Field and Modeling Study on Saving Mineral Fertilizers, Increasing Farm Income and Improving Soil Fertility Using Bio-Irrigation with Drainage Water from Fish Farms

: The reuse of new and non-traditional sources of water for the purpose of irrigation is the primary goal of all countries that are located in dry areas and suffer from water scarcity, including Egypt in particular. This study was conducted to determine the appropriateness and quantify the benefits of using fish farm wastewater (DWFF), as an alternative to fresh irrigation water (IW), for the irrigation of barley. Two types of water quality were tested for the irrigation of barley, namely DWFF and IW, in addition to four levels of fertilization rates, 100% N, 80% N, 60% N, and 40% N, where 100% N represents 156 kg of nitrogen per hectare. The results showed a positive effect of increasing the nitrogen fertilization rate with irrigation water on the crop with the use of DWFF and IW for irrigating barley in two seasons: 2017/2018 and 2018/2019. The yield when using DWFF for the irrigation of barley was higher than the yield when using IW, which was in the range of 5.1% and 25.9% in 2017/2018 and between 9.8% and 33.3% in 2018/2019. This was due to the additional amount of dissolved biological nitrogen and other nutrients contained in DWFF. Notably, an additional amount of dissolved nitrogen is inherent in DWFF (12.81 kg nitrogen ha −1 in 2017/2018 and 12.43 kg nitrogen ha −1 in 2018/2019) and other elements, such as phosphorus and potassium, which are two macronutrients for crops. The SALTMED model was used to simulate soil moisture content, water application efficiency, nitrogen concentration in the soil layer in the effective root zone, N uptake, the dry matter of grown barley, and yield and water productivity for all treatments, with R 2 values of 0.94, 0.89, 0.99, 0.916, 0.89, 0.915, and 0.919 respectively. The research concluded that the use of DWFF is an effective alternative to IW for irrigating barley. It also helped to achieve higher yields while applying lower amounts of IW and chemical fertilizers. There are also additional benefits, such as reducing the drainage to the drainage network and increasing the income of farmers. Investigation, I.A.A., M.E.-Z.; and


Introduction
Arid regions are generally characterized by having high population densities and limited fresh water available. These factors put substantial pressure on the agricultural sector to reduce the consumption of limited amounts of fresh water for irrigation to allow water to be available to other The current version of the model, which was released in 2015, allowed simultaneous simulations of 20 fields each with different irrigation systems, soils, crops, irrigation strategies, and N fertilizers. This model simulates soil moisture, dry matter, crop productivity, soil salinity, soil nitrogen dynamics, the requirements for salinity filtration, nitrate filtration, and drainage, soil temperature, evaporation, water absorption, water salinity, groundwater level, and effluent flow. This model was calibrated and validated using observed field data by [26][27][28][29][30][31][32][33][34][35][36][37]. They demonstrated the high predictability of field-measured yield, soil moisture, dry matter, and salinity.
This studyaimed to determine the suitability and maximum benefit ofusing fish farm drainage water to reduce the need formineral fertilizers and the total costs of production requirements in addition to improving soil fertility under dry conditions in Egypt through a field study and modeling using the SALTMED model.

Location and Climatic Data of the Experimental Site
Field experiments were conducted during the winter season of 2017/2018 and 2018/2019 at the research farm of the National Research Centre (NRC) (latitude 30°30′1.4′′ N, longitude 30°19′10.9′′ E, and 21 m above mean sea level at Nubaryia Region, Al Buhayrah Governorate, Egypt. The experimental area has an arid climate with cool winters and hot dry summers. The maximum and minimum temperature, relative humidity, and wind speed were obtained from the local weather station at El-Nubaryia Farm, as shown in Figure 1 and Table 1.

Chemical and Physical Properties of the Studied Soil and Water Quality of Two Types of Irrigation
The primarychemical and physical properties of the soil samples were identified and measured in situ and in the laboratory at the start of the field experiment ( Table 2). The average values for the chemical, physical, and biological properties of fresh IW and fish farm wastewater were also measured, particularly during the effective period of barley fertilization,when a sample was taken every week until the number of samples reached seven samples during that period. The results for the analysis of the samples of fresh IW and DWFF showed that there were no significant differences between the fresh IW samples during the sampling period (effective fertilization period), while there were significant differences between the fish farm wastewater samples. These findings would be expected for differentin fish activity, feeding, and temperature, which change the content of this water from the dissolved elements, organic components, algae, and other vital components (Tables 3 and 4).

Treatments and Experimental Design
The experimental design included eight treatments: two different types of IW and four fertigation rates with three replicates. The two types of water used were fresh irrigation water (IW) and drainage water from fish farms (DWFF). These water types were crossed with four nitrogen application rates: 100%, 80%, 60%, and 40%. For the 100% chemical N-fertilizer treatment, N was applied at the rate of 156 kg N ha −1 per season (Table 5) in ammonium nitrate form (33.5%N). We used 24 experimental plots,each with an area of 720 m 2 . The statistical design was a split design for this experiment. Soil moisture probe access tubes were placed in each pilot plot to measure the moisture content of the soil (Figure 2). Table 3 shows that DWFF was richerthan IW in the major elements phosphorous, nitrogen, and potassium, which are essential nutrients for plants. It was also distinguished by micronutrients, such as nickel, zinc, and copper. In addition, it contained more microorganisms and OM ( Table 4). The sprinkler irrigation system with fish farm wastewater did not include a filtering unit to ensure that all the beneficial biological organic materials and algae were added to the sandy soil to improve the fertility of these poor sandy soils. The planting and harvesting dates for barley were the 20th of November and 14th of April for seasons 2017/2018 and 2018/2019, respectively. The growth period for barley was 146 days, with 20 days for the initial stage, 32 days for the development stage, 57 days for the middle stage,and 37 days for the late stage.
The type of fish used was tilapia (Oreochromis niloticus) and the specific variety is tilapia niloticus (Nile tilapia),raised in 12 ponds, and the dimensionsof the pondswere 5 × 5 × 2 m, meaning that the volume of the ponds was 50 cubic meters, and the density of fish was 50 fish per cubic meter, meaning that the number of fish in one pond was 2500 fish, and the total number of fish in the experiment was 30,000 fish. IW: fresh irrigation water; DWFF: drainage water of fish farms, N: Nitrogen.

Irrigation Requirements for Barley
The water applied using sprinkler irrigation, which was obtained from Equation (1), was 2680 m 3 /ha for the 2017/2018 season and 2600 m 3 /ha for the 2018/2019 seasonafter deducting the amount of rainwater that fell during the two growing seasons for barley, Hordeum vulgare, variety Giza 123, where it was 681 m 3 ha −1 and was 617 m 3 ha −1 for the first and second season, respectively.
where IRg is the gross irrigation requirements in mm/day, ETOis the reference evapotranspiration in mm/day (estimated by the Central Laboratory for Climate, Agricultural Research Center, Egyptian Ministry of Agriculture at El-Nubaryia farm, following the Penman-Monteith equation), Kc is crop factor [38], IEis irrigation efficiency in %, R israinfall in mm, and LRis the amount of water required for the leaching of salts in mm.

Observed and Simulated Soil Moisture Content
Thesoil moisture content was determined using aprofile probe at a depth of 0-30 cm and soil moisture was also simulated.

Observed and Simulated Water Application Efficiency
Water application efficiency (WAE) is the proportion of waterstored in the root zone of the water applied in the field. The WAE was calculated using Equation (2): where WAE is the water application efficiency (%), and Ds is the depth of stored water in the root zone (cm), calculated using Equation (3)

mm valve
Orange color is the drainage water from fish farms where Da is the depth of applied water (mm), d is the soil layer depth (mm), θ1 is the average of soil moisture content after irrigation (g/g) in the root zone, θ2 is the average soil moisture content before irrigation (g/g) in the root zone and ρ is the bulk density of soil (g/cm 3 ), while the simulated WAE was calculated using Equation (2) for soil moisture content variables, Ds and Da,before and after irrigation at different soil depths.

Observed and Simulated Dry Matter and Grain Yield of Barley
The dry matter and grain yield of barley was measured at harvesting time, and random samples for a sample area of 1 m 2 cm were examined from each plot to estimate dry matter and grain yields in kg/ha for the entire area of the experimental unit, which was then converted to yield per tonper ha. The simulated dry matter and grain yield of barley was measured using the different variables that are calculated and saved as output variables for dry matter and yield at harvesting time. Notably, dry matter was estimated every 15 days from the beginning of planting until harvest.

Observed and Simulated Water Productivityof Barley
The water productivity of barley was estimated following James (1988): where WPbarley is the water productivity of barley (kgbarley/m 3 water), Ey is the primary yield (kg/ha). Ir is the applied amount of the IW (m 3 waterha −1 per season). The water productivity of barley was also simulated using Equation (4) for different yields and levels of irrigation.

SALTMED Model
The SALTMED model [25] is freely downloadable from the Water4Crops EU funded project web site: http://www.water4crops.org/saltmed-2015-integrated-management-tool-water-crop-soil-nfertilizers/ and from the International Commission on Irrigation and Drainage, ICID, web site: http://www.icid.org/res_tools.html#saltmed_2015. During model calibration, the relevant SALTMED model parameters were fine-tuned against the observed data for soil moisture, dry matter, and crop yield. For calibration, IW + 100% N was selected. Various soil parameters, such as the hydraulic properties of sandy soils, were adjusted until a very close match was achieved between the actual and simulated soil moisture values. The parameters of the cultivated crop under study were also adjusted ( Table 6). After a good fit for sandy soil moisture was achieved, only precise and specific control of the photosynthetic efficiency of barley dry matter and the actual yield of the crop were needed. The quality of the appropriate evaluation criteria used were the coefficient of determination (R 2 ), root mean square error (RMSE), and residual mass factor (CRM). The RMSE, R 2 and CRM valueswerecalculated using Equations (6)-(8) respectively. = ∑( − ) (6) where = predicted value, = observed value, and = total number of observations. The R 2 statistics demonstrate the ratio between the scatter of simulated values to the average value of measurements: where = averaged observed value, = averaged simulated value, = observed data standard deviation, and = simulated data standard deviation.
The coefficient of residual mass (CRM) is defined in Equation (8): Whenever these evaluation criteria are related to the following values, this indicates the high ability of this model to simulate the actual values. As for the full compatibility between the observed data and the simulation, the RMSE, CRM, and R2 values should be equal to 0.0, 0.0, and 1.0, respectively. Residual water content, m 3 m −3 0. 22 Lambda pore size 11 Bubbling pressure, cm 52 Max. depth for evaporation, mm

Statistical Analysis
Most of the data were subjected to statistical analysis to clarify the discrepancy between the different treatments under study as described by Snedecor and Cochran [39]. A joint statistical analysis of the results for both seasons was conducted following the method adopted by Steel and Torrie [40], where the average values of the recorded data are compared using least significant differences (L.S.D with significance at <0.05). When measuring the SMC under all the transactions under study, it was observed that they did not differ under the treatments that were irrigated with IW while the values were slightly higher under the transactions that irrigated using DWFF, and this may be due to the the OM added with the DWFF keeping the water in the root zone in the sandy soil.

Soil Moisture Content
SMC did not change significantly with an increase in nitrogen fertigation rates. Perhaps this was due to the amount of water added being the same for all treatments for the same growing season.
Initially  6). Overall, there was a close fit between the data simulated by the SALTMED model and the observed data during the calibration and validation processes. These results are similar to those obtained by [25,27,[35][36][37]41,42].

Water Application Efficiency
For the two studied seasons, WAE was investigated using different water qualities for irrigation and different rates of nitrogen fertilizer under sandy soil conditions. WAE increased under irrigation with DWFF in comparison to IW (Figure 7). This is likely due to the increase in the OM that was added as a result of irrigating with DWFF and the OM playing a vital role in increasing the water holding capacity and preventing nutrients leaching from the soil.
The effect of the fertigation rate of nitrogen on the WAE is shown in Figure 7. WAE decreased with a decreasing fertigation rate, which could be due to the increasein the fertigation rate increasing the size of the plants roots, enabling the absorbtion of most of the added IW and fertilizers.
There was a positive correlation between the simulated and observed values for both the 2017/2018 and 2018/2019 seasons (Figures 7 and 8), where the R 2 was 0.89, indicating the accuracy of the SALTMED model for simulating WAE.   There was an improvement in the N-CSL with DWFF irrigation compared with IW irrigation in both the 2017/2018 and 2018/2019 seasons. This is due to the additional amount of biological nitrogen and other nutrients already present in the DWFF.

Nitrogen Concentration in the Soil Layer of the Effective Root Zone
The N-CSL increased with increasing fertigation of N, which seems logical, as increasing the rate of N-Fertigation increases the N concentration in the root zone,leading to an increase in N.
After making a comparison between the actual N concentration values in the root diffusion region of the barley intakes measured using the field and simulated values given by the model (Figures 9 and 10

Nitrogen Uptake
Nitrogen up take changed for the 100%N, 80%N, 60%N and 40%N treatments for both the IW and DWFF treatments ( Figure 11).
The nitrogen uptake values were higher for the 2017/2018 season than the 2018/19 season. This finding is likely due to the increase in the amount of rain water falling during the effective period of barley fertilization in the second growing season (2018/2019) when the amount of rainfall was 34.48 mm during January, which translates to 344.8 m 3 ha −1 and led to an increase in the rate of leaching of fertilizer elements, including N, from the root zone, which led to a decrease in the concentration of N. Previous influences led to a negative effect on the decrease of nitrogen uptake compared with the first season (2017/2018). For the 2018/2019 season, the amount of rain that fell at the end of the growing season and outside of the effective period of fertilization did not affect the concentration of fertilizer elements during the effective period of fertilization.
Total N uptake improved with the use of DWFF irrigation for barley compared with when irrigation with IW was used during growing seasons. This was due to the presence of excess and additional amounts of biological N in addition to other nutrients that were already present in the DWFF, with the additional N being estimated to be 12.81 kg of N per ha in the 2017/2018 season and 12.43 kg of N per ha in the 2018/2019 season compared with that for IW irrigation.
The total N uptake increased with an increasing fertigation rate of N, which seems logical, as increasing N-fertigation increases the concentration of N in the root zone, leading to increased nitrogen.  . Figure 13. Observed versus simulated accumulated-nitrogen uptake for all treatments for two seasons. Table 8 shows that the average dry matter weight of barley was affected by the 100%N, 80%N, 60%N, and 40%N treatments for both the IW and DWFF treatments. The average weight of barley dry matter was higher when irrigating with DWFF than when irrigating with IW in the 2017/2018 and 2018/2019 seasons. This is due to the increasein N uptake under DWFF, because this water had additional bio-nitrogen and other bio-nutrients.

Barley Dry Matter
The average weight of barley dry matter increased with increasing N uptake, which increased with increasing N-fertigation. The observed and simulated dry matter weights showed agood fit for all age stages of barley plants for most treatments. The SALTMED model simulated the dry matter values with an R 2 of 0.896 for all treatments for 2017/2018 and 0.82 for all treatments during 2018/2019 (Table 8).  Simulated AccN_uptake -gN m -2 Observed AccN_uptake -gN m -2  Figure 14 and Table 9

Water Productivity of Barley
Water productivity of barley (WPbarley) was estimated as the amount of grain yield in kg per m 3 of applied IW. Figure

Conclusions
This pilot study investigated the suitability and optimization of irrigation using DWFF and IW for barley production through a field study using the SALTMED model.
The WAE increased under irrigation with DWFF compared to IW, which was likely due to the increase in the OM added with the DWFF to sandy soil, keeping the water in the root zone. WAE decreased by decreasing the fertigation rate of nitrogen, and this was likely due to an increase in the size of the roots of plants as a consequence of increasing the fertigation rate, enabling the absorbtion of most of the added IW and fertilizers. Total N uptake was improved with the use of DWFF compared with IW for the irrigation of barley during the 2017/2018 and 2018/2019 growing seasons. This is likely due to the presence of additional biological N in addition to other nutrients that were already present in the DWFF. The total N uptake has been increased by increasing the N-fertigation rate and this seems logical, as increasing the rate of Nfertigation increases the concentration in the root zone and increases N uptake.
The obtained results indicate that there was a positive effect with increasing the rate of N-fertilization with irrigation water on crop yield using DWFF and IW in both seasons. However, the yield with irrigation using DWFF was higher than the yield under the IW treatment, for which the percentage increase in productivity ranked between 5.1% and 25.9% in 2017/2018 and between 9.8% and 33.3% in 2018/2019. The biggest difference was found for the treatment with the least nitrogen. Of note is an additional amount of dissolved nitrogen inherent in DWFF (12.81 kg N ha −1 in 2017/2018 and 12.43 kg N ha −1 in 2018/2019) along with more phosphorous and potassium. These findings are in agreement with the results of other reports that show that the integrated cultivation of rice and fish is environmentally sound and healthy, as fish farming improves soil quality and fertility by increasing the availability of N and phosphorous. In general, crop yields and total N uptake increased by increasing the rate of nitrogen fertilization. There was a positive In summary, the SALTMED model simulated soil moisture, WAE, N concentration within the barley plant root zone for sandy soil, N uptake, the weight of barley dry matter,and the productivity and water productivity of the barley crop and their R 2 values were 0.94, 0.89, 0.99, 0.916, 0.89, 0.915, and 0.919, respectively. The results obtained from the field study and the modeling data indicated that the yield from the use of DWFF has many benefits, which include improving yields and limiting the excessive use of mineral chemical fertilizers. These benefits have increased the income of ordinary farmers, in addition to reducing the volume of wastewater that must be disposed of in local sewage networks. Therefore, this study recommends the use fish culture wastewater for irrigation as a suitable alternative or supplement to limited freshwater resources and as a vital alternative to fertilization. Funding: This research received no external funding.