By Carinus De Kock, Stellenbosch University

Disaster Risk Management, Sustainability and Urban Resilience Centre (DiMSUR) and UN-Habitat, with strong support from several University partners, developed CityRAP which is a participatory planning tool used for enabling city managers and municipal technicians to implement participatory urban resilience planning http://dimsur.org/3-cityrap-tool/). It is aimed at small to intermediate sized cities in sub-Saharan Africa to understand and plan actions aimed at reducing risk and building resilience through the elaboration of a City Resilience Action Plan. CityRap is an enabling initiative rather than prescriptive, which therefore also focusses on the ownership of local governments.

Fathum WP2 in the field – Langeberg, South Africa

Vegetation in the Kogmans River

Proper clearing in the Keisie River + river bed and bank stabilisation

The above two pictures speak to one flood risk driver in the municipality – (alien) vegetation in major river which exacerbate damage downstream when flooding occurs. The Keisie River (tributary of the Kogmans), successfully cleared from vegetation is clearly visible, along with groynes to strengthen embankments. The Kogmans River on the other hand still remains overgrown. The identification of these flood risk root causes proved critical to identify most at-risk areas, in order to improve and focus early warnings.

Experiences from Fathum assisting CityRAP in South Africa

CityRAP is being piloted in three South African cities (George, Port Alfred and Potchefstroom). The United Nations Human Settlements Programme (UN-Habitat) in collaboration with the National Disaster Management Centre (NDMC) of South Africa therefore hosted a Training of Trainers and Academics Exchange workshop on urban resilience planning, held from 24-28 February 2020 in George, South Africa. The workshop was attended by approximately 50 people, which included municipal officials and academics from across Africa, where CityRap has been/will be implemented.

Group of participants at Garden Route District Municipality in George

The meeting also served to strengthen the involvement of the academic community in the implementation of CityRAP in southern Africa, to further develop the CityRAP tool, and to reinforce partnerships and collaboration between southern African Universities on disaster risk reduction, urban resilience and adaptation to climate change. Experiences from the Fathum study was shared during this workshop. The critical importance of in-depth risk analysis alongside the local community is highlighted by the FATHUM WP2 work; hazards, vulnerabilities and exposures are unique to a certain setting or community (such as the alien vegetation in the Kogman’s River example) and therefore are often unrecognised by municipal managers. Such studies provide an important bridge that simultaneously informs local development, humanitarian action and resilience-building (such as informed Early Warning Systems).

Municipal officials identifying flood risks in Potchefstroom

UN-Habitat introducing the CityRAP tool

By Dr Liz Stephens, University of Reading

Photo: Denis Onyodi | Red Cross Red Crescent Climate Centre

Humanitarian action for natural hazards is evolving, learning from catastrophic events such as the 2011 Horn of Africa famine, and the Mozambique floods of 2000.

While the full extent of the impact of Cyclone Idai is still unknown, we do know that even with the airport closed and roads impassable, the Mozambique Red Cross were already on the ground in Beira having been preparing communities by disseminating early warning messages and prepositioning non-food items such as emergency shelter kits, blankets, and mosquito nets.

The preparedness activities led by the Mozambique Red Cross were facilitated by an innovative humanitarian system known as ‘Forecast-based Action’, whereby early action plans are triggered when a specific forecast of a natural hazard is made. These early action plans are supported by evidence from academics, with research contributing to early pilot projects in Uganda and Peru, and ongoing research under NERC / DfID’s SHEAR programme providing the tools and evidence to support the scale-up from these initial pilot projects to systematic international financing mechanisms for approving and funding early action on the basis of a forecast.

As part of the Forecasts for Anticipatory Humanitarian Action (FATHUM) project, my team of researchers has been supporting the Mozambique Red Cross and German Red Cross in the development of early action protocols for preparing for tropical cyclones. This work has addressed the relationship between tropical cyclone magnitude and associated impacts in the country, and has prompted discussions about the most appropriate actions to take to reduce the risk. For example, while installation of temporary roofing is a valuable early action for heavy rainfall caused by El Nino events in northern Peru, this is inappropriate during tropical cyclones where strong winds and poorly-secured roofing create an additional hazard.

On March 14th 2019, a day before Idai hit the coast of Mozambique, 340,000 swiss francs were released from the IFRC’s Disaster Relief Emergency Fund to directly assist 7500 people believed to be at risk. At this point it was estimated that just over 100,000 people would be affected by this storm, but even with the full extent of the disaster yet to emerge we can already see that this number is now much higher, with reports that at least 400,000 people have been made homeless in Mozambique alone, with many more affected in Zimbabwe and Malawi.

As with any innovation there is always room for improvement. Our research in understanding tropical cyclone impacts in Mozambique has been hampered by the lack of comprehensive historical data of rainfall, winds, and storm surge impacts as well as vulnerability data that you might expect to find for countries affected by Atlantic hurricanes, for example. But in addition to that, more research and development is clearly needed to create more impact-relevant forecasts to support decision-making.

For example, while the wind strength of the storm as it made landfall on the 15th was forecasted well as far ahead as March 10th, uncertainty over the magnitude and location of both the storm surge and heaviest rainfall makes the impactful flooding much harder to predict. Put another way, while we were sure that Cyclone Idai would make contact with Mozambique as a “very dangerous system” many days ahead, the Global Flood Awareness System hovered around a 20% chance of severe flooding on the Buzi River, only giving certainty once the storm had made landfall. We now know that flooding has hit 5000 people in the Buzi District alone (personal communication with the Mozambique Red Cross).

Over the coming days and weeks, we will see better information emerge about the impacts of the storm, and unfortunately we are also likely to hear about the secondary impacts, such as outbreaks of disease, which happen after news crews have left the region.

Acting on early warnings is one part of the picture when it comes to building resilience to weather extremes, and Forecast-based Action is just one cog in the humanitarian financing machinery. Successfully avoiding impacts on this scale requires linking together long-term disaster reduction with better early action and faster response, by both supporting local capacity and mobilising the international community when the local capacity is overwhelmed.

Behind the scenes the humanitarian community will be working hard to identify lessons learned and improve plans and processes. Researchers from across the interdisciplinary FATHUM project will be using this storm as an opportunity to better understand the skill of flood forecasts during tropical cyclones, to improve our understanding of the complexity of impacts driven by both wind and rain, and to better identify appropriate early actions. One thing already emerging is the suggestion of using this Forecast-based Action mechanism to establish emergency communication systems before a disaster; something that would no doubt improve the humanitarian response by enabling earlier assessment of impacts.

Top: forecast on the Buzi River on March 11th showing a small probability of extreme flooding (a few forecast scenarios are within the pink colour); Bottom: forecast on the Buzi River on March 16th, showing certainty of extreme flooding (all forecast scenarios are within the pink colour). Source: www.globalfloods.eu

Author: Erin Coughlan de Perez, Red Cross Red Crescent Climate Centre and FATHUM team member 

In order to get the most time for early warning early action (e.g. for Forecast-based Financing projects), we are particularly interested in seasonal rainfall forecasts, which are issued before a rainy season starts and can give an indication of whether the next three months are likely to be unusually wet or dry. However, in our guidance materials for the use of seasonal forecasts, we clearly mention that these are not flood forecasts (see IFRC maproom), because they are only forecasting total rainfall over a three-month period. The rain could come all at once, or it could be nicely spread out over the three months.

In 2008 in West Africa, we successfully used a seasonal rainfall forecast to prepare for flooding, and Red Cross teams were grateful to have taken action before widespread regional flooding happened that year. Therefore, we decided to analyze whether there is a clear link between total seasonal rainfall and flooding – maybe these seasonal forecasts can be a good indicator for flood preparedness.

In our recent paper, we use the Global Flood Awareness System hydrological model to compare the total seasonal rainfall that went into the model with the “floodiness” that came out of it. We analyzed all the modeled rivers in sub-Saharan Africa above a certain size, therefore we did not consider flash flooding.

Our results showed that in many places, the association between total seasonal rainfall and floodiness is not very strong. Especially in wetter regions of West and Central Africa, it seems that more total rainfall in a season only marginally increases the chances of flooding that season. In East Africa and Southern Africa, the relationship was slightly stronger.

Recommendations for disaster managers

Our results show that:

• even if forecasters could say with complete certainty that rainfall would be in the “above normal” tercile category, this doesn’t mean that there is certainty over whether it will flood
• forecasts can usually only give slight increases in the likelihood of “above normal” rainfall, this means that in reality there is even lower certainty over where it will flood.
• so in practice, current seasonal rainfall forecasts are unlikely to ever provide high enough confidence in whether there will be flooding for disaster managers to be comfortable taking flood preparedness actions.
• future work may change that though! Some scientists are working to get more useable information on flood likelihood during El Nino / La Nina events, and based on the current temperature anomalies in the Indian Ocean. Others have been developing seasonal hydrological forecasting systems.

Recommendations for seasonal forecasters

In our paper we explored several alternative options to make seasonal rainfall forecasts more informative for flood preparedness. We have have the following recommendations for institutions providing operational seasonal forecasts:

• Make seasonal total rainfall forecasts for river basins, in addition to the current gridbox forecasts.
• Develop skillful seasonal forecasts of the upper extremes of total rainfall (e.g. top 10th percentile) would already be more useful than terciles for anticipating floods.
• Forecast the number of 3-day extreme rainfall events and seasonal rainfall intensity is more informative for flood preparedness than total rainfall.
• Invest further into seasonal hydrological forecasts, in which seasonal rainfall forecasts are passed through a hydrological model to forecast seasonal flooding, would also be a very useful step forward (see the freely available new GloFAS model for this).
• Continue to provide the valuable information on forecast skill

Coughlan de Perez, E., Stephens, E., Bischiniotis, K., van Aalst, M., van den Hurk, B., Mason, S., Nissan, H., and Pappenberger, F.: Should seasonal rainfall forecasts be used for flood preparedness?, Hydrol. Earth Syst. Sci., 21, 4517-4524, https://doi.org/10.5194/hess-21-4517-2017, 2017.

Available to read for free here: https://www.hydrol-earth-syst-sci.net/21/4517/2017/

Project Management

Liz joined the first meeting of the SHEAR Studentship Cohort at Imperial College London on April 26th. There are some interesting PhD projects that are now underway that will be working on FATHUM-related topics, for example flood forecasting in Kenya, citizen science for early warning systems and visualisation of forecasts. There will be a SHEAR studentship conference in September which Early Career Scientists from across the 4 main SHEAR consortia will be invited.

Liz has also collated all of the abstracts for the Adaptation Futures session on Forecast-based Financing and sent them all to the organising committee. The session will last 1hr45 minutes. There will be a short (5 minute) introduction to FbF for those in the audience who are not familiar in it. Then the rest of the session will be split into two. The first half will showcase FbF case studies, the second half FbF thematic topics. There are 5/6 abstracts in each section, and so there will be 5/6 groups rotating around the room. You will therefore have 8 minutes with each group. I would suggest preparing a flexible presentation – be prepared to speak for up to 8 minutes if there are no questions, but ideally audience members will interject or ask questions and direct the flow of conversation. That way the audience members become an active part of your presentation and research, rather than it being a passive listening exercise.

Project management activities this month have also included arrangements for the meeting in South Africa in June. Joy’s efforts have been indispensable, thanks Joy! 

Work Package 1

Flood seasonality and ENSO/IOD teleconnections:

Andrea has been focusing on the changes of seasonality of flooding in Sub-Saharan Africa with ENSO and Indian Ocean Dipole. He has calculated some circular statistics based on the streamflow reanalysis data-set (GloFAS driven by Era-Interim/Land), in order to identify the flood timing for different ENSO/IOD conditions. These statistics have been mapped over Africa and show significant changes of seasonality in some regions of Eastern and Southern Africa.

Andrew led a joint FATHUM/ForPAC team Skype discussion (19 April):

Participants: Erin Coughlan de Perez, Arlindo Meque, Liz Stephens, and Andrea Ficchì, from FATHUM (WP1); Martin Todd, Dave MacLeod, Maurine Ambani, Mary Kilavi and Augustine Kiptum from ForPAC.

We have discussed about the seasonality analysis and the impact of ENSO/IOD on flooding likelihood changes (EGU presentation). Some of the valuable comments and suggestions received from the team concern:

• The need for an uncertainty assessment of the reliability of forecasts of flooding with ENSO/IOD.
• The possible inclusion of the Subtropical Indian Ocean Dipole in the analyses, because of its expected influence on rainfall in Southern Africa;
• The interest of deeper investigations on particular case studies, such as the Shabelle River (Somalia). The results of flood likelihood changes with ENSO/IOD for this river are relevant for decision-making. Results, so far, highlight some distinct predictions for the region, contrasting some current Early-Action policies for El-Niño preparedness.

EGU presentation (11 April): Liz presented Andrea’s work at the European Geophysical Union conference, including recent results on the reliability of the statistical forecasts of flooding likelihood based on ENSO/IOD teleconnections. This part highlighted some interesting differences between the reliability of predictions based on either ENSO or IOD. The feedback from the talk was good, in particular there were requests to do the same analysis for drought hazard.

Work Package 2

For Uganda:
They are continuing with their desktop review, including a review of past flood events in Teso (from 2007) and review of sociodemographic info for Amuria and Katakwi districts. They are trying to set up an interview with their disaster risk contact person at OPM.

For South Africa:
Busy mostly with admin and logistics for the upcoming June meeting. Setting up and conducting interviews with key stakeholders in our field site – this process also critical for finalising the fieldtrip plan for June, and selecting at-risk communities (in the towns and farming communities) for household-level interviews. Ongoing desktop review of past flood events in the Langeberg and collecting reports/historical info on the case study area.

For Mozambique:

• Evaluation of flood triggers;
• Desktop review: data collection on past floods in Chibuto region (initiated);
• Study of the contribution of Limpopo Basin human formations to the flood warning system
• Desktop review: DRM/EWS legislation (initiated).

Work Package 3

Currently Sara is transcribing the recordings from both Dialogue Platforms (Berlin, October 2017 & Nairobi, March 2018), analysing interviews and refining the interview protocol. This preliminary research data forms the basis for the observational analysis document that Sara is currently also drafting. In order to analyse the conferences and interviews, Sara is developing a coding system that enables her to distil dominant themes and understand plural perceptions.

Furthermore, Sara has been reading and summarizing international reports and reviews of existing early action/ and cash transfer projects and literature on the use of probabilistic forecasts in decision-making.

Sara is also actively approaching potential interviewees to follow up on some of the talks that I had during the Dialogue Platform in Nairobi. In particular Sara is keen to interview more people on the periphery of the FbF community, please let her know any suggestions you might have.

Work Package 4

Tobias is due to begin at Reading University at the beginning of June, and will be attending the meeting in South Africa.

Topics for discussion on May 3rd

• Question on progress: does anybody have any comments or questions about the progress updates?
• Does everyone understand the format and what they need to prepare for the Adaptation Futures FbF session?
• Question on June meeting: what needs to be discussed in Stellenbosch in June?

Joy comments: I think that the first half of the first day should be work package presentations and updates on their progress – so everyone knows what everyone else is doing. The morning of Day 2 should involve each work package breaking into separate groups to plan/discuss their work and plans for the upcoming months (while we are fresh and full of ideas!). Those are two ‘sessions’ that I think are critical.

• Open question: What are the challenges of setting up FbF for transboundary rivers?

Author: Liz Stephens

Thanks to everyone who joined our monthly progress call, the minutes are below. Please add your thoughts on the discussion point in the comments section at the bottom of this post.

Work Package 1 report (Liz on behalf of Hannah)

• Andrea has been continuing his work on the influence of ENSO / Indian Ocean Dipole on flood likelihood in sub-saharan Africa.
• He made a presentation to the Tropical Research Group in the Department of Meteorology at Reading and gained some valuable feedback. Liz will be presenting this work for Andrea at the European Geophysical Conference in Vienna next week.
• Arlindo Meque, based in Mozambique, has been researching tropical cyclone impacts and the precipitation and wind thresholds which lead to different impacts.

Andrea Ficchi presenting his research to the University of Reading Tropical Meteorology group

Work Package 2 report (Joy)

• Joy is in the process of finalising the research tools which have been co-produced among all the case study teams, final versions will be shared soon. From this the teams will be working on desktop reviews and begin the fieldwork.
• The WP2 team had the opportunity to meet up at the Dialogue Platform in Nairobi.
• Joy is currently working on a paper on the methodological approach used in FATHUM.

Rui da Maia leading a discussion at the Africa FbF Dialogue Platform on the challenge of understanding human behaviour for FbF in Mozambique

Work Package 3 report (Sara)

• Sara has been refining the interview protocol and will be preparing a report based on her initial observations, identifying key themes and looking at factors that inhibit / enable the use of uncertain forecasts.

Work Package 4 report (Emily W)

• This work is about to start with Toby joining the University of Reading in June. Toby has been working with the Start Network on FbF-related research.
• Toby introduced himself to the team. All of the WP4 team will be in Stellenbosch / Cape Town in June.

Other collaborations (Liz)

• We welcome Arielle Tozier de la Poterie and Anne-Catherine Vanhove as collaborators with the FATHUM team. They are working on research related to FbF in Mozambique.

Report on FbF Activities (Erin)

• A new DREF mechanism for early action will be launched in May in Geneva. Following the Africa FbF Dialogue Platform in Nairobi there will be similar Dialogue Platforms in Latin America and Asia in the coming months where FATHUM work will be featured.
• FATHUM work will also be presented by the Climate Centre team at the Understanding Risk conference in Mexico City in May.

Communication Channels (Liz)

• I’m keen for a shift from email to shared comments on the blog posts (obviously where appropriate).
• Next month’s progress reporting will be done via the blog before the Skype call, giving more time for discussion during the call.
• I also welcome contributions to the blog, however short, even just a photo of your activities!
• Anything on the blog will be syndicated via Twitter (to our external audience) and via Slack (to our compatriots in other SHEAR projects).
• There is a FATHUM shared OneDrive folder that hasn’t been well used, please upload relevant reports, publications, photos etc.

Annual meeting in Stellenbosch / Adaptation Futures

• I have a final deadline for all information related to the Adaptation Futures session of April 20th.
• Please could everyone register for the conference by end of the day on Monday. Reading will pick up the invoice for registration fees for those of you under a subcontract if registration has been carried out by then.
• Joy has been managing accommodation and shuttle bookings for South Africa, please let Joy know if you think you are missing from these arrangements.

Discussion point:

• What presentations / discussions / activities do you want to see at the annual meeting in Stellenbosch?

NEXT CALL: 2pm GMT+1, Thursday May 3rd

Author: Liz Stephens

Many members of the FATHUM team came together for the Africa Forecast-based Financing Dialogue Platform in Nairobi to hear the latest on FbF and related topics across Africa. The German Red Cross will no doubt be writing a full report of the session, but here is some ‘I like I wish I wonder’ feedback from the FATHUM team.

I like –

• that the discussions were rich and interdisplinary – Harriet Lowalem, Makerere University
• the diversity of the current Fbf pilot approaches and the genuine excitement building up among the field’s growing group of practitioners. – Paris Kasiz, WFP
• learning how people around Africa are approaching FbF and seeing how the concept has evolved. – Arielle Tozier, German Red Cross Mozambique
• the diversity of participants and high interest on FbF – Irene Amuron, RCCC
• the diversity of ideas on how to set up effective systems and take action at scale! – Erin Coughlan de Perez
• the paradigm shift to Impact based forecasting and not just forecasting for the sake of forecasting – Beauty Shamboko
• the variety of topical themes discussed regarding forecast based financing and actions – Chris Garimoi Orach, Makerere University
• finding out about all the FBF activities and research that is happening across the African continent – Joy Waddell, Stellenbosch University
•  I like the thematic break-out sessions as they form a more intimate and fruitful platform to exchange ideas and challenges – Sara de Wit, Oxford University
• I would like to see more participants being aware of the huge size of African river basins and the need to support them with basic forecasting infra-structure. For example, the Limpopo basin has a size of 40 million hectares, some 4 times the size of England and includes highlands in South Africa and lowlands at the Indian Ocean, Mozambique.  River flooding under these circumstances can have different genesis year by year.  If we understand well the hazards we will have more time to act and the FbF mechanism will be fulfilling the role of reducing suffering of affected populations. Massive rains in Mozambique have started in the year 2000, on February 3, caused by the cyclone Eline originating from Australia.The resulting massive flooding of the Limpopo river happened in fact much later, on February 26. We had 23 days of lead time to act but we didn’t act because we did not understand well the “hazard”, the link between Cyclone Eline and the Limpopo flooding. – Rui da Maia, UDM
  

  Erin Coughlan de Perez leading a session on forecast triggers. Photo: Liz Stephens

• all agencies and actors in the field talked more with each other, recognizing and taking advantage of the tremendous potential for complementarities and synergies in their work! – Paris Kasiz, WFP
• there was better coodination among research initiatives pertaining to FbF (and I´m happy that the dialogue platform opened the lines of communication) – Arielle Tozier, German Red Cross Mozambique
• that the topics were all inclusive rather than selecting a few topics to attend – Harriet Lowalem, Makerere University
• we had given government representatives more centre stage to share practical examples on how FbF can be institutionalised within DRM government agency – Irene Amuron, RCCC
• that was more government representation including at national and sub-national levels – Chris Garimoi Orach, Makerere University
• Zambia will be able to have presentation on the progress made from the resolutions made at the Dialogue Platform – Beauty Shamboko
• we had more information on forecast verification, to know which actions should go with which forecasts. – Erin Coughlan de Perez
• that there was better collaboration and communication about FbF research across partners, organisations, and research projects.  – Joy Waddell, Stellenbosch University
• I wish that there was more emphasis placed on the importance of community-level participation in the implementation of FbF – Sara de Wit, Oxford University

I wonder –

• to what degree and timeframe will Fbf be scaled in whole countries and regions, fully integrated within national systems as well as humanitarian operations. – Paris Kasiz, WFP
• whether lessons learned from fbf interventions shall necessarily apply in other local context during the scale up phase – Harriet Lowalem, Makerere University 
• how best to use the research (and what research would be most valuable) to influence practice and build government support for early action – Arielle Tozier, German Red Cross Mozambique
• what the future of FbF will be without donor funding – Irene Amuron, RCCC
• about scale up of forecast based interventions – Chris Garimoi Orach, Makerere University
• if there will be more time allocaed for the next Dialogue Platform – Beauty Shamboko
• how the world might be different if we had open data. – Erin Coughlan de Perez
• what impact or difference our FATHUM research will have on the scaling up of FbF in countries and the application of FbF across the world.  – Joy Waddell, Stellenbosch University
• I wonder how local forecasting systems (or environmental knowledge) and scientific models can complement and strengthen each other – Sara de Wit, Oxford University

Liz Stephens presenting progress in FATHUM forecasting research. Photo: Becky Hemingway, Met Office

Author: Emily Wilkinson, Overseas Development Institute

A range of large scale social protection and risk finance initiatives as well as promising innovations in forecast-based action (FbA) have emerged in recent years. This field of forecast-based early action is rapidly expanding, and consolidating the evidence, experience, and lessons from early efforts to develop forecast-based action and finance tools can help improve the impacts and effectiveness of future investments.

The Scoping and Design for Taking Forecast-Based Early Action to Scale project funded under the WISER programme will do just that. ODI and partners including the University of Sussex, the START Network, the Red Cross Red Crescent Climate Centre, and Tetra Tech, will work together to review and share existing forecasting and decision-making capabilities and links to finance for early action that can be triggered in a range of different contexts. The project team will identify opportunities for institutionalising early action through modifications to existing delivery systems, carefully designed in partnership with relevant stakeholders.

Three focal studies will be selected to reflect a variety of early action financing and delivery mechanisms. The goal of the focal studies will be to (a) uncover the political economy and other factors influencing the formalisation of FbA mechanisms; and (b) build interest and dialogue between key stakeholders (providers of forecasts and potential users of that information). Research will cover up to three types of ‘systems’: an international financing mechanism; a national delivery system (in a stable institutional context); and through non-state actors (in a fragile context). The completed Focal Studies report will set out the context and rationale for scaling up in each system and describe the process of co-production used to bring together the necessary stakeholders to develop a sound concept.

Author(s): Liz Stephens (University of Reading)

An interesting comment piece from Madhab Uprety, Disaster Risk Reduction consultant, and Sumit Dugar, Research Associate at Practical Action Consulting South Asia on the recent floods in Nepal has been posted here.

It is no surprise that a similar situation has been encountered in Nepal as elsewhere in the world:

You may be interested in connecting with the Practical Action project team to compare and contrast how Forecast-based Financing is being implemented in Nepal compared to our own pilot projects in Uganda and Mozambique:

Sumit Dugar is part of the project team of the SHEAR LANDSLIDE-EVO project, and is keen to connect with the FATHUM team. We will shortly be advertising a PhD project to work alongside the FATHUM team, focussed on supporting Forecast-based Financing in Nepal. The student would link with Practical Action Nepal and the Nepal Department of Hydrology and Meteorology.