Communicating the probabilistic seasonal forecast for a better farming management and decisions. By : Dr Ousmane Ndiaye, Senegalese National Weather Agency (ANAMS) Place : Mission catholique, Kaffrine Dates : 04-05 June 2011 Partners : CCAFS, ANAMS, SDDR, WV, JAPANDOO, ISRA and ANCAR.  Agenda : Session Saturday June 4th, 2011 Sunday June 5th 2011 Morning • Official opening local authority • Presentation of the agenda • Climate variability and farming • Indigenous indicators of climate variability • Basis of the seasonal forecasting • El Niño phenomenon and the monsoon • Group exercise: interpreting probability of exceedance forecast. • Closing ceremony After-noon • Survey activities at 5 villages • Installation of rain gauge in “Tousne mosquée” Stakeholders : • Senegalese National Weather Agency –ANAMS- (06 participants) • Farmers organization: JAPANDOO, Woman farmers organization, growing peanuts seeds organization, individual farmers. (31 participants) • Agricultural Research Institute of Senegal –ISRA- (01 participant) • Local governmental agricultural extension –SDDR- at Kaffrine (05 participants) • National agency for agricultural and rural advice –ANCAR- (03 participants) • World Vision (04 participants) Location of the sites in Kaffrine region in Senegal (indented), map courteously by CSE.  Introduction This is a combined report on the workshop interaction with framers on probabilistic seasonal forecasting on June 4th and 5th, and the follow up workshop on communicating the actual seasonal forecasting on June the 8th, 2011 both held in Kaffrine (Senegal). The first workshop was designed to expose farmers on probabilistic seasonal forecasting and also to establish dialogue and trust between farmer’s organizations and experts working on climate forecasting and farming in general. Emphasis was made to listen the framers first and better understand their decision system. This is a first step toward helping small-scale farmers to make appropriate management decisions for improved agricultural based on probabilistic seasonal climate forecast. The second workshop was to deliver the actual seasonal forecast and to ensure they can interpret it. We also distributed some rain gauges and offer training on their installation and use. We chose the site of Kaffrine in Senegal for many reasons. Kaffrine is located in the midst of the transition zone from the Sahelian towards the Sudan Savannah zone with annual rainfall averages of around 500 mm in the northern, 600 mm at Kaffrine, and around 800 mm in the southwestern part of the area. Dominant soils are deep sandy (“Dior”). Predominant cropping systems are based on pearl millet, peanut and cowpea, all generally not intensified and cropped without agricultural input. In the south, peanut is intensified using inputs, and maize, sorghum, lowland rice and sesame are also cropped. Main constraints to agricultural production are high rainfall variability, poor soil fertility, no attractive markets and high poverty levels with low access to capital. I. Organization of the Workshop A preparatory meeting involving Dr. Jim Hansen, experts on climate, agro-economist and agriculture expert was held on June 2nd 2011 at the Senegalese national weather Agency to revisit the materials that will be presented to the farmers, discuss the workshop agenda, the structure and methodology for attaining our objective. Mr Seck the local agriculture extension in Kaffrine gave us a brief report on the logistic and selected farmers for the workshop. The actual training workshop took two (2) days June 4th and 5th 2011. We had 31 farmers amongst them 13 were women as well as 10 experts coming from local NGOs and governmental services working with farmers. There was training discussing session as well field trips where a survey questionnaire was conducted in 6 villages. The main objective of this workshop was to know how farmers use currently climate information, what kind of forecast they used, and to present them example of probabilistic seasonal forecasting. The workshop began with an official opening by the local administrative authority, the prefet, of the district of Kaffrine. Who welcomed everyone and acknowledged the importance of rainfall forecast on farmer’s livelihood as well the climate change. Dr Ousmane Ndiaye and Dr Jim Hansen presented respectively the objective of the workshop as well as CCAFS program. After we go around the table and each participant introduced himself in a lively manner. Dr ousmane Ndiaye started by asking farmers how they manage to forecast the trend of the upcoming seasonal forecast, its beginning and its end. What are their indicators?  II. Indigenous knowledge on weather and climate : The question asked was to describe the indicators for climate forecast specifically : good or bad season. Farmers were able to open up and express themselves without difficulties. Here are some indicators : Imminent onset of the rainy season : When the wind changes direction to fetch the rainfall Apparition of stars shaped as elephant Birds crying as if it calls men to go to field and woman to stay at home early flouring of many trees species: Néré, dimb, tamarinier, sone butterflies and libellees are numerous Some persons feel heavy in their body Hot night time Empirical sign of a major rainy event : When wind is shifting direction When dark clouds become white Sign for a good rainy season : When snakes and frogs are more numerous than usual The shooting star direction indicates which zone will receive excess rain this year Net appearance of seven stars in the sky Sign of a good cropping season : When the rain is settled in June the 24th and we start the millet around the 14th of July we can expect good harvest Sign of the end of the season : When frogs start chanting When the sky is high When we observed dew in the morning Then Dr ousmane Ndiaye started to link some of their signs with known climate phenomena. The idea is to build confidence.  Table comparing some empirical signed and scientific facts: Empirical knowledge Scientific explanation When the wind change direction to go to fetch for the rain Before the rainy season blows the dry harmattan wind from the North-East, the rainy season coincide with the monsoon flow which blow from the South-West Flouring of many species High temperature and high humidity preceding the rainy season can explain plant behavior Sense of heavy body in certain people Hot night time The high humidity and high temperature give a bad comfort index When dark cloud turns white Rainy clouds are cumulo-nimbus they are announced first by low dark stratocumulus then shows the icy white part When the sky is high Lack of low cloud with the retreat of the monsoon Dew during dawn End of the rainy season, temperatures become lower and allow easy saturation It also emerged that farmers differentiate clearly a good rainy season from good crop season. Farmer Babacar Diaw went through explaining the change in variety due to the decrease of the length of the rainy season. He said in the past we used to plant varieties of 110 days cycle like the peanuts 28/206 and sorghum. Now we used peanuts of 90 days cycle (55/4737). And during dry season we cultivate short cycle of sorghum (thialakh in local language). Building from some similarities and scientific explanation farmers start getting a glimpse on general circulation in west Africa. The facilitator build confidence by emphasizing the proximity of farmers indicators and indicators used to make the forecast that will be presented to them. He said “we used quite the same indicator but our only difference is we use better tools, better observing system and more systemized reporting”. He mentioned observing through satellites, large and comprehensive network of sharing information around the world, capacity of storing and analyzing data using computers. Farmers were very open to the new forecast and at least could understand why this new way combine both their usual indicator but in a more modern and better way. The facilitator emphasis a lot why the scientific system presented to them today should be considered or at least deserved a trial. Building on that Dr Ndiaye started explaining the basis forecasting at seasonal time scale months ahead.  III. Basis of probabilistic seasonal forecasting : Dr Ndiaye in very comprehensible terms explained farmers the reason why seasonal forecasting is possible. He started first by differentiated weather and the climate. Weather: is just what’s happening now or during few days, it can be described precisely by what they can see or used to characterize the atmosphere for example just by looking through the window like : clouds, sunshine, wind, hot, cold etc. The facilitator added weather is just what they see on TV or hear to the radio. Farmers are very familiar with the weather bulletin on radio. Climate: Climate is the way you describe a long period of “weathers” for example what did you relocate from last season 2009. How can you describe last month in general. So the climate is the cumulative effect of weather in a particular area or region over a long period of time. The facilitator pointed out what the subject matter here is not the weather but the climate. We are not predicting what’s happening exactly tomorrow or a particular day but the cumulative effect of the climate over a season and precisely the rainy season To set the basis of seasonal forecast the facilitator starts with a dialogue by asking a simple question “When it is hot, why do people go to the beach?”. All Farmers responded because sea breeze brings fresher air. Then he took it further: “Is it not the same sun that heats both land and ocean? Why then does the ocean get cooler in summer?”. He then explained that ocean has better memory of the past compare to the continent. The ocean remembers the heat of the past days and weeks. That’s why, on a very hot day you go to the beach to benefit from ocean memory of the past weeks. Same thing when it is cold, the ocean still remembers warm days. He mentions some phenomenon that few know like el Niño and la Niña as perfect examples how ocean can stay on one state for months. As ocean has longer memory and occupies larger area at the surface of the earth they constrain the climate in the tropics. This is the basis of seasonal forecasting: using ocean heat memory. He added rain comes from water vapor, most of the water vapor comes from ocean, so the ocean temperature can control rain. Upon hearing this, farmers were less skeptical and actually start asking questions which indicate their understanding. After he pointing out the difficulty of estimating exactly how much it will rain, the difficulty of having perfect forecast. There are always errors. He challenges farmers whether their traditional indicators are 100% accurate; they all say no. One farmer commented, “Only God knows what is happening 100% in the future”. They understood the uncertainties. By the end of the first session, trust has been built by connecting common indicators, the basis for seasonal forecasting established, and probabilistic nature of the forecast introduced.  IV. Practical session with farmers : This exercise objective is to have farmers getting used to rainfall amount, its variability, being able to recognized normal year, wet years and dry years. They first asked to take the seasonal rainfall over 6 years (2005 to 2010) in Kaffrine and to draw into bar chat their classification from the highest to the lowest. Some volunteers to come front and to show to their peers how to do such classification. They was asked afterwards to try to compare the classification with their memory did they perceive these years to be dry or wet ? After such exercise farmers can now understand what is a normal, below normal or dry and above normal or wet year mean. They are now introduced to some language in the probabilistic forecast. After classifying the years from the driest to the wettest they were asked to indicate what is the amount of rainfall total having a chance to be observed at 1 year out of 4 (25%), 1 out of 2 (50%) and 3 out of 4 (75%). The facilitator explained first what these probabilities mean intuitively 1 out of 4 means if you have in a bag 4 balls 3 are red and one is black what is the chance to pick up the black ball. That is 1 out of 4. Farmers picked up quickly the probability and were able to draw lines at the cumulative graph of 25%, 50% and 75% probability. Then a full climatological probability of exceedance calculated from 30 years (1961-1990) was displayed on the board. Sometimes were spend then by choosing farmers to go front and show some probability of exceedance. When the probability of exceedance was understood, the facilitator shows an hypothetical forecast on top of the climatology. From that the facilitator indicated how a forecast can be shifted above or below the climatology observed curve. Many discussion of wet and dry forecasted followed and how to associate probability of exceedance to an amount of rainfall and vice versa. Amy ndiaye a farmers from Gniby summarized it quite nicely saying a shift up is la Niña and a shift down is el Niño. Exercise of familiarizing with rainfall and seasonal forecast.  V. Field trip and survey : During the afternoon of the first day (June the 4th 2011), the experts were divided into 5 multi- disciplinary groups to conduct in 5 villages around kaffrine : Santhiou Galgoné (-15.5, 14.0), Katakel (-15.4, 13.9), Ngodiba (-15.5 , 14.0), Tousne Mosquée et Sorokogne (-15.5, 14.2). The target was 30 farmers in total, over 6 per village. The objective is first to visit farmer in their own environment and to gather data related to their farming activities : description of the household, surface cultivated, access to and used of climate information. One rain gauge was installed in the village of tousne mosquée, training were conducted on how to use the rain gauge and to understand collected rainfall amount. Rainfall amount was shown to the farmers on the ground : 1 mm is distributing 1 litter of water over a square meter surface. Training farmers on how to read the rain gauge record in tousne mosquée village. To fix the ideas and relate the probability of exceedance forecast to concrete decision on farm management, farmers were divided into 4 groups in the next day. Each group was given a map of an hypothetical forecast along with the climatological forecast. Each group was asked to choose a coordinator and a reporter. We affected each group an expert from each institution (ANCAR, World Vision, Agriculture and Climate) as supervisor. They help only if farmers don’t understand or they do need guidance. Questions asked were first to interpret the given forecast and then to deduce all decisions and actions they would take from this forecast.  VI. Discussion and decisions from probability forecast: Two hypothetical seasonal forecast on rainfall total and number of rainy days were given to them two were dry and two wet. Each group presented the summary of their discussion by the group reporter. • They were able to differentiate between a good forecast in term of rainfall exceeding certain threshold and a good cropping year where the rainfall is well distributed in time to allow the crop to finish its cycle. From their experience even though 2010 is the wettest year, 2008 was better in term of yield. • They all interpreted correctly the forecast : dry forecast below the climatology and wet forecast above the climatology. And two of the groups did mention el Niño for dry forecast and la Niña for the wet forecast. • They were able to select different strategies according to the forecast. • For a wet forecast : o Select a crop variety with longer growth cycle : peanuts 73/30, sorghum, maize, sesame o Build water dams for water storage to be used later during the dry season or for a second crop o Increase the cultivated surface o Hire more people to help o Use more fertilizer and pesticide o Delay peanuts (cash crop) planting date to avoid damages due to late rain : peanuts re- germinated during 2010 as it keeps on raining beyond the rainy season • For a dry forecast : o Plant a crop with shorter growth cycle : water melon, peas, okra, millet variety “madio”, sesam variety “Jalgon”, peanuts 55/437 or “fleur 11” o Use less fertilizer o Use less paid workers to avoid being in debt o Plant at the first rain to allow the crop to finish its cycle It follows afterwards a long discussion on the number of rainy days and how their distribution in the season can affect the crop.  Farmer’s interpretation of wet (right) and dry (left) forecast of probability of exceedance.  VII. Evaluation of the workshop and the process : The last activity was dedicated to the evaluation of the overall process building relationship between farmers and climate expert on using probabilistic seasonal forecasting. The evaluation concerned the training agenda, training material, and workshop logistics. Weaknesses :  No breakfast was offered to people coming far away  the transportation reimbursed was less than that some people paid (actually three persons and we paid them the difference afterwards)  the program was so condensed for a short time  the room was small specially in this hot season  receive the seasonal forecast well in time before farmers start buying seeds, and working on the field positive points :  the knowledge learnt allow us to better take action in the future  like a lot the session on the rain gauge and what “1 mm” is actually representing  trained people can relate the climate forecast at village level  knowing the weather is crucial for planning to go to the farm or not  this training is the first of its kind for all participants  necessity of repeating this training in all the districts of Kaffrine  workshop well organized (local coordinator) as well as the selection of attendees  all the sessions were on time  we better understand climate forecast now  new and useful knowledge were acquired  we better understood the uncertainties related to the forecast  reviewing the indigenous traditional knowledge was very important for us  in the framework of globalization farmers need to be kept updated with new tools recommendations : ⇒ need follow up and receive this year seasonal forecast (we did deliver it a week later) ⇒ experts from other discipline (not climate) need to be trained in climate to be able to incorporate climate information in their activities ⇒ keep in touch with farmers because many organizations come to give trainings then we won’t hear from them ⇒ ask for more rain gauges (the met service did give more one week after) ⇒ women asked to be trained in order to train their fellows in their local organization After the evaluation prayers did follow and the workshop was closed.  On the June the 8th 2011, a week later, a workshop was held in Kaffrine with the same farmers (22 actually did attend) and the actual seasonal forecasting of July-Aug-Sep 2011 was given. Four experts from the national weather agency (ANAMS) and some local technical partners (World Vision, ANCAR and Agriculture Department) were also there. The program is as follow :  Official opening  Training on how to install and read the rain gauge  Delivering the seasonal forecast of JAS 2011 The workshop was opened by the same local authority (prefet) in the presence of the district director of rural development. The prefet was glad that we kept our promise and came back to deliver the actual seasonal forecast after the training of last week. He reminded us the importance of the climate information on agricultural activities. He himself learnt a lot from last workshop and welcomed our collaboration in the future. I. Training on installing and reading the rain gauge : Due to interest expressed during the probabilistic seasonal forecasting workshop, the ANAMS has brought 11 rain gauges to be distributed to the farmers. We first train the farmers on how to install the rain gauge : a remote place no close to any obstacle, the normal height of the rain gauge, …. We discussed about a “significant” rainfall event. It was understood that 20 mm was enough for millet and sorghum but maize needs 35 to 30 mm. We discussed also how they used to estimate when it is time to plant. They provided two techniques: 1. Digging the ground after a rainy event and to see whether the depth of the wetness is more than a man-hand span, this seems to be a well known and accepted practices amongst farmers 2. Take the first layer of the soil and to see if the humidity in it is enough They all agree that if they have a rain gauge it would be much practical. Then we installed a rain gauge in the yard and explain them how to read it. The training covered: How to read the rain gauge : avoid error of parallax When to read the rain gauge : at 08:00am each day except when the rain gauge is full during an extreme event, How to record the rainfall amount in a book : we gave them copy of a book where to record the rainfall and explain to them the meaning of each column. We let the local agriculture extension to select farmers whom to give the rain gauge. As some rain gauges where already installed in some villages by other projects and they know best each farmers. It was ask to ensure there is a good spatial coverage of the district. II. Delivering the seasonal forecast of JAS 2011 In the second half of the morning we reviewed the probabilistic workshop that we did last week. What they understood from it and what they recollected. After a short discussion we presented hypothetical forecast that they exercise on last workshop and reviewed all detailed of the forecast. Then we divided them into groups and gave each the actual forecast of JAS 2011 in term of number of rainy days and rainfall total. Each group was asked to interpret the forecast and derive actual action they can take in term of farming management and to nominate a reporter. The facilitator was available for eventual questions to each group. Then after we had an open discussion on the seasonal forecast. Most of them did understand that the forecast will be above normal. From the probability of exceedance they said the seasonal rainfall amount will be between 600 and 900mm. The facilitator concluded by emphasizing clearly what this forecast did tell and what it did not tell.  Seasonal forecast for JAS 2011 as it was presented to farmers : top seasonal total observed in grey, hindcasted in red and the actual forecasted intervals for 2011 in green (left) and the probability of exceedance (right) and in the bottom the forecast as number of rainy days greater than 1 mm. It follows many recommendations on which crop to plant and what strategies need to be taken. The planting date was a big issue for farmers : should they anticipate and plant early or should they wait for the first event. The spatial inhomogeneity throughout the district of Kaffrine was also pointed out, should all of them adopt the same planting date or should it be different according to the place. Another issue was the availability of seeds, fertilizer and also the lack of resources. At the end of the training we discussed how to keep in touch, to collect the rainfall collected by farmers. We gave phone numbers where to get in touch with us and promised to give an update of the seasonal forecast as the indicators may change.  Acknowledgments : Funding for this workshop was provided by Climate Change and Food Security (CCAFS). We want to thank Dr Jim Hansen theme Leader on “Adaptation through Managing Climate Risk” who provided technical as well as advisory supports and Dr Robert Zougmore West-Africa regional CCAFS coordinator, who facilitated lots of administrative arrangements. The Senegalese National Weather Agency (ANAMS) for coordinating, and providing rain gauges. We also recognize the support and participation of local experts based in Kaffrine working with farmers : volunteers from World Vision (WV) , agricultural advisers from the national agency for agricultural and rural advice (ANCAR), agronomist from the local representative of Ministry of agriculture (SDDR), agro-economist from the national agricultural research institute (ISRA). The special efforts of Elhaji Malick Seck local coordinator of the workshop and his help to select and interface with farmers are also acknowledged with great appreciation. We want to thank individual farmers as well as farmer’s organizations for dedicating much effort and time to attend the workshop and lively participate in the discussions, answer questionnaires, and make this workshop useful. We also acknowledge the availability of chief of villages of Santhiou Galgoné, Katakel, Ngodiba, Tousne Mosquée and Sorokogne for welcoming us and helping on the questionnaire. We finally thank the International Research Institute for Climate and Society (IRI) at columbia for providing data and software (Climate Predictability Tool) used to produce the forecast. A full list of participants can be found in the Appendix.  Acronyms : ANAMS Agence Nationale de la Meteorologie du Senegal ANCAR Agence Nationale du Conseil Agricole et Rural CCAFS Climate Change Agriculture and Food Security CGIAR Consultative Group on International Agriculture Research COPROSA Coopératives de Producteurs de Semences d'Arachide GPF Groupement de Promotion Féminine ISRA Institut Sénégalais de Recherches Agricoles IRI International Research Institute for Climate and Society JAPANDOO Le Syndicat national des agriculteurs, éleveurs et pêcheurs du Sénégal SDDR Service Départemental du Développement Rural WV World Vision  List of participants N° Prénom et Nom Structure Localité 1. Mansour SARR Syndicat JAPANDOO Gniby 2. Cheikh FALL Syndicat JAPANDOO Boulel 3. Arona NDIAYE Syndicat JAPANDOO Kahi 4. Khady DIENG Syndicat JAPANDOO Kaffrine 5. Mame Diarra TOURE Syndicat JAPANDOO Boulel 6. Ablaye SEGNANE Syndicat JAPANDOO Diockoul Mbelbouck 7. Fatou SECK Syndicat JAPANDOO Pathé Thiangaye 8. Pape Samba DIANE Syndicat JAPANDOO Nganda 9. Aliou WILANE Syndicat JAPANDOO Diamagadio 10. Satou TOP GPF Diockoul Mbelbouck 11. Amy NDIAYE Productrice individuelle Gniby 12. Aïssatou NDAO GPF Ngodiba 13. Khady MBAYE GPF Kaffrine 14. Magatte NDIAYE Productrice individuelle Thiamène 15. Aly DIAW COPROSA Djigui 16. Ibrahima SALANE Fédé GIE de Kathiote Sinthiou Galgoné 17. Lamine SEGNANE Producteur individuel Keur Lahine 18. Talla NDAO (Père) Producteur individuel Sorocogne 19. Babou Fana NDIAYE Producteur individuel Nguer Mandakh 20. Badou DIOP Producteur individuel Boulel 21. Modou BADIANE Producteur individuel Boulel 22. Aly BADIANE Coop. Prod. Sem. d’Arac Nganda 23. Abdou SECK Producteur individuel Dankou 24. Omar NDIAYE Producteur individuel Nganda 25. Arame Cissé Productrice individuelle Darou kahi 26. Baye GUISSE Producteur individuel Médina Kaffrine 27. Ismaïla DIABY Producteur individuel Sikilo 28. Bacar THIALL Producteur individuel Walalane 29. Babacar Ba BADIANE Producteur individuel Ndodj 30. Satou MBENGUE Syndicat JAPANDOO 31. Momar 1 NDAO Secr. COPROSA Ndioudiène 32. Momar NDAO Productrice individuelle Ndioudiène 33. El haji Malik NDAO Productrice individuelle Tousne mosquee 34. Mor KA Productrice individuelle Tousne mosquee 35. Khady NDIAYE Productrice individuelle Kaffrine 36. Fode DIOUF Productrice individuelle Diakhao Saloum 37. Babacar THIAM Productrice individuelle Kaffrine 38.  List of technical experts : N° Prénom et Nom Structure Localité 1. Ousmane Ndiaye ANAMS Dakar 2. Aïda Diongue ANAMS Dakar 3. Moustapha Ciss ANAMS Dakar 4. Amy Sal Thiam ANAMS Dakar 5. Serigne Ndiaye ANAMS Dakar 6. Diabel Ndiaye ANAMS Dakar 7. Mohammed Lamine Diop ANAMS Dakar 8. Babacar Badiane Journaliste Dakar 9. Moussa Sall ISRA Dakar 10. Ibou Sagna SDDR Kaffrine 11. Ndiamé Gueye SDDR Kaffrine 12. Cheikh Diouf SDDR Kaffrine 15. El Haji Moussa Seck SDDR Kaffrine 16. Abdoulaye Kharma Adj Prefet Kaffrine 17. Modou ndiaye SDDR Kaffrine 18 Ibrahima diop SDDR Kaffrine 19 Pape Amadou DIA WV Kaffrine 20 Babou GUEYE ANCAR Kaffrine 21 El haji Seny SAMB WV Kaffrine 22 Abdoulaye SECK ANCAR Kaffrine 23 Badara NDAO WV Kaffrine 24 Abdou LOUM ANCAR Kaffrine