THE INTERNATIONAL RICE COMMISSION

Twentieth Session

Bangkok, Thailand, 23-26 July 2002

THE DEVELOPMENT AND USE OF INTEGRATED CROP MANAGEMENT FOR RICE PRODUCTION

Warwick S. Clampett, Van Nguu Nguyen and Dat Van Tran

Table of Contents

I. INTRODUCTION

II. RICE CROP MANAGEMENT PROGRAMMES

2.1 Evolution in the Formulation of Technological Recommendations for Crop Management

2.2 Evolution in the Transferring of Technological Recommendations for Crop Management

III. THE AUSTRALIAN Ricecheck PROGRAMME

3.1 The Ricecheck System

3.2 Farmer Discussion Groups

3.3 The Specific Technology Program

IV. GUIDELINES FOR THE DEVELOPMENT OF THE RICE INTEGRATED CROP MANAGEMENT SYSTEM

4.1 Understanding the Concept of Rice Integrated Crop Management

4.2 Formulation of the RICM system

4.3 Extension for the RICM system

V. CONCLUSIONS

REFERENCES


I. INTRODUCTION

Rice is the staple food for over half the world's population. Since the 1960s considerable progress has been made in increasing production - largely by raising yields - to improve world food security (Nguyen, 2001a). More recently, however, it has become obvious that the current rate of increase in yield (down to around 1% per annum) and the small increase in rice area will not be sufficient to meet growing demand. By 2030, it is estimated that the world's demand for rice will be about 533 million tonnes of milled rice (FAO, 2001a). How can this be achieved? The most obvious answer is increased yield. While varietal improvement through hybrids, traditional plant breeding and biotechnology will undoubtedly help, the fact remains that there is huge untapped potential for increasing yields: the yield gap.

The FAO Expert Consultation on Yield Gap and Productivity Decline in Rice Production in 2000 (FAO, 2001b) indicated that the yield gap - the gap between average farmer yields and the yields of the best farmers or those obtained in research plots - was approximately 46 percent. In the irrigated ecology of Asia alone, halving this yield gap would produce an extra 65 million tonnes of paddy rice each year. The gap is usually caused by inappropriate crop management, it is manageable and can be narrowed by enhancing research and extension services and with appropriate government intervention, particularly with regards to institutional issues (Tran, 2001). In the past, much of the effort to transfer technology to farm level focused on problem-solving through a single issue approach. In more recent years, the need for an integrated approach fostered the development of Integrated Pest Management, Integrated Weed Management and Integrated Nutrient Management systems. In the 1990s, the need for more holistic and integrated management on a whole crop basis was increasingly recognized as essential for optimizing production. At the same time, extension strategies were modified and had an ever-increasing degree of integration. The trend has been to move from a top-down approach to a more participatory and facilitative approach involving farmers in strategy development and implementation.

This paper discusses the evolution of Rice Integrated Crop Management (RICM) systems and the process of extending them to farmers. There is particular emphasis on the successful development and transfer of the Australian Ricecheck programme as an example of RICM for yield improvement. The paper also proposes guidelines for the development and transfer of RICM systems to assist other national rice programmes in their efforts to optimize productivity and achieve environmentally friendly and sustainable rice production.

II. RICE CROP MANAGEMENT PROGRAMMES

Rice growing is a production system involving a wide range of component factors (Clampett, 2001). Rice farmers carry out a large number of management activities, including: development and maintenance of the rice field infrastructure (field levels and grade; the system of bunds with supply channels and drains); selection of variety and seed source; determination of the sowing and cropping calendar; land preparation practices; plant establishment techniques; protection from weeds, insects, diseases and other pests; nutrition supply to meet growth needs; management of water supply and depth control; and harvesting and grain storage. All these activities, singularly and collectively, affect the production of biomass in all phases of crop development, which ultimately determines the parameters of plant growth, yield components and yield. Crop development phases include germination, seedlings, vegetative development and reproductive phase from panicle initiation through flowering to grain ripening and maturity. Rice crop management programmes involve the formulation and transfer of technological recommendations for rice production throughout the entire growing season and often also include rotation factors.

The system to formulate technological recommendations for rice crop management has evolved slowly over the last 20 to 30 years. It has developed concurrently with changes in extension strategies and methodology. Technological recommendations for rice crop management have evolved from a single-problem or component focus, to a broader integrated system, with emphasis on the interaction and relationships between and among the system components. The extension approach is evolving from a top-down, directional and authoritarian "this is what you must do" to a more participatory, facilitative and knowledge-processing approach involving farmers, researchers, extension officers and related stakeholders in the development of recommendations and technology sharing.

2.1 Evolution in the Formulation of Technological Recommendations for Crop Management

During the 1970s and 1980s, various technologies for managing different activities in growing rice crops were formulated into packages (Nguyen, 2001a). Technological recommendations were usually developed by research experts with little or no involvement of extension officers, who had the task of delivering the packages to farmers. Also, recommendations were usually very simplistic - "use X kg fertilizer per hectare at Y stage on variety Z" - and involved little promotion of farmers' understanding of why or how. Too often, productivity increases were disappointing because factors such as rice plant stand and pests limited the rice response to fertilizer application. Moreover, there was emphasis on agrochemicals, which were heavily used.

Increased fertilizer or pesticide use was not necessarily an improvement. Apart from the immediate and often variable effects on rice productivity and profitability, there were other problems. Resistant biotypes of many insects developed as a result of indiscriminate use of insecticides causing ecological imbalances. There were similar observations concerning weed resistance to herbicides, increased use of nitrogen fertilizer led to the limitation of other nutrient deficiencies, and environmental pollution became an issue.

The concept of integration was, therefore, introduced in the 1980s in the formulation of technological recommendations for rice crop management, with the development of programmes, such as Integrated Pest Management (IPM), Integrated Weed Management (IWM) and Integrated Nutrient Management (INM) (Shastry et al., 1996). These approaches attempted to broaden the understanding of the range of factors affecting pest or weed growth and development or fertilizer response of rice, and to involve these factors in the farmer's decision-making process. While these programmes created improvements and benefits, the technology focus was still relatively narrow, generally involving only specific areas of crop management - insect management, weed management or nutrient provision. However, the concept of integrated management was beginning to develop and influence attitudes towards crop management (Way and Heong, 1994).

Over the last 10 to 15 years, a broader system of rice crop management, the Rice Integrated Crop Management (RICM) system, has been developed for irrigated rice production in many areas, covering diverse climates and socio-economic environments, such as Burkina Faso (Nguyen et al., 1994), Australia (Lacy et al., 1993; Clampett et al., 2001), Egypt (Badawi, 2001), Korea (Moon, 2001), Senegal (Nguyen, 2001b) and the whole Sahel zone of West Africa (Wopereis et al., 2001). The RICM system seeks to develop a management approach at farm level that manages the growing of rice crops as a total production system, taking into account all factors affecting yield and quality. It takes into account the interactions and interdependencies among management technologies. It also provides a framework that helps farmers to evaluate their management skills and to recognize their strengths and weaknesses in order that the management of subsequent rice crops may be improved. Among the RICM systems, the Australian Ricecheck system, which will be discussed in detail, is outstanding in terms of its success in raising productivity and its innovations.

2.2 Evolution in the Transferring of Technological Recommendations for Crop Management

Concurrent with the changes in the system to formulate technological recommendations, there has been a change in the system to transfer technological recommendations for crop management or extension methodology. The key changes are related to the changing attitude towards farmers and their involvement in the programmes: from being regarded as passive recipients and beneficiaries, to being active participants and an integral part of the process of technology development and evaluation.

When the emphasis was on problem-solving with a single-issue focus (such as nitrogen, weeds or pests) the approach tended to be directive, rigid and prescriptive. Extension attitudes such as: "we know best", "this is your problem" and "this is what you must do", often prevailed. This top-down approach had severe limitations. The training and visit ("T and V") system fostered by the World Bank became popular for transferring rice technologies during the 1970s and 1980s. The important feature of this system is the systematic training of village extension workers and frequent extension visits to farmers' fields (Benor and Harrison, 1977). T and V was expensive, but more importantly, farmer particiaption in the process was minimal; it was often too prescriptive.

The development of integrated programmes, such as Integrated Pest Management, introduced a broader approach: other factors were considered, options for action were sought and a decision-making tree was sometimes included. They emphasized greater understanding and knowledge as important factors in making decisions and solving problems. This resulted in a greater level of farmer involvement (Kenmore et al., 1995). The Farmer Field Schools were then used as a vehicle for transferring IPM to farmers. A Farmer Field School lasts one rice-cropping season, during which 12 to 15 weekly classes are held. The field is the classroom and farmers are taught how to observe rice growth, natural enemies and insect populations, and how to make management decisions. Experiments on sprayed and unsprayed rice crops may be simultaneously conducted in the field where the class is held (Kenmore et al., 1995). In Australia, farmer discussion groups were used to transfer the Ricecheck system (a Rice Integrated Crop Management system) (Clampett, 2001).

III. THE AUSTRALIAN Ricecheck PROGRAMM

This successful and innovative programme is represented diagrammatically in Figure 1. It has three broad-linked components: the Ricecheck system, the Discussion Groups and the Specific Technology.

3.1 The Ricecheck System

From 1973 to 1985, rice yield in Australia stagnated at around 6 t/ha (Fig. 2). After the development and the transfer of the Ricecheck system in 1986, the Australian national yield increased rapidly and steadily from about 6 t/ha in 1987 to 9.65 t/ha in 2000 (Fig. 2). The Australian rice yield in 2000 was the world's highest national yield (FAOSTAT, 2001). Australian rice scientists considered that half of the observed yield increase since 1986 can be attributed to the adoption of new rice varieties and another half to the adoption of the Ricecheck system (Clampett, 2001). Moreover, results obtained in Australia also showed that not only rice yield, but also the efficiency of nitrogen fertilizer application in rice production were obtained with the application of the Ricecheck system (Batten et al., 1994). The increase in the efficiency of nitrogen due to the application of the Ricecheck system, in turn, reduced not only production costs but also the negative effects of nitrogen losses on the environment.

The development of the Ricecheck system involved major changes to the way technological recommendations were formulated and the approach through which the recommended technologies were transferred to farmers. It evolved through of a number of trends that occurred in the planning, development and delivery of extension (Clampett, 1994). These trends are summarized in Table 1. The Ricecheck system has two key elements: a technical framework, which provides objective recommendations and the basis for managing the crop, and an action framework to use the recommendations as criteria or checks to evaluate the effectiveness of management.

3.1.1. The technical framework

The Ricecheck system provides recommendations that are the Best Management Practices for rice-growing, based on knowledge from research, latest known technologies and farming experiences. It works on the principle that to obtain optimal yield, the management of production inputs must achieve optimum output levels at every crop growth phase or in every management area. It provides both input and output recommendations. Input recommendations are the recommended technologies and materials for use in managing rice crops at different growth phases. Output recommendations are the optimal results that crop management should achieve at different growth phases (Table 2). For example, a seed rate (e.g. 120 kg/ha of good seed) is a recommended input in order to achieve a desired plant stand of rice at 20 days after seeding (e.g. 150 plants/m2) - an output recommendation. The output recommendation (e.g. 150 plants/m2 ) serves as a criterion or benchmark for the evaluation of farmers' success or failure in managing the crop establishment.

In the formulation of technological recommendations, the Ricecheck system takes into account the fact that each single practice affects a range of management practices and outputs, thus reinforcing the importance of the need for integrated management. For example, land preparation will affect a wide range of outputs, such as seedling number, weed population, nitrogen uptake and pest population. The seedling number (an output of land preparation) will in turn affect the technological recommendations for weed management and nitrogen fertilizer application. The outputs at different phases of crop growth, when combined and allowed to interact together, result in yield, grain quality, environmental impacts and economic return. Examples of inputs and outputs are provided in Table 2.

The input and output recommendations in the Ricecheck system are, as far as possible, objective, providing numbers that can be measured and compared. For example, water depths at specific crop growth stages are presented in numbers and sowing times for each variety are presented as actual dates (Lacy et al., 2001). Subjective descriptions, such as early or late, high or low, deep or shallow, short or tall, are avoided. Also, recommendations are the most up to date and the best, and are continually improving. They are revised each year by rice farmers, extension workers and researchers working together. The changes are based on experience with the current recommendations and the integration of new technologies.

While numerous factors affect yield and other outcomes, attention is focused on key factors. These recommendations are called "Key Checks". The Key Checks for yield optimization cover seven Management Areas. A Management Area may include one or more Key Checks, including benchmarks and/or objective targets, as well as minor targets. The Management Areas are as follows:

3.1.2. The action framework

The other innovative aspect of the Ricecheck system is the use of Key Checks to evaluate the application of recommended technologies. It encourages farmers to use the targeted (or recommended) outputs for comparison with actual results in order to identify management strengths and weaknesses. The identification of strengths would enable farmers to correct poor management in future crops. The steps in using the Ricecheck Action Framework are:

The concept of achieving management benchmarks or checks to improve yields is outlined in Table 3. The higher the level of achievement of the key management areas or the more Key Checks achieved, the higher the yields. Evidence of this is provided by the survey results from 2 100 commercial rice crops over 7 years (1994-2000) (Fig. 3).

This "checking" may occur in a variety of ways, from informal and relatively unstructured to very formal and structured. At one end of the scale, checking within the action framework can occur at a simple level involving very few - if any - paper records: the processes occur but there is little formal evidence. Further along the line, an individual may keep written records to various standards. At the formal, structured end are Computer Database Systems providing detailed recording formats, data entry and analysis. In New South Wales a Ricecheck Crop Evaluation Program each year compares the performance of some 600 to 800 commercial crops with the benchmark targets and with the performance of the peer group. This programme is outlined by Lacy et al. (2002).

3.2 Farmer Discussion Groups

The success of the Ricecheck system owes much to a pro-active extension team approach, involving farmers through farmer discussion groups. These groups provide a forum for both the presentation of relevant technology, farmer-centered discussion, collaborative learning and feedback on the applicability of the technology and an evaluation in the real world of commercial experiences (Clampett et al., 2001). These are neighborhood groups, commonly having 10 to 20 members, although attendance can vary greatly as a result of seasonal farming commitments. The farmer groups perform a number of functions with the extension officer acting as both an expert/resource person and a facilitator. Through the groups, farmers are able to learn as much from each other as they do from the extension officer. The extension officer also learns and can complete the link by sharing the new knowledge with research colleagues. Groups also provide the opportunity to survey particular practices or topical developments. Groups may also conduct demonstrations to compare current and new technology. Once groups are established and a rapport among group members is developed, these groups provide an excellent basis for successful extension programmes. Normally there are four to five meetings of a rice discussion group each season, timed to fit in with the patterns of the rice-growing system and the needs of the farmers. A typical sequence of group meetings would be:

3.3 The Specific Technology Program

Within the rice crop management system there are some component areas where recommended technologies are poorly adopted and have a major impact on yield, quality or the environment. These technology areas may relate to a new technology or to a poorly adopted old technology. These subject areas require special attention. The Specific Technology Program is conducted on a pilot basis in order to identify solutions to these problem areas aimed at further improvement of the Ricecheck system.

IV. GUIDELINES FOR THE DEVELOPMENT OF THE RICE INTEGRATED CROP MANAGEMENT SYSTEM

The development and adoption of the RICM system requires a wide range of factors to be considered: from an understanding of the concept of Integrated Crop Management to the formulation of technological recommendations based on best management practices and the importance of farmer experience in using the system to evaluate management ability.

4.1 Understanding the Concept of Rice Integrated Crop Management

It is important for those who develop the RICM system to understand and be committed to the approach as a management and extension tool for yield improvement, or at least to be prepared to objectively evaluate the concept. A number of factors are important:

4.2 Formulation of the RICM System

The input and output recommendations should be based on the Best Management Practice (BMP) for the particular agro-ecological conditions and local level of management.

"Best Management Practices can be defined as those practices best able to achieve the results required given the available technology and knowledge, the practical experience on commercial farms, and the resources, understanding and capabilities of the rice farmers. They should reflect what is perceived as achievable and desirable. They are not necessarily those practices that the most recent research results would support (Clampett, 2001)"

BMPs can be developed from the practices of the best farmers and from the knowledge and experience of research and extension officers. Six major steps to formulate recommendations in a particular area or season are outlined in Box I. The Key Management Areas may vary, depending upon the local agro-ecological conditions and the level of crop management in the local area. It is also important to consider recommendations in the context of inputs (what you want the farmer to do) and outputs (how the practices aim to achieve optimum growth and yield, quality and environmental conservation). In the development of the RICM system, it should be remembered that farmers do not start at the same level of management - it is important to start where farmers are and not where you want them to be.

4.3 Extension for the RICM System

The RICM system can be seen as a complex system, and it is. Although the system can be broken down into the component technologies, it is nevertheless essential to examine the whole, if the RICM system is to be fully utilized and is to cope with the complexity of the whole integration system. It thus becomes very important to use an extension approach that involves working with farmers in a participatory and collaborative way that sees farmers as partners in a process to improve the outcomes of growing rice. The ultimate objectives are food security and socio-economic well-being. A forum, such as the farmer discussion group, for hands-on, participatory, collaborative learning, and a practical and effective environment for the delivery and integration of improved rice-growing practices into the farm system, is essential for the effective transfer of the RICM system.

The development of farmer discussion groups requires considerable effort. There needs to be a clear understanding of the objectives and strategies underlying the operation of the group. The resources needed to support the group must be clearly identified and provided. The operation of the group, particularly in a pilot scheme, needs to be monitored. It is important to keep the group and the extension officer interested, involved and motivated. Farmer groups can provide the ultimate participation of rice farmers at the frontline/grassroots level. Key issues in developing farmer discussion groups are outlined in Box II.

V. CONCLUSIONS

REFERENCES

Badawi, A.T. 2001. "Yield gap and productivity decline in Egypt". In Yield gap and productivity decline in rice production, p. 429-442, Proceedings of the Expert Consultation, Rome, 5-7 Sept. 2000. Rome, FAO.

Batten, G.D., Blakeney, A.B. & Ciaverella, S. 1994. "An interactive database for use with the rice tissue test service". In E. Humphreys, E.A. Murray, W.S. Clampett & L.G. Lewin, eds. Temperate rice - achievements and potential, p. 473-476, Proceedings of Temperate Rice Conference, Yanco, NSW, Australia, Feb. 1994.

Benor, D. & Harrison, J.Q. 1977. Agricultural extension: the training and visit system. World Bank.

Clampett, W.S., Williams, R.L. & Lacy, J.M. 2001. "Major achievements in closing yield gaps of rice between research and farmers in Australia". In Yield gap and productivity decline in rice production, p. 441-428, Proceedings of the Expert Consultation, Rome, 5-7 Sept. 2000. Rome, FAO.

Clampett, W.S. 1994. "Extension programs and technology transfer in the New South Wales rice industry". In E. Humphreys, E.A. Murray, W.S. Clampett & L.G. Lewin, eds. Temperate rice - achievements and potential, p. 243-346, Proceedings of Temperate Rice Conference, Yanco, NSW, Australia, Feb. 1994.

Clampett, W. 2001. Report submitted for a FAO consultancy mission to West Africa, Aug. 2001.

FAO. 2001a. World agriculture: towards 2015/2030.

FAO. 2001b. Yield gap and productivity decline in rice production, Proceedings of the Expert Consultation, Rome, 5-7 Sept. 2000. Rome.

Kenmore, P.E., Gallagher, K.D. & Ooi, P.A.C. 1995. Empowering farmers: experiences with Integrated Pest Management. Entwicklung + Landlicher raum 1.

Lacy, J.L., Clampett, W. & Nagy, J. 2002. "Development and use of a crop management database to evaluate rice crop performance in New South Wales, Australia". In Proceedings of Second Temperate Rice Conference, Sacramento, CA, 13-17 June 1999. (in press)

Lacy, J.L., Clampett, W., Lewin, L., Reinke, R., Williams, R., Beale, P., Fleming, M., Murray, A., McCaffery, D., Lattimore, M., Schipp, A. & Salvestro, R. 2001. 2001 Ricecheck recommendations. Australia, NSW Agriculture and Rice Research and Development Committee. 16 pp.

Lacy, J., Clampett, W.S., Lewin, L., Reinke, R., Batten, G., Williams, R., Beale, P., McCaffery, D., Lattimore, M., Schipp, A., Salvestro, R. & Nagy, J. 1993. 1993 Ricecheck recommendations. Australia, NSW Agriculture and Rice Research and Development Committee. 16 pp.

Moon, H.P. 2001 "Yield gap and rice productivity in the Republic of Korea". In Yield gap and productivity decline in rice production, p. 345-56, Proceedings of the Expert Consultation, Rome, 5-7 Sept. 2000. Rome, FAO.

Nguyen, V.N. 2001a. Rice integrated crop management for food security. Paper presented at the WARDA PEAT Workshop, Senegal, 2-4 April 2001.

Nguyen, V.N. 2001b. Productive and environmental friendly rice integrated crop management systems. IRC Newsl., 51. (in press)

Nguyen, V.N., Tran, D.V., Bautista, R.C., Maiga, M. & Weerapat, P. 1994. "Thriving with rice" technologies for small farmers in irrigated systems in sub-Saharan Africa. IRC Newsl., 43: 33-39.

Shastry, S.V., Tran, D.V., Nguyen, V.N. & Nanda, J.S. 1996 "Sustainable integrated rice production". In Progress assessment and orientation in the 1990s, p. 45-58, Proceedings of the 18th Session of the International Rice Commission, Rome, 5-9 Sept. 1994. Rome, FAO.

Tran, D.V. 2001. "Closing the rice yield gap for food security. Rice Research for Food Security and Poverty Alleviation". In Proceedings of the International Rice Research Conference, p. 27-41, Los Baños, Philippines, 31 Mar. -3 April 2000. Los Baños, International Rice Research Institute.

Way, M.J. & Heong, K.L. 1994. The role of biodiversity in the dynamics and management of insect pests of tropical irrigated rice - a review. Bull. Entomol. Res., 84: 567-587.

Wopereis, M.C.S., Hafele, S.M., Kebbeh, M., Miezan, K. & Diack, B.S. 2001. "Improving the productivity and profitability of irrigated rice production in Sahelian West Africa". In Yield gap and productivity decline in rice production, p. 117-142, Proceedings of the Expert Consultation, Rome, 5-7 Sept. 2000. Rome, FAO.

Table 1: Trends in rice extension programme development since the early 1980s (modified from Clampett, 1994)

REACTIVE TO FARMER WANTS & NEEDS TO PRO-ACTIVE TO ACHIEVE CHANGE
LESS STRUCTURED & TARGETED TO MORE STRUCTURED & TARGETED
EMPHASIS ON SIMPLER MORE READILY ADOPTED CHANGES TO MORE BALANCED APPROACH TO SIMPLE AND COMPLEX CHANGES
MORE INDIVIDUAL EXTENSION OFFICER EFFORT TO MORE COORDINATED EXTENSION TEAM APPROACH
MORE FARM VISITS TO MORE GROUP WORK AND BALANCED METHODOLOGY
LESS FARMER INVOLVEMENT TO MORE FARMER INVOLVEMENT
FOCUS ON SOLVING PROBLEMS TO FOCUS ON INTEGRATED CROP MANAGEMENT

Table 2: Examples of relationships between input practices, growth and management outputs, and yield, grain quality and environmental outcomes

INPUTS
- the practices
OUTPUTS
- the results
OUTCOME
- the final impacts
Cultivation, vegetation control, leveling, seed rates, sowing techniques, water depths, time of sowing, pest control PLANTS/M2
100% UNIFORM PLANT COVERAGE
YIELD
GRAIN QUALITY
Cultivation, nitrogen rates, timing, application, water management, pest and weed control, time of sowing PI NITROGEN UPTAKE KgN/ha YIELD
GRAIN QUALITY
Field irrigation layout, banks (bunds), channels, water supply rates and depth control, drainage SPECIFIC WATER DEPTHS @ GROWTH STAGE YIELD
Cultivation, sowing time, establishment, plant stand & uniformity, pre-harvest, draining, nitrogen supply, harvest time GRAIN MOISTURE @ HARVEST GRAIN QUALITY
Field irrigation layout, land preparation, time of sowing, establishment, field drainage, water management, pesticide use and timing, PESTICIDE RESIDUE LEVELS IN FIELD DRAINAGE WATER ENVIRONMENTAL IMPACT

Table 3: The concept of achieving checks in Ricecheck Integrated Crop Management

>
KEY MANAGEMENT
AREA
Check
Achievement
  Check
Achievement
Field Layout OR
Sowing Time OR
Establishment OR
Crop Protection OR
Nutrition OR
Water Management OR
Harvest Management OR
YIELD RESULT HIGH   LOW

Figure 1: Key elements of the Australian Ricecheck Program (modified from Clampett, 1994).

Undisplayed Graphic

Figure 3: Relationship between yield and the achievement of Key Checks of 2 100 commercial Amaroo rice crops over 7 seasons 1994 - 2000. ( Lacy et al., 2001)

Undisplayed Graphic