The Use of Cyanobacteria Biofertilizers to Increase Crop Productivity, Improve Soil Health, and Agricultural Sustainability in Florida

Figure 3 Figure 4 Figure 5

Support letter from AECOM

Approaches and Methods:

The USACE has an ongoing project (Harmful Algal Bloom Interception, Treatment, and Transformation Systems; HABITATS) in collaboration with AECOM (an International company; website: https://aecom.com/) to remove cyanobacterial blooms from Lake Okeechobee and adjacent canals. The USACE and AECOM agreed to assist in collecting cyanobacteria from the Lake and a support letter from Mr. Dan Levy (Vice president, AECOM) is attached along with this proposal. Cyanobacterial scums will be collected using ‘float it up and skim it off’(micro-flotation) technique. An overview of cyanobacteria collection, processing, storage, field application, and extension outreach is presented in Figure 3. Detailed description of collection and processing are discussed below:

1.Cyanobacteria collection (AECOM) and processing:

The AECOM will deploy a custom-built, mobile algae harvesting unit (Figure 4) that separates algae biomass from water. Even a small-scale harvester can process between 100-175 gallons per minute and produces up to 1,000 gallons of algae biomass in 8 hours. Skimmers will capture thick mats of algae floating on the water’s surface. Collected algae biomass slurry will be dewatered to produce semi-solid mats. Composite cyanobacterial mats will be collected from various locations of the Lake and adjacent river area. Collected cyanobacteria mats will be transported to FIU greenhouse for drying and processing. Dried and ground (in powder form) cyanobacteria biofertilizer will be stored in colored plastic bags for field application. Storage of the dried cyanobacteria biofertilizer in colored plastic bags can increase its shelf life (Jha and Prasad, 2006). In this phase we will invite our agroecology students to experience cyanobacteria collection and processing for biofertilizer preparation.

A preliminary study was conducted by our research team where we (a) collected cyanobacteria samples (mats; n = 8) from Lake Okeechobee, (b) dried those samples at 45⁰C for 72 hours, and (c) ground those dried samples using mortar and pestle (Figure 5A). Commercial grinding machine (Model# BL-230) will be used to process the biofertilizer required for greenhouse study and field trials.

We found the dry matter content of the cyanobacterial mat was 17 to 19%.   

UPDATE - Annual report (2022)

Cyanobacteria collection and processing

Figure 2022-1 Figure 2022-2 Figure 2022-3

As we mentioned in our statement of work (SOW), we communicated several times with the AECOM company, who helped us collecting the cyanobacteria from Lake Jesup (28⁰43’N, 81⁰13’W) located in Central Florida. We also visited the floating cyanobacteria collection station of AECOM company (Figure 2022-1) and learned more about the working principles of the unit. The custom-built, mobile algae harvesting unit of AECOM was separating algae biomass from the lake Jesup water. Generally, a small-scale harvester can process between 100-175 gallons per minute and produces up to 1,000 gallons of algae biomass in 8 hours. Cyanobacterial slurries were collected using ‘float it up and skim it off’(micro-flotation) technique and later coagulated in the unit. Cyanobacterial slurry was collected in five-gallon buckets (Figure 2022-2) for easy transporting and brought back to the FIU campus for further processing. One small glass jar of sample was brought in ice cooler and stored at -20⁰C at FIU laboratory for microcystin analysis. This sample will give us the microcystin concentration of the cyanobacteria sample immediately after harvesting from the lake. At FIU campus, the rest of the slurry was then air dried (after spreading on large plastic trays) in the greenhouse for 3 to 4 weeks (Figure 2022-3) to obtain dried cyanobacteria materials. Slurry on the plastic trays were stirred up occasionally to expedite the drying process. Dried cyanobacteria then powdered using a mechanical grinder and stored in colored plastic bags for future applications.

Update 2024 - Scaled cyanobacteria processing approach for field trials

To prepare for our field trials, we sought a means to dry the remaining 200 gal ( 757 L) of cyanobacteria slurry collected from Lake Jesop on a larger scale. We arranged a shallow pool of approximately 2.4m x 4.6m x 0.3m) long using standard cinder blocks. The blocks were lined with a plastic pond liner to keep slurry from leaking. The entire pool was placed beneath a high tunnel with an impermeable cover to prevent rain from entering the pool (Figure 2024-1A). Approximately 100 gal (378 L) were added per drying batch, keeping the slurry level at about 4 cm high. After about 2 weeks, the slurry is dry enough for collection and has lost about 80% of it’s mass from water evaporation. At that point the remaining slurry is easier to condense and collect. We dried the remaining cyanobacteria in an oven overnight at 70°C to dry off any remaining moisture. Dried cyanobacteria was then grinded in a designated blender to process into a fine biofertilizer powder. We were able to transform the aproximate 200 gal of slurry into 15.9 Kg of biofertilizer powder ready to be applied to fields.

2. Timeline development for safe use of the biofertilizer:

This project also aims to develop the timeline for safe use of the cyanobacteria biofertilizer produced in the field for crop production. As commented by one of the reviewers about the safety concern of the biofertilizer, we will measure the microcystin, a common cyanotoxin found in cyanobacteria during first project year.  

Microcystin content of cyanobacteria will be measured following the USEPA method 546 (ADDA- Enzyme-Linked Immunosorbent Assay technique) from collection time (T0) till 50 days after collection (T50) at 10 days interval to develop a decay rate curve (Birbeck et al., 2019). Krishnaswamy Jayachandran (Co-PI of this project) and colleagues conducted an experiment on cyanobacteria samples collected from Lake Okeechobee to measure microcystin degradation over time (Ramani et al., 2012). They found that more than 75% of the cyanobacteria microcystin degraded within 3-weeks after harvesting from the Lake. Cyanobacteria powder do not cause airborne problems or human health issues. However, we will take proper pre-cautions (gloves, masks) during biofertilizer processing and field application.

We will conduct safety training and knowledge dissemination workshops for growers and the farm workers for safe and better handling of the biofertilizer.

3.Objective # 1: Nutrient and chemical analyses of cyanobacteria biofertilizer

Subsamples of cyanobacteria biofertilizer will be analyzed at the FIU Soil-Plant Microbiology Laboratory for nutrient (macro and micro) contents and other chemical properties including OM content, pH, and cation exchange capacity (CEC). The pH of the biofertilizer will be measured in 2:1 (water: biofertilizer) slurry using ‘Denver Instrument’ pH meter. Cation exchange capacity (CEC) of the biofertilizer will be measured by sodium acetate extraction following method USEPA 9081. A subsample of cyanobacteria will be combusted in 600⁰C for 6 hours in Fisher Scientific isotemp muffle furnace to obtain organic matter content of the biofertilizer. Total C and N contents will be analyzed in a Thermo-Scientific Flash EA (Dumas combustion method). For multi-element analysis (macro and micro elements), pre-weighed cyanobacteria samples will be combusted at 550°C and then extracted using concentrated hydrochloric acid (HCl). The extracted samples will be analyzed in ICP-OES (Perkin Elmer Optima 8300). Total P will be extracted by combusting samples in the presence of MgSO₄ at 550°C and solubilizing the ash with 0.6 N HCl to convert all P to ortho-P forms followed by colorimetric analysis (antimony-phospho-molybdate complex) in TechniconTM AutoanalyzerTM Sampler IV.  

Agroecology graduate students and student interns will be involved to gain experimental and experiential learning about chemical analyses of cyanobacteria biofertilizer in the laboratory.

Preliminary results (fertilizer grade): Physicochemical parameters of processed cyanobacteria biofertilizer samples were analyzed at the extension soil testing laboratory of the University of Florida (Table 1).

Table 1. Nutritional properties of cyanobacteria biofertilizer used for pilot experiment. Values are presented as mean ± SD of n = 8 samples.

The C:N ratio (mol:mol) of test cyanobacteria biofertilizer averaged 12.7:1, lower than other organic nutrient sources. Therefore, we expect increased N mineralization in soil when cyanobacteria biofertilizer is applied. Another important component was Fe content (average 2005 ppm) in our biofertilizer mix. South Florida soils generally lack Fe because of low SOM content, high temperature, and high soil pH (Ozores-Hampton, 2013). Vegetable farmers often face problems due to Fe deficiency in these soils (Liu et al., 2012). Our collaborative growers informed us of the need to apply chelated Fe (an expensive option) to meet the crop requirements. Thus, cyanobacteria biofertilizer would serve as an excellent source of Fe and other micronutrients. Visual observation from preliminary experiments indicate that the interveinal chlorosis of young okra leaves in control plots possibly resulted from Fe deficiency (Figure 5B). Leaf samples were collected and submitted for micronutrient analyses (analysis in progress at the time of this proposal submission).

Typical N% for commonly used organic fertilizers in South Florida such as poultry manure is 1.5 to 2.3% (Hochmuth et al., 2009). Our test cyanobacterial biofertilizer falls within this range. However, cyanobacteria biofertilizer provides micronutrients and offer ecosystem services which makes it a high potential biofertilizer for vegetable production.  

4.Greenhouse experiment at FIU:

Greenhouse experiment will be conducted during first project year (2021-2022) to evaluate the effect of cyanobacteria biofertilizer on crop growth and soil properties (pH, EC, total P, total N, total C content, OM content, aggregate stability, and nutrient loss). Representative South Florida soils will be collected from LNB groves farm, Homestead, FL (courtesy of Mr. Marc Ellenby, our collaborative grower) for this greenhouse study. Seven-gallon and five-gallon pots will be used to grow tomato (Lycopersicum spp variety: BHN 975) and okra (Abelmoschus esculentus variety: Cajun Delight), respectively. These tomato and okra varieties perform well in Florida’s climate and are highly resistant to insect pests and were selected after discussing with Dr. Qingren Wang (extension specialist for commercial vegetables; our team member) and Mr. Marc Ellenby. Fertilizer application rates will be 224 kg N ha^-1, 112 kg P₂O₅ ha^-1, and 168 kg K₂O ha^-1 for tomato, and 135 kg N ha^-1, 112 kg P₂O₅ ha^-1, and 112 kg K₂O ha^-1 for okra. Biofertilizers will be applied along with synthetic fertilizers in different combinations. Seven treatments combinations [T0: control (no fertilizer applied); T1: 100% biofertilizer; T2: 75% biofertilizer + 25% synthetic; T3: 50% biofertilizer + 50% synthetic; T4: 25% biofertilizer + 75 % synthetic; T5: only 50% synthetic fertilizer and no biofertilizer; and T6: 100% synthetic fertilizer] will be assigned in a randomized complete block design (RCBD) with 20 replications (5 main plot x 4 within plot = 20) per treatment.

Because of the highly reactive nature of N fertilizers and the porous sandy soils of Florida, the N fertilizer will be applied in two splits. Urea (H₂N-CO-NH₂; 46-0-0) will be used as the primary source of inorganic N. Triple super phosphate [Ca(H₂PO₄)₂. H₂O; 0-45-0] and muriate of potash (KCl; 0-0-62) will be used as inorganic source of P and K, respectively. We will conduct demonstration training on shade house experiments to the local farmers and students during organic farming workshops and field days.

Preliminary results (pot experiment): We arranged the pot experiment in RCBD at FIU shade house using 2-gallon pots for each plant. Four treatments (with six replications for each treatment) used in this study including control (no fertilizer), full synthetic N fertilizer (urea; 46-0-0), full cyanobacteria biofertilizer, and 50% synthetic + 50% cyanobacteria biofertilizer. Soils were collected from LNB Grove Farm, Homestead, FL (website: https://www.lnbgrovestand.com/).

Table 2. Effect of cyanobacteria biofertilizer on plant height, stem diameter, and leaf chlorophyll content (SPAD readings) of okra. Like letters (lowercase) indicate not significant at α = 0.05 level (SAS 9.4 was used)

* SPAD: Soil Plant Analytical Development represents chlorophyll contents of the leaves (no unit)

Overall, plant height (69 cm) and SPAD (51) readings at week-seven were significantly (P<0.05) higher in full biofertilizer treatment than unfertilized control. Stem diameter (9.2 and 11.2 mm at week 4 and week 7, respectively) was always significantly higher in fertilizer treatments than control (Table 2). However, no significant differences were found between fertilizers, indicating that cyanobacteria biofertilizer can achieve same crop physiological parameters as synthetic fertilizers.

UPDATE - Annual Report (2022)

Experimental set up in the greenhouse

Okra (Abelmoschus esculentus Var: Cajun Delight) and tomato (Lycopersicum spp Var: BHN 975) crops are growing at three- and five-gallon plastic pots, respectively, at FIU greenhouse. Plastic pots were filled up with soils collected from Homestead area of South Florida. Physicochemical properties of the background soil are presented in Table 2022-1. Okra and tomato pots were organized in separate bench tops with each pot was spaced about 3 ft from each other. Walking rows are available between each bench tops for watering the plants, regular maintenance, and recording the experimental data. Biofertilizer was applied along with synthetic fertilizers in different combinations. Seven treatment combinations were a) T0: control (no fertilizer applied); b) T1: 100% biofertilizer; c) T2: 75% biofertilizer + 25% synthetic; d) T3: 50% biofertilizer + 50% synthetic; e) T4: 25% biofertilizer + 75 % synthetic; f) T5: only 50% synthetic fertilizer and no biofertilizer; and g) T6: 100% synthetic fertilizer. The experimental set up followed a randomized complete block design (RCBD) with five replications for each treatment. At this point fruit setting just started for both okra and tomato. Okra seeds were planted in January 2022. We grew tomato seedlings in nursery trays from December, 2021 to January, 2022. One month tomato seedlings were then transplanted in the pots in 3^rd week of January, 2022. Both okra and tomato plants were watered in regular intervals in the greenhouse.  

Table 2022-1. Physicochemical properties of the soils collected from the field and used to fill up the pots in the greenhouse

5.Objective # 2: Efficacy of cyanobacteria biofertilizers (crop yield and soil health) in field settings

Agricultural field plots from our collaborative growers will be used to test the effect of cyanobacteria biofertilizer on tomato and okra production and soil health analysis. Physicochemical analysis of background soil will be analyzed before conducting the experiment. Yield parameters (shoot weight, root weight, plant height, and fruit yield) and compositional analyses (leaf chlorophyll content, plant height, plant shoot diameter, and nutrient uptake) of tomato and okra will be recorded during plant growth stages (early transplanting/seeding stage, flowering, early fruit set/pod-filling) and at harvesting. The yield parameters for our study were selected after discussing with Mr. Robert Barnum and Mr. Marc Ellenby. Routine physicochemical parameters (pH, EC, TP, TN, TC, POM, biomass C, N and P, Mehlich-3 extractable plant essential nutrients, ammonium, nitrate, nitrite, OM%, water filled pore space WFPS%, and aggregate stability) of soil samples will be analyzed in regular intervals throughout crop production. Soil pH will be measured in 1:1 soil to water slurry. Soil EC would be measured using Oakton conductivity meter. Analysis of TC, TN, and TP content will follow methods discussed earlier for soil samples. Particulate organic matter (POM) is a powerful indicator of soil health and quality (Chan, 2001; Wilson et al. 2001). Particulate organic matter (POM) will be measured in dry soil using a method described by Nciizah and Wakindiki, 2012 and will be calculated using the formula:

POM g/g dry soil= [(Dry soil weight - Combusted soil weight)/(Dry soil weight)] x 100

Soil aggregate stability will be measured using dry and wet sieving techniques (Kemper and Rosenau, 1986). Water soluble aggregates (WSA) is generally determined using:

WSA = {(Wac - Wc) / Wf}*100

Where, WSA (%) is the water-soluble aggregate in particular class or range of aggregate class, Wac (gm) id the dry weight of aggregates and coarse material; Wc is the dry weight of coarse material and Wf is the dry weight of the soil under each fraction of aggregate class.

Paid student (assistantships/scholarships) will assist in sample analysis where they will gain knowledge about sustainable agricultural research projects.

UPDATE 2024- Field Trial Experimental Design

Field trials for okra (Abelmoschus esculentus; variety Clemson Spineless) production were conducted across three South Florida farm locations: FIU Organic Garden (Sweetwater, FL), Rama Farm (Homestead, FL), and Lion Fruit Farm (Homestead, FL). Each location provided approximately 30 m² of farm area for the trial setup. Soil samples were collected from each plot at all three farms, and sent to the UF/IFAS Analytical Services Laboratories in Gainesville,FL. The physicochemical properties of the baseline soil are summarized in Table 2024-1.

Table 2024-1. Soil properties of each farm site collected from all plots on Day 0. N=12

The trial area at each farm was divided into 12 plots, each measuring approximately 0.7 m x 3 m. Fourteen okra plants were planted per plot, spaced 23 cm apart. Four experimental treatments were tested: (a) control (C; no fertilizer applied), (b) biofertilizer (B), (c) vermicompost (V), and (d) standard hot compost (H). Treatments were assigned using a randomized complete block design (RCBD), with each treatment replicated across 3 plots.

The vermicompost was sourced from Lion Fruit Farm, the standard hot compost from Fertile Earth Farms, and the biofertilizer was prepared using cyanobacteria collected and processed according to previously outlined procedures.

Recommended nutrient doses for okra production are 135 kg N ha⁻¹ and 112 kg K₂O ha⁻¹. Fertilizer application rates for each treatment were calculated based on the nutrient composition (N) of each fertilizer, as analyzed by the LECO Carbon Nitrogen Analyzer. The organic fertilizer composition percentages and the calculated fertilizer amounts applied per plot are detailed in Table 2024-2.  UF/IFAS Analytical Services Laboratories in Gainesville,FL analyzed all parameters except Total N and C.

Table 2024-2. Physicochemical properties of all three compost treatments. Calculated N percentage values were used to determine needed compost quantities to achieve an application rate of 135 kg N ha⁻¹ per plot. All analyses had an n=3.

Okra seeds were germinated in the FIU Shadehouse and transplanted into field plots on June 6, 2024. Plants were monitored biweekly for 12 weeks, measuring SPAD values and plant height. Once fruiting began, okra pods were harvested weekly and weighed to track total production. At the end of the 12-week period, three randomly selected plants from each plot were harvested, measured, and weighed for individual biomass. Biomass samples from each plot were subsequently analyzed for total nitrogen (TN), total phosphorus (TP), and other essential nutrients. Soil samples were also collected from each plot at the end of the trial to compare with the baseline soil samples collected at the trial’s initiation.

6.Objective # 5: Economic assessment of biofertilizer application

We assume that if farmers were to adopt the new biofertilizer, they would experience the following types of changes and/or opportunity costs: (a) changes (increase or decrease) in the annual costs of cultivation; (b) changes in crop yields (increase or decrease) and products sold; and (c) change in product market price (due to product quality changes). Therefore, the “true” or “incremental” benefits of integrating the proposed biofertilizer will be equal to the difference in the net profits between with and without implementation. The following methodology captures the economic effects of all the above three factors on the incremental benefits of applying new fertilizer. Assume that the per acre cost of crop production is c ($/ac). Per acre crop yield is y (lb/ac) and market price is p ($/lb). Let π ($/ac) denote the net profit. Subscripts pre and pos denote before and after implementation.

The expected net profit (π_pre) before the implementation of the new fertilizer is,

π_pre = (y_pre · p_pre – c_pre)

Similarly, the expected net profit after the implementation of a given N-I combination (π_pro) is,

π_pos = (y_pos · p_pos – c_pos)

Note that both quantity and quality of the product could vary from the current to the new practice. The annual per acre incremental benefits of biofertilizer application (IB) is the difference in the net profit between without implementation and with implementation. Formally,

            IB = π_pos - π_pre

Crop yield, costs, and price parameters used in the above economic model are stochastic in nature and may have certain underlying probability distributions. Based on the field and lab experimental trials (sample mean and standard deviation), we will construct appropriate probability distributions (e.g., normal distribution) for yield parameters. Historical market data (USDA) on prices and costs will be used to develop appropriate uniform or other probability distribution reflecting market uncertainty. Then, the incremental benefits (IB) will be estimated by running Monte Carlo simulations, which accounts for uncertainty in model parameter values (yields, costs, etc.) by expressing them as probability distributions rather than fixed values.

We will first need to gather the data for the economic enterprise budgets for existing tomato and okra crops in Florida. Using these as starting points, we will then develop mixed-crop budgets for alternative systems that involve tomatoes and okra production and seasons.  The information on the output side of the budget (e.g., market price, season, product quality, value addition, etc.) will directly come from secondary sources such as USDA. Monte Carlo stochastic analysis will be conducted around each mixed-crop budget in order to capture the effects of multiple scenarios reflecting alternative market channels, degree of value addition, output and input price variances. The final recommendation will be based on alternative production scenarios of biofertilizers, crops, seasons, and others.