields within a two year time frame. When plants were grown paddy style in the greenhouse there was no flowering the first year and only 2% of stems were reproductive the second year. Plants were moved outdoors when they were 2.5 year old and maintained under saline paddy style agronomic conditions. Heavy flowering was first seen mid-February 2013 and resulted in a grain yield of 310 g m-2 followed by a yield of 231 g m-2 in 2014. Our results indicate that when cultivated under saline paddy style conditions, D. palmeri requires four years to mature. Thereafter it produces high grain yields with crude protein, digestible carbohydrates and other nutritional quality similar to traditional annual crops of: rice, wheat and quinoa (Table 2).

The major finding of this study is that when allowed to grow for several years, D. palmeri reaches it full reproductive potential, a characteristic common for perennial grasses under consideration as new grain crops (Glover et al., 2010).

Once established, perennial grains do not require replanting each year, offsetting

53 the disadvantage of having to wait several years for the first grain harvest.D. palmeri stem material can be cut for forage until reproductive maturity is reached, but cutting too close to the ground appears to result in slower re-growth with a potential decrease in grain yield. The grain yields obtained here, 231-310 g m-2, compare to estimates of global grain yields of 290 g m-2 (SD = 177) from 2000-2013

(Word Bank data at http://data.worldbank.org/indicator/AG.YLD.CREL.KG).

The yields in this experiment cannot be extrapolated directly to potential field yields because plants were hand-harvested and were not grown in closed-canopy conditions. However, results do confirm the potential of D. palmeri to produce high grain yields when allowed to reach full reproductive potential under saline paddy style irrigation. Tub salinities ranged from 26-34 g L-1, typical of global open-ocean salinities but lower than those in the esteros in the native range of D. palmeri indicating that this could be a true seawater crop.

In conclusion, Distichlis palmeri is a perennial salt water plant with high agricultural potential that flourishes under hydric soils in arid climates. It produces a large nutritious grain with yields similar to rice when cropped at a 1:3 (or less) male to female ratio. It is a prime candidate for domestication as a constructive use of waterlogged salinized lands as well as volume reduction of saline waste streams. Spikelets from both harvests have been donated to

Pronatura Noroeste for their use in Colorado River Delta re-vegetation work and with our ongoing expansion of stock; we will deposit spikelets with Native Seed

Search for preservation and continue to donate to esteros re-vegetation efforts.

54 V. Acknow ledgments

This research and manuscript partially fulfills requirements for the Degree of Doctor of Philosophy, University of Arizona.

55 V. References

Bell, H., Columbus, J. T., (2008). Proposal for an expanded Distichlis (Poaceae, Chloridoideae): Support from molecular, morphological and anatomical characters. Systematic Botany 33: 536-551. Briere, P. R., (2000). Playa, playa- lake, sabkha: Proposed definitions for old terms. Journal of Arid Environments 45: 1-7. Brusca, R. C., Cudney-Bueno, R., Moreno-Baez, M., (2006). Gulf of California Esteros and Estuaries: Analysis, State of Knowledge and Convervation Priority Recommendations. Final Report to the David and Lucile Packard Foundation, Arizona-Sonora Desert Museum, Tucson, AZ. Available on-line at: http://www. rickbrusca.com/http___www.rickbrusca.com_index.html/Blank_files/Packard%20 Final%20Report%20SMALL.pdf Castetter, E. F., Bell, W . H., (1951). Yuman Indian Agriculture: Primitive Subssistance on the Lower Colorado and Gila Rivers. University of New Mexico Press, Albuquerque. Felger, R. S., (1979). Ancient crops for the 21st century. In: G. A., Ritchie (Ed.), New Agricultural Crops, Westview Press, Boulder, CO, pp. 5-20. Felger, R. S., (2000). Flora of the Gran Desierto and Río Colorado of Northwestern Mexico. University of Arizona Press, Tucson. Felger, R. S., (2007). Living resources at the center of the Sonoran Desert: native American plant and animal utilization. In: R. S. Felger, B. Broyles (Eds.), Dry Borders: Great Natural Reserves of the Sonoran Desert. University of Utah Press, Salt Lake City, pp. 147-192. Glenn, E. P., Nagler, P. L., Brusca, R. C., Hinojosa-Huerta, O., (2006). Coastal wetlands of the northern Gulf of California: inventory and conservation status. Aquatic Conservation: Marine and Freshwater Ecosystems 16: 5-28. Glover, J. D., Reganold, J. P., Bell, L. W ., Borevitz, J., Brummer, E. C., Buckler, E. S., Cox, C. M., Cox, T. S., Crews, T. E., Culman, S. W ., DeHaan, L. R., Eriksson, D., Gill, B. S., Holland, J., Fu, F., Hulke, B. S., Ibrahim, A. M. H., Jackson, W ., Jones, S. S., Murray, S. C., Paterson, A. H., Ploschuk, E., Sacks, E. J., Snapp, S., Tao, D., Van Tassel, D., Wade, L. J., Wyse, D. L., Xu, Y., (2010). Increased food and ecosystem security via perennial grains. Science 328: 1638-1639. Leake, J., (2004). NyPa “W ild Wheat” Proving Trials, Final Report, NyPa Australia Limited, Adelaide, South Australia.

56 Pearlstein, S. L., Felger, R. S., Glenn, E. P., Harrington, J., Al-Ghanem, K. A., Nelson, S. G., (2012). Nipa (Distichlis palmeri): A perennial grain crop for saltwater irrigation. Journal of Arid Environments 82: 60-70. Van Tassel, David, DeHaan, Lee, (2013). W ild Plants to the Rescue. American Scientist, 101:3 218-223. Yechieli, Y., Wood, W . W ., (2002). Hydrogeologic processes in saline systems: playas, sabkhas, and saline lakes. Earth Science Reviews 58: 343-365. Yensen, S. B., Weber, C. W ., (1986). Composition of Distichlis palmeri grain, a saltgrass. Journal of Food Science 51: 1089-1090. Yensen, S. B., Weber, C. W ., (1987). Protein quality of Distichlis palmeri, a salt grass Nutrition Reports International 35: 863-872. Yensen, N. P., (2006). Halophytes uses for the twenty-first century. In: M.A. Khan (Eds.), Ecophysiology of High Salinity Tolerant Plants, Springer, New York, pp. 36-39.

57 VI. Supplemental Material

Table 1. Tabulation of number of female and male flower producing stems per pot for Crop 1. Pot #5 is not considered because it was used to resupply plant material for the greenhouse and the 35 g L-1 tests.

Crop 1, 2013 Crop 2, 2014

Pot# Female Male # Stems Grain (g) 1 508 0 1416 62.082 2 0 399 1065 0 3 463 29 1766 79.440 4 279 132 1270 44.382 6 529 5 1124 55.933 7 352 42 1065 69.548

Total 2131 607 7706 332.7

58 a Table 2. Proximate analysis of Distichlis palmeri grain and biomass from crops grown outdoors under paddy style agronomic conditions with 10 g L-1 synthetic sea water compared to values published (USDA nutritional database, http://ndb. nal.usda.gov/ndb) for traditional grains.

Constituent 2013 2014 USDA Nutrition Database grain grain biomass Quinoa Rice Wheat Crude protein (%) 7.22 12.54 6.04 14.12 7.94 13.68 Acid detergent fiber (%) 4.61 5.68 37.98 - - - Crude fiber (%) 3.68 4.54 30.38 7.00 3.50 - Digestible carbohydrates (%) 68.22 71.29 45.85 64.16 77.24 71.13 Ash (%) 1.06 2.38 4.29 - - - Fat (%) 1.68 1.33 1.26 6.07 2.92 2.47 Total digestible nutrients (%) 77.88 85.75 67.33 - - - Effective net energy (%) 67.27 75.11 56.81 - - - Digestible energy (Mcal/kg) 0.71 0.78 0.61 0.37 0.37 0.34 Phosphorus (%) 0.26 0.36 0.15 0.46 0.33 0.51 Calcium (%) 0.04 0.06 0.33 0.05 0.23 0.03 Potassium (%) 0.68 0.79 0.47 0.56 0.22 0.43 Sodium (%) 0.09 0.1 0.49 - - - Magnesium (%) 0.07 0.09 0.18 0.20 0.14 0.14 Iron mg kg-1 116 28 107 - - -

59 A

S

B

Figure 1. A) Esteros of the Colorado River Delta, U.S.A. are crucial areas where freshwater mingles with tidal saltwater. The image was taken in 2012, after a pilot channel had been dug to reunite the river and ocean after the 2010 earthquake had shifted topography. credit: Sonoran Institute/LightHawk. http:// www.lighthawk.org/what-we-do/blog/happy-accidents-and-restoring-colorado- river-delta. B) Esteros at high tide. source: http://sonoranjv.org/wp-content/ uploads/2014/01/610x250-delta-osvel-hinojosa.jpg.

60 Figure 2. Growth characteristics of D. palmeri at the Colorado River delta. Ben W ilder collecting specimens on Montague Island. Photo by Richard Felger, May 13, 2009.

61 Figure 3. Plant material from outdoor stock grown paddy-style under 26-34 g L-1. A) panicle B) male flower C) female flower D) spikelets of panicle E) caryopses F) outdoor paddy setup showing the light green algal mat three months after Crop 1 harvest in May 2013. Photos by Cylphine Bresdin

62 Paddy Water Concentration During Reproduction: Crop 2 40

35

30 ) ) 1 - 25

20

Paddy Water 15 Concentration (g L Concentration (g

10

5

0 65 72 79 86 93 100 107 114 121 128 135 142

Day-of-Year

Figure 4. Concentration of paddy water through the reproductive period from mid-February to harvest in mid-May 2014. The slight increase in salinity over the reproductive period may be due to decreased water volume as a result of increasing temperatures. Water volume was not measured.

63 Appendix C: Design Concept

Design Concept of a Reverse Osmosis Reject Irrigated

Landscape: Connecting Source to Sabkha in

Sabkha ecosystems V: The Americas © 2016 Elsevier Inc. All rights reserved.

Cylphine Bresdin, Margaret Livingston, Edward P. Glenn

Abstract

Feasibility studies in Arizona (U.S.A.) have determined that ocean delivery is a viable disposal option for saline waste water when sourced from near coastline regions. Use of open canals to transport waste water and use of evaporation ponds to reduce waste water volume are standard engineering practices.

Engineered designs tend to focus on practicality and efficiency without regard to principles of landscape ecology. The concept of a saline ecosystem with landscape pattern incorporated as the vehicle for an evapotranspiration induced sequence of ecotopes along a directional saline gradient is proposed. This model will serve as a constructive, ecologically-based method to reduce reverse osmosis concentrate waste volume while increasing salinity during transport from source

(RO facility) to sink (sabkha). In the process, biota is allowed to self-organize into marsh habitat and the system of pattern creates potential for plant and microbial

64 crops. Potential for research use of the ecosystem is illustrated in light of a conceptual plan for the Santa Clara Slough, located at the northern end of the Sea of Cortez in the Gulf of California.

I. Introduction

As humans use fresh water it becomes more saline and as the broad salinity level of fresh water increases, there is demand to desalinate and return potable water for human consumption. As the number of desalination operations increase, the volume of concentrated waste also increases, but desalination is not the only source of brine waste streams. According to Lefebvre and Moletta

(2006) the leather, textile, petroleum, agro-food, and chemical industries generate large volumes of saline waste. Industrial waste brine has been successfully used to irrigate ornamental landscapes and agricultural crops in arid regions of the Southwest (Gerhart, et al., 2006; Glenn, et al., 2009; Riley, et al., 1997). In his 2010 manuscript, van der Gaag proposes that we can responsibly deal with industrial waste brine by creation of remedial saline wetlands (van der Gagg, et al., 2010), but system salinity increases due to evaporation and consumptive water use by plants leading to increased soil salinity problems especially at inland locations. The movement of water from source to sink via an open canal system also subjects water to evaporation and a consequent increase in salinity.

Alternative ideas for management and disposal of saline water in central

Arizona were investigated during 2003-2006 by the Central Arizona Salinity

Study (CASS) (CASS, 2003, 2006) and the Central Arizona Salinity Interceptor

65 (CASI) (CASI, 2004). Both Arizona studies concluded that salts should be delivered to the ocean. Three routes of disposal for brine waste generated from the Yuma Desalting Plant in Yuma, Arizona were suggested in the reports: 1) gravity flow to the Salton Sea, 2) supplement flow to Cienega de Santa Clara, a natural saline wetland filtering agricultural runoff, or 3) transport via open canal to Puerto Penasco, Mexico for ocean delivery into the Sea of Cortez, which was determined to be the best option. Since concentration of brine waste from desalting source is 3-10 g L-1 Total Dissolved Solids (TDS) and concentration at ocean sink is 35 g L-1 (TDS), the issue is how to move waste volume from source in an ecologically sensitive manner so that it is delivered to sink in site context while providing ecosystem services. An ecological solution to this problem might be creation of coastal depressions akin to evaporation ponds in the intertidal zone, essentially creating synthetic tide pools or tidal marshes to dispose of brine waste in a manner confluent with natural tidal regimes (Yechieli & Wood, 2002).

Results of the CASI study and success of Cienega de Santa Clara (Baeza, et al.,

2013; Mexicano, et al., 2013) combined with the salinity gradient that underlies sequential cell structure of traditional salterns and the idea of saline wetlands functioning as concentrators was the thrust for design concept.

II. Methods / Approach

The goal of this work was to generate a conceptual design for a saline system; an ecological-based landscape along a directional gradient of increasing salinity which uses halophytic vegetation and sequential pools in accordance

66 with a salinity continuum (Silvestri, et al., 2005; Ungar, 1998) to reduce volume and increase salinity of waste brine through evapotranspiration (Figure 1).

It is assumed that correctly structured ecosystems will resort to self- organization or allogenic succession (Mitsch & Gosselink, 2007) where energy flows are causal in saline zonation patterns. Self-organization is defined here as, a lateral diffusion of biota through the landscape matrix creating heterogeneity in the ecosystem. Structural landscape ecology, the foundation of complex patterns in landscapes, will allow and encourage evolution of ecological process.

If landscape is the spatial pattern of the environment, ecology is the temporal processes. In his abstract, Franklin, (1993) communicates that matrix and an appropriate biotic system is the concern of ecological landscape design. He presents an argument that the role of the matrix is to maintain diversity; highly functional ecosystems are generally associated with high species diversity with ample connectivity among edged patches.

This work designs an idea of ecosystem structure where the principle of landscape design and environmental sensitivity suggests that site analysis be an investigation of contextual pattern designed into existing structure (Makhzoumi,

2000). In this sense, site location becomes important for this work. The project site is in the delta of the Colorado River in the Mexican state of Sonora, located in the arid southwest region of the North American continent south of Yuma, Arizona.

Sonora has a coastal desert environment and is separated from Baja California by the Salton Slough and Colorado River Delta at the northern boundary of the Gulf

67 of California (Figure 2).

The CASI proposed canal and route from Yuma to the Gulf of California is used as guidance to propose an alternative route for regional considerations and design location. Figure 2 shows a regional route which parallels the existing

MODE canal that feeds Cienega de Santa Clara until Highway 3, where it becomes the Bypass Canal that parallels the Highway until the southern end of the delta and enters Santa Clara Slough tidal basin; the sabkha will serve as sink.

This location was chosen to serve as sink due to its geologic relation to tidal flux, proximity to proposed canal, and existing infrastructure of an abandoned shrimp farm (Figure 3). This site serves as vehicle for design of an idea which employs ecotope pattern as basis for a serial evapotranspiration induced saline gradated ecosystem for sustainable disposal of brine waste.

Site Analysis (Figure 4): Abiotic components analyzed are substrate, topography and tidal flux. Basin soil is quaternary sediment, sand and clay which have been delivered onto the estuarine bed from river flow and tidal action. Encrusted salt layers residual in the basin are approximately 250-300 g L-1. Soil structure within the shrimp farm boundary is expected to be rich in nitrogenous nutrients and harbor dormant microbes. Topographic quality of Santa Clara Slough basin ranges from approximately minus one meter to six meters. Its eastern border is an escarpment which increases in grade from an elevation of four meters to fifty meters along a north-south transect in line with the Cerro Prieto fault (Nelson, et al., 2013). Elevation of the design site is

68 approximately four meters. It drops below one meter eight kilometers northward to the northwestern border where it abuts a low point in the slough. The southwestern border abuts a low point marine-side where there is potential for tidal interchange. Low point abutments are potential outflow locations. Tidal flux resulting in basin flushing occurs when tides exceed five meters, approximately nine times a year at monthly intervals.

The site has full solar exposure, an annual precipitation of fifty four millimeters and evaporation of two meters. The average humidity is forty four percent, average high temperature is 32°C and average low is 17°C.

Biotic components analyzed mainly consist of vegetation systems: Sonoran

Desert, intertidal saltgrass meadows dominated by Distichlis palmeri, brackish wetlands (Cienega de Santa Clara), and fresh water spring wetlands (El Doctor); a corridor vegetated with over twenty-nine wetland species which buffers slough and escarpment. Santa Clara Slough is a crusted mudflat with a mosaic of bacteria and a variety of shore birds; the Colorado River Delta is a major Pacific

Flyway stop-over for migrating birds (Hinojosa-Huerta, et al., 2013). For species detail the reader is referred to, Conservation Priorities in the Colorado River

Delta Mexico and the United States (Zamora-Arroyo, 2005) available at www. sonoran.org. Protected vaquita (an endemic species) habitat is located off shore in the northern Sea of Cortez.

III. Results / Design

Design implications and applications are focused on development of

69 flow between source and sink. When viewed at a fine grain, this situation becomes a saline system with a source of channeled waste brine, flow within the shrimp farm boundary, and sink to sabkha. Presence of pre-existing structure is advantageous because earthwork for gravity flow and containment from east to west has been done. Use of this engineered structure as a framework for a productive sequential eco-evaporator of brine waste and biosphere research would be a beneficial repurpose for human education and wildlife.

III.A. Concept

Premise of concept is to inflow channeled water from Yuma Desalting

Plant and flow it through a series of saline ecotopes that are spatially arranged to create a sequence of saline based communities with each community being more saline than previous (Figure 3C).

Ecotopes are in a heterogeneous linear pattern along a saline gradient.

W ithin the confines of shrimp farm boundary, the pathway folds back onto itself and displays some sinuosity prior to sink. Brackish ecotopes (~35 g L-1) are situated adjacent to marine-side low points to allow discharge of excess flow and potential ocean interchange of equivalent salinities during high tides.

This requires design to consider tidal marsh stratification with the most saline ecotope situated adjacent to Santa Clara Slough where high saline outflow into the basin can be flushed out by monthly high tides that flood the basin, mixing sediment then recedes carrying salts out to open ocean leaving lagoons that evaporate. Geological positioning and climactic conditions of Santa Clara tidal

70 basin defines it as a sabkha, a supratidal lagoon/playa where salt deposits (250-

300 g L-1) remain in contact with alluvial mud. Ecological design of the idea of a linear flow of saline ecotope pattern from source to sink is dependent on an appropriate natural saline sink; a sabkha. The system as a whole is dependent on supply of source water. Central Arizona Salinity Interceptor study reported an expected value of 26.5 million gallons per day (MGD) at a salinity of 4.5 g L-1 total dissolved solids. This is design inflow salinity. If a microbial pre-filter or combination of pre-filter and RO are expected, they should be implemented prior to landscape system design. Salinity input into the landscape system would be at a greater concentration following wetland and mechanical treatment. Calculation to quantity water loss required within each successive ecotope is based upon difference of salinity inflow and desired outflow per ecotope. Area necessary to evapotranspire determined quantity of water can be calculated from known variables. Creative aspect of design is embedded in ecotope pattern; placement and proportion of units: size and shape of areas, depth of open water versus vegetation, and chromomorphic behavior of microbes.

III.B. Calculations

Premise of calculations was a simple sequential mass balance model built with Microsoft Excel and was based upon chosen salinity range spanned per ecotope. Hence, area of each ecotope required to increase salt concentration from inflow to outflow for each salinity range was mathematically deduced using equation 5 (Table 1 and Figure 6). Outflow salinity of preceding ecotope equals

71 inflow salinity of next ecotope and volume of water out of preceding ecotope equals volume of water into next ecotope (Figure 5). This is the basis of sequence in the saline ecosystem pattern. Exploration of possible salinity ranges asw the variable used to control area required per ecotope so that total area dedicated to ecotope pattern space is less than total area of project site which allows for human and mechanical space.

Assumptions were: 1) there is no water loss during canal transport, 2) there is no biotic uptake of salt, 3) there is no water loss due to biotic retention,

4) all water consumed by vegetation is transpired, 5) total water consumption equals pan evaporation: 8.268 af y-1 (acre feet per year), and 6) calculations are per annum and based on CASI reported values of flow and salinity.

Site situation of the abandoned shrimp farm allows for perimeter expansion or reduction when accommodation of volume variability is needed.

Calculated values for monthly fluctuations are reported at the end of Appendix

C. They will need to be considered in future design development. Future construction calculations which address assumption 2 and 3 may need to consider zonal mechanism of plant tolerance: exclusion, sequestration and secretion. Secretion in the intertidal zone will cycle salts through a vegetation/ environment interface. Sequestration in the supratidal zone will remove a small amount of salt but release it upon decay. Detailed calculations should help enumerate balance between open pools and vegetation area to seed. Microbial water consumption is considered negligible for theoretical design related

72 calculations.

III.C. Design

Abiotic focus is on pattern of open pool network created by subtle elevation changes within the microtopography of each ecotope. Since design related abiotic pattern has a foundation in elevation, raised human circulation is included in abiotic pattern, not land use. Land use is dedicated to ecologically constructive consumption of saline water (Figure 7). It is understood that larger sinuous pools have islands to accommodate wildlife needs.

Elevation affects soil moisture; low equates to hydric soils, high equates to dry sandy soils that wick water upwards causing salts to accumulate in substrate pores. Consequently, soil salinities may be higher than open pool salinities which means that resident halo-xerophytes or psammophiles require a higher tolerance to salt induced stress than halo-hydrophytic plants because soil moisture level and amount of soil wicking will vary with seasonal variation of evapotranspiration. Annual per month mass balance spreadsheet results for the system are reported in Appendix C. Average MGD outflow derived from monthly calculations correlate closely with those derived by an annual approach. Figure 1 is the graphical representation of design elements under design restrictions of area, volume and salinity input. Seasonal variation in water consumption, outflow and resultant salinity define the designed ecosystem as a poikilohaline environment which will demand seasonally adaptive management.

73 Biotic focus of this work is pattern structure of a heterogeneous saline ecosystem for purpose of controlled volume reduction. Spatial relationship of area designation accomplished in a logical and environmentally sensitive manner does not require design specifics of each pattern block or ecotope. Generalities can be implemented. Design is not based upon intent to create wildlife habitat.

However, design is sensitive to needs of migratory birds. Ecotopes are treated as self-realization units, broad spectrum planting of vegetation and microbial inoculation can be warranted, but should be relinquished to self-organization under adaptive management techniques. Suggested biota in Table 2 is arranged by unit type, salinity range, then alphabetically by genus. Suggested microbial organisms, and their related salinity and light intensity are shown in Figure 8.

Ecotope 1 - 898.2 acre; 4.5-6 g L-1. Saline water enters an open pool just below surface level causing turbulence which will hasten evaporation. Deeper pools (1.37 m minimum) can serve as reservoirs, buffering against drought, increasing zonation and heterogeneity. W icking will cause soil salinity to rise beyond the salinity of pools. Grading will be required to direct water eastward so that it can gravity flow down-slope through the system.

Ecotope 2 - 898.2 acre; 6-9 g L-1. Similar to e1 but requires a fan shaped slope to spread water going into e3. Consequently, meandering streams are more plentiful and pools are flat bottomed.

Ecotope 3 - 718.6 acre; 9-15 g L-1. Vegetation pattern begins to shift from low to mid salinity. Pools are shallower (1 m minimum) and represent a higher

74 percentage of ecotope surface area. Water flow is diverted in two directions to accommodate system continuance and potential research volumes.

Ecotope 4 - 135.6 acre; 15-20 g L-1. Similar to e3, but area shape has begun to transform into geometric confines and pool to surface area has increased.

Attempt to control Salt Cedar’s invasive nature is accomplished through area of water greater than 1 m deep. Vegetation pattern is overlapped.

Ecotope 5 - 135.6 acre; 20-30 g L-1. Vegetation pattern has fully shifted from low to mid salinity. Reeds are no longer present. The upper salinity level of e5,

30 g L-1 equates to the upper level of brackish environments. Pools continue to increase in percentage of surface area as shape becomes more geometric.

Ecotope 6 - 108.5 acre; 30-50 g L-1. Contains sea water salinity and pools have become shallower (1 m maximum) and more geometric with a structure that resembles holding ponds of Aviero Salinas complex (Rodrigues, et al., 2011).

Ecotope 7 - 54.2 acre; 50-75 g L-1. Constitutes a broader salinity range at where vegetation can no longer survive. Microbial populations increase and begin to display chromomorphic behavior; diatoms appear yellowish brown, anaerobic bacteria can lend a purplish hue, and Dunaliella, if present, is green.

Pools have been conformed into straight sided cells.

Ecotope 8 - 27.1 acre; 75-100 g L-1. Fish are no longer present and brine shrimp is first seen. Pools are large rectilinear evaporation cells consuming total ecotope surface area. Water retention time becomes an engineering issue to address. Chromatic dynamism of microbes becomes a dominant visual feature.

75 Ecotope 9 - 27.1 acre; 100-150 g L-1. Highest salinity range as endpoint ecotope within the designed saline ecosystem. It supplies output flow into

Santa Clara tidal basin at a projected rate of 0.4 MGD. Chromatic dynamism of microbes remains a dominant visual feature.

Salinity increases as volume and area decrease, open pool depth decreases, diversity of vegetative form and distribution decreases, vegetative zonation becomes more distinct and microbial presence becomes prominent.

It is the view of the author that salinity range-appropriate seeding of biota can and should be initiated, but then left to an adaptive style of management that will allow self-organization to mature. It is anticipated that ecotope 1 will develop emerging marshes while ecotope 6 will develop into a submerging marsh. The site affords an offset dedicated area for experimental research where there is potential for tidal interaction. Since the idea design site lies within Gulf

Biosphere Reserve boundary, research could potentially be managed by that entity. Appropriate research is vast, some possibilities are: halophyte agriculture, nursery stock production for coastal rehabilitation, microbial farming for polymer production, solar pond and energy research, and ornithological research. Human circulation should be limited to confined pathways atop sturdy rock trestles built to withstand and drain tidal surges.

IV. Discussion

Based upon current knowledge of zonation pattern, its abiotic causes and plant stress coping mechanisms, an image of self-directed maturation could be

76 suggested. However, this requires knowledge of contour details and hydrologic regime within the site context. Design development to this detail is beyond the scope of this work. To accommodate flow dynamics, grading will be necessary.

Human investment is required for seasonal consideration and maintenance of ecological services. These include: wildlife habitat, halophytic agriculture for food, fiber or nursery stock, fish and brine shrimp production, or polymer extraction from microbes. Space for human related circulation and activities has been considered, and 12 acres have been allotted for the design. Four MGD in the salinity range of 10-30 g L-1 on an approximate 271 acre allotment satisfies research needs. Research ecotope 5 is located adjacent to research ecotope 4 and sea level to allow for 2.0 MGD leak-out of 30 g L-1 into tidal streams. CASI proposed solution was to flow 26.5 MGD of 4.5 g -1L directly into the Sea of

Cortez. Certainly there would be consequential biotic effects from a fresh aterw

(4.5 g L-1) plume alteration on the abiotic conditions of the sea (35 g L-1). 3274 acres was required to reduce 4.5 g L-1 saline waste from 26.5 MGD to the sum of

0.4 MGD at 150 g L-1 and 2.0 MGD at 30 g L-1. 0.4 MGD of 150 g L-1 gravity flows into Santa Clara tidal basin. This translates into: 0.4 MGD x 694.4 = 277.8 gallons per minute | 0.4 MGD x 1121 = 448.4 af y-1 delivered to the sabkha.

Primary goal of this work was to derive an environmentally compatible and feasible landscape approach to mitigate anthropogenic saline waste.

Volume is reduced from 26.5 MGD inflow to 2.4 MGD total outflow. Pathway of the flow through sequential ecotopes is divided into two directions which

77 provide an option for research applications and wildlife habitat. Along the main path, salinity (TDS) is increased from an input salinity of 4.5 g L-1 to a discharge of 150 g L-1 into a tidal basin for periodic tidal flushing. If research dedicated flow is not consumed in experimentation, it discharges 2 MGD of 30 g L-1 into bidirectional tidal streams. Design is sensitive to existing contextual pattern. Vegetation and biota are seeded in a manner that should be conducive to maturation of self-organization. Variety in pool structure results in variety of vegetation diversity (Soga, 2013) and zonation patterns. Varieties in zonation patterns provide diversity in unit associations; the greater the diversity, the greater the heterogeneity, the greater the heterogeneity, the greater the stability.

The system is designed to be dynamic and to establish its own equilibrium over time. The approach of using landscape pattern based upon salinity range allows for alterations in inflow and outflow salinity and volume, making this approach directly transferable to other shoreline disposal sites.

The proposed ecosystem was based upon annual data. Calculations used a pan evaporation of 8.268 af y-1, this exceeds customary Yuma area ETo of 6 af y-1.

Future design development will require extensive analysis of seasonal conditions and engineering aspects of hydrologic flow. Exact details of pool design and structure must be clarified and developed in accordance with substrate type, seasonal fluctuations, and available management practices. How will this saline waste water driven eco-evaporation system available for research and habitat be managed? What potential ecosystem services and goods could be harvested?

78 Future considerations should focus on design of a small nearby pilot system that can mature into a public road-side park accessible to visitors.

V. Conclusion

we have presented a concept to dispose of inland sourced anthropogenic saline waste water by use of landscape pattern and structure to enhance evapotranspiration and reduce volume in a contextual and environmentally sensitive manner with a favor toward self-design. An abandoned shrimp farm in the Colorado River Delta was repurposed into a tidal ecology with potential economic and research value as it was used to communicate general ecological design implications of pattern structure in relation to a sequential saline gradient.

Each salinity stop-point along a salinity continuum in a heterogeneous horizontal flow ecosystem serves as a node or ecotype of the saline continuum with a pre- existing sabkha as the sink and method to convey concentrated brine to the ocean.

VI. Acknow ledgements

Cylphine would like to acknowledge and thank Dr. Charles Moody at the

Yuma Desalting Plant for testing of the model. This manuscript partially fulfills requirements for the Degree of Doctor of Philosophy, University of Arizona and was originally published in its entirety as a Masters of Landscape Architecture thesis in 2013 from the College of Architecture, Planning and Landscape

Architecture at the University of Arizona; Committee Chair, Elizabeth Scott.

79 VII. References

Baeza, K., Lopez-Hoffman, L., Glenn, E. P., & Flessa, K., (2013). Salinity Limits of Vegetation in Cienega de Santa Clara, an Oligotrophic Marsh in the Delta of the Colorado River, Mexico: Implications for an Increase in Salinity. Ecological Engineering, 59: 157-166. CASI, U.S. Department of the Interior, (2004). Reverse Osmosis Treatment of Central Arizona. Report No. 36, Desalination Research and Development Program. CASS I, U.S. Bureau of Reclamation and the Sub-Regional Operating Group, (2003). Central Arizona Salinity Study Phase I Report. Phoenix. CASS II, U.S. Bureau of Reclamation and the Sub-Regional Operating Group, (2006). Central Arizona Salinity Study Phase II. Phoenix. Franklin, J. F., (1993). Preserving Biodiversity: Species, Ecosystems, or Landscapes? Ecological Applications, 3:2 202-205. Gerhart, V. J., Kane, R., & Glenn, E. P., (2006). Recycling Industrial Wastewater for Landscape Irrigation in a Desert Urban Area. Journal of Arid Environments, 67: 473-486. Glenn, E. P., McKeon, C., Gerhart, V., Nagler, P. L., Jordan, F., & Artiola, J., (2009). Deficit Irrigation of a Landscape for Reuse of Saline Waste Water in a Desert City. Landscape and Urban Planning, 89: 57-64. Hinojosa-Huerta, Osvel, Soto-Montoya, Eduardo, Gomez-Sapiens, Martha, Calvo-Fonseca, Alejandra, Guzman-Olachea, Ricardo, Burtron-Mendez, Juan, Burtron-Rodriguez, Jose Juan, Roman-Rodriguez, Martha, (2013). The Birds of the Cienega se Santa Clara, a wetland of international importance within the Colorado River Delta. Ecological Engineering, 59: 61-73. Lefebvre, O., & Moletta, R., (2006). Treatment of Organic Pollution in Industrial Saline Wastewater: a literature review. Water Research, 40: 3671-3682. Makhzoumi, J. M., (2000). Landscape Ecology as a Foundation for Landscape Architecture: application in Malta. Landscape and Urban Planning, 50: 167-177. Mexicano, Lourdes, Glenn, Edward P., Hinojosa-Huerta, Osvel, Garcia- Hernandez, Jaqueline, Flessa, karl, Hinojosa-Corona, Alejandro, (2013). Long- term sustainability of the hydrology and vegetation of Cienega de Santa Clara, an anthropogenic wetland created by disposal of agricultural drain water in the delta of the Colorado River, Mexico. Ecological Engineering, 59: 111-120. Mitsch, W . L., & Gosselink, J. G., (2007). Wetlands. Hoboken: John W iley & Sons.

80 Mirjam Foti, Dimitry Y. Sorokin, Bart Lomans, Marc Mussman, Elena E. Zacharova, Nikolay V. Pimenov, J. Gijs Kuenen, and Gerard Muyzer, (2007). Diversity, Activity, and Abundance of Sulfate-Reducing Bacteria in Saline and Hypersaline Soda Lakes. Applied and Environmental Microbiology, 73:7 2093- 2100. Nelson, Steven M., Fielding, Eric J., Zamora-Arroyo, Francisco, Flessa, Karl, (2013). Delta dynamics: Effects of a major earthquake, tides, and river flows on Cienega de Santa Clara and the Colorado River Delta, Mexico. Ecological Engineering, 59: 144-156. Riley, J. J., Fitzsimmons, K. M., & Glenn, E. P., (1997). Halophyte Irrigation: an Overlooked Strategy for Management of Membrane Filtration Concentrate. Desalination, 110: 197-211. Rodrigues, C. M., Bio, A., Amat, F., & Vieira, N., (2011). Artisinal Salt Production in Aveiro/Portugal - an Ecofriendly Process. DOI:10.1186/1746-1448-7-3, Retrieved from Saline Systems: http://www.salinesystems.org. Silvestri, S., Defina, A., & Marani, M., (2005). Tidal regime, salinity and salt marsh plant zonation. Estuarine Coastal and Shelf Science, 62: 119-130. Soga, Masashi, Ishiyama, Nobuo, Sueyoshi, Masanao, Yamaura, Yuichi, Hayashida, Kazufumi, Koizumi, Itsuro, Negishi, Junjiro, N., (2013). Interaction Between Patch area and Shape: Lakes with Different Formation Processes have Contrasting Area and Shape Effects on Macrophyte Diversity. Landscape and Ecological Engineering, 10 pp. DOI 10.10007/s11355-013-0216-9. Ungar, I. A., (1998, April 1). Are Biotic Factors Significant in Influencing the Distribution of Halophytes in Saline Habitats? The Botanical Review. van der Gagg, J. J., Paulissen, M., & Slim, P. A., (2010). Halophyte Filters as Saline Treatment Wetlands applications and constraints. Wageningen, The Netherlands: Alterra. Yechieli, Y. & Wood, W . W ., (2002). Hydrologic Processes in Saline Systems: playas, sabkhas, and saline lakes. Earth-Science Reviews, 58: 343-365. Zamora-Arroyo, Francisco, Jennifer Pitt, Steve Cornelius, Edward Glenn, Osvel Hinojosa-Huerta, Marcia Moreno, Jaqueline García, Pamela Nagler,Meredith de la Garza, and Iván Parra, (2005). Conservation Priorities in the Colorado River Delta, Mexico and the United States. Prepared by the Sonoran Institute, Environmental Defense, University of Arizona, Pronatura Noroeste Dirección de Conservación Sonora, Centro de Investigación en Alimentación y Desarrollo, and World W ildlife Fund—Gulf of California Program. 103pp.

81 VI. Supplemental Material

Table 1. Design of excel based model provides definitions, units and equations used behind the interface of the model shown in Figure 6.

Equation Variable Units Constants ETo = pan evaporation af y-1 ETo = 8.268 af y-1 af y-1 = acre feet per year af y-1 = MGD x 1121 Asite = area of site acre ATe = total ecotope area acre Ar = area remaining acre Si = salinity inflow g L-1 So = salinity outflow g L-1 W i = volume water in MGD Eq 1 W i x (Si x So) Wo = volume water out MGD Eq 2 Wi - Wo Wc = volume water consumed MGD Eq 3 Wc x ET o Ae = area needed per ecotope acre

Eq 4 Ar = Asite - ATe

Eq 5 Ae = (W i -(W i x (Si x So))) x 1121 x ETo

Eq 6 ATe = ∑Ae < Asite

82 Table 2. Short list of suggested species and vegetation forms for seeding or inoculation of ecotope ranges; a: shrub, b: grass, c: semi-succulent prostrate, d: reed.

83 Design Units

Surface Area

Saline Water

Vegetation

Microbes

Figure 1. Graphical representation depicting area versus salinity relationships of major design units constituting the concept. As salinity increases, surface area and area of water and vegetation decrease in a linear fashion. There is minimal increase in surface area of halophilic microbes with a steep increase in system salinity.

84 Figure 2. Regional location of project site located at the southern end of the Santa Clara Slough on the eastern side of the Colorado River Delta, just north of El Golfo, Mexico. A surface conveyance canal from Yuma, AZ, shown as a solid grey line parallel to the pre-existing Bypass Canal, that carries farm runoff to Cienega de Santa Clara, turns and follows Highway 3 south to the abandoned shrimp farm indicated by the backward letter F.

85 Figure 3. A) Imagery of the local environment showing physical relationship of the shrimp farm boundary to Santa Clara Slough, Cienega de Santa Clara and conveyance canal. B) Simple schematic that shows ecotope alignment from input canal to outputs; ocean and basin.

86 Figure 4. A) Land use map of existing conditions shows that the project site is located within the Upper Gulf Biosphere Reserve and that the Pinacate Biosphere Reserve is adjacent. Fishing and Agriculture dominate the economy of the region. B) Vegetation map shows that the project site is west of desert and abutted to tidal marsh and down-grade from spring wetland habitat (El Doctor). Elevation drops from 4 m at inflow (a) to -1m at outflow (b) relative to sea level. C) Ecosystem map shows that desert dominates the region and the project site is ocean-side between desert and sabkha where ecotourism has a viable potential.

87 Figure 5. Schematic of flow through the system showing how output of a previous ecotope become the input of the next ecotope, hence, the process of serial flow where water is continually consumed in sequential fashion.

Figure 6. Screen snip of the excel model interface. Outlined cells are user input cells used to calculate surface area required per ecotope based upon assigned salinity range. Area values are dependent on: total system area, input water volume and salinity, and evapotranspiration rate.

88 Figure 7. A) Diagram of basic landscape and process concept where input water flows sinuously down-grade through the system to outflow points. The table shows salinity range and area per ecotope of the system and indicates that ocean-side outflow is at brackish seawater salinity while outflow into sabkha is hypersaline at greater than 150 g L-1. B) Diagram of potential layout of ponds indicating spatial relationship of size to ecotope area. As ecotope salinity increases, ecotope area, and single pond area and depth decrease while ponds become the dominant landscape unit. C) Biotic concept plan uses a general approach to salinity tolerance mirroring that used in horticulture because the system, as a whole, is intended to be self-organizing.

89 g L-1 Light Nutrient >250 high low >200 low high 100 - 200 low low <150 any medium 50 - 120 any high

Figure 8. Potential halophilic microbes that can be seeded at specified salinities, light intensities and nutrient levels are suggestions to get the system started, but since microbes are ubiquitous, self-organization will occur, regardless. Mirjam et al., (2007) reported that some strains of sulfur reducing bacteria are found at salinities exceeding 450 g L-1 under sodic conditions.

90 Monthly Tide Values o 26.28 26.28 -59.71 W 808.46 846.54 808.46 777.99 482.40 133.74 133.74 723.48 673.43 723.48 673.43 633.37 435.58 357.49 435.58 357.49 295.02 762.77 248.20 232.60 263.81 755.16 613.36 747.55 603.36 593.35 846.54 -119.34 -145.62 -124.14 -102.66 1958.33 1441.11 1547.09 1051.89 1763.72 1027.32 2143.96 1812.37 1971.05 2223.30 1769.25 590.09 1556.01 1266.78 7.61 7.61 c 21.48 38.08 21.48 38.08 38.08 30.47 85.99 50.06 42.95 50.06 50.06 50.06 40.05 78.08 78.08 78.08 78.08 62.48 15.22 15.61 15.61 31.21 20.01 10.00 10.00 38.08 W 517.21 517.21 265.28 711.82 711.82 569.49 107.46 107.46 107.46 107.46 413.79 331.59 331.59 252.24 252.24 201.81 Total outflow = Total outflow = Total outflow = Total outflow = in 26.28 -59.71 846.54 884.62 846.54 808.46 773.54 241.20 133.74 241.20 133.74 723.48 773.54 723.48 673.43 513.66 435.58 513.66 435.58 357.49 777.99 263.81 248.20 295.02 633.37 762.77 755.16 613.36 603.36 884.62 W -124.14 -102.66 2475.54 1958.33 1812.37 1763.72 2475.54 1051.89 1441.11 2475.54 2143.96 2223.30 2475.54 1971.05 e1 e5 e9 Ecotope e2 Re4 e3 Re5 e6 e2 e1 e8 e4 e3 e4 e5 Re4 Re5 e6 e7 e5 Re4 Re5 e6 e3 e4 e5 Re4 Re5 e6 e7 e8 e9 e1 e7 e2 e7 e8 e9 e2 e8 e9 e1 e3 e4 Oct Sep Dec Nov 3.37 6.91 4.43 9.51 7.49 6.98 7.74 7.16 -8.55 -9.91 -8.55 -8.18 -8.96 -7.80 -7.47 -7.17 o 22.24 15.55 27.70 17.48 S -17.05 -11.81 -29.65 -26.08 -17.05 -12.73 -12.52 -15.05 -44.10 -21.50 -44.10 -21.50 -15.25 -13.31 -10.58 -15.05 -10.58 -38.56 -18.98 -12.59 -18.98 -12.59 -11.30 -10.69 -10.15 -29.65 861.06 -113.51 o 12.94 -98.14 501.00 716.57 402.20 637.23 W -730.54 -326.60 -471.46 -187.83 -427.19 -326.60 -437.63 -444.93 -370.10 -126.31 -259.08 -126.31 -259.08 -365.32 -418.39 -526.60 -526.60 -370.10 -651.83 -288.86 -293.48 -293.48 -442.53 -442.53 -561.78 -493.10 -651.15 -680.93 -621.36 -714.39 -520.83 -548.56 -875.16 -745.66 -776.94 -187.83 1488.27 1596.05 1438.87 1556.38 -1123.46 -1303.55 c 26.54 26.54 53.07 55.46 29.79 29.79 59.57 62.56 27.73 27.73 31.28 31.28 W 987.27 138.76 138.76 987.27 829.38 138.76 111.03 879.49 879.49 156.51 132.78 132.78 703.63 132.78 132.78 106.24 156.51 156.51 156.51 125.23 789.86 149.05 149.05 149.05 149.05 119.26 919.16 735.37 919.16 138.76 1036.67 1036.67 Total outflow = Total outflow = Total outflow = Total outflow = 6.47 6.47 in -49.07 -49.07 402.20 716.57 501.00 637.23 W -444.93 -187.83 -418.39 -187.83 -326.60 -365.32 -213.59 -437.63 -126.31 -126.31 -259.08 -370.10 -621.36 -651.15 -370.10 -213.59 -526.60 -144.43 -144.43 -293.48 -293.48 -442.53 -561.78 -651.83 -493.10 -520.83 -714.39 -745.66 2475.54 1488.27 1596.05 2475.54 2475.54 1438.87 1556.38 2475.54 e1 e9 e3 e5 e8 Ecotope Re4 e2 Re5 e6 e3 e2 e1 e7 e3 e4 e7 e4 e5 Re4 Re5 e6 Re5 e8 e9 e5 Re4 e6 Re4 e4 e5 Re5 e6 e7 e1 e2 e7 e8 e9 e8 e3 e9 e2 e1 e4 July Aug May June 13.85 12.28 11.75 13.19 5.18 7.77 7.68 6.11 6.46 5.74 5.05 7.60 7.13 6.77 7.13 6.77 7.13 7.43 9.08 9.23 7.61 7.61 8.16 8.16 8.66 8.94 5.67 7.65 6.23 o 10.63 80.04 32.16 80.04 12.46 15.04 15.04 12.46 18.05 20.04 21.22 20.12 22.54 10.10 32.16 S -73.82 -57.92 -418.08 -101.75 o 69.59 69.59 -13.32 -54.74 -75.45 -96.16 -26.57 W 717.00 725.09 173.21 733.17 447.19 822.16 781.71 822.16 781.71 749.34 370.23 613.38 603.54 370.23 447.19 308.66 731.61 731.61 682.34 682.34 642.92 623.23 277.90 262.52 553.66 247.14 173.21 2149.20 1822.85 1048.28 1725.23 1939.62 2207.58 1561.76 1965.81 1456.08 1789.17 1102.79 1498.709 1285.879 617.38 8.08 8.08 9.85 9.85 c 82.91 16.17 76.95 41.42 40.45 40.45 40.45 40.45 32.37 76.95 76.95 76.95 61.57 49.27 49.27 49.27 49.27 39.42 19.69 30.76 20.71 20.71 15.38 15.38 W 326.35 103.62 326.35 407.81 103.62 214.38 267.96 267.96 261.09 103.62 509.73 509.73 549.13 686.37 686.37 103.62 Total outflow = Total outflow = Total outflow = Total outflow = in 69.59 -13.32 -54.74 -75.45 725.09 276.83 173.21 749.34 733.17 524.14 173.21 862.62 822.16 862.62 822.16 781.71 447.19 447.19 524.14 370.23 780.88 780.88 731.61 731.61 682.34 642.92 623.23 613.38 308.66 277.90 262.52 276.83 W 2475.54 2149.20 1456.08 1939.62 2207.58 2475.54 1822.85 2475.54 1965.81 1102.79 2475.54 1789.17 e1 e9 Re4 e2 e3 e5 e3 e2 e1 e7 e8 e3 e4 Ecotope Re5 e6 e7 e4 e5 Re4 Re5 e6 Re5 e5 Re4 e6 Re4 e4 e5 Re5 e6 e7 e8 e9 e1 e2 e7 e8 e9 e8 e3 e9 e1 e2 e4 Jan Feb 3.58 4.36 9.17 6.81 April March A 27.1 54.2 27.1 718.6 898.2 898.2 135.6 135.6 135.6 135.6 108.5 in S 4.5 in W 26.5 o o o S S e9 Et 7.1 7.8 9.2 8.2 8.3 6.9 9.4 7.5 -8.2 -7.2 8.27 0.28 0.57 0.36 0.30 0.37 0.38 0.76 0.98 1.10 1.15 1.02 0.79 0.58 22.5 80.0 15.0 15.6 23.9 -57.9 -11.8 -10.6 -21.5 -12.6 -17.1 -10.2 -38.3 212.0 Re5 1.98 (MGM) (MGM) o o W W 3.37 6.81 4.36 3.58 255.22 234.10 4.43 9.17 11.75 13.19 13.85 12.28 9.51 6.91 197.05 222.78 -31.40 219.87 263.96 -153.93 -222.32 80.69 22.72 -171.93 -84.59 -144.48 120.88 116.72 8.58 -106.63 -253.67 75.94 193.73 -179.10 244.07 -47.54 W @ af/month Average Jul Jul Jul Jan Jan Jan Jun Jun Jun Apr Oct Apr Apr Oct Oct Feb Sep Feb Feb Sep Sep Dec Aug Dec Aug Aug Dec Nov Nov Nov Mar Mar Mar May May May Month output Re5 output Basin output Average MGD MGD Average Monthly Tide Values Tide Monthly

91