Calculations of normalized RMSE, as computed by equation [6], yielded 1.8 and 2.3 when comparing SWAP-simulated ETC with weather-based maximum ETC and water-balance based ETC. According to Timsina and Humphreys (2006), the values for normalized RMSE can be considered as relatively high, indicating unreliable modelling performance. Considering only simulated and observed pairs with the water-balance measured ETC below 30 mm yields a normalised RMSE of 0.9, almost three times lower than considering the higher ETC values.

According to the simulations, actual sugarbeet evapotranspiration was close to maximum evapotranspiration throughout the crop season. Furthermore, the irrigation managements at both plots were enough to keep the soil water content above the field capacity, as pointed out above. Accordingly, the simulated relative yields were 0.98 at the A plot and 0.92 at the Z plot. This means that the yields obtained with the said irrigation managements are close to the maximum, i.e. 87 and 82 t/ha on the A and Z plots, respectively. However, the simulated water loss due to percolation was relatively high: approximately 10% of the irrigation water applied. This over-irrigation could be useful in the case of saline soils. However, soil salinisation is not a main concern in the Duero basin (JCYL, 1987). The results indicate that sugarbeet irrigation management can still be improved in the area, as pointed out by Playan and Mateos (2005) and many others.

4.5.4 Simulating Current and Future Sugarbeet Water-Use

The baseline 1960-1990 and the 2010-2040 Climate Change scenarios were taken from the CGCM2 model outputs, provided by the Canadian Centre for Climate Modelling and Analysis (Flato et al., 2000; Flato and Boer, 2001). The IPCC SRES A2 scenario for greenhouse gases emissions (IPCC, 2001) was considered. Despite other global circulation models, CGCM2 provides free internet access to daily simulation data in a text format. Hence, this model is more suitable for simple agricultural applications anywhere. According to Merrit et al (2006), results considering CGCM2 are similar than those obtained through other general circulation models.

A historical meteorological series of Valladolid (41.7° N, 4.85° W), comprising daily data from 1970 to 2005 of maximum and minimum temperatures, sunshine hours and precipitation; was used in combination with the LARS-WG weather generator (Semenov and Barrow, 2002) to generate 100 realizations of local weather corresponding to 2025, approximately.

The weather generator realizations were perturbed according to the CGCM2 results corresponding to the study site, i.e., Northeast of Iberian Peninsula. The relative change in wet and dry series lengths, as affected by global change, was done following the approach recommended by Semenov and Barrow (2002), based on the daily CGCM2 outputs for each ten-year range. The relative changes in temperature standard deviations, as well as relative changes in mean temperature, precipitation amount and solar radiation were obtained from the CGCM2 daily estimations, as suggested by Semenov and Barrow (2002).

Besides, 100 realizations were also obtained, without perturbing the weather generators. Such data was representative of current climate conditions, for the 1970-2005 period. The Priestley and Taylor (1972) equation for computing the maximum evapotranspiration was used instead of the recommended Penman-Monteith approach. Penman-Monteith computations involve meteorological variables that are not included in GCM and downscaling assessments. The Priestely and Taylor approach, however, needs only maximum and minimum temperatures, as well as global radiation. Those variables can be obtained from GCM and are usually considered in the available weather generators. Besides, many crop-growths oriented models, as DSSAT, uses this approach to compute evapotranspiration (Ritchie, 1998) and most of the studies addressed to estimate climate-change effects on Spanish agriculture (Guereña et al., 2001; Minguez et al., 2005). Furthermore, according to Utset et al. (2004), both approaches are statistically equivalent when considered to simulate water managements through SWAP.

Both climate data, representing current and 2025 climate conditions, were used as input in the SWAP model, considering the sugarbeet calibration parameters described above and shown by Utset et al. (2007b). A typical irrigation management, as conducted by farmers in the zone (Utset et al., 2007b) was considered. The soil hydraulic properties estimated for a Cambisol at Valladolid province were used in SWAP simulations.

Figure 11 depicts the average components of the simulated water balance in the sugarbeet plot, according to the irrigation management considered, for the current and the 2025 climate conditions.

Building Your Own Greenhouse

Building Your Own Greenhouse

You Might Just End Up Spending More Time In Planning Your Greenhouse Than Your Home Don’t Blame Us If Your Wife Gets Mad. Don't Be A Conventional Greenhouse Dreamer! Come Out Of The Mould, Build Your Own And Let Your Greenhouse Give A Better Yield Than Any Other In Town! Discover How You Can Start Your Own Greenhouse With Healthier Plants… Anytime Of The Year!

Get My Free Ebook

Post a comment