水资源1

e w MK te B a g r i c u l t u r a l wa t e r man a g em en t 9 6 ( 2 0 0 9 ) 2 2 6 – 2 3 2 Received 10 July 2007 Accepted 5 August 2008 Published on line 26 September 2008 uction in Botswana is irrigated, as average rainfall is less than average reference evapotranspiration in all parts of the country for all of the year, and most of the irrigation in the country is used for horticultural crops. In the main, the water used for irrigation comes from groundwater, and much of this can be classified as moderately saline. avai lable at www.sc iencedi rec t .com .e l 1. Introduction Botswana is a landlocked country located between 18–278S and 20–308E in southern Africa with an average elevation of 1000 m above sea level. The climate is semi-arid with a cool dry season between May and September and a hot wet season between October and April. The average annual precipitation of 450 mm is unreliable and unpredictable, whilst the average annual reference evapotranspiration is 1400 mm and exceeds average precipitation in every month (FAO, 1984). About 85% of the land is covered by the Kalahari sands and shrub savannah with the driest region towards the south characterised by active sand dunes and very sparse vegeta- tion. Water resources are extremely scarce in many parts of the country. The potential of using surface water for irrigation development in Botswana is limited as all the rivers within the country are either dammed, or planned for damming, for domestic water use or livestock watering. In 2002, only 1439 ha were irrigated, of which only 620 ha were irrigated in the dry season (FAO, 2005). Almost all irrigation is for horticultural crops (FAO, 2005) – mainly varieties of brassica, tomatoes (Lycopersicon esculentum) and onions (Allium cepa) – and irrigation is essential for vegetable cropping (Bok et al., 2006). Most irrigation uses groundwater from deep (>10 m below ground level) aquifers (Department of Water Affairs, Pers. Commun.) with poor recharge and moderate levels of salinity (Tahal Consulting Engineers, 2000). The country has the potential to produce 75% of its national demand of horticultural produce but produces only 20% (Tahal Consulting Engineers, 2000). Cabbage (Brassica oleracea var. capitata) is grown in two cropping seasons. A crop may be planted between March and May so that it grows through the winter months (May–August) and a summer crop may be planted between October and Keywords: Irrigation Scheduling Salinity Cabbage Botswana A survey was carried out of 60 cabbage farmers in five regions of the country to identify the level of salinity of irrigation water and the irrigation practices used. It was concluded that there was no relationship between the irrigation rate (average irrigation depth divided by average irrigation interval) and the salinity of the water used or the salinity of the soil, such that some farmers were over-irrigating and others under-irrigating. This means that water is being wasted and yields are likely to be reduced. In addition, there is a risk of long-term build-up of salinity and soil degradation. # 2008 Elsevier B.V. All rights reserved. * Corresponding author. Tel.: +44 1234 750111; fax: +44 1234 752970. E-mail address: t.hess@cranfield.ac.uk (T.M. Hess). 1 Tel.: +267 328780. 0378-3774/$ – see front matter # 2008 Elsevier B.V. All rights reserved. Irrigation management practic Botswana using saline ground T.M. Hess a,*, G. Molatakgosi b,1 aSchool of Applied Sciences, Cranfield University, Cranfield, Bedford bDepartment of Agricultural Research, Ministry of Agriculture, Priva a r t i c l e i n f o Article history: a b s t r a c t All horticultural prod journa l homepage: www doi:10.1016/j.agwat.2008.08.005 s of cabbage farmers in ater 43 0AL, UK ag 0033, Gaborone, Botswana sev ier .com/ locate /agwat Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 打字机 只有22%的农民在不同的季节改变灌溉措施 November. Most that grow cabbage also grow other vegetables alongside or in rotation. Cabbage is moderately sensitive to the salinity of irrigation water. Ayers and Westcot (1985) suggest a threshold electrical conductivity of irrigation water, ECw, of 1.2 dS m �1 for maximum yield and Beltra˜o et al. (2000) found that yields decreased linearly with increasing ECw above 2 dS m �1. Maggio et al. (2005) and Jamil et al. (2005) both observed significant reductions in head fresh weight with irrigation water with an ECw around 4 dS m �1 associated with reduced leaf area and above ground dry matter. and Grattan, 1999). FAO and Ministry of Agriculture (1998) brated against the other four stations (see Allen et al., 1998 for methods). The crop water requirement of cabbage at full ground cover, ETc, was estimated using assuming a crop coefficient of 1.05 (Allen et al., 1998). 2.2. Farmer survey An on-farm survey was conducted between June and Septem- ber 2004 in five of the six agricultural regions of the country. The farmers selected all grew cabbage using irrigation water drawn from boreholes. Information was collected on of water collected was at least 50 ml. The time taken to collect ltu Me a g r i c u l t u r a l wa t e r man a g em en t 9 6 ( 2 0 0 9 ) 2 2 6 – 2 3 2 227 concluded that inadequate investigation and/or poor design has resulted in inefficient operation and the use of saline water in some irrigation schemes. Many farmers have limited irrigation management experience and are poorly informed about irrigated agriculture. This paper reports on a survey of cabbage farmers in Botswana that aimed to identify whether, and how, cabbage farmers in Botswana adapt their irrigation management according to the salinity of the water used. 2. Methodology 2.1. Climate Climate data from Gaborone, Jwaneng, Mahalapye, Francis- town and Maun were used to represent the climate of each region (Table 1). Average monthly reference evapotranspira- tion, ETo, was calculated from average monthly met data from the FAO CLIMWAT database with the Penman–Monteith method, except for Jwaneng, where average monthly max- imum and minimum temperatures were used with the Hargreaves method for reference evapotranspiration, cali- Table 1 – Number of farmers, dates during which each agricu five agricultural regions in Botswana Agricultural region Dates of visit (2004) Number of farmers Name Gaborone 14–26 June 10 Gaborone Southern 5–10 July 10 Jwaneng Central 2–14 August 20 Mahalapye Francistown 16–21 August 10 Francistown If salts added to the soil through irrigation water are not adequately leached, the level of soil salinity will gradually increase. The salinity of the soil is usually expressed as the electrical conductivity of the saturation extract, ECe, and Maas and Grattan (1999) suggest a threshold value of 1.8 dS m�1 for maximum yield of cabbage. Above this threshold, head fresh weight was estimated to decline at a rate of 9.7% dS m�1 (Maas and Grattan, 1999). In Botswana, rainfall is insufficient to leach salts from the soil profile. For sustainable production, additional irrigation water over and above the crop water requirements should be applied according to the salinity of the irrigation water in order to avoid excessive accumulation of salts in the soil (Maas North West 6–11 September 10 Maun ral region was visited for interview and average climate for t station Annual rainfall (1989–2002) Annual ETo Location Alt (m) Mean (mm) CV (%) Mean (mm) 24.408S, 25.558E 994 525.7 36 1536.6 24.608S, 24.668E 1189 443.5 34 1593.6 23.058S, 28.488E 1006 448.2 29 1567.1 21.138S, 27.308E 1000 459.5 40 1689.8 and 19.5 the last laterals. Cans were left to stand until the amount wer Where the farmers were using drip irrigation, catch-cans e placed under the first and last emitters of the first, middle take n to collect the water in each can was recorded. wat 1) Farm identity: name, size, location and years of operation. 2) Crop: crops grown, depth and methods of cultivation and area allocated to cabbage. 3) Irrigation: method, duration and interval. 4) Soil and water salinity: Electrical conductivity of irrigation water, ECw and root zone soil water, ECs. Farmer’s perception of the salinity. 2.3. Irrigation practices The typical application rate and depth applied was assessed in the field according to the irrigation system used. Hand irrigation refers to a system whereby farmers apply water using portable containers (e.g. watering cans, buckets). The depth of water applied was estimated from the volume of the container used for irrigation, the area of the bed and the number of containers applied at each irrigation. Where the farmer was using a number of sprinklers on a lateral, five catch-cans were placed at equal intervals (of atmost 3 m) depending on the sprinkler spacing, between two sprinklers. The catch-cans were left until each one of them had collected at least 50 ml. The volume of water collected in each can and the time taken were recorded. Where the farmer was using a single sprinkler and moving it around, the radius of the wetted circle was measured. Five catch-cans were placed at equal intervals in the direction of the sprinkler move from the sprinkler to the end of the wetted soil. The cans were left in place and the sprinkler moved as normal until the sprinkler had past all the cans and no more er was collected in any one of them. The volume and time 98S, 23.258E 994 407.5 33 1671.8 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 打字机 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 打字机 注意分析水质 Administrator 打字机 Administrator 打字机 Administrator 打字机 ce a g r i c u l t u r a l wa t e r mana g em en t 9 6 ( 2 0 0 9 ) 2 2 6 – 2 3 2228 the water was noted and the irrigation depth at each irrigation was estimated. Inorder tocompare irrigation systems, the average irrigation rate was estimated as the amount of water applied at each irrigation divided by the average interval between irrigations. 2.4. Soil and water salinity ECw was measured using a portable conductivity meter. For hand irrigation, water was collected from the water source whereas for drip and sprinkler irrigation, the water collected in the catch-cans was tested. Three different positions in the field were chosen for soil sampling. The first position was within the plant row, the second one between the rows and the last one at the edge of the cropped area. At each position, a hole was dug to 0.5 m (estimated as the effective rooting depth). Soil samples were taken at the soil surface, 0.12, 0.25, 0.37, and 0.5 m below the Fig. 1 – Average (1989–2002) monthly rainfall and referen soil surface. About 300 g of soil from each depth was collected in a sampling bag. The soil textural class was identified by feel. The soil samples were taken to the laboratory, a saturation extract prepared (U.S. Salinity Laboratory Staff, 1954) and ECe was measured using a conductivity meter. 2.5. Statistical analysis All attribute data analysed in the project was tested using the chi-square test. Unless otherwise stated, all the measurement data was analysed using the Duncan multiple range test (DMRT), and all the error bars in the figures represent the range of 95% confidence interval. All tests were done at the alpha level of 0.05. 3. Results 3.1. Climate Average long-term (1989–2002) monthly rainfall and reference evapotranspiration are shown in Table 1. All the regions are characterised by a summer rainy season from October to March and a dry winter. At all the stations, average precipitation is lower than average reference evapotranspira- tion throughout the year (Fig. 1). 3.2. Cabbage cropping In all regions, most farmers were growing cabbage in the winter and only 18% of the farmers interviewed also grew cabbage in the summer. 85% of the farmers had been growing cabbage for 5 years or less. With an average farm size of 2.8 ha (ranging from 0.1 to 30 ha), the area of farm land planted with cabbage varied from 0.006 to 5 ha with an average of 0.66 ha per farm. Soil texture varied from clay to sand and there was no significant difference between the soil types used for growing cabbage in the different regions. All the farmers interviewed produced crops using only light cultivation and none were using mulch after the crops were evapotranspiration (ETo) for five stations in Botswana. transplanted. No farm had a subsurface drainage system. 3.3. Irrigation systems The irrigation systems used varied significantly between the regions. In Gaborone, North West and Southern regions, more farmers used sprinklers whereas in Francistown region and Central region hand irrigation and drip irrigation respectively were most common (Fig. 2). 3.4. Irrigation management Farmers based their irrigation scheduling on observation of the soil or crops. They did not schedule irrigation indepen- dently for the different crops they grew and they irrigated different crops with the same irrigation depth and interval. Only 22% of the farmers changed their irrigation schedule between growing seasons. The average irrigation interval was 3 days (range 1–7 days) and the average application depth was 13 mm (range 3– 45 mm) (Fig. 3). On average, farmers using sprinklers used larger applications ( p � 0.001) and longer intervals (p � 0.001) Administrator 高亮 Administrator 打字机 卡方测验 Administrator 打字机 Administrator 打字机 Administrator 打字机 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 打字机 只有22%的农民在不同的季节改变灌溉措施 Administrator 高亮 Administrator 高亮 Fig. 2 – Proportion of farms using different irrigation systems in each region. Fig. 5 – Average irrigation rate and average summer and winter crop evapotranspiration (ETc) in five agricultural regions of Botswana. a g r i c u l t u r a l wa t e r man a g em en t 9 6 ( 2 0 0 9 ) 2 2 6 – 2 3 2 229 than those using drip irrigation. The applications and intervals used by those irrigating by hand were not significantly different from those using the other two methods. On average, the cabbage farmers surveyed were irrigating at an average of 5.3 mm/day and there was no significant difference in the average application rates between irrigation systems (Fig. 4) or region (Fig. 5). Fig. 3 – Irrigation depth and interval for different irrigation methods. Fig. 4 – Average irrigation rate (mm/day) by irrigation system for cabbage farmers in Botswana. 3.5. Soil and water salinity There was no significant difference in irrigation water salinity between the regions (Fig. 6). Although most of the farmers said that they knew the salinity status of their water, none of them had done any scientific test—they estimated the salinity status either by tasting the water or observing white deposits on the soil. Most of the farmers in Central region believed that the water they used for irrigation was salty whereas in Southern and in North West regions most farmers believed that it was not. In the Gaborone region, most farmers said that they did not know the salt status of their irrigation water. Fig. 7 shows that most farmers did not have a good understanding of the salinity status of the irrigation water. The average ECw of the 23 farmers who believed that they had saline water is only slightly higher than that of the 29 farmers who believed that their water was not saline. However the large ranges in Fig. 7 suggest that many farmers had an incorrect perception of Fig. 6 – Average irrigation water salinity for five regions in Botswana. Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Fig. 7 – Actual irrigation water salinity for the farmer’s perception groups. Fig. 8 – Average soil salinity for five regions in Botswana. Fig. 9 – Average soil salinity by soil texture class. a g r i c u l t u r a l wa t e r mana g em en t 9 6 ( 2 0 0 9 ) 2 2 6 – 2 3 2230 Fig. 10 – Actual soil salinity for the farmer’s perception their water status. None of the farmers who believed that their water was salty was doing anything about it. The average ECe was 3.7 dS m �1 and was significantly ( p < 0.001) higher in the Central region (6.3 dS m�1) than Francistown, Gaborone and Southern regions (Fig. 8). There was no significant difference in root zone salinity between the soil types (Fig. 9). From Fig. 10, it is clear that most of the farmers did not know the salinity status of their soil. Only four farmers believed that their soil was saline but there was no significant difference in average soil salinity between the groups. As with irrigation water, those who believed that their soil was salty, said that they were doing nothing about it. There was no significant relationship between the amount of irrigation water applied per day and the electrical conductivity of the soil water extract (Fig. 11). ECe was significantly correlated with ECw (p < 0.001) and farmers using higher salinity irrigation water had higher soil salinity (Fig. 12) although the predictive power of the regression is weak (r2 = 0.27). There was no significant groups. Fig. 11 – Relationship between average irrigation rate (mm/ day) and soil salinity, ECe. Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 Administrator 高亮 a g r i c u l t u r a l wa t e r man a g em en t 9 6 ( 2 0 0 9 ) 2 2 6 – 2 3 2 231 4. Discussion None of the cabbage farmers surveyed used any form of scientific method to schedule their irrigation and their relationship between the soil salinity and the number of years the farms had been cropped except in Francistown region, where the correlation was weak (p = 0.050), as most farmers had not been growing irrigated vegetables for many years. Fig. 12 – Relationship between electrical conductivity of the irrigation water, ECw, and soil water extract, ECe, showing classification for irrigation water suitability (Ayers and Westcot, 1985) and yield thresholds for soil salinity (after Maas and Grattan, 1999). practices have been based on experience. Although those using sprinklers tended to put on larger applications at longer intervals than those using drip irrigation, the differences in average application rate (depth � interval) do not appear to be related to any of the factors studied. During the summer and winter seasons, the average crop water requirements of cabbage at full ground cover are 5.7 and 3.5 mm/day respectively. 40% of those surveyed were applying, on average, less and 38% more water than the winter and summer crop water requirements respectively. There was a lot of variability in the irrigation depth within each region as shown by the error bars and the average irrigation depth did not vary significantly between agricul- tural regions. Sixty-eight percent of those surveyed were growing cabbage in soils with an ECe higher than the 1.8 dS m �1 threshold for maximum yield (Maas and Grattan, 1999). Similarly, 68% of the water sources tested fell within the ‘‘slight to moderate restriction’’ range of 0.7–3 dS m�1 (Ayers and Westcot, 1985) and the average ECw was 1.1 dS m �1. Only one site, of the 60 tested, fell into the ‘‘severe’’ category. A significant relationsh