Expl Agric. (2012), volume 48 (2), pp. 222–237 C© Cambridge University Press 2011

doi:10.1017/S0014479711001153

UNRAVELLING CROP WATER PRODUCTIVITY OF TEF(ERAGROSTIS TEF (ZUCC.) TROTTER) THROUGH

AQUACROP IN NORTHERN ETHIOPIA

By ALEMTSEHAY TSEGAY†, ‡§, DIRK RAES‡, SAM GEERTS‡,ELINE VANUYTRECHT‡, BERHANU ABRAHA†, JOZEF DECKERS‡,

HANS BAUER¶ and KINDEYA GEBREHIWOT††

†Department of Dryland Crop and Horticultural Sciences, Mekelle University, P.O. Box 231,

Mekelle, Ethiopia, ‡Division of Soil and Water Management, K.U. Leuven University,

Celestijnenlaan 200E-2411, B-3001 Leuven, Belgium, ¶VLIR-UOS Ethiopia,

P.O. Box 80522, Addis Ababa, Ethiopia/P.O. Box 231, Mekelle, Ethiopia and ††Department of

Land Resource Management and Environmental Protection, Mekelle University, P.O. Box 231,

Mekelle, Ethiopia

(Accepted 26 October 2011; First published online 25 November 2011)

SUMMARY

At various locations in North Ethiopia (Tigray), field experiments were conducted from 2006 to 2009 toassess the crop response to water stress of tef (Eragrostis tef (Zucc.) Trotter) under rainfed, fully irrigated anddeficit irrigation conditions. Observed soil water content (SWC), canopy cover (CC), biomass production (B)and final grain yield (Y) were used to calibrate and validate AquaCrop for tef. Data from an experiment in acontrolled environment in 2008 were also considered in the calibration process. Simulations of SWC, CC, B

and Y were evaluated by determining the index of agreement, the root mean square error, the coefficient ofdetermination and the Nash–Sutcliffe efficiency. The statistical parameters showed an adequate fit betweenobservations and simulations. The model was able to simulate for tef growing under rainfed condition theobserved fast drop in SWC and CC when the rains ceased. The overall goodness of fit between the observedand simulated CC and SWC indicated that the thresholds for root zone depletion at which water stress(i) affects canopy development, (ii) induces stomata closure and (iii) triggers early canopy senescence werewell selected. The normalised biomass water productivity (WP ∗) for tef was 14 g m−2 for the local varietyand 21 g m−2 for the improved variety, which is a lot smaller than the WP ∗ expected for C4 plants (30–35g m−2). The results revealed an increase of 27% in reference harvest index (HIo) of tef in response to mildwater stress during the yield formation of up to 33%. However, severe water stress causing stomata closurehad a negative effect on HIo. Once it is properly calibrated, AquaCrop can provide room to improve thewater productivity of tef by developing guidelines for good agricultural management strategies.

I N T RO D U C T I O N

Tef (Eragrostis tef (Zucc.) Trotter) originated in Ethiopia (Vavilov, 1951) and is one ofthe important cereal crops that grows in diverse climatic and edaphic zones of thecountry. The crop occupies 28% of the total cultivated area for cereals and accountsfor about 21% of the total cereal production (Central Statistical Agency of Ethiopia(CSA), 2010). Tef grain is predominantly produced by smallholder farmers and about62% of the population depend on it as staple food (Kebebew Assefa, 2011, personal

§Corresponding author. Email: alemtsehay.tsegay@gmail.com; alemtsehaytsegay@yahoo.com

1

Shorten title: Water productivity of tef revealed through AquaCrop 1

Full title: Unraveling crop water productivity of tef (Eragrostis tef (Zucc.) 2

Trotter) through AquaCrop in Northern Ethiopia. 3

4

Names of Authors: Alemtsehay Tsegay*1, 2, Dirk Raes2, Sam Geerts2, Eline Vanuytrecht2 5

Berhanu Abraha1, Seppe Deckers2, Hans Bauer3, and Kindeya Gebrehiwot4 6

Adress of Authors: 7 8 *1 Mekelle University, Department of Dryland Crop and Horticultural Sciences, P.O.BOX. 231, Mekelle, Ethiopia. 9 10 2 K.U.Leuven University, Division of Soil and Water Management, Celestijnenlaan 200E - 2411, B-3001 Leuven, 11 Belgium 12

3 VLIR-UOS Ethiopia, P.O.BOX 80522, Addis Abeba, Ethiopia / P.O.Box 231 Mekelle, Ethiopia 13

4 Mekelle University, Department of Land Resource Management and Environmental Protection, P.O.Box 231, 14

Mekelle, Ethiopia. 15

16

*Corresponding author: 17

Present address: 18

Mekelle University, Department of Dryland Crop and Horticultural Sciences, P.o.box 231, Tigray, 19

Ethiopia. 20

Email: alemtsehay.tsegay@gmail.com 21 alemtsehaytsegay@yahoo.com22

ManuscriptClick here to download Manuscript: Revised Manuscript .doc

2

SUMMARY 23

At various locations in North Ethiopia (Tigray), field experiments were conducted from 2006 to 24

2009 to assess the crop response to water stress of tef (Eragrostis tef (Zuccu.) Trotter) under 25

rainfed, fully irrigated and deficit irrigation conditions. Observed soil water content (SWC), 26

canopy cover (CC), biomass production (B) and final grain yield (Y) were used to calibrate and 27

validate AquaCrop for tef. Data from an experiment in a controlled environment in 2008 were 28

also considered in the calibration process. Simulations of SWC, CC, B and Y were evaluated by 29

determining the index of agreement, the root mean square error, the coefficient of 30

determination and the Nash-Sutcliffe efficiency. The statistical parameters showed an adequate 31

fit between observations and simulations. The model was able to simulate for tef growing under 32

rainfed condition the observed fast drop in SWC and CC when the rains ceased. The overall 33

goodness of fit between observed and simulated CC and SWC indicated that the thresholds for 34

root zone depletion at which water stress (i) affects canopy development, (ii) induces stomata 35

closure and (iii) triggers early canopy senescence, were well selected. The normalized biomass 36

water productivity (WP*) for tef was 14 g/m2 for the local variety and 21 g/m2 for the improved 37

variety, which is a lot smaller than the WP* expected for C4 plants (30-35 g/m2). The results 38

revealed the increase of the 27 % reference harvest index (HIo) of tef in response to mild water 39

stress during yield formation up to 33 %. However, sever water stress causing stomata closure 40

had a negative effect on HIo. Once it is properly calibrated, AquaCrop can provide room to 41

improve the water productivity of tef by developing guidelines for good agricultural 42

management strategies.43

3

Alemtsehay Tsegay et al. 44

INTRODUCTION 45

Tef (Eragrostis tef (Zucc.) Trotter) originated in Ethiopia (Vavilov, 1951) and is one of the 46

important cereal crops that grow in diverse climatic and edaphic zones of the country. The crop 47

occupies 28% of the total cultivated area for cereals and accounts about 21% of the total cereal 48

production (CSA, 2010). Tef grain is predominantly produced by smallholder farmers, where 49

about 62% of the population depend on it as staple food (Kebebew Assefa, 2011, personal 50

communication). It provides about two-thirds of the dietary protein intake for most Ethiopians 51

who use tef as staple food (Seyfu, 1997). The grains are highly nutritious and have a protein 52

content of 9 to 11%, which is slightly higher than that of sorghum (Sorghum bicolour), maize 53

(Zea mays L.) or oats (Avana sativa L.) (NRC, 1996). Moreover, the grains are gluten free, making 54

them a valuable food source for people with celiac disease. Although tef follows the C4 55

photosynthetic pathway (Seyfu, 1997), its low light utilization, which could be associated with its 56

small leaf size and specific leaf orientation leads to limited photosynthetic efficiency (Dejene, 57

2009). 58

59

In semiarid areas like in the northern highland of Tigray agriculture is dependent on rainfed 60

where the spatial and temporal variability of rainfall and occurrences of dry spell during seasons 61

and in between seasons is common. As a result of drought stress and other stress factors, the 62

regional average yield of cereals and tef is oscillating around 1.4 and 0.93 t ha-1 (CSA, 2000-2010) 63

respectively. The water stress particularly at the later development stage of crops affected tef 64

productivity as the result of earlier cessation of the rainfall. Dejene (2009) reported that water 65

stress during this period caused greater reduction in net C02 assimilation and transpiration rates 66

suggesting the need of supplementary irrigation to make up for the deficit in the seasonal 67

rainfall. 68

69

Crop water productivity models can assist in understanding the effect of different physiological 70

mechanisms to water stress (Geerts and Raes, 2009). AquaCrop (Steduto et al., 2009; Raes et al., 71

2009) can be used for this purpose as a tool to predict yield loss due to water stress. 72

Furthermore it supports to design management strategies like developing guidelines related to 73

74

4

Water productivity of tef revealed through AquaCrop 75

water management and developing scenarios for the future climatic conditions. The model 76

keeps a good balance between accuracy and relatively limited input requirements and was 77

successfully calibrated for common crops such as maize and cotton (Heng et al., 2009; Garcia 78

Vila et al., 2009) and for under-utilized crops such as quinoa (Geerts et al., 2009). 79

80

The research for this study was carried out in Tigray, the northernmost region of Ethiopia, 81

regarded as one of the most important regions for tef production in the country. The rainfall 82

pattern in Tigray is unimodal with a clear-cut rainy season that occurs principally in July and 83

August (Nyssen et al., 2005). The crop water productivity, which refers to the ratio of biomass 84

produced over water transpired, of this the rather unknown crop was modelled to gain insight 85

into its physiological response to water stress. In this paper, we discuss the calibration and 86

validation of AquaCrop for tef. In four successive years, field experiments were carried out in 87

three locations in Tigray region, under rainfed and irrigation treatments, with both local and 88

improved tef varieties. Preliminary yield response of tef to water stress has been investigated by 89

Araya et al. (2010). They analyzed only two growing seasons and used all the observed data for 90

both calibration and validation of the AquaCrop model. By analyzing our own data, this paper 91

confirms, partly adjusts and completes the findings of Araya et al. (2010) and discusses some of 92

the physiological uncertainties that remained. 93

MATERIALS AND METHODS 94

Model description 95

AquaCrop simulates green canopy (for water transpiration) and root growth (for water uptake) 96

under the governing environmental conditions (Raes et al., 2009). In exchange for water 97

transpired, biomass is produced. Cumulative aboveground biomass production (B) is a function 98

of the daily ratio of crop transpiration (Tr) and reference evapotranspiration (ETo) (Eq. 1). Yield 99

(Y) is calculated by multiplying the final B by the harvest index (HI) (Eq. 2). 100

101

102

103

5

Alemtsehay Tsegay et al. 104

n

i

oii ETTrWPB1

* [1] 105

Y = HI ∙ B [2] 106

107

where B is the cumulative aboveground biomass production (g m-2); WP* is the normalized crop 108

water productivity (g m-2); Tri is the daily crop transpiration (mm day-1); EToi is the daily 109

reference evapotranspiration (mm day-1); n is the sequential days spanning the period when B is 110

produced; Y is the yield production (g m-2); and HI is the harvest index (Raes et al., 2009). 111

112

The proportional factor WP* in Eq. 1 is the water productivity normalized for the local climate 113

(expressed by ETo) and the CO2 concentration (Steduto et al., 2007). The main features of the 114

AquaCrop model that differentiate it from other crop models is its focus on water, the use of 115

ground cover instead of leaf area index and the use of water productivity values normalized for 116

evaporative demand and carbon dioxide concentration (Raes et al., 2009). Simulations in 117

AquaCrop model need site specific (climate, soil and management) and crop specific (plant 118

density, crop development, rooting depth, harvest index, crop coefficient (KC) and water 119

productivity) input parameters to make yield estimation. A detailed description of the model can 120

be found in Steduto et al. (2009) and Raes et al. (2009) who describe the underlying principles 121

and distinctive components as well as the structural details and algorisms of AquaCrop model, 122

respectively. 123

124

125

126

6

Water productivity of tef revealed through AquaCrop 127

128

Determination of water productivity (WP*) for tef in micro-lysimeters 129

130 The experiment was carried out at Mekelle University (MU) campus (13.30° N, 39.29° E, 2212 m 131

above sea level (a.s.l)) in 21 aboveground free-draining mini-lysimeters. The mini-lysimeters had 132

a diameter of 0.21 m and a height of 0.33 m. Local variety of tef was broad casted at a seed rate 133

of 30 kg ha-1, which equals a population density of approximately 2000 plants m-2. The crop was 134

grown from 22 November 2008 to 3 March 2009 in the mini-lysimeters containing cambisol. 135

Urea and DAP (Diamonium Phosphate) were applied at a rate of 60 kg ha-1 as a source of 136

nitrogen and phosphorus. All the mini-lysimeters were placed under a rain shelter and arranged 137

in a completely randomized design. Water was applied every other day to keep the water 138

content in the lysimeters close to field capacity (FC), i.e., 36.6 vol%. If water drained out of the 139

lysimeters, it was re-applied. The required amount of water was determined by calculating a soil 140

water balance according to Allen et al. (1998). 141

142

In six out of the 21 lysimeters, tef was well watered, actively growing, completely shading the 143

ground and regularly clipped to keep it at a uniform height of 0.12 m. As such it was used as 144

hypothetical reference crop for reference evapotranspiration as mentioned in Allen et al. (1998). 145

The clipped plants were kept aside and no micro-climate, distorting ETo, was generated. Since 146

the crop covered most of the lysimeter, evaporation was assumed negligible.These lysimeters 147

were used as a reference to determine the evaporating power of the atmosphere (ETo) for the 148

environment 149

150

On the other hand the actual evapotranspiration (ETa) was determined from tef freely growing 151

in the other fifteen mini-lysimeters. To keep the transpiring surface exactly the same as the 152

surface of the mini-lysimeter and thus to minimize the error, overhanging tef plants were forced 153

to grow straight with supporting sticks and slightly tied threads. Both ETa and ETo were 154

determined every other day by measuring the mass difference of the lysimeters and the volume 155

of water applied. The cumulative ratio of ETa over ETo ((ETa/ETo)) was calculated as a proxy 156

for the cumulative ratio of transpiration normalized for climate, assuming the evaporation 157

negligible. Three replicates of intermediate B harvests taken at five successive points in time 158

from the fifteen lysimeters were used to determine the normalized water productivity (WP*) of 159

7

Alemtsehay Tsegay et al. 160

the crop, as the linear regression between (ETa/ETo) and the corresponding B (Steduto et al., 161

2007). For these calculations, ETa was assumed equal to Ta. Since nearly all the evaporation 162

losses were limited to the initial season before canopy closure, which occurred very early, the 163

slope of the curve between B and (ETa/ETo) is a valid proxy for WP*. 164

Field experiments 165

Study site and experimental design 166

Field experiments (Table 1) were performed during the main rainy season for four years (i) in 167

2006 and 2007 at MU and (ii) in 2008 and 2009 at two sites located in east (Maiquiha, 13°48’ N, 168

39°27’ E, 2078 m a.s.l) and south-east (Dejen, 13°20’ N, 39°22’ E , 2128 m a.s.l) Tigray. An 169

improved (Dz-Cr-37 in 2007) and local (in 2006, 2008 and 2009) varieties of tef were sown by 170

broadcast method with a seed rate of 30 kg ha-1. Recommended doses of nitrogen (60 kg ha-1) 171

and phosphorous (60 and 40 kg P205 ha-1 for heavy and light soils respectively) (EARO, 2002) 172

were applied in the form of urea and DAP. The urea application was split: 50% was given at 173

sowing and the remaining 50% at tillering. DAP was applied at sowing. The crops were regularly 174

weeded and protected against pests and diseases throughout the growing seasons. 175

176

Rainfed (RF), fully irrigated (FI) and deficit irrigation (DI, where irrigation was only applied during 177

flowering) treatments were arranged in randomized complete block design with three (MUFI07, 178

MURF07, DEFI08, DERF08, DEFI09, DERF09, MAFI08, MARF08, MAFI09 and MARF09) or four 179

(MURF06 and MUDI06) replications (see Table 1 for codes). The size of the experimental plots 180

was 9 m2 and 18 m2 in MU in 2006 and 2007 respectively, 4 m2 and 6 m2 in Dejen in 2008 and 181

2009 respectively, and 4 m2 and 9 m2 in Maiquiha in 2008 and 2009 respectively. The spacing 182

between the experimental plots was 0.5 m and the replications were 1.0 m apart from each 183

other. 184

Measurements 185

Climate data 186

Daily maximum and minimum air temperature, average wind speed at 2 m height, average 187

percentage of relative humidity, and hours of bright sunshine were recorded at the 188

189

8

Water productivity of tef revealed through AquaCrop 190

experimental sites and at a nearby agro-meteorological station to calculate ETo with the FAO 191

Penman-Monteith equation (Allen et al., 1998). Daily rainfall was recorded on each site. 192

Soil data 193

The initial soil water content (SWo) of the experimental plots was measured gravimetrically 194

(three replicates) on the respective sowing dates. The soil water content in the root zone was 195

measured gravimetrically every week at two depths between 0 and 0.4 m (Maiquiha; MU in 196

2006) or three depths between 0 and 0.6 m (Dejen; MU in 2007) from sowing till harvest, in 197

each experimental plot (three replicates, or four in MU in 2006). 198

199

Soil samples (three replicates) of the various fields were analyzed in the laboratory to determine 200

particle size distribution (by the hydrometer method). Soil water content at saturation (SAT), 201

field capacity (FC) and permanent wilting point (PWP) was determined using a pedotransfer 202

function (Saxton and Rawls, 2006). The soil physical characteristics are reported in Table 2. 203

204

Crop data 205

The time (in days after sowing) to reach different development stages (emergence, maximum 206

canopy cover, flowering, senescence and physiological maturity) was recorded during each 207

growing season. The plant population density was counted using a 0.25 m by 0.25 m quadrant. 208

The plant density for all experiments and all treatments was approximately 2000 plants m -2. 209

210

Green crop canopy cover (CC) was determined every 10 to 15 days by taking digital pictures 211

perpendicular to the experimental plot at an approximate height of 1.5 m. The pictures of an 212

area (2.52 m2) were first resized to confirm uniform pixel size (1000*1000 pixels). To reduce the 213

error that could occur at the four corners due to deformation (viewing angle), the centre square 214

was selected as most representative part of each pictures and was subjected to the pixel 215

analysing software SigmaScan Pro 5.1 software (SPSS Inc., 1990-2000). With the help of the Turf-216

analysis macro (Karcher and Richardson, 2005) the software counted the amount of ‘green’ 217

pixels against the total amount of pixels. The three replicates of each treatment were used to 218

determine canopy cover in percentage. 219

220

221

9

Alemtsehay Tsegay et al. 222

B during the season was measured at an interval of 10 to 15 days by cutting the plant from a 223

quadrant of 0.25 m*0.25 m for the 4 m2 and 6 m2 plot sizes, and a quadrant of 0.5 m by 0.5 m 224

for the 9 m2 and 18 m2 plot sizes and was dried for 48 hours at 65°C in a ventilated oven. In 225

Maiquiha in 2008, only the final B and Y were recorded. Final Y and final B at harvest were taken 226

from a quadrant of 2.5 m2 and 3.68 m2 respectively for the smaller (4 or 6 m²) and 6 m2 for the 227

larger plot size (9 m² or 18 m2). The harvest index (HI) was calculated as the ratio of Y to final B. 228

229

Management 230

Supplementary irrigation for full and deficit irrigation treatments was applied by flooding the 231

small basins. The timing and depth of the irrigation applications were determined by means of a 232

soil water balance (Allen et al., 1998). In the calculation, the crop evapotranspiration (ETc) was 233

determined with the Kc × ETo approach. A crop coefficient (Kc) of 1.1 was used, which is a good 234

indicative value for small cereals with good soil cover after development. 235

Calibration and Validation of AquaCrop for tef 236

The determined WP* within the mini-lysimeters and the observations of the field experiments in 237

2007 and 2008 in three locations were used to calibrate AquaCrop model. In the calibration 238

process the observed time (in calendar days) to developmental stages (emergence, maximum 239

CC (100 %), flowering, senescence and physiological maturity) under well watered conditions (FI 240

treatments) were used to describe the crop development under non-limiting conditions in 241

AquaCrop. Subsequently, water stress coefficients (Ks) for leaf expansion, stomata closure and 242

early canopy senescence were calibrated based on the physiological characteristics of the crop 243

and in an iterative way by comparing observed CC and SWC with the simulated outputs of 244

AquaCrop for the FI and RF treatments. The default shape of the Ks curves, which determines 245

the magnitude of the effect of soil water stress factors, was kept (Raes et al., 2009). The 246

reference HI and parameters describing the effect of water stress on the HI were calibrated with 247

data from FI and RF treatments. The observations of CC, SWC, B and Y were used as benchmarks 248

during the calibration process. Calibrations started with tef under FI (no water stress). 249

Subsequently, tef under RF conditions was considered. 250

251

252

10

Water productivity of tef revealed through AquaCrop 253

Calibration and validation were evaluated for SWC, CC, B and Y by comparing simulated outputs 254

with data collected from the field using the following statistical parameters: i) the coefficient of 255

determination (R²), which quantifies the portion of the total observed variance that can be 256

explained by the model; ii) the root mean squared error (RMSE), which expresses the overall 257

mean deviation between observed and simulated value as a measure for the relative model 258

uncertainty (Loague and Green (1991); iii) the index of agreement (d, Willmott, 1982), which 259

shows the differences in the observed and predicted means and variances and iv) the Nash-260

Sutcliffe efficiency (EF, Nash and Sutcliffe, 1970), which assesses the overall deviation between 261

observed and simulated values to the variability of the observations as a measure of the model 262

performance over the whole simulation period. Better model performance is achieved when R², 263

EF and d approach unity and when RMSE approaches zero. 264

265

RESULTS 266

Water productivity (WP*) for tef derived from lysimeter experiments 267

The water productivity of tef in the lysimeter experiment was determined from the linear 268

regression between the cumulative ratio of observed dry aboveground biomass and cumulative 269

ratio of transpiration over ETo. As clearly presented in Figure 1 there is linear increase of 270

observed biomass in function of the cumulative ratio between transpiration adjusted for the 271

reference evapotranspiration (R2= 0.98). The experimentally determined WP* for tef was 15.5 g 272

m-2 (Figure 1). 273

Calibration of AquaCrop for Tef 274

Examples of simulated and observed CC and SWC for RF and FI treatments are presented in 275

Figures 2 and 3. Regardless of the year and the location of the experiment, the observed and 276

simulated CC development fitted well (Figure 2a and 2b) and followed the standard logistic 277

growth curve used by AquaCrop for non stressed conditions (Raes et al., 2009). The maximum 278

CC of about 100% was reached 49 days after sowing in the experiments of 2007 and 2008. 279

280

281

11

Alemtsehay Tsegay et al. 282

In the RF treatment, both simulated and observed CC (Figure 2) and SWC (Figure 3) dropped 283

very quickly after the rains ceased indicating the shorter crop cycle of RF tef due to early 284

senescence. Table 3 shows the statistical parameters for the fit with high R2, EF and d values 285

except for the low EF of the yield. The low values of the RMSE indicate that the overall deviation 286

between the model simulations and observations was limited for all the computed parameters. 287

288

Observed values for B from field experiments were plotted against the cumulative ratio between 289

simulated transpirated water by the crop (Ta) and adjusted reference evapotranspiration (ETO) 290

in Figure 4. The slope of the linear fit allowed to draw the normalize water productivity (WP*) of 291

both improved and local varieties of tef valued 13.3 g m-2 and 18.5 g m-2, respectively. The 292

coefficient of determination (R2) for the WP* determined in the field for both varieties was 0.92, 293

indicating that there is excellent correlation between the cumulative normalized transpiration 294

and B. 295

296

A WP* of 14 g m-2 was selected for the local variety by considering the results from the 297

lysimeters (WP* = 15.5 g m-2) and the field experiments (13.3 g m-2). A WP* of 21 g m-2 was 298

selected for the improved varieties, assuming (i) a similar deviation as for the local varieties 299

between the realistic WP* and the value from field experiments (WP* = 18.53 g m-2) and (ii) by 300

giving more weight to the final B (where sample sizes are larger than for intermediate B) during 301

the final calibration of the model. After the selection of WP* and a reference HI (HIo) of 27 %, 302

good simulation for B and Y could be obtained. Figure 5 shows examples of simulated and 303

observed B for RF and FI treatments. The simulations fitted the observed data very well with 304

higher B attained from irrigated treatment in both locations and years. The regression between 305

the observed and simulated final B and Y for all data set used for calibration are given in Figure 6 306

indicating R2=0.84 and 0.96 for Y and B respectively. The crop parameters for tef obtained after 307

calibration are listed in Table 4. 308

12

Water productivity of tef revealed through AquaCrop 309

Validation of the calibrated model 310

The validation of the model was carried out by considering the calibrated crop parameters 311

(Table 4) and data sets from field observations in 2006 and 2009 on three sites and three 312

treatments (FI, RF and DI). Observed soil water content in the root zone, green canopy cover, B 313

and Y were compared with the validated result of simulation. Table 5 gives an overview of the 314

statistical parameters for the validation of the AquaCrop model. The parameters confirmed the 315

good match between simulations and observations. 316

DISCUSSION 317

The experimentally determined WP* of tef from the lysimeter (Figure 1) and field experiments 318

(Figure 4) demonstrates the linear increase of observed biomass in function of the cumulative 319

ratio between transpiration and reference evapotranspiration, which confirms the hypothesis of 320

the conservative behaviour of WP* as formulated by Steduto et al. (2007). The normalized WP* 321

of the crop derived as the slope of the linear regression between (ETa/ETo) and the 322

corresponding B sampled periodically from both experiments did not level off during yield 323

formation, as is the case for sunflower by the production of highly nutritious seeds (Steduto and 324

Albrizio, 2005). However, the estimated WP* of tef found from both experiments was very low 325

for a C4 plant and contrasts with the WP* for other C4 crops: 32.9 g m-2 for sorghum (Steduto 326

and Albrizio, 2005), 33.7 g m-2 for maize (Hsiao et al., 2009; Heng et al., 2009), and the indicative 327

range (30-35 g m-2) given in AquaCrop (Raes et al., 2010). The low WP* could be the 328

consequence of a higher protein content of tef in comparison with other C4 crops, due to which 329

the crop requires considerably more energy per unit dry weight to convert the assimilated CO2 330

into carbohydrates (Azam-Ali and Squire, 2002). Its lower light use efficiency compared to other 331

C4 plants, which is related to leaf size and orientation, as mentioned by Dejene (2009) could also 332

explain the low WP*. 333

334

The evolution of the observed and simulated soil water content in the root zone is plotted for 335

the irrigated and rainfed treatment as shown in Figure 3, which indicate well simulated crop 336

transpiration as a result of a very good match between simulations and observations for both 337

treatments. This was confirmed by adequate values of the parameters used to evaluate the 338

goodness of fit of the model (Table 3). 339

13

Alemtsehay Tsegay et al. 340

A calibration on crop characteristics was carried out to improve the simulations. Input values for 341

different crop parameters given in Table 4, were based on observations with minimal 342

adjustment. The evolution of the simulated values for CC over time for RF and FI tef showed a 343

very good comparison with observed values (Figure 2a and 2b). This result shows that the 344

thresholds for root zone depletion at which water stress affects canopy development (pexp), 345

induces stomata closure (psto) and triggers early canopy senescence (psen) were well selected. 346

347

By making use of the observed harvest index and a selection of WP* for both local and improved 348

varieties of tef, satisfactory grain yield simulation was observed (Figure 6) suggesting that 349

AquaCrop correctly simulated the effect of water stress during different phenological stages on 350

the HI of tef. Depending on the magnitude of water stress during yield formation, HIo in 351

AquaCrop might be adjusted up or downward. Stresses affecting only leaf expansion have a 352

positive effect on HI, while more severe stresses inducing stomata closure have a negative effect 353

on HI. In AquaCrop, HIo can also be adjusted to water stress occurring before and during 354

flowering. 355

For tef, water stress affected the HIo in the following way: 356

- As there was no or very mild water stress before flowering for all experimental years, the 357

positive effect of pre-flowering water stress on HIo of tef could not be considered; 358

- The data analysis revealed that tef is very tolerant to water stress during flowering. A high 359

depletion of TAW was selected (Pupper = 0.92) for the water stress affecting pollination during 360

flowering; 361

- Simulations with AquaCrop indicated for tef that slight to moderate water stresses during 362

yield formation increased HIo as a result of water stress affecting leaf expansion. This is due 363

to the restriction of excessive vegetative growth during this growth stage, which means that 364

more energy can be used for the production of yield; 365

- The drop of HI due to strong water stress during yield formation as reported among many 366

studies by Farré and Faci (2006) for maize and sorghum and by Karunaratne et al. (2010) for 367

Bambara groundnut is a common physiological process. Simulations indicated that this is 368

also present in tef for severe water stresses, although to a smaller extent. 369

370

371

14

Water productivity of tef revealed through AquaCrop 372

The manifest positive effect of water stress on HI during yield formation and the relatively small 373

negative effect during flowering and yield formation render tef to be a crop with unique 374

characteristics. This may be the reason why farmers in areas prone to moisture stress use tef as 375

a rescue crop in case of failure of other cereals (Seyfu, 1997). It indicates its ability to withstand 376

water stress, which is likely to occur when planting late in the rainy season after a crop failure. 377

From the statistical analysis of the simulation results (Table 3) it could be concluded that the 378

main features of tef as affected by water stress were well modelled by AquaCrop. R2 for all 379

variables was ≥ 0.84 and d approached 1. The relative small RMSE and the good range of EF (≥ 380

0.90, except for yield) confirmed the goodness of fit between the observed and simulated 381

results. The low EF value for the yield could be attributed due to the under-estimation of the 382

yield by the model for RF tef in Maiquiha 2008. This may indicate that the model is less 383

satisfactory in simulating crop growth and production in severe water stress conditions as was 384

the case at this site in the dry year 2008 or it could also be an error in the measurements. 385

386

The validated SWC in the root zone of the crop was in a good agreement with the observed 387

values. This indicated that the calibration of AquaCrop for tef was satisfactory and ETc was well 388

estimated. The model was able to appropriately simulate the development of CC and B. A 389

correct calibration of the effect of water stress on reference HI (HIo), by which the HI increased 390

from 27% in conditions without water stress to 33% in conditions with mild water stress, 391

resulted in an accurate simulation of Y (Figure 7). 392

CONCLUSION 393

The simulations of the soil water balance and the development of green canopy cover gave a 394

good indication that AquaCrop is able to calculate the crop transpiration (Ta) quite well. A good 395

estimation of Ta is crucial to estimate yields, since biomass yield follows directly from the 396

product of the WP* with the cumulative Ta adjusted for ET0. Notwithstanding the fact that tef is 397

a rather unknown crop and by considering the physiological complexity of the crop responses to 398

water stress, adequate results were obtained with AquaCrop with a relatively limited amount of 399

crop parameters. In the current study, it was found that the common biomass water 400

productivity for a C4 crop is far from reached in tef. 401

402

15

Alemtsehay Tsegay et al. 403

The simulation of tef reported in this paper extended the preliminary calibration of Araya et al. 404

(2010) for AquaCrop because it made a clear distinction between calibration and validation, and 405

used an extensive dataset of four years observations with three water treatments, in three 406

different locations and on different soil types. Further on, a distinction was made between local 407

varieties close to the wild type used by most farmers and improved tef varieties. 408

409

Given that AquaCrop keeps a good balance between accuracy and input requirements, it could 410

be used to improve tef crop water productivity in its centre of origin through developing 411

guidelines on the effective way of cultivating crops in terms of water management strategies. 412

Moreover as it is a useful tool to develop scenarios for future climatic conditions, estimated 413

yield gaps between the potential and actual yield of a particular crop under its specific climatic 414

and soil conditions, can be useful to increase awareness for timely and quantitative information 415

related to food relief measures. Additionally, AquaCrop can serve as an explanatory model in 416

breeding programs of tef. Nevertheless further studies have to be carried out to test the model 417

to other management practices including fertility stress effect, weed infestation effect and 418

effect of sowing dates, as these are also important factors that limit tef productivity. 419

420

Acknowledgements: The authors are grateful to Reggers, R., Viaene, N., Raes, W. and Garcia-Vila 421

for their valuable effort in the collection and analysis of field data. This research project was 422

funded by the Mekelle University-Institutional University Cooperation Program (MU-IUC, VLIR 423

UOS). 424

16

References 425

Allen, R. G., Pereira, L. S., Raes, D., and Smith, M. (1998). FAO Irrigation and Drainage 426

Paper 56. FAO, Rome: Italy. 427

428

Araya Alemie, Keesstra, S. D., and Stroosnijder, L. (2010). Simulating yield response to 429

water of Teff (Eragrostis tef) with FAO’s AquaCrop model. Field Crops Research 116:196–430

204. 431

432

Azam-Ali, S. N., and Squire, G. R. (2002). Principles of Tropical Agronomy. CABI Publishing, 433

CAB International, Wallingford: UK. 434

435

CSA (Central Statistical Agency of Ethiopia) (2010). Agricultural Sample survey 2009/2010. 436

Report on area and production of major crops (Meher season), Statistical Bulletin 446. 437

Addis Ababa: Ethiopia. 438

439

CSA (Central statistical Agency of Ethiopia) (2000-2010). Agricultural Sample survey 2000-440

2010. Report on area and production of major crops (Meher season), Statistical Bulletin 441

227, 245, 302, 331, 361, 388, 417 and 446. Addis Ababa: Ethiopia. 442

443

Dejene K. Mengistu. (2009). The influence of soil water deficit imposed during various 444

developmental phases on physiological processes of tef (Eragrostis tef). Agriculture, 445

Ecosystems and Environment 132:283–289. 446

447

EARO (Ethiopian Agricultural Research Organization). (2002). Crop Research Directorate: 448

Research Recommendations for Improved Crop Production. Addis Ababa: Ethiopia. 449

450

Farré, I. and Faci, J. M. (2006). Comparative response of maize (Zea mays L.) and sorghum 451

(Sorghum bicolor L. Moench) to deficit irrigation in a Mediterranean environment. 452

Agricultural Water Management 83:135–143. 453

454

17

Garcia Vila, M., Fereres, E., Mateos, L., Orgaz, F., and Steduto, P. (2009). Deficit Irrigation 455

Optimization of Cotton with AquaCrop. Agronomy Journal 101:477-487. 456

457

Geerts, S., Raes, D., Garcia, M., Miranda, R., Cusicanqui, J., Taboada, C., Mendoza, J., 458

Huanca, R., Mamani, A., Octavio, C., Mamani, J., Morales, B., Osco, V., and Steduto, P. 459

(2009). Simulating yield response of quinoa (Chenopodium quinoa Willd.) to water 460

availability with AquaCrop. Agronomy Journal 101:499-508. 461

462

Geerts, S., and Raes, D. (2009). Deficit irrigation as an on-farm strategy to maximize crop 463

water productivity in dry areas. Agricultural Water Management 96:1275-1284. 464

465

Heng, L. K., Hsiao, T., Evett, S., Howell, T., and Steduto, P. (2009). Validating the FAO 466

AquaCrop Model for Irrigated and Water Deficient Field Maize. Agronomy Journal 467

101:488–498. 468

469

Hsiao, T.C., Heng, L., Steduto, P., Rojas-Lara, B., Raes, D., and Fereres, E. (2009). AquaCrop-470

The FAO Crop Model to Simulate Yield Response to Water: III. Parameterization and 471

Testing for Maize. Agronomy Journal 101:448–459. 472

473

Karcher, D. E., and Richardson, M. D. (2005). Batch Analysis of Digital Images to Evaluate 474

Turfgrass Characteristics. Crop Science 45:1536–1539. 475

476

Karunaratne, A.S., Azam-Ali, S.N., Al-Shareef, I., Sesay, A., Jørgensen, S.T., Crout, N.M.J. 477

(2010). Moddeling the canopy development of bambara groundnut. Agricultural and 478

Forest Meteorology 150 (7-8):1007-1015. 479

480

Loague, K., and Green, R. E. (1991). Statistical and graphical methods for evaluating solute 481

transport models: Overview and application. J. Contam. Hydrol 7:51–73. 482

Nash, J.E. and Sutcliffe, J.V. (1970). River flow forecasting through conceptual models. Part 483

1- a discussion of principles. Journal of Hydrology 10(3):282-290. 484

18

485

NRC (National Research Council), Board on Science and Technology for International 486

Development (1996). Lost Crops of Africa. Volume I: Grains. National Academy Press, 487

Washington DC: USA. 488

489

Nyssen, J., Vandenreyken, H., Poesen, J., Moeyersons, J., Deckers, J., Mitiku Haile, Salles, 490

C., and Govers, G. (2005). Rainfall erosivity and variability in the Northern Ethiopian 491

Highlands. Journal of Hydrology 311:172-187. 492

493

Raes, D., Steduto, P., Hsiao, T. C., and Fereres, E. (2010). Reference manual AquaCrop 494

Version 3.1. Web site http://www.fao.org/nr/water/aquacrop.html 495

496

Raes, D., Steduto, P., Hsiao, T. C., and Fereres, E. (2009). AquaCrop-The FAO crop model to 497

predict yield response to water: II. Main algorithms and software description. Agronomy 498

Journal 101:438–447. 499

500

Saxton, K. E., and Rawls, W. J. (2006). Soil water characteristic estimates by texture and 501

organic matter for hydrologic solutions. Soil Sci. Soc. Am. J 70:1569–1578. 502

503

Seyfu Ketema. (1997). Tef (Eragrostis tef (Zucc.) Trotter). Promoting the conservation and 504

use of underutilized and neglected crops. 12. Institute of Plant Genetics and Crop Plant 505

Research, Gatersleben/International Plant Genetic Resources Institute, Rome: Italy. 506

507

SPSS Inc. (1990-2000). Sigma Scan Pro 5.0. SPSS Science Marketing 508

Department.Chicago:USA. 509

510

Steduto, P., Hsiao, T. C., Raes, D., and Fereres, E. (2009). AquaCrop-The FAO crop model to 511

predict yield response to water: I. Concepts and Underlying Principles. Agronomy Journal 512

101:426-437. 513

514

Steduto, P., Hsiao, T. C., and Fereres, E. (2007). On the conservative behavior of biomass 515

water productivity. Irrigation Science 25:189–207. 516

19

517

Steduto, P. and Albrizio, R. (2005). Resource use efficiency of field-grown sunflower, 518

sorghum, wheat and chickpea II. Water use efficiency and comparison with radiation use 519

efficiency. Agricultural and Forest Meteorology 130:269–281. 520

521

Vavilov, N.I. (1951). The Origin, Variation, Immunity and Breeding of Cultivated plants. The 522

Cronica Botanica Co., Stechert-Hafner, Inc., New York: USA. 523

524

Willmott, C.J. 1982. Some comments on the evaluation of model performance.Bull. 525 Am Meteorol. Soc. 63:1309–1313. 526

20

FIGURE CAPTIONS 527

528

Figure 1. Water productivity of tef (local variety) from the lysimeter experiment based on 529

cumulative normalized evapotranspiration versus dry aboveground biomass. Error bars 530

represent standard deviations (n=3). 531

532

Figure 2. Observed (symbols with error bars of standard deviation (n=3)) and simulated (lines) 533

canopy cover for rainfed (solid squares, full line) and fully irrigated (open circles, dotted line) 534

treatments for (a) Mekelle University campus 2007, (b) Dejen 2008. 535

536

Figure 3. Observed (symbols with error bars of standard deviation (n=3)) and simulated (lines) 537

soil water balance for rainfed (solid squares, full line) and fully irrigated (open circles, dotted 538

line) treatments for Mekelle University campus 2007. 539

540 Figure 4. Water productivity of tef grown under field conditions based on cumulative normalized 541

transpiration versus dry aboveground biomass for improved variety (open circles, dotted trend 542

line) and local variety (solid squares, full trend line). Error bars represent standard deviations 543

(n=3). 544

545

Figure 5. Observed (symbols with error bars of standard deviation (n=3)) and simulated (lines) 546

dry aboveground biomass for rainfed (solid squares, full line) and fully irrigated (open circles, 547

dotted line) treatments for (a) Mekelle University campus 2007 and (b) Dejen 2008. 548

549

Figure 6. Regression between observed and simulated yield (solid diamonds, R2= 0.84) and final 550

dry aboveground biomass (open circles, R2=0.96) with error bars of standard deviation (n=3)) for 551

all calibrated treatments. 552

553

Figure 7. Regression between observed and simulated yield (solid diamonds, R2=0.87) and final 554

dry aboveground biomass (open circles, R2= 0.77) for the validation fields. Error bars represent 555

standard deviations (n=3). 556

557 558 559

(a) (a) (a)

21

TABLE CAPTIONS 560

Table 1. Field experiments for the calibration and validation of AquaCrop for tef. 561

Table 2. Soil physical parameters at various depths for the experimental sites. 562

Table 3. Goodness of fit analysis for the simulation of the soil water content (SWC), the canopy. 563

cover (CC), the biomass (B; which include intermediate and final biomass), final biomass (Final B) 564

and yield (Y) after calibration of AquaCrop for tef. 565

Table 4. Calibrated crop parameters for tef in AquaCrop. 566

Table 5. Goodness of fit analysis for the simulation of the soil water content (SWC), the canopy 567

cover (CC), the biomass (B; which include intermediate and final biomass), final biomass (Final B) 568

and yield (Y) for the validation of AquaCrop for tef. 569

570

LIST OF TABLES 571

572

Table 1. Field experiments for the calibration and validation of AquaCrop for tef. 573

574 1 FI: full irrigation; RF: Rainfed; DI: deficit irrigation: irrigated during flowering 575

2 Improved variety 576 577 578 579

Year Location Treatment1 Code Variety Sowing date

Harvesting date Rainfall growing season (mm)

Irrigation (mm)

Calibration 2007 Mekelle University

Campus FI MUFI07 DZ-cr-372 July 30 November 16 242 120

2007 Mekelle University Campus

RF MURF07 DZ-cr-372 July 30 November 6 242 0

2008 Dejen FI DEFI08 Local July 25 November 10 275 100

2008 Dejen RF DERF08 Local July 25 October 29 275 0

2008 Maiquiha FI MAFI08 Local August 18

December 5 142 180

Validation

2006 Mekelle University Campus

DI MUDI06 Local August 7 November 22 309 66

2006 Mekelle University Campus

RF MURF06 Local August 7 November 22 309 0

2009 Dejen FI DEFI09 Local August 3 November 12 174 139 2009 Dejen RF DERF09 Local August 3 October 28 174 0 2009 Maiquiaha FI MAFI09 Local July 31 November 10 180 124 2009 Maiquiaha RF MARF09 Local July 31 November 3 180 0

22

Table 2. Soil physical parameters at various depths for the experimental sites. 580 Experimental site Soil depth SAT FC PWP Ksat Texture class

m Vol% Vol% Vol% mm/day

Dejen 2008; Irrigated fields

0.2 50.0 36.9 8.5 275 Silt loam 0.4 50.0 39.9 8.5 275

0.6 50.0 36.9 8.5 275 Dejen 2008; Rainfed fields

0.2 41.3 28.9 8.0 241 Loam 0.4 41.3 28.9 8.0 241

0.6 41.3 28.9 8.0 241 Mekelle University campus 2007; All fields

0.2 45.8 36.6 20.0 46 Clay loam 0.4 45.8 36.6 20.0 46

0.6 45.8 36.6 20.0 46 Maiquiha 2008/2009; All fields

0.2 60.0 38.0 16.9 150 Silt loam 0.4 60.0 38.0 16.9 150

Dejen 2009; All fields

0.2 0.4 0.6

41.0 46.0 46.0

27.0 22.5 20.4

9.1 12.6 12.0

500 250 250

Sandy loam Loam Loam

Mekelle University campus 2006; All fields

0.1 50.7 34.6 21.8 89 Silty clay 0.2 51.0 35.5 23.3 74

0.3 51.7 36.1 24.1 71

581 Table 3. Goodness of fit analysis for the simulation of the soil water content (SWC), the canopy 582 cover (CC), the biomass (B; which include intermediate and final biomass), final biomass (Final B) 583 and yield (Y) after calibration of AquaCrop for tef. R² is the coefficient of determination, EF the 584 Nash-Sutcliffe efficiency, d the index of agreement, and RMSE the root mean squared error (all 585 values averaged over different fields and for the time series of SWC, CC and B). 586 Parameter R2 EF d RMSE

SWC 0.92 0.91 0.98 14.42 (mm)

CC 0.91 0.90 0.97 10.75 (%)

B 0.90 0.90 0.97 0.89 (t/ha)

Final B 0.96 0.95 0.99 0.72 (t/ha)

Y 0.84 0.16 0.88 0.54 (t/ha)

587 Table 4. Calibrated crop parameters for tef in AquaCrop 588

Description value Unit

Days to emergence 7 Days after sowing(DAS)

Initial canopy cover (CCo) 2.6 %

Plant density 2000 m2

Canopy growth coefficient (CGC) 14.6 %/day

Time to maximum canopy cover 49 DAS

Maximum canopy cover (CCx) 73-99 %

Maximum rooting depth 0.6 M

Time to reach maximum rooting depth 50 DAS

Time to flowering 52 DAS

Duration of flowering 11 days

Length to build up HI 40 DAS

Time to start of canopy senescence 75 DAS

23

Canopy decline coefficient (CDC) 11.6 %/day

Time to maturity 95 DAS

Crop transpiration coefficient (Kcbx) 1.1 -

Water stress response factors

Soil water depletion threshold for leaf expansion, upper limit 0.32 fraction TAW

Soil water depletion threshold for leaf expansion, lower limit 0.66 fraction TAW

Shape factor for water stress coefficient for leaf expansion 3 -

Soil water depletion threshold for stomatal control, upper limit 0.60 fraction TAW

Shape factor for water stress coefficient for stomatal control 3 -

Soil water depletion threshold for canopy senescence, upper limit 0.58 fraction TAW

Shape factor for water stress coefficient for canopy senescence 3 -

Soil water depletion threshold for failure of pollination, upper limit 0.92 fraction TAW Coefficient describing positive impact of restricted vegetative growth during yield formation on HIo

0.5 (very strong effect) -

Coefficient describing negative impact of stomatal closure during yield formation on HIo

10 (small effect) -

Reference harvest index (HIo) 27 % Water productivity normalized for climate (WP*) 14 (local varieties)

21 (improved varieties) g/m2

589 Table 5. Goodness of fit analysis for the simulation of the soil water content (SWC), the canopy 590 cover (CC), the biomass (B; which include intermediate and final biomass), final biomass (Final B) 591 and yield (Y) for the validation of AquaCrop for tef. R² is the coefficient of determination, EF the 592 Nash-Sutcliffe efficiency, d the index of agreement and RMSE the root mean squared error (all 593 values averaged over different fields and for the time series of SWC, CC and B). 594 595

Parameter R2 EF d RMSE

SWC 0.85 0.84 0.95 18.27(mm)

CC 0.75 0.75 0.93 16.75 (%)

B 0.65 0.53 0.86 1.16 (t/ha)

Final B 0.77 0.63 0.91 0.76 (t/ha)

Y 0.87 0.27 0.89 0.45 (t/ha)

FigureClick here to download high resolution image

FigureClick here to download high resolution image

Fig 5 and 6Click here to download high resolution image

fig 7Click here to download high resolution image

1

1

LIST OF FIGURES 2

3

y = 15.501x - 24.172

R2 = 0.9852

0

200

400

600

800

1000

1200

1400

0 20 40 60 80

∑(ETa/ETo)

Dry aboveground biomass (g/m

2)

4

Figure 1. Water productivity of tef (local variety) from the lysimeter 5

experiment based on cumulative normalized evapotranspiration versus dry 6

aboveground biomass. Error bars represent standard deviations (n=3). 7

8

9

0

20

40

60

80

100

0 20 40 60 80 100

Days after sowing

Canopy cover (%

) (a)

0

20

40

60

80

100

0 20 40 60 80 100

Days after sowing

Canopy cover (%

) (b)

10

Figure 2. Observed (symbols with error bars of standard deviation (n=3)) 11

and simulated (lines) canopy cover for rainfed (solid squares, full line) and 12

fully irrigated (open circles, dotted line) treatments for (a) Mekelle University 13

campus 2007, (b) Dejen 2008. 14

15

List of figures in pdf format

2

0

50

100

150

200

250

300

0 20 40 60 80 100

Days after sowing

Soil water content in the total root

zone of 0.6m (mm)

Wilting point

Field capacity

16

Figure 3. Observed (symbols with error bars of standard deviation (n=3)) 17

and simulated (lines) soil water balance for rainfed (solid squares, full line) 18

and fully irrigated (open circles, dotted line) treatments for Mekelle 19

University campus 2007. 20

21

y = 13.31x

R2 = 0.92

y = 18.53x

R2 = 0.92

0

300

600

900

1200

1500

1800

0 30 60 90

∑(Ta/ETo)

Dry aboveground biomass (g/m

2)

22

Figure 4. Water productivity of tef grown under field conditions based on 23

cumulative normalized transpiration versus dry aboveground biomass for 24

improved variety (open circles, dotted trend line) and local variety (solid 25

squares, full trend line). Error bars represent standard deviations (n=3). 26

27

3

0

2

4

6

8

10

12

14

16

0 20 40 60 80 100

Days after sowing

Dry

abovegro

und b

iom

ass

(ton/h

a)

(a)

0

2

4

6

8

10

12

14

16

0 20 40 60 80 100

Days after sowing

Dry abovegro

und biom

ass

(ton/h

a)

(b)

28

Figure 5. Observed (symbols with error bars of standard deviation (n=3)) 29

and simulated (lines) dry aboveground biomass for rainfed (solid squares, full 30

line) and fully irrigated (open circles, dotted line) treatments for (a) Mekelle 31

University campus 2007 and (b) Dejen 2008. 32

33

0

3

6

9

12

15

0 3 6 9 12 15

Observed (t/ha)

Simulated (t/ha)

34

35

Figure 6. Regression between observed and simulated yield (solid diamonds, 36

R2= 0.84) and final dry aboveground biomass (open circles, R2=0.96) with 37

error bars of standard deviation (n=3)) for all calibrated treatments. 38

4

0

3

6

9

0 3 6 9

Observed (t/ha)

Simulated (t/ha)

39

Figure 7. Regression between observed and simulated yield (solid diamonds, 40

R2=0.87) and final dry aboveground biomass (open circles, R2= 0.77) for the 41

validation fields. Error bars represent standard deviations (n=3). 42