Abstract

Eija Bakken

Soil erosion under the Mediterranean conditions has been mainly attributed to the climate and the vegetative cover. This is because the vegetation affects both the amount and the frequency of the overland flow occurrence, and the resultant soil erosion.

Under the hilly and semi-arid areas of the eastern part of the Mediterranean, intense rain events are not unusual. Such rain events can cause appreciable runoff and soil erosion. In that area, farmers are adopting an intense network of stonewalled terraces to counteract the effects of such erratic rain events, in addition to conserve more soil moisture for attaining proper plant growth with higher production.

The main objectives of this study were to quantify the runoff and the associated soil loss under a terraced system compared to a non-terraced system. The effect of plant canopy as well as the socio-economic impacts of such soil-moisture conservation was also studied. Due to the lack of basic data on soil erosion in the study area, the 2-years experimental data were used for a preliminary assessment of the risk of erosion, both on micro-scale (plot level) as well as macro-scale (watershed level), utilizing the Revised Universal Soil Loss Equation (RUSLE) and the Geographic Information System (GIS).

To achieve the aforementioned objectives, a field plot experiment was conducted to study the effect of old stonewalled terraces on soil and moisture conservation compared to the noterraces as a control. The effects of the wheat canopy were considered as a second treatment. The experiment was conducted during the winter seasons of 2000/ 2001. The soil is classified according to USDA as Lithic Xerorthent loamy, mixed, thermic with the prevalence of Thermic and Xeric soil temperature and moisture regimes, respectively.

The results showed that erosion processes were mostly transport-limited, only 6.5% of the rain events were detachment-limited due to their erratic nature. In addition, rainfall events with amount > 60 mm and with I30 ≥ 12 mm h^-1 would shift erosion process from the transport to the detachment-limited phase. It was also noticeable that, under saturation conditions, rainstorms with intensity ≥ 4 mm h^-1 and rainfall ≥ 10 mm are more likely to cause runoff and erosion.

Significant reduction (P<0.05) in soil erosion was obtained in the terraced plots than in the non-terraced (182 kg ha^-1 and 3525 kg ha^-1 during 2000, 1769 kg ha^-1 and 5057 kg ha^-1 during 2001 for terraced and non-terraced plots, respectively). Similar trend was applied for runoff in areas under the same treatments. The effect of wheat canopy was not significant, although it showed lower runoff and erosion in most of the cases and during both seasons. As a result of more soil and water conservation, significant increase in the total plant dry matter was observed in the terraced plots than the non-terraced plots (1570 and 630 kg ha^-1 in 2000, 2545 and 889 kg ha-1 in 2001 for terraced and non-terraced treatment, respectively). This increase in productivity was associated with a higher net profit for the terraces than areas with no terraces (3.5-6 times higher net profit).

Analysis of the cause of the differences in soil erosion and runoff between terraced and nonterraced plots showed that organic carbon, Ca and K had positive effects on soil aggregate stability and hence, on soil erosion and runoff. The effect of organic matter on aggregate stability was following up a threshold level (> 1.3%) below which the effect was negligible. It was also found that runoff, interrill and splash erosion can be estimated by the application of different stepwise multiple regression relationships (r= 0.87, 0.87, 0.93 for runoff, interrill, 2 and splash erosion, respectively), using rainfall kinetic energy, storm duration, EI30 and I30, with more pronounced effect of kinetic energy (r=0.93) than the other factors.

Application of the Revised Universal Soil Loss Equation (RUSLE), using the available experimental data and on a micro scale (plot) level, showed a higher estimation of soil loss by RUSLE than the actual soil loss (7.8 and 2.6 Mg ha^-1 , for RUSLE and actual measured soil loss, respectively). The reason for this difference in soil loss was attributed to the differences between some of the RUSLE factors and those related to the measured ones. Specifically, theRUSLE K-factor showed three times the magnitude of the measured K-factor, the RUSLE Cfactor underestimated the measured C-factor, and the RUSLE P-factor overestimated the measured P-factor by three times. Adjustment of these factors according to the measured ones increased the model’s predictability and efficiency by three times (model efficiency was 0.26 and 0.86 before and after adjustment of the model respectively). Application of the RUSLE on a watershed scale, using the Geographic Information System (GIS), showed a
good efficiency of predicting the average annual soil loss (R2= 0.68), where the model revealed that 24% of the watershed area has an annual soil loss that exceeded the soil loss tolerance of the area (5 Mg ha-^-1 ). When all the RUSLE factors were kept constant, the potential for soil loss was the highest for the olive plantation (20-30 Mg ha^-1 ), whereas the wheat and barley had the lowest soil loss potential (5 Mg ha^-1 ). This necessitated changes in the current practices used in the olive plantation, i.e. reducing the number of tillage and/ or time of tillage practices. Generally speaking, application of the RUSLE, whether on micro
(plot) scale or macro (watershed) scale provided a strong basic tool for future planning, management and conservation of the soil resource. The model also comprises an efficient, eproducible, easy to apply, time and money conserving method of assessment, which can be easily applied in areas with similar agro-climatic conditions. Nevertheless, and whether the RUSLE will be used on micro or macro scale, it is necessary to conduct long-term soil loss experimentation and easurements for more accurate and reliable validation of the RUSLE under the Mediterranean conditions.

Correlation analysis indicated significant relationships (P<0.05) between the farmers’ perceptions, land ownership, geomorphology and the incentives to farmers from one side,
with the farmers’ willingness to adopt soil conservation measures from the other side, although the correlations were in the negative direction. Smallholder farmers (52% of the
interviewed farmers) were involved in selling their agricultural land for urban uses, mainly because of the good price and the immediate economic returns offered. The associated land
use changes warranted the need for intensive involvement of both the private and public sectors. This interference can be done through the introduction of a long-term agricultural
loan system, the development of proper legislations accompanied with a comprehensive and durable infrastructure and service system, aiming at the reduction of the negative impact of land use changes and sustainable use of the available resources.

Keywords: Terraces; Soil erosion; Rainfall intensity; Wheat canopy, Interrill erosion; Splash;
Aggregate stability; RUSLE; model efficiency; Socio-economy; Incentives; Tenure size.





Oppdatert: 02.12.04
Utskriftsvennlig versjon : Arkiv : Tips en venn